MapServer Documentation

Release 6.0

The MapServer Team

May 12, 2011

 Contents

1 About 3

2 An Introduction to MapServer 5
2.1 MapServer Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Anatomy of a MapServer Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 Installation and Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 Introduction to the Mapfile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.5 Making the Site Your Own . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.6 Enhancing your site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.7 How do I get Help? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3 MapServer Tutorial 25
3.1 Tutorial Timeframe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2 Tutorial Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.3 Before Using the Tutorial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.4 Windows, UNIX/Linux Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.5 Other Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.6 Section 1: Static Maps and the MapFile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.7 Section 2: CGI variables and the User Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.8 Section 3: Query and more HTML Templates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.9 Section 4: Advanced User Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4 Installation 31
4.1 Compiling on Unix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.2 Compiling on Win32 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.3 PHP MapScript Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.4 .NET MapScript Compilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.5 IIS Setup for MapServer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.6 Oracle Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

5 Mapfile 63
5.1 Cartographical Symbol Construction with MapServer . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.2 CLASS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.3 CLUSTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

i
 5.4 Display of International Characters in MapServer . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.5 Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.6 FEATURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
5.7 FONTSET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
5.8 GRID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5.9 INCLUDE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.10 JOIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
5.11 LABEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.12 LAYER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.13 LEGEND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.14 MAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.15 OUTPUTFORMAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
5.16 PROJECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
5.17 QUERYMAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
5.18 REFERENCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
5.19 SCALEBAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
5.20 STYLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
5.21 SYMBOL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
5.22 Symbology Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
5.23 Templating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
5.24 Union Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
5.25 Variable Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
5.26 WEB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
5.27 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

6 MapScript 171
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
6.2 SWIG MapScript API Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
6.3 PHP MapScript . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
6.4 Memory Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
6.5 Python MapScript Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
6.6 Python MapScript Image Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
6.7 Mapfile Manipulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
6.8 Querying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
6.9 MapScript Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253

7 Data Input 261

7.1 Vector Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
7.2 Raster Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

8 Output Generation 339

8.1 AGG Rendering Specifics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
8.2 AntiAliasing with MapServer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
8.3 Dynamic Charting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
8.4 Flash Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
8.5 HTML Legends with MapServer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
8.6 HTML Imagemaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
8.7 OGR Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
8.8 PDF Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
8.9 SVG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
8.10 Tile Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
8.11 Template-Driven Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
8.12 Kml Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389

9 OGC Support and Configuration 395

ii
 9.1 WMS Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
9.2 WMS Client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415
9.3 WMS Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
9.4 Map Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
9.5 WFS Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434
9.6 WFS Client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
9.7 WFS Filter Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
9.8 SLD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
9.9 WCS Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 462
9.10 WCS Use Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
9.11 SOS Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
9.12 MapScript Wrappers for WxS Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494

10 Optimization 503
10.1 Debugging MapServer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503
10.2 FastCGI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514
10.3 Mapfile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517
10.4 Raster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
10.5 Tile Indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
10.6 Vector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522

11 Utilities 525
11.1 legend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525
11.2 msencrypt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525
11.3 scalebar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527
11.4 shp2img . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527
11.5 shptree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529
11.6 shptreetst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 530
11.7 shptreevis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 531
11.8 sortshp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532
11.9 sym2img . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
11.10 tile4ms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535
11.11 Batch Scripting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539
11.12 File Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539

12 CGI 541
12.1 MapServer CGI Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541
12.2 mapserv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542
12.3 Map Context Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542
12.4 MapServer CGI Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543
12.5 Run-time Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547
12.6 A Simple CGI Wrapper Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549

13 Community Activities 551

13.1 IRC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551
13.2 Mailing Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552
13.3 MapServer Wiki Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
13.4 MapServer Service Providers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553

14 Development 555
14.1 Sponsors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555
14.2 MapServer Release Plans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556
14.3 Announcements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559
14.4 Bug Submission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
14.5 Subversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562

iii
 14.6 Documentation Development Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
14.7 Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 570
14.8 Request for Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 576
14.9 Mapfile Editing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775
14.10 External Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 777

15 Download 779
15.1 Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 779
15.2 Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 779
15.3 Binaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 780
15.4 Demo Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 781

16 Glossary 783

17 Errors 787
17.1 drawEPP(): EPPL7 support is not available . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 787
17.2 loadMapInternal(): Given map extent is invalid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 787
17.3 msSaveImageGD(): Unable to access file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 788
17.4 msLoadMap(): Failed to open map file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789
17.5 msQueryByPoint: search returned no results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789
17.6 msLoadFontset(): Error opening fontset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789
17.7 msGetLabelSize(): Requested font not found . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 790
17.8 loadLayer(): Unknown identifier. Maximum number of classes reached . . . . . . . . . . . . . . . . 790
17.9 msReturnPage(): Web application error. Malformed template name . . . . . . . . . . . . . . . . . . 790
17.10 Unable to load dll (MapScript) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 790
17.11 msProcessProjection(): Projection library error.major axis or radius = 0 not given . . . . . . . . . . . 791
17.12 msProcessProjection(): no options found in ‘init’ file . . . . . . . . . . . . . . . . . . . . . . . . . . 791
17.13 msProcessProjection(): No such file or directory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 792
17.14 msQueryByPoint: search returned no results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 792

18 FAQ 795
18.1 Where is the MapServer log file? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 795
18.2 What books are available about MapServer? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 795
18.3 How do I compile MapServer for Windows? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 795
18.4 What do MapServer version numbers mean? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 795
18.5 Is MapServer Thread-safe? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 796
18.6 What does STATUS mean in a LAYER? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 797
18.7 How can I make my maps run faster? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 797
18.8 What does Polyline mean in MapServer? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 797
18.9 What is MapScript? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798
18.10 Does MapServer support reverse geocoding? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798
18.11 Does MapServer support geocoding? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798
18.12 How do I set line width in my maps? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798
18.13 Why do my JPEG input images look crappy via MapServer? . . . . . . . . . . . . . . . . . . . . . . 798
18.14 Which image format should I use? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 799
18.15 Why doesn’t PIL (Python Imaging Library) open my PNGs? . . . . . . . . . . . . . . . . . . . . . . 799
18.16 Why do my symbols look poor in JPEG output? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 800
18.17 How do I add a copyright notice on the corner of my map? . . . . . . . . . . . . . . . . . . . . . . . 800
18.18 How do I have a polygon that has both a fill and an outline with a width? . . . . . . . . . . . . . . . 801
18.19 How can I create simple antialiased line features? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 802
18.20 Which OGC Specifications does MapServer support? . . . . . . . . . . . . . . . . . . . . . . . . . . 802
18.21 Why does my requested WMS layer not align correctly? . . . . . . . . . . . . . . . . . . . . . . . . 803
18.22 When I do a GetCapabilities, why does my browser want to download mapserv.exe/mapserv? . . . . 804
18.23 Why do my WMS GetMap requests return exception using MapServer 5.0? . . . . . . . . . . . . . . 804

iv
 18.24 Using MapServer 6.0, why don’t my layers show up in GetCapabilities responses or are not found
anymore? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805
18.25 Where do I find my EPSG code? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805
18.26 How can I reproject my data using ogr2ogr? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805
18.27 What’s with MapServer’s logo? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 806

19 License 807

20 Credits 809

Bibliography 811

v
vi
 MapServer Documentation, Release 6.0

Note: The entire documentation is also available as a single PDF document and ePub document
If you plan on upgrading to the MapServer 6.0 release, be sure to review the MapServer Migration Guide.

Table 1: Quick Links

An Introduction to MapServer Installation Mapfile
MapScript Data Input Output Generation
OGC Support and Configuration Optimization Utilities
Development Glossary Errors
genindex About Community Activities

Contents 1
MapServer Documentation, Release 6.0

2 Contents
 CHAPTER 1

About

MapServer is an Open Source geographic data rendering engine written in C. Beyond browsing GIS data, MapServer
allows you create “geographic image maps”, that is, maps that can direct users to content. For example, the Minnesota
DNR Recreation Compass provides users with more than 10,000 web pages, reports and maps via a single application.
The same application serves as a “map engine” for other portions of the site, providing spatial context where needed.
MapServer was originally developed by the University of Minnesota (UMN) ForNet project in cooperation with
NASA, and the Minnesota Department of Natural Resources (MNDNR). Later it was hosted by the TerraSIP project,
a NASA sponsored project between the UMN and a consortium of land management interests.
MapServer is now a project of OSGeo, and is maintained by a growing number of developers (nearing 20) from
around the world. It is supported by a diverse group of organizations that fund enhancements and maintenance,
and administered within OSGeo by the MapServer Project Steering Committee made up of developers and other
contributors.
• Advanced cartographic output
– Scale dependent feature drawing and application execution
– Feature labeling including label collision mediation
– Fully customizable, template driven output
– TrueType fonts
– Map element automation (scalebar, reference map, and legend)
– Thematic mapping using logical- or regular expression-based classes
• Support for popular scripting and development environments
– PHP, Python, Perl, Ruby, Java, and .NET
• Cross-platform support
– Linux, Windows, Mac OS X, Solaris, and more
• Support of numerous Open Geospatial Consortium <OGC (OGC) standards
– WMS (client/server), non-transactional WFS (client/server), WMC, WCS, Filter Encoding, SLD, GML,
SOS, OM
• A multitude of raster and vector data formats
– TIFF/GeoTIFF, EPPL7, and many others via GDAL

3
MapServer Documentation, Release 6.0

– ESRI shapfiles, PostGIS, ESRI ArcSDE, Oracle Spatial, MySQL and many others via OGR
• Map projection support
– On-the-fly map projection with 1000s of projections through the Proj.4 library

4 Chapter 1. About
 CHAPTER 2

An Introduction to MapServer

Revision $Revision: 10486 $

Date $Date: 2010-08-30 09:24:42 -0700 (Mon, 30 Aug 2010) $
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Author David Fawcett
Contact david.fawcett at moea.state.mn.us
Author Howard Butler
Contact hobu.inc at gmail.com

Contents
• An Introduction to MapServer
– MapServer Overview
– Anatomy of a MapServer Application
– Installation and Requirements
– Introduction to the Mapfile
– Making the Site Your Own
– Enhancing your site
– How do I get Help?

2.1 MapServer Overview

MapServer is a popular Open Source project whose purpose is to display dynamic spatial maps over the Internet. Some
of its major features include:
• support for display and querying of hundreds of raster, vector, and database formats
• ability to run on various operating systems (Windows, Linux, Mac OS X, etc.)
• support for popular scripting languages and development environments (PHP, Python, Perl, Ruby, Java, .NET)
• on-the-fly projections

5
MapServer Documentation, Release 6.0

• high quality rendering

• fully customizable application output
• many ready-to-use Open Source application environments
In its most basic form, MapServer is a CGI program that sits inactive on your Web server. When a request is sent to
MapServer, it uses information passed in the request URL and the Mapfile to create an image of the requested map.
The request may also return images for legends, scale bars, reference maps, and values passed as CGI variables.
See Also:
The Glossary contains an overview of many of the jargon terms in this document.
MapServer can be extended and customized through MapScript or templating. It can be built to support many different
vector and raster input data formats, and it can generate a multitude of output formats. Most pre-compiled MapServer
distributions contain most all of its features.
See Also:
Compiling on Unix and Compiling on Win32
Note: MapScript provides a scripting interface for MapServer for the construction of Web and stand-alone appli-
cations. MapScript can used independently of CGI MapServer, and it is a loadable module that adds MapServer
capability to your favorite scripting language. MapScript currently exists in PHP, Perl, Python, Ruby, Tcl, Java, and
.NET flavors.
This guide will not explicitly discuss MapScript, check out the MapScript Reference for more information.

2.2 Anatomy of a MapServer Application

A simple MapServer application consists of:

• Map File - a structured text configuration file for your MapServer application. It defines the area of your map,
tells the MapServer program where your data is and where to output images. It also defines your map layers,
including their data source, projections, and symbology. It must have a .map extension or MapServer will not
recognize it.
See Also:
MapServer Mapfile Reference
• Geographic Data - MapServer can utilize many geographic data source types. The default format is the ESRI
shapefile. Many other data formats can be supported, this is discussed further below in Adding data to your site.
See Also:
Vector Input Reference and Raster Input Reference
• HTML Pages - the interface between the user and MapServer . They normally sit in Web root. In it’s simplest
form, MapServer can be called to place a static map image on a html page. To make the map interactive, the
image is placed in an html form on a page.
CGI programs are ‘stateless’, every request they get is new and they don’t remember anything about the last time
that they were hit by your application. For this reason, every time your application sends a request to MapServer,
it needs to pass context information (what layers are on, where you are on the map, application mode, etc.) in
hidden form variables or URL variables.
A simple MapServer CGI application may include two html pages:

6 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

Figure 2.1: The basic architecture of MapServer applications.

2.2. Anatomy of a MapServer Application 7

MapServer Documentation, Release 6.0

– Initialization File - uses a form with hidden variables to send an initial query to the http server and
MapServer. This form could be placed on another page or be replaced by passing the initialization infor-
mation as variables in a URL.
– Template File - controls how the maps and legends output by MapServer will appear in the browser. By
referencing MapServer CGI variables in the template html, you allow MapServer to populate them with
values related to the current state of your application (e.g. map image name, reference image name, map
extent, etc.) as it creates the html page for the browser to read. The template also determines how the user
can interact with the MapServer application (browse, zoom, pan, query).
See Also:
Templating
• MapServer CGI - The binary or executable file that receives requests and returns images, data, etc. It sits in the
cgi-bin or scripts directory of the http server. The Web server user must have execute rights for the directory that
it sits in, and for security reasons, it should not be in the web root. By default, this program is called mapserv
• HTTP Server - serves up the html pages when hit by the user’s browser. You need a working HTTP (Web)
server, such as Apache or Microsoft Internet Information Server, on the machine on which you are installing
MapServer.

2.3 Installation and Requirements

2.3.1 Windows Installation

OSGeo4W is a new Windows installer that downloads and/or updates MapServer, add-on applications, and also other
Open Source geospatial software. The following steps illustrate how to use OSGeo4W:
1. Download OSGeo4W http://download.osgeo.org/osgeo4w/osgeo4w-setup.exe
2. Execute (double-click) the .exe
3. Choose “Advanced” install type

8 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

Note: Express contains options for higher-level packages such as MapServer, GRASS, and uDig. Advanced
gives you full access to choosing commandline tools and applications for MapServer that are not included in the
Express install
4. Select packages to install

2.3. Installation and Requirements 9

MapServer Documentation, Release 6.0

Note: Click on the “Default” text beside the higher-level packages (such as Web) to install all of Web’s sub-
packages, or click on the “Skip” text beside the sub-package (such as MapServer) to install that package and all
of its dependencies.
5. Let the installer fetch the packages.

10 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

6. Run the apache-install.bat script to install the Apache Service.

Note: You must run this script under the “OSGeo4W Shell”. This is usually available as a shortcut on your
desktop
Note: A apache-uninstall.bat script is also available to remove the Apache service installation.
7. Start Apache from the OSGeo4W shell and navigate to http://127.0.0.1
apache-restart.bat

2.3. Installation and Requirements 11

MapServer Documentation, Release 6.0

8. Verify that MapServer is working

12 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

2.3.2 Hardware Requirements

MapServer runs on Linux, Windows, Mac OS X, Solaris, and more. To compile or install some of the required pro-
grams, you may need administrative rights to the machine. People commonly ask questions about minimum hardware
specifications for MapServer applications, but the answers are really specific to the individual application. For devel-
opment and learning purposes, a very minimal machine will work fine. For deployment, you will want to investigate
Optimization of everything from your data to server configuration.

2.3.3 Software Requirements

You need a working and properly configured HTTP (Web) server, such as Apache or Microsoft Internet Information
Server, on the machine on which you are installing MapServer. OSGeo4W contains Apache already, but you can
reconfigure things to use IIS if you need to. Alternatively, MS4W can be used to install MapServer on Windows.
If you are on a Windows machine, and you don’t have a HTTP server installed, you may want to check out MS4W,
which will install a pre-configured HTTP server, MapServer, and more. The FGS Linux Installer provides similar
functionality for several Linux distributions.
This introduction will assume you are using pre-compiled OSGeo4W Windows binaries to follow along. Obtaining
MapServer or Linux or Mac OS X should be straightforward. Visit Download for installing pre-compiled MapServer
builds on Mac OS X and Linux.
You will also need a Web browser, and a text editor (vi, emacs, notepad, homesite) to modify your html and mapfiles.

2.3. Installation and Requirements 13

MapServer Documentation, Release 6.0

2.3.4 Skills

In addition to learning how the different components of a MapServer application work together and learning Map File
syntax, building a basic application requires some conceptual understanding and proficiency in several skill areas.
You need to be able to create or at least modify HTML pages and understand how HTML forms work. Since the
primary purpose of a MapServer application is to create maps, you will also need to understand the basics of geographic
data and likely, map projections. As your applications get more complex, skills in SQL, DHTML/Javascript, Java,
databases, expressions, compiling, and scripting may be very useful.

2.4 Introduction to the Mapfile

The .map file is the basic configuration file for data access and styling for MapServer. The file is an ASCII text file,
and is made up of different objects. Each object has a variety of parameters available for it. All .map file (or mapfile)
parameters are documented in the mapfile reference. A simple mapfile example displaying only one layer follows, as
well as the map image output:
MAP
NAME "sample"
STATUS ON
SIZE 600 400
SYMBOLSET "../etc/symbols.txt"
EXTENT -180 -90 180 90
UNITS DD
SHAPEPATH "../data"
IMAGECOLOR 255 255 255
FONTSET "../etc/fonts.txt"

#
# Start of web interface definition
#
WEB
IMAGEPATH "/ms4w/tmp/ms_tmp/"
IMAGEURL "/ms_tmp/"
END

#
# Start of layer definitions
#
LAYER
NAME ’global-raster
TYPE RASTER
STATUS DEFAULT
DATA b l u e m a r b le.gif
END
END

Note:
• Comments in a mapfile are specified with a ‘#’ character
• MapServer parses mapfiles from top to bottom, therefore layers at the end of the mapfile will be drawn last
(meaning they will be displayed on top of other layers)
• Using relative paths is always recommended
• Paths should be quoted (single or double quotes are accepted)

14 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

Figure 2.2: Rendered Bluemarble Image

2.4.1 MAP Object

MAP
NAME "sample"
EXTENT -180 -90 180 90 # Geographic
SIZE 800 400
IMAGECOLOR 128 128 255
END

• EXTENT is the output extent in the units of the output map

• SIZE is the width and height of the map image in pixels
• IMAGECOLOR is the default image background color

2.4.2 LAYER Object

• starting with MapServer 5.0, there is no limit to the number of layers in a mapfile
• DATA parameter is relative to the SHAPEPATH parameter the MAP object
• if no DATA extension is provided in the filename, MapServer will assume it is an ESRI shapefile (.shp)

Raster Layers

LAYER
NAME "bathymetry"
TYPE RASTER
STATUS DEFAULT
DATA "bath_mapserver.tif"
END

2.4. Introduction to the Mapfile 15

MapServer Documentation, Release 6.0

See Also:
Raster Data

Vector Layers

Vector layers of TYPE point, line, or polygon can be displayed. The following example shows how to display only
lines from a TYPE polygon layer, using the OUTLINECOLOR parameter:
LAYER
NAME "world_poly"
DATA ’shapefile/countries_area.shp’
STATUS ON
TYPE POLYGON
CLASS
NAME ’The World’
STYLE
OUTLINECOLOR 0 0 0
END
END
END # layer

See Also:
Vector Data

Figure 2.3: Rendered Bluemarble image with vector boundaries

2.4.3 CLASS and STYLE Objects

• typical styling information is stored within the CLASS and STYLE objects of a LAYER
• starting with MapServer 5.0, there is no limit to the number of classes or styles in a mapfile
• the following example shows how to display a road line with two colors by using overlayed STYLE objects

16 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

CLASS
NAME "Primary Roads"
STYLE
SYMBOL "circle"
COLOR 178 114 1
SIZE 15
END #style1
STYLE
SYMBOL "circle"
COLOR 254 161 0
SIZE 7
END #style2
END

Figure 2.4: Rendered Bluemarble image with styled roads

2.4.4 SYMBOLs

• can be defined directly in the mapfile, or in a separate file

• the separate file method must use the SYMBOLSET parameter in the MAP object:
MAP
NAME "sample"
EXTENT -180 -90 180 90 # Geographic
SIZE 800 400
IMAGECOLOR 128 128 255

2.4. Introduction to the Mapfile 17

MapServer Documentation, Release 6.0

SYMBOLSET "../etc/symbols.txt"
END

where symbols.txt might contain:

SYMBOL
NAME "ski"
TYPE PIXMAP
IMAGE "ski.gif"
END

and the mapfile would contain:

LAYER
...
CLASS
NAME "Ski Area"
STYLE
SYMBOL "ski"
END
END
END # layer

Figure 2.5: Rendered Bluemarble image with skier symbol

See Also:
Cartographical Symbol Construction with MapServer, Symbology Examples, and SYMBOL

18 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

2.4.5 LABEL

• defined within a LAYER object

• the LABELITEM parameters in the LAYER object can be used to label by a specific column in the data refer to
a FONTSET file, that is set in the MAP object, that contains a reference to the available font names
An example LABEL object that references one of the above fonts might look like:
LABEL
FONT "sans-bold"
TYPE truetype
SIZE 10
POSITION LC
PARTIALS FALSE
COLOR 100 100 100
OUTLINECOLOR 242 236 230
END # label

Figure 2.6: Rendered Bluemarble image with skier symbol and a label

See Also:
LABEL, FONTSET

2.4.6 CLASS Expressions

MapServer supports three types of CLASS expressions in a LAYER:

2.4. Introduction to the Mapfile 19

MapServer Documentation, Release 6.0

1. String comparisons
(EXPRESSION "africa")

2. Regular expressions
(EXPRESSION /^9|^10/)

3. Logical expressions
([POPULATION] > 50000 AND ’[LANGUAGE]’ eq ’FRENCH’)

Note: Logical expressions should be avoided wherever possible as they are very costly in terms of drawing time.
See Also:
Expressions

2.4.7 INCLUDE

Added to MapServer 4.10, any part of the mapfile can now be stored in a separate file and added to the main mapfile
using the INCLUDE parameter. The filename to be included can have any extension, and it is always relative to the
main .map file. Here are some potential uses:
• LAYER s can be stored in files and included to any number of applications
• STYLE s can also be stored and included in multiple applications
The following is an example of using mapfile includes to include a layer definition in a separate file:
If ‘shadedrelief.lay’ contains:
LAYER
NAME ’shadedrelief’
STATUS ON
TYPE RASTER
DATA ’GLOBALeb3colshade.jpg’
END

therefore the main mapfile would contain:

MAP
...
INCLUDE "shadedrelief.lay"
...
END

The following is an example of a mapfile where all LAYER s are in separate .lay files, and all other objects (WEB,
REFERENCE, SCALEBAR, etc.) are stored in a “.ref” file:
NAME "base"
#
# include reference objects
#
INCLUDE "../templates/template.ref"
#
# Start of layer definitions
#
INCLUDE "../layers/usa/usa_outline.lay"
INCLUDE "../layers/canada/base/1m/provinces.lay"
INCLUDE "../layers/canada/base/1m/roads_atlas_of_canada_1m.lay"

20 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

INCLUDE "../layers/canada/base/1m/roads_atlas_of_canada_1m_shields.lay"
INCLUDE "../layers/canada/base/1m/populated_places.lay"
END # Map File

Warning: Mapfiles must end with the .map extension or MapServer will not recognize them. Include files can
have any extension you want, however.

See Also:
INCLUDE

2.4.8 Get MapServer Running

MapServer version 5.2 OUTPUT=GIF OUTPUT=PNG

OUTPUT=JPEG OUTPUT=WBMP SUPPORTS=PROJ SUPPORTS=FREETYPE
SUPPORTS=WMS_SERVER INPUT=SHAPEFILE

You can also send a HTTP request directly to the MapServer CGI program without passing any configuration vari-
ables (e.g. http://your.domain.name/cgi-bin/ms4/mapserv.exe). If you receive the message, ‘No query information to
decode. QUERY_STRING not set.’, your installation is working.

2.4.9 Get Demo Running

Download the MapServer Demo. UnZip it and follow the directions in ReadMe.txt. You will need to move the demo
files to their appropriate locations on your HTTP server, and modify the Map File and html pages to reflect the paths
and URLs of your server. Next, point your browser to init.html and hit the ‘initialize button’. If you get errors, verify
that you have correctly modified the demo files.

2.5 Making the Site Your Own

Now that you have a working MapServer demo, you can use the demo to display your own data. Add new LAYERs
to your Map file that refer to your own geographic data layers. (You will probably want to delete the existing layers or
set their status to OFF.)
Unless you are adding layers that fall within the same geographic area as the demo, modify MAP EXTENT to match
the extent of your data. To determine the extent of your data, you can use ogrinfo. If you have access to a GIS, you
could use that as well. The MAP EXTENT needs to be in the units of your output projection.
If you add geographic data layers of different projections, you will need to modify your Map File to add a PROJEC-
TION block to the MAP (output projection ) and each of the LAYER (existing layer projection).

2.5.1 Adding Data to Your Site

MapServer supports several data input formats ‘natively’, and many more if it is compiled with the open source
libraries GDAL and OGR.

2.5. Making the Site Your Own 21

MapServer Documentation, Release 6.0

2.5.2 Vector Data

Vector data includes features made up of points, lines, and polygons. MapServer supports ESRI shapefiles by de-
fault, but it can be compiled to support spatially enabled databases such as PostgreSQL-PostGIS, Geography Markup
Language (GML), MapInfo, delimited text files, and more formats with OGR.
See the Vector Data reference for examples on how to add different geographic data sources to your MapServer project.

2.5.3 Raster Data

Raster data is image or grid data. By default, MapServer supports Tiff/GeoTiff, and EPPL7. With GDAL, it supports
GRASS, Jpeg2000, ArcInfo Grids, and more formats. If you do compile MapServer with GDAL, which includes tiff
support, do not compile with native tiff support, as this will cause a conflict. More specific information can be found
in the Raster Data reference.

2.5.4 Projections

Because the earth is round and your monitor (or paper map) is flat, distortions will occur when you display geographic
data in a two-dimensional image. Projections allow you to represent geographic data on a flat surface. In doing
so, some of the original properties (e.g. area, direction, distance, scale or conformity)of the data will be distorted.
Different projections excel at accurately portraying different properties. A good primer on map projections can be
found at the University of Colorado.
With MapServer, if you keep all of your spatial data sets in the same projection (or unprojected Latitude and Longi-
tude), you do not need to include any projection info in your Map File. In building your first MapServer application,
this simplification is recommended.
On-the-fly projection can be accomplished when MapServer is compiled with Proj.4 support. Instructions on how to
enable Proj.4 support on Windows can be found on the Wiki.

2.6 Enhancing your site

2.6.1 Adding Query Capability

There are two primary ways to query spatial data. Both methods return data through the use of templates and CGI
variable replacement. A QUERYMAP can be used to map the results of the query.
To be queryable, each mapfile LAYER must have a TEMPLATE defined, or each CLASS within the LAYER must have a
TEMPLATE defined. More information about the CGI variables used to define queries can be found in the MapServer
CGI Reference.

2.6.2 Attribute queries

The user selects features based on data associated with that feature. ‘Show me all of the lakes where depth is greater
than 100 feet’, with ‘depth’ being a field in the shapefile .dbf or the spatial database. Attribute queries are accomplished
by passing query definition information to MapServer in the URL (or form post). Mode=itemquery returns a single
result, and mode=itemnquery returns multiple result sets.
The request must also include a QLAYER, which identifies the layer to be queried, and a QSTRING which contains
the query string. Optionally, QITEM, can be used in conjunction with QSTRING to define the field to be queried.
Attribute queries only apply within the EXTENT set in the map file.

22 Chapter 2. An Introduction to MapServer

 MapServer Documentation, Release 6.0

2.6.3 Spatial queries

The user selects features based on a click on the map or a user-defined selection box. Again the request is passed
through a URL or form post. By setting mode=QUERY, a user click will return the one closest feature. In
mode=NQUERY, all features found by a map click or user-defined selection box are returned. Additional query
options can be found in the CGI.

2.6.4 Interfaces

See: OpenLayers http://openlayers.org

2.6.5 Data Optimization

Data organization is at least as important as hardware configuration in optimizing a MapServer application for perfor-
mance. MapServer is quite efficient at what it does, but by reducing the amount of processing that it needs to do at the
time of a user request, you can greatly increase performance. Here are a few rules:
• Index Your data - By creating spatial indexes for your shapefiles using shptree. Spatial indexes should also be
created for spatially aware databases such as PostGIS and Oracle Spatial.
• Tile Your Data - Ideally, your data will be ‘sliced up’ into pieces about the size in which it will be displayed.
There is unnecessary overhead to searching through a large shapefile or image of which you are only going
to display a small area. By breaking the data up into tiles and creating a tile index, MapServer only needs to
open up and search the data files of interest. Shapefile data can be broken into smaller tiles and then a tileindex
shapefile can be created using the tile4ms utility. A tileindex shapefile for raster files can also be created.
• Pre-Classify Your Data - MapServer allows for the use of quite complex EXPRESSIONs to classify data.
However, using logical and regular expressions is more resource intensive than string comparisons. To increase
efficiency, you can divide your data into classes ahead of time, create a field to use as the CLASSITEM and
populate it with a simple value that identifies the class, such as 1,2,3, or 4 for a four class data set. You can then
do a simple string comparison for the class EXPRESSION.
• Pre-Process Your Images - Do resource intensive processing up front. See the Raster Data reference for more
info.
• Generalize for Overview - create a more simple, generalized data layer to display at small scales, and then
use scale-dependent layers utilizing LAYER MINSCALE and LAYER MAXSCALE to show more detailed data
layers as the user zooms in. This same concept applies to images.
See Also:
Optimization

2.7 How do I get Help?

2.7.1 Documentation

• Official MapServer documentation lives here on this site.

• User contributed documentation exists on the MapServer Wiki.

2.7. How do I get Help? 23

MapServer Documentation, Release 6.0

2.7.2 Users Mailing List

Register and post questions to the MapServer Users listserv. Questions to the list are usually answered quickly and
often by the developers themselves. A few things to remember:
1. Search the archives for your answer first, people get tired of answering the same questions over and over.
2. Provide version and configuration information for your MapServer installation, and relevant snippets of your
map and template files.
3. Always post your responses back to the whole list, as opposed to just the person who replied to your question.

2.7.3 IRC

MapServer users and developers can be found on Internet Relay Chat. The channel is #mapserver on irc.freenode.net.

2.7.4 Gallery

See examples of existing MapServer applications.

2.7.5 Tutorial

Perry Nacionales built a great Tutorial on how to build a MapServer application. You are invited to extend the collec-
tion of examples if you see cases that are missing.

2.7.6 Test Suite

Download the MapServer Test Suite for a demonstration of some MapServer functionality.

2.7.7 Books

Web Mapping Illustrated , a new book by Tyler Mitchell that describes well and provides real-world examples for the
use of Web mapping concepts, Open Source GIS software, MapServer, Web services, and PostGIS.
Mapping Hacks , by Schuyler Erle, Rich Gibson, and Jo Walsh, creatively demonstrates digital mapping tools and
concepts. MapServer only appears in a handful of the 100 hacks, but many more are useful for concepts and inspiration.
Beginning MapServer: Opensource GIS Development , by Bill Kropla, is a new book focusing on MapServer. So new,
I haven’t seen it yet. According to the publisher, it covers installation and configuration, basic MapServer topics and
features, incorporation of dynamic data, advanced topics, MapScript, and the creation of an actual application.

24 Chapter 2. An Introduction to MapServer

 CHAPTER 3

MapServer Tutorial

Author Pericles S. Nacionales

Contact pnaciona at gmail.com
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Updated 2010-04-07
This tutorial was designed to give new users a quick (relatively speaking) introduction to the concepts behind
MapServer. It is arranged into four sections with each section having one or more examples and increasing in com-
plexity. Users can jump to any section at any time although it is recommended that absolute beginners work on the
first three sections sequentially.
Section one focuses on basic MapServer configuration concepts such as layer and class ordering, using vector and
raster data, projections and labeling. Section two provides examples on how to use HTML templates to create a simple
interface for an interactive web mapping application. Section three introduces the use of HTML templates to provide
a “query” interface. Finally, section four introduces some advanced user interface concepts.

3.1 Tutorial Timeframe

While some users can go through this tutorial in one day, those who work on each example in detail can probably
expect to finish in one week.

3.2 Tutorial Data

The dataset used in this tutorial was taken from the U.S. Department of the Interior’s National Atlas of the United
States. You can visit their web site at http://www.nationalatlas.gov. The dataset was clipped to the upper great lakes
region (Minnesota, Michigan, and Wisconsin) to reduce storage size. Additional raster images were added courtesy
of the TerraSIP project at the University of Minnesota. When using this tutorial, you are encouraged to use your own
dataset.
Like MapServer itself, this tutorial is open and customizable to anyone. This was done in the hope that someone (or
some folks) will help design and develop it further.
Download the data (and all html files) for this tutorial at http://demo.mapserver.org/tutorial/tutorial.zip.

25
MapServer Documentation, Release 6.0

3.3 Before Using the Tutorial

There are some prerequisites to using this tutorial:

1. Users will need to have a web server installed and running on your computer. This web server has to have
support for common gateway interface (CGI) programs.
2. Users should have a basic understanding of web servers and internet security. A poorly configured web server
can easily be attacked by malicious people. At the very least your software installation will be corrupted and
you’ll lose hours of productivity, at worst your computer can be used to attack other computers on the internet.
3. It is recommended that users of this tutorial read the Introduction to MapServer before proceeding with this
tutorial.
4. To use this tutorial, users will need to have a web server and a MapServer CGI program (mapserv or
mapserv.exe) installed in their systems. MapServer source code is available for download here. Documentation
exists on how to compile and install MapServer:
• for UNIX users, please read the MapServer UNIX Compilation and Installation HOWTO.
• Windows users should read the MapServer Win32 Compilation and Installation HOWTO
In addition, Windows users can also download precompiled binaries from here.

3.4 Windows, UNIX/Linux Issues

3.4.1 Paths

This tutorial was created in Linux/UNIX but should work with minimal changes on Windows platform. The main
differences are the paths in the map files. Windows users need to specify the drive letter of the hard disk where their
tutorial files reside. Here’s an example:
A UNIX map file might include a parameter like this:
SHAPEPATH "/data/projects/tutorial/data"

In Windows, the same parameters might look like this:

SHAPEPATH "C:/data/projects/tutorial/data"

or:
SHAPEPATH "C:\data\projects\tutorial\data".

Notice that either slash or backslash works in Windows. The usual backslash may work well for you if you want to
make a distinction between virtual (as in URLs or web addresses) and local paths in your map file. However, if you
plan to move your application to UNIX at some point, you’ll have the tedious task of switching all backslashes to
slashes.
While we’re on the subject of paths, keep in mind that paths in mapfiles are typically relative to the system’s root
directory: the slash (“/”) in UNIX or some drive letter (“C:”) in Windows. This is true except when specifi-
cally asked to enter a URL or when referencing a URL. When working with HTML template files, paths are rel-
ative to the web server’s root directory. i.e., “/tutorial/” is relative to “http://demo.mapserver.org/“. Please read
http://www.alistapart.com/articles/slashforward/ for a few insights on URLs.

26 Chapter 3. MapServer Tutorial

 MapServer Documentation, Release 6.0

3.4.2 Executable

Another issue is that UNIX executable files don’t require a .EXE or .COM extensions, but they do in Windows. If
you are using Windows, append .exe to all instances of “/cgi-bin/mapserv” or “/cgi-bin/mapserv50” to make it “/cgi-
bin/mapserv.exe” or “/cgi-bin/mapserv50.exe”.

3.5 Other Resources

Other documentation exist to give you better understanding of the many customizations MapServer offer. Please visit
the MapServer documentation page at http://www.mapserver.org. There you will find several HOWTO documents,
from getting started to using MapScript, a scripting interface for MapServer.

Back to Tutorial home | Proceed to Section 1

3.6 Section 1: Static Maps and the MapFile

• Take a shapefile. Any shapefile. We can use MapServer to display that shapefile on a web browser. Look...

– Example 1.1 - A map with a single layer

• We can display the same shapefile repeatedly. We can display the polygon attributes on one LAYER and and the line attrib

– Example 1.2 - A map with two layers

• And we can select which parts of the shapefile to display. We do this using the CLASS object...
– Example 1.3 - Using classes to make a “useful” map
• We can also label our maps...
– Example 1.4 - Labeling layers and label layers
• Or add raster data such as satellite images, aerial photographs, or shaded reliefs...
– Example 1.5 - Adding a raster layer
• We can reproject our data from just about any projection to just about any... Yeah, check it out!
– Example 1.6 - Projection/Reprojection <example1-6
• And we can use layers from other map servers on the internet (as long as they are WMS servers)...
– Example 1.7 - Adding a WMS layer
• MapServer can output to various formats such as PDF and GeoTIFF.
– Example 1.8 - A different output format
• MapServer not only generates static maps, it can also create interactive maps...
– Example 1.9 - The difference between map and browse mode

Back to Tutorial home | Proceed to Section 2

3.5. Other Resources 27

MapServer Documentation, Release 6.0

3.7 Section 2: CGI variables and the User Interface

So far we have only looked at the mapfile when creating maps. In creating web mapping applications, it is usually
our intention to make maps that can be changed by the user (of the application) interactively. That is, a user should be
able to change the content of (or the information in) the map. To accomplish this interactivity, we use the MapServer
HTML templates.

3.7.1 HTML Templates

A MapServer HTML template is essentially an HTML file with a few MapServer specific tags. These tags are the
MapServer CGI variables and are enclosed in square brackets “[]”. When the MapServer CGI program processes an
application, it first parses the query string and the mapfile, and produces the necessary output. Some of this output
will need to be written to the HTML template file which you would have to also specify in the mapfile using the web
template keyword (or in a separate HTML initialization file). The CGI program will replace all the variables in the
HTML template with the proper value before sending it back to the web browser. If you are to directly view an HTML
template on a web browser, there won’t be any maps rendered and you will instead get blank images and other junk.

Variables

MapServer provides several variables for web mapping: the “img” variable which you’ve seen in Example 1.9 is but
one example. There area few core CGI variables originally designed as part of the mapping interface but practically
all the mapfile parameters can be defined as variables. The definitive reference to the CGI variables can be found at
here.
We can also define our own variables, which MapServer will pass along to our application. For example, we can create
a variable called “root” to represent the root directory of this tutorial, the value for “root” will then be “/tutorial”.
When the MapServer CGI program processes our HTML template, it will replace every instance of he “[root]” tag
with “/tutorial”. You will see this in action for each of the following examples.

3.7.2 Examples

So, let’s build an interactive interface for our application...

• Users of a web mapping application should be able to pan and zoom on the map: Example 2.1 - Pan and Zoom
Controls
• They also should be able to turn on and off layers on a map: Example 2.2 - Layer Control
• A map should always include a scalebar. Example 2.3 - Adding a Scalebar
• If users are to navigate through the map, a reference map should be provided: Example 2.4 - Adding a Reference
Map
• The map should include a legend. Example 2.5- Adding a Legend

Back to Section 1 index | Proceed to Section 3

28 Chapter 3. MapServer Tutorial

 MapServer Documentation, Release 6.0

3.8 Section 3: Query and more HTML Templates

To learn more about query and HTML templates with MapServer, see examples 3.1 to 3.4 in the Tutorial Viewer.

Back to Section 2 index | Proceed to Section 4

3.9 Section 4: Advanced User Interfaces

To learn more about advanced navigation such as pan and rubber-band zoom with Javascript and MapServer CGI, see
examples 4.1 to 4.4 in the Tutorial Viewer.

Back to Section 3 index | Tutorial home

Begin tutorial

3.8. Section 3: Query and more HTML Templates 29

MapServer Documentation, Release 6.0

30 Chapter 3. MapServer Tutorial

 CHAPTER 4

Installation

4.1 Compiling on Unix

Author J.F. Doyon

Contact jdoyon at nrcan.gc.ca
Author Howard Butler
Contact hobu.inc at gmail.com
Revision $Revision: 8463 $
Date $Date: 2009-01-28 07:37:36 -0800 (Wed, 28 Jan 2009) $

Table of Contents
• Compiling on Unix
– Introduction
– Obtaining the necessary software
– libgd
– Anti-Grain Geometry Support
– OGC Support
– Spatial Warehousing
– Compiling
– Installation

4.1.1 Introduction

The University of Minnesota’s MapServer is an open-source and freely available map rendering engine for the web.
Due to its open-source nature, it can be compiled on a wide variety of platforms and operating systems. We will focus
on how to obtain, compile and install MapServer on UNIX-like platforms.
You might also check the MapServerCompilation wiki page for additional information.

31
MapServer Documentation, Release 6.0

4.1.2 Obtaining the necessary software

You can obtain the MapServer source code as well as the demo package from the Download section.
You can also get the latest MapServer source code from Subversion.

Required External Libraries

• libpng: libpng should be on your system by default. 1.2.12 is the current release with security patches, although
versions all the way back to 1.2.7 should work.
• freetype: Version 2.x or above is required by GD.
• GD: libgd is used by MapServer for rendering images. Version 2.0.28 or greater required. Version 2.0.29 or
later is required to use curved (following) labels, and version 2.0.34 is required for antialiasing (1 pixel wide
lines/outlines).
• zlib: Zlib should be on your system by default. 1.2.1 is the current release with security patches.

Highly Recommended Libraries

• libproj: libproj provides projection support for MapServer. Version 4.4.6 or greater is required.
• libcurl: libcurl is the foundation of OGC (WFS/WMS/WCS) client and server support. Version 7.10 or greater
is required
• OGR: OGR provides access to at least 18 different vector formats.
• GDAL: GDAL provides access to at least 42 different raster formats.
• AGG: AGG (Anti-Grain Geometry) is an optional dependency to enable high quality antialiased output for
vector data. Currently versions 2.4 and 2.5 are identical featurewise, and only vary in their licence (2.4 is BSD,
2.5 is GPL)

Optional External Libraries

• libtiff: libtiff provides TIFF (Tagged Image File Format) reading support to MapServer.
• libgeotiff libgeotiff provides support to read GeoTIFF files (TIFF files with geographic referencing).
• libjpeg: libjpeg allows MapServer to render images in JPEG format. A sufficient version should be installed by
default on your system. Version 6b is the current version and dates back to 1998.
• GEOS: GEOS allows MapServer to do spatial predicate and algebra operations (within, touches, etc & union,
difference, intersection). Requires version 4.10 or greater.
• libxml: libxml is required to use OGC SOS support in MapServer (versions 4.10 and greater).
• SDE Client Library: The client libraries for your platform should be part of the ArcSDE media kit. They are not
publicly available for download.
• Oracle Spatial OCI: The client libraries for your platform are available for download from Oracle’s website.
Ideally, your client library matches the database you are querying from, but this is not a hard requirement.
• libpq: libpq is required to support the use of PostGIS geometries within the PostgreSQL database. Ideally, your
client library matches the database you are querying from.
• pdflib (lite): PDFlib Lite is the Open Source version of PDFlib that allows MapServer to produce PDF output.
Version 4.0.3 or greater is required.

32 Chapter 4. Installation
 MapServer Documentation, Release 6.0

• libming: libming provides Macromedia Flash output to MapServer. Version 0.2a is required. Later versions are
not known to work.

4.1.3 libgd

There are a number of issues that you should be aware of when using GD in combination with MapServer.

Minimum libgd versions

MapServer aggressively takes advantage of new features and bug fixes in the latest versions of libgd. The minimum
required version to run MapServer is 2.0.29. Upgrading to at least 2.0.34 is advised as it includes an important bug
fix for antialiased lines. Configure should detect which version of libgd you have installed, but you can quickly check
yourself by issuing the following command:
gdlib-config --version

libiconv

If you intend to use international character sets, your version of libgd must be compiled against the GNU iconv
libraries. If you are using a pre-packaged version, it is very likely that this is the case. To check for yourself, issue the
following command and look for ‘-liconv’ in the output:
gdlib-config --libs

Pre-packaged/system libraries

If you intend to use your system’s libgd, ensure that you have the development package also installed so MapServer
can find and use the appropriate headers.

MacOSX

A useful FAQ on for libgd on OSX is available at http://www.libgd.org/DOC_INSTALL_OSX

FreeType support

The GD you compile MapServer against MUST be compiled against the FreeType library in order to use TrueType
fonts. MapServer no longer uses it’s own interface to FreeType, using it through GD instead.
When you run your “configure” script, look for the following output:
using GD ( -DUSE_GD_GIF -DUSE_GD_PNG -DUSE_GD_JPEG
-DUSE_GD_WBMP -DUSE_GD_TTF -DGD_HAS_GDIMAGEGIFPTR) from system libs.

If your GD is built against FreeType, you will see either “-DUSE_GD_TTF” (Or “-DUSE_GD_FT” for Freetype 2.x)
part. If it’s missing, you will need to recompile your GD to make sure you include FreeType support. See the GD
documentation for more information.
Also note that the configure script looks for the FreeType library separately as well, generating output looking some-
what like this:

4.1. Compiling on Unix 33

MapServer Documentation, Release 6.0

checking where FreeType is installed...

checking for FT_Init_FreeType in -lfreetype... yes
using libfreetype -lfreetype from system libs.

Even though you have FreeType installed on your system and the configure script finds it, does NOT mean you will
have TrueType font support. GD MUST be compiled against FreeType either way.

1px Anti-Aliasing and segfaults

Versions of libgd earlier than 2.0.34 contain a one very significant bug and will always cause a segfault if you attempt
to do one pixel wide antialiasing. You can manually patch older gd’s, or better yet upgrade to at least GD 2.0.34.
In gd.c, function gdImageSetAAPixelColor() change:
int dr,dg,db,p,r,g,b;
p = gdImageGetPixel(im,x,y);

to
int dr,dg,db,p,r,g,b;
if (!gdImageBoundsSafeMacro (im, x, y)) return;
p = gdImageGetPixel(im,x,y);

More detail about this patch (if you need any) was described by Steve Lime in a post to mapserver-users.

Curved label support

ANGLE FOLLOW, a new feature that allows MapServer to draw curved labels about a linear feature like a road,
requires libgd 2.0.29 and TrueType font support. Configure should autodetect if you have a sufficient libgd and
TrueType support to be able to use this feature.

4.1.4 Anti-Grain Geometry Support

Since version 5.0 MapServer supports the AGG rendering backend. Download the 2.4 tarball from the antigrain
website and just type make in the root directory. If you intend on using mapscript, you must beforehand tweak the agg
makefile to add -fPIC to the compiler options.

4.1.5 OGC Support

MapServer provides support for many OGC specifications. At 4.2.3, it provides support for WMS (Web Mapping
Service), SLD (Styled Layer Descriptor), WFS (Web Feature Service), and experimental support for WCS (Web
Coverage Service).

WMS support

WMS Server

Support for this specification is automatically enabled when you include PROJ.4 support. (–with-proj) You can check
this yourself by looking for the following in your “configure” output:
checking whether we should include WMS support...
OGC WMS compatibility enabled (-DUSE_WMS).

34 Chapter 4. Installation
 MapServer Documentation, Release 6.0

If, for some reason you DON’T want WMS support, you can force it off by passing “–without-wms” to your configure
script.
More information on using this feature is available in the WMS Server HOWTO available on the MapServer website.

WMS Client

Cascading is also supported. This allows mapserver to transparently fetch remote layers over WMS, basically acting
like a client, and combine them with other layers to generate the final map.
In order to enable this feature, you will need to pass the “–with-wmsclient” option to the configure script. MapServer
will automatically look for libcurl, which is also required.
To verify that the WMS Client feature is enabled, check the output from the configure script:
checking whether we should include WMS Client Connections support...
OGC WMS Client Connections enabled (-DUSE_WMS_LYR).

Note that this feature is disabled by default, you have to specifically request it.
More information on using this feature is available in the WMS Client HOWTO available on the MapServer website.

WFS support

WFS Server

Support for this specification is enabled by passing the configure script the “–with-wfs” option. OGR and PROJ.4
support is required.
You can check this yourself by looking for the following in your “configure” output:
checking whether we should include WFS Server support...
OGC WFS Server support enabled (-DUSE_WFS_SVR).

Note that this feature is disabled by default, you have to specifically request it.
More information on using this feature is available in the WFS Server HOWTO available on the MapServer website.

WFS Client

MapServer can also act as a WFS client. This effectively means that MapServer reads it’s data from a remote server’s
WFS output and renders it into a map, just like it would when reading data from a shapefile.
In order to enable this feature, you will need to make sure you include OGR (Built with Xerces support) and PROJ.4
support, and pass the “–with-wfsclient” option to your configure script. MapServer will automatically look for libcurl,
which is also required.
To verify that the WFS Client feature is enabled, check the output from the configure script:
checking whether we should include WFS Client Connections support...
OGC WFS Client Connections enabled (-DUSE_WFS_LYR).

Note that this feature is disabled by default, you have to specifically request it.
More information on using this feature is available in the WFS Client HOWTO available on the MapServer website.

4.1. Compiling on Unix 35

MapServer Documentation, Release 6.0

4.1.6 Spatial Warehousing

MapServer can use a wide variety of sources of data input. One of the solutions growing in popularity is to use spatially
enabled databases to store data, and to use them directly to draw maps for the web.
Here you will find out how to enable mapserver to talk to one of these products. Please refer to the MapFile reference
for more details on how to use these. This section only details how to compile MapServer for their use.

PostGIS

PostGIS adds support for geographic objects to the PostgreSQL object-relational database. In effect, PostGIS “spa-
tially enables” the PostgreSQL server, allowing it to be used as a backend spatial database for geographic information
systems (GIS), much like ESRI’s SDE or Oracle’s Spatial extension. PostGIS is included in many distributions’
packaging system, but you can also roll your own if needed.
MapServer can use PostGIS as a data source. In order to do so simply use “–with-postgis” when running your configure
script.
--with-postgis=/usr/local/pgsql/bin/pg_config

ArcSDE

MapServer allows you to use SDE as a data source both for geometry and attributes. In order to achieve this, you must
have the SDE client librairies at your disposition, and have them installed on the machine running MapServer.
In order to enable SDE support in MapServer, you have to compile it with two options specified:
--with-sde=/opt/sdeexe90
--with-sde-version=90

Oracle Spatial

Oracle’s Spatial Warehousing cartridge is also supported by MapServer. In order to connect to it, you will need to
compile MapServer against the Oracle libraries by passing the “–with-oraclespatial” argument to your configure script.
You will very likely need an ORACLE_HOME environment variable set to have it configure things correctly.
--with-oraclespatial=/opt/oracle

4.1.7 Compiling

First prepare the ground by making sure all of your required and/or recommended libraries are installed before at-
tempting to compile MapServer. This will make your life much less complicated ;). Here is the order that I usually
use:
1. Compile GD. This often means acquiring libjpeg, libpng, zlib, and freetype before actually compiling the library.
You shouldn’t have too much trouble finding binaries of the libraries that GD requires, and often, they will
already be installed with your system. On unix, I’ve had very little luck finding pre-compiled binaries of the
required GD library. See libgd section for notes about patching libgd if you plan to use antialiasing.
2. Compile GDAL/OGR. Describing how to compile GDAL/OGR is beyond the scope of this document. If you
have requirements for lots of different formats, make sure to install those libraries first. I often find that building
up a GDAL/OGR library often takes as long as compiling MapServer itself!
3. Compile Proj.4. Proj.4 is a straight-forward configure/make/make install library.

36 Chapter 4. Installation
 MapServer Documentation, Release 6.0

4. Compile libcurl. libcurl is a straight-forward configure/make/make install library.

5. Compile/install optional libraries. These might include SDE, PostGIS, Oracle Spatial, AGG, Ming, PDFlib, or
MyGIS. Mix and match as you need them.
6. Unpack the MapServer tarball and cd into the mapserver directory:
[user@host user]$ tar -zxvf mapserver-X.Y.Z.tar.gz

7. Configure your environment using “configure”. I often place my configure command in its own file and changes
its mode to be executable (+x) to save typing and have a record of how MapServer was configured.
./configure --with-sde=/usr/sde/sdeexe90 \
--with-sde-version=90 \
--with-ogr=/usr/local/bin/gdal-config \
--with-gdal=/usr/local/bin/gdal-config \
--with-httpd=/usr/sbin/httpd \
--with-wfsclient \
--with-wmsclient \
--enable-debug \
--with-curl-config=/usr/bin/curl-config \
--with-proj=/usr/local \
--with-tiff \
--with-gd=/usr/local/ \
--with-jpeg \
--with-freetype=/usr/ \
--with-oraclespatial=/usr/oracle \
--with-threads \
--with-wcs \
--with-postgis=/usr/local/database/bin/pg_config \
--with-libiconv=/usr \ # new in 4.8
--with-geos=/usr/local/bin/geos-config \ # new in 4.8
--with-libiconv=/usr \ # new in 4.8
--with-xml2-config=/usr/bin/xml2-config \ # new in 4.10
--with-sos \ # new in 4.10
--with-agg=/path/to/agg-2.4

8. Now that you have configured your build options and selected all the libraries you wish mapserver to use, you’re
ready to compile the source code into an executable.
This is actually quite simple, just execute “make”:
[user@host mapserver]$ make

9. There is no make install step in the installation of MapServer. The output of the compilation of MapServer is a
binary executable that you can use in a CGI execution environment.
To make sure all went well, look for the file called mapserv
[user@host mapserver]$ ls -al mapserv
-rwxr-xr-x 1 user user 351177 Dec 21 11:38 mapserv

A simple test is to try and run it:

[user@host mapserver]$ ./mapserv
This script can only be used to decode form results and
should be initiated as a CGI process via a httpd server.
[user@host mapserver]$

The message above is perfectly normal, and means exactly what it says. If you get anything else, something
went terribly wrong.

4.1. Compiling on Unix 37

MapServer Documentation, Release 6.0

4.1.8 Installation

MapServer binary

The MapServer program itself consists of only one file, the “mapserv” binary executable. This is a CGI executable,
meant to be called and run by your web server.
In this section, we will assume you are running Apache under its default directory structure in /usr/local/apache. You
may need to have privileges to edit your httpd.conf (the main apache configuration file), or have someone (such as
your webmaster) help you with the configuration details.
The main goal is to get the “mapserv” binary installed in a publicly accessible directory that is configured to run CGI
programs and scripts.

The basic install

Under a default configuration, the CGI directory is “/usr/local/apache/cgi-bin” (RedHat users will use
“/home/httpd/cgi-bin”). Placing the mapserv file in this directory makes it accessible by the following URL:
“http://yourhostname.com/cgi-bin/mapserv“. When accessing this URL through your web client, you should expect
the following output if all has worked well: “No query information to decode. QUERY_STRING is set, but empty.”
If you get this message, you’re done installing MapServer.

Common problems

File permissions

The most common problem one is likely to encounter when attempting to install the binary are permissions issues:
• You do not have write permissions into your web server’s CGI Directory. Ask your webmaster to install the file
for you.
• The web server gives you a “403 Permission denied” error. Make sure the user the web server runs as (usually
“nobody”) has execute permission on the binary executable. Making the file world executable is perfectly fine
and safe:
[user@host cgi-bin]$ chmod o+x mapserv

Apache errors

You may receive a few different type of errors as well if your web server configuration isn’t right:
• 500 Internal server error: This is a fairly generic error message. All it basically tells you is that the web server
was unsuccessful in running the program. You will have to consult the web server’s error log to find out more,
and may need to enlist the help of your webmaster/system administrator.

Where to go once you’ve got it compiled

The An Introduction to MapServer document provides excellent coverage of getting started with MapServer.

38 Chapter 4. Installation
 MapServer Documentation, Release 6.0

4.2 Compiling on Win32

Author Pericles Nacionales

Contact pnaciona at gmail.com
Revision $Revision: 8380 $
Date $Date: 2009-01-02 10:27:46 -0800 (Fri, 02 Jan 2009) $

Table of Contents
• Compiling on Win32
– Introduction
– Compiling
– Set up a Project Directory
– Download MapServer Source Code and Supporting Libraries
– The MapServer source code
– Set Compilation Options
– Compile the Libraries
– Compile MapServer
– Compiling MapServer with PostGIS support
– Common Compiling Errors
– Installation
– Other Helpful Information
– Acknowledgements

4.2.1 Introduction

This document provides a simple set of compilation procedures for MapServer on Win32 platforms.
If you’ve made it this far, chances are you already know about MapServer and are at least tempted to try compiling it
for yourself. Pre-compiled binaries for MapServer are available from a variety of sources. Refer to Windows. Building
MapServer for win32 platforms can be a daunting task, so if existing binaries are sufficient for your needs, it is strongly
advised that they be used in preference to trying to build everything from source.
However, there can be a variety of reasons to want to build MapServer from source on win32. Reasons include the
need to enable specific options, to build with alternate versions of support libraries (such as GDAL), the desire for
MapScript support not part of the core builds, the need to debug and fix bugs or even to implement new features in
MapServer. To make it easy for users and developers, I’ve made a list of steps to compile MapServer. Background
information is provided in each step, along with examples. Each example is a continuation of the previous one and in
the end will produce the MapServer DLL (libmap.dll), the CGI program (the mapserv.exe), and utility programs.

Warning: This document may refer to older library versions. You may want to try to use more recent library
versions for your build.

4.2.2 Compiling

If you are new to Windows programming, please follow this document carefully. The compilation steps are fairly
simple but I’ve added a few blurbs in each step to help you understand how MapServer compiles. For the more
experienced programmers, perhaps reading the README.Win32 that accompanies the MapServer source code would
be more useful. For those who are antsy, compiling MapServer involves download and unpacking the source codes,

4.2. Compiling on Win32 39

MapServer Documentation, Release 6.0

editing the make files, and invoking Microsoft’s Visual C++ compiler from the command prompt. The resulting
mapserv.exe is the CGI program that installs in the cgi-bin directory of your web server.
For those who are willing to take the time, the compilation steps follow.

4.2.3 Set up a Project Directory

Before you start to compile MapServer, I recommend creating a directory called “projects” where you can put the
source code for MapServer and its supporting libraries. Since you will be working with DOS-style commands, you
might as well get used to the Windows command prompt. For Windows 95/98 users the command processor would be
called command.com. For Windows NT/2000/XP, it would be cmd.exe. So fire up the old command prompt and go to
the drive where you want to create the project directory.
Here is an example of how to create a directory called projects on the C: drive:
C:\Users> mkdir C:\Projects

To go to that directory:
C:\Users> cd \Projects
C:\Projects>

From the projects directory, you can extract the source codes for MapServer and its libraries. Now you’re ready to
download the source codes.

4.2.4 Download MapServer Source Code and Supporting Libraries

After creating a project directory, download the MapServer source code and the codes for the supporting libraries and
save the source code packages in the newly created “projects” directory. These source codes are usually packaged as
ZIP, or as UNIX TAR and GZIP files. You’ll need a software that can unzip these packages. 7-Zip is an example of
software that can handle these files.
Cygwin is a free, open-source software package which is a port of these tools on Windows. You can use the gzip and
tar utilities from this tool collection. Cygwin is available from http://www.cygwin.com.
In order to compile the MapServer CGI program, you must download a few required and optional libraries. At
its simplest configuration, MapServer only requires the GD (to provide the image output) and REGEX (to provide
regular expression support) libraries. This configuration allows the developer/data provider to use shapefiles as input
and, depending on the version of GD library used, GIF or PNG images as output. Additional libraries are needed for
input data in alternative formats. The libraries that work with MapServer are listed below.

4.2.5 The MapServer source code

The MapServer source code can be downloaded from the download page. If you’d like to get the current development
version of the software, following the nightly snapshot link under the Interim Builds title. The absolute latest copy of
the source code can be obtained from SVN; however, the SVN respository does not contain several important source
files (maplexer.c, mapparser.c and mapparser.h) normally generated on unix, so if possible, using a nightly snaphot is
substantially easier than working directly from Subversion.

Required Libraries

GD Library: MapServer uses the GD graphics library for rendering map images in GIF, PNG and JPEG format.
These map images are displayed in web browser clients using the MapServer CGI. The current official version
of GD is 2.0.33. The distributed makefiles are setup to use the prebuilt GD Win32 DLL binaries which include

40 Chapter 4. Installation
 MapServer Documentation, Release 6.0

GD, libjpeg, libpng, libz, libgif and FreeType 2 all within one DLL. This package is generally listed as “Windows
DLL .zip” and the latest version is normally available at http://www.boutell.com/gd/http/gdwin32.zip.
Regex: Regex is the regular expression library used by MapServer. It can be downloaded at http://ftp.gnu.org/old-
gnu/regex/regex-0.12.tar.gz

Optional Libraries

JPEG library: This library is required by GD to render JPEG images, if building GD from source. You may down-
load this library at http://www.ijg.org/files/jpegsrc.v6b.tar.gz
PNG library: This library is required by GD to render PNG images, if building GD from source. You may download
this library at http://sourceforge.net/projects/libpng/
Zlib: This library is required by libpng to provide graphics compression support. It can be downloaded along with
the PNG library, or at http://www.gzip.org/zlib.zip .
FreeType 2: FreeType provides TrueType support in MapServer via GD. We only need to build FreeType seperately
if building GD from source. It can be downloaded at http://gnuwin32.sourceforge.net/packages/freetype.htm .
PROJ.4: Proj.4 provides on-the-fly projection support to MapServer. Users whose data are in different projection
systems can use this library to reproject into a common projection. It is also required for WMS, WFS or WCS
services.
GDAL/OGR: The GDAL/OGR library allows MapServer to read a variety of geospatial raster formats (GDAL) and
vector formats (OGR). It can be downloaded at http://www.gdal.org/.
ArcSDE: ArcSDE is an ESRI proprietary spatial database engine. Most users will not have access to it but if you
have ArcSDE license, you can use its libraries to give MapServer access to SDE databases.
EPPL7: This library allows MapServer to read EPPL7 datasets, as well as the older Erdas LAN/GIS files. This library
is set as a default library in MapServer so there’s no special source code to download.
Now that you have reviewed the libraries that provide support to MapServer, it is time to decide which ones to compile
and use. We will work with the pre-built GD distributed on Boutell.com with PNG, GIF, JPEG, and FreeType “built
in”. If you want to provide OGC Web Services (ie. WMS, WFS) or want to perform on the fly reprojection then the
PROJ.4 library will be needed. If you need additional raster and vector data sources consider including GDAL/OGR
support. GDAL is also required for WCS service.
Our example calls for the required libraries and on-the-fly projection support so we need to download GD, regex, and
Proj.4 libraries. Go ahead and get those libraries.

4.2.6 Set Compilation Options

MapServer, like many of it’s support libraries, comes with a Visual C++ makefile called Makefile.vc. It includes the
file nmake.opt which contains many of the site specific definitions. We will only need to edit the nmake.opt file to
configure the build for our local site options, and support libraries. The Makefile.vc, and nmake.opt template file have
been provided by Assefa Yewondwossen, and the DM Solutions folks.
As of MapServer 4.4, the default MapServer build options only include GD, and regex. MapServer is built using the
/MD option (which means MSVCRT.DLL should be used), so if any support libraries are being built statically (rather
than as DLLs) we need to use /MD when building them as well. By default modern PROJ.4 builds use /MD so we
should be able to use the default PROJ.4 build without tweaking.
The example will compile with the GDWin32 pre-built DLL as well as regex-0.12, and PROJ.4. The PROJ.4 support
will ensure we can enable MapServer OGC-WMS compatibility. Use notepad or another text editor to open the
nmake.opt file and make the following changes.

4.2. Compiling on Win32 41

MapServer Documentation, Release 6.0

Comments

Use the pound sign ( # ) to comment out the lines that you want to disable, or remove the pound sign to enable an
option for NMAKE.
A. Enable PROJ.4 support, and update the path to the PROJ.4 directory. Uncomment the PROJ= line, and the
PROJ_DIR= line as follows, and update the PROJ_DIR path to point to your PROJ build.
# Reprojecting.
# If you would like mapserver to be able to reproject data from one
# geographic projection to another, uncomment the following flag
# Proj.4 distribution (cartographic projection routines). PROJ.4 is
# also required for all OGC services (WMS, WFS, and WCS).
#
# For PROJ_DIR use full path to Proj.4 distribution
PROJ=-DUSE_PROJ -DUSE_PROJ_API_H
PROJ_DIR=c:\projects\proj-4.4.9

If you look down later in the file, you can see that once PROJ is enabled, MapServer will be linked with proj_i.lib, the
PROJ.4 stub library, meaning that MapServer will be using the PROJ.DLL as opposed to statically linking in PROJ.4.
2. Uncomment the WMS option.
# Use this flag to compile with WMS Server support.
# To find out more about the OpenGIS Web Map Server Specification go to
# http://www.opengis.org/
WMS=-DUSE_WMS_SVR

3. Update to use GD. Here’s what it should look like in our example.
GD_DIR=c:/projects/gdwin32
GD_LIB=$(GD_DIR)/bgd.lib

Note: As distributed the GDWin32 binary build does not include the bgd.lib stub library. It is necessary to run the
makemsvcimport.bat script in the gdwin32 directory first.
D. Make sure the regex path is set correctly. In order for the “delete” command in the “nmake /f makefile.vc clean”
target to work properly it is necessary to use backslashes in the REGEX_DIR definition.
# REGEX Libary
#
# VC++ does not include the REGEX library... so we must provide our one.
# The following definitions will try to build GNU regex-0.12 located in the
# regex-0.12 sub-directory.
# If it was not included in the source distribution, then you can get it from:

# ftp://ftp.gnu.org/pub/gnu/regex/regex-0.12.tar.gz
# Provide the full path to the REGEX project directory
# You do not need this library if you are compiling for PHP mapscript.
# In that case the PHP regex library will be used instead
!IFNDEF PHP
REGEX_DIR=c:\projects\regex-0.12
!ENDIF

Your Makefile is now set.

42 Chapter 4. Installation
 MapServer Documentation, Release 6.0

4.2.7 Compile the Libraries

Before compiling MapServer, you must first compile its supporting libraries. How this is done varies for each library.
For the PROJ.4 library a nmake /f makefile.vc command in the proj-4.4.9src directory should be sufficient. The
regex-0.12 code is actually built by the MapServer build process, so you don’t need to do anything there.

Compiling libcurl

Previously, curl libraries can be compiled using the following command:

nmake /f makefile.vc6 CFG=release

This creates a static library, libcurl.lib, to which you compile against. Versions newer than version 7.10.x should be
compiled as dynamic library. This is accomplished using the command:
nmake /f makefile.vc6 CFG=release-dll

You will then need to edit MapServer’s nmake.opt to replace the CURL_LIB variable with this line:
CURL_LIB = $(CURL_DIR)/lib/libcurl_imp.lib

4.2.8 Compile MapServer

Once you have compiled the supporting libraries successfully, you are ready to take the final compilation step. If
you have not already done so, open a command prompt and set the VC++ environment variables by running the
vcvars32.bat usually located in C:Program FilesMicrosoft Visual StudioVC98binvcvars32.bat.
C:\Users> cd \projects\mapserver
C:\Projects\mapserver&> C:\Program Files\Microsoft Visual Studio\VC98\Bin\vcvars32.bat"
C:\Projects\mapserver>

Setting environment for using Microsoft Visual C++ tool.

C:\Projects\mapserver>

Now issue the command: nmake /f Makefile.vc and wait for it to finish compiling. If it compiles successfully, you
should get mapserver.lib, libmap.dll, mapserv.exe, and other .EXE files. That’s it for the compilation process. If you
run into problems, read section 4 about compiling errors. You can also ask for help from the helpful folks in the
MapServer-dev e-mail list.

4.2.9 Compiling MapServer with PostGIS support

To compile PostGIS support into MapServer, here’s what you need to do:
1. download the PostgreSQL 8.0.1 (or later) source from: ftp://ftp.heanet.ie/pub/postgresql/source/
2. I extracted them to C:projectspostgresql-8.0.1
3. download the Microsoft Platform SDK otherwise you get link errors on shfolder.lib.
4. compile libpq under C:projectspostgresql-8.0.1srcinterfaceslibpq using the win32.mak makefile
5. copy everything from C:projectspostgresql-8.0.1srcinterfaceslibpqrelease to C:projectspostgresql-
8.0.1srcinterfaceslibpq as the MapServer makefile will try to find it there
6. Define the following in the nmake.opt for MapServer: POSTGIS =-DUSE_POSTGIS POSTGIS_DIR
=c:/projects/postgresql-8.0.1/src

4.2. Compiling on Win32 43

MapServer Documentation, Release 6.0

7. nmake /f makefile.vc
8. don’t forget to copy libpq.dll (from C:projectspostgresql-8.0.1srcinterfaceslibpqrelease) into a location where
MapServer can find it.

4.2.10 Common Compiling Errors

Following are a few common errors you may encounter while trying to build MapServer.
• Visual C++ Tools Not Properly Initialized.
C:\projects\mapserver> nmake -f /makefile.vc
’nmake’ is not recognized as an internal or external command,
operable program or batch file.

This occurs if you have not properly defined the path and other environment variables required to use MS
VisualC++ from the command shell. Invoke the VCVARS32.BAT script, usually with the command C:Program
FilesMicrosoft Visual StudioVC98binvcvars32.bat or something similar if visual studio was installed in an
alternate location. To test if VC++ is available, just type “nmake” or “cl” in the command shell and ensure it is
found.
• Regex Build Problems.
regex.obj : error LNK2001: unresolved external symbol _printchar
libmap.dll : fatal error LNK1120: 1 unresolved externals
NMAKE : fatal error U1077: ’link’ : return code ’0x460’
Stop.

This occurs if you use the stock regex-0.12 we referenced. I work around this by commenting out the “extern”
statement for the printchar() function, and replacing it with a stub implementation in regex-0.12regex.c.
//extern void printchar ();
void printchar( int i ) {}

• GD Import Library Missing.

LINK : fatal error LNK1104: cannot open file ’c:/projects/gdwin32/bgd.lib’
NMAKE : fatal error U1077: ’link’ : return code ’0x450’
Stop.

If you are using the pre-built GD binaries, you still need to run the makemsvcimport.bat script in the gdwin32
directory to create a VC++ compatible stub library (bgd.lib).

4.2.11 Installation

The file we are most interested in is mapserv.exe. The other executable files are the MapServer utility programs.
See Also:
MapServer Utilities
to learn more about these utilities.
To test that the CGI program is working, type mapserv.exe at the command prompt. You should see the following
message:
This script can only be used to decode form results and
should be initiated as a CGI process via a httpd server.

44 Chapter 4. Installation
 MapServer Documentation, Release 6.0

You may instead get a popup indicating that a DLL (such as bgd.dll) is missing. You will need to copy all the required
DLLs (ie. bgd.dll, and proj.dll) to the same directory as the mapserv.exe program.
Now type mapserv -v at the command prompt to get this message:
MapServer version 4.4.0-beta3 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER INPUT=SHAPEFILE
DEBUG=MSDEBUG

This tells us what data formats and other options are supported by mapserv.exe. Assuming you have your web server
set up, copy mapserv.exe, libmap.dll, bgd.dll, proj.dll and any other required DLLs to the cgi-bin directory.
You are now ready to download the demo application and try out your own MapServer CGI program. If you wish,
you can also create a directory to store the utility programs. I’d suggest making a subdirectory called “bin” under
the directory “projects” and copy the executables to that subdirectory. You might find these programs useful as you
develop MapServer applications.

4.2.12 Other Helpful Information

The MapServer Unix Compilation and Installation HOWTO has good descriptions of some MapServer compilation
options and library issues. I will write more about those options and issues on the next revision of this HOWTO.
The README documents of each of the supporting libraries provide compilation instructions for Windows.
The MapServer User community has a collective knowledge of the nuances of MapServer compilation. Seek their
advice wisely.

4.2.13 Acknowledgements

Thanks to Assefa Yewondwossen for providing the Makefile.vc. I would not have been able to write this HOWTO
without that file.
Thanks to Bart van den Eijnden for the libcurl and PostGIS compilation info.
Thanks to the Steve Lime for developing MapServer and to the many developers who contribute time and effort in
order to keep the MapServer project successful.

4.3 PHP MapScript Installation

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Revision $Revision: 8459 $
Date $Date: 2009-01-27 10:13:56 -0800 (Tue, 27 Jan 2009) $

Table of Contents
• PHP MapScript Installation
– Introduction
– Obtaining, Compiling, and Installing PHP and the PHP/MapScript Module
– FAQ / Common Problems

4.3. PHP MapScript Installation 45

MapServer Documentation, Release 6.0

4.3.1 Introduction

The PHP/MapScript module is a PHP dynamically loadable module that makes MapServer’s MapScript functions and
classes available in a PHP environment.
The original version of MapScript (in Perl) uses SWIG, but since SWIG does not support the PHP language, the
module has to be maintained separately and may not always be in sync with the Perl version.
The PHP module was developed by DM Solutions Group and is currently maintained by Mapgears.
This document assumes that you are already familiar with certain aspects of your operating system:
• For Unix/Linux users, a familiarity with the build environment, notably make.
• For Windows users, some compilation skills if you don’t have ready access to a pre-compiled installation and
need to compile your own copy of MapServer with the PHP/MapScript module.

Which version of PHP is supported?

PHP MapScript was originally developed for PHP-3.0.14 but after MapServer 3.5 support for PHP3 has been dropped
and as of the last update of this document, PHP 4.3.11 or more recent was required (PHP5 is well supported).
The best combinations of MapScript and PHP versions are:
• MapScript 4.10 with PHP 5.2.1 and up
• MapScript 4.10 with PHP 4.4.6 and up

How to Get More Information on the PHP/MapScript Module for MapServer

• For a list of all classes, properties, and methods available in the module see the php reference document.
• More information on the PHP/MapScript module can be found on the PHP/MapScript page on MapTools.org.
• The MapServer Wiki also has PHP/MapScript build and installation notes and some php code snippets.
• Questions regarding the module should be forwarded to the MapServer mailing list.

4.3.2 Obtaining, Compiling, and Installing PHP and the PHP/MapScript Module

Download PHP and PHP/MapScript

• The PHP source or the Win32 binaries can be obtained from the PHP web site.
• Once you have verified that PHP is installed and is running, you need to get the latest MapServer source and
compile MapServer and the PHP module.

Setting Up PHP on Your Server

Unix
• Check if you have PHP already installed (several Linux distributions have it built in).
• If not, see the PHP manual’s “Installation on Unix systems” section.
Windows
• MS4W (MapServer For Windows) is a package that contains Apache, PHP, and PHP/MapScript ready to use in
a simple zipfile. Several Open Source applications are also available for use in MS4W.

46 Chapter 4. Installation
 MapServer Documentation, Release 6.0

• Windows users can follow steps in the Installing Apache, PHP and MySQL on Windows tutorial to install
Apache and PHP manually on their system.
• Window users running PWS/IIS can follow php.net’s howto for installing PHP for PWS/IIS 3, PWS 4 or newer,
and IIS 4 or newer.
Note: When setting up PHP on Windows, make sure that PHP is configured as a CGI and not as an Apache module
because php_mapscript.dll is not thread-safe and does not work as an Apache module (See the Example Steps of a Full
Windows Installation section of this document).

Build/Install the PHP/MapScript Module

Building on a Linux Box

NOTE: For UNIX users, see the README.CONFIGURE file in the MapServer source, or see the Compiling on Unix
HowTo.
• The main MapServer configure script will automatically setup the main makefile to compile php_mapscript.so
if you pass the –with-php=DIR argument to the configure script.
• Copy the php_mapscript.so library to your PHP extensions directory, and then use the dl() function to load the
module at the beginning of your PHP scripts. See also the PHP function extension_loaded() to check whether
an extension is already loaded.
• The file mapscript/php3/examples/phpinfo_mapscript.phtml will test that the php_mapscript module is properly
installed and can be loaded.
• If you get an error from PHP complaining that it cannot load the library, then make sure that you recompiled
and reinstalled PHP with support for dynamic libraries. On RedHat 5.x and 6.x, this means adding “-rdynamic”
to the CLDFLAGS in the main PHP3 Makefile after running ./configure Also make sure all directories in the
path to the location of php_mapscript.so are at least r-x for the HTTPd user (usually ‘nobody’), otherwise dl()
may complain that it cannot find the file even if it’s there.
Building on Windows
• For Windows users, it is recommended to look for a precompiled binary for your PHP version on the MapServer
download page or on MapTools.org.
• If for some reason you really need to compile your own Windows binary then see the README.WIN32 file in
the MapServer source (good luck!).

Installing PHP/MapScript

Simply copy the file php4_mapscript.dll to your PHP4 extensions directory (pathto/php/extensions)
Using phpinfo()
To verify that PHP and PHP/MapScript were installed properly, create a ‘.php’ file containing the following code and
try to access it through your web server:
<HTML>
<BODY>

<?php
if (PHP_OS == "WINNT" || PHP_OS == "WIN32")
{
dl("php_mapscript.dll");
}
else
{

4.3. PHP MapScript Installation 47

MapServer Documentation, Release 6.0

dl("php_mapscript.so");
}
phpinfo();
?>

</BODY>
</HTML>

If PHP and PHP/MapScript were installed properly, several tables should be displayed on your page, and ‘MapScript’
should be listed in the ‘Extensions’ table.

Example Steps of a Full Windows Installation

Using MS4W (MapServer for Windows)

1. Download the latest MS4W base package.
2. Extract the files in the archive to the root of one of your drives (e.g. C:/ or D:/).
3. Double-click the file /ms4w/apache-install.bat to install and start the Apache Web server.
4. In a web browser goto http://127.0.0.1. You should see an MS4W opening page. You are now running PHP,
PHP/MapScript, and Apache.
5. You can now optionally install other applications that are pre-configured for MS4W, which are located on the
MS4W download page.
Manual Installation Using Apache Server
1. Download the Apache Web Server and extract it to the root of a directory (eg. D:/Apache).
2. Download PHP4 and extract it to your Apache folder (eg. D:/Apache/PHP4).
3. Create a temp directory to store MapServer created GIFs. NOTE: This directory is specified in the IMAGEPATH
parameter of the WEB Object in the Mapfile reference. For this example we will call the temp directory
“ms_tmp” (eg. E:/tmp/ms_tmp).
4. Locate the file httpd.conf in the conf directory of Apache, and open it in a text viewer (eg. TextPad, Emacs,
Notepad).
In the Alias section of this file, add aliases to the ms_tmp folder and any other folder you require (for this
example we will use the msapps folder):
Alias /ms_tmp/ "path/to/ms_tmp/"
Alias /msapps/ "path/to/msapps/"

In the ScriptAlias section of this file, add an alias for the PHP4 folder.
ScriptAlias /cgi-php4/ "pathto/apache/php4/"

In the AddType section of this file, add a type for php4 files.
AddType application/x-httpd-php4 .php

In the Action section of this file, add an action for the php.exe file.
Action application/x-httpd-php4 "/cgi-php4/php.exe"

5. Copy the file php4.ini-dist located in your Apache/php4 directory and paste it into your WindowsNT folder (eg.
c:/winnt), and then rename this file to php.ini in your WindowsNT folder.

48 Chapter 4. Installation
 MapServer Documentation, Release 6.0

6. If you want specific extensions loaded by default, open the php.ini file in a text viewer and uncomment the
appropriate extension.
7. Place the file php_mapscript.dll into your Apache/php4/extensions folder.
Installation Using Microsoft’s IIS
(please see the IIS Setup for MapServer document for uptodate steps)
1. Install IIS if required (see the IIS 4.0 installation procedure).
2. Install PHP and PHP/MapScript (see above).
3. Open the Internet Service Manager (eg. C/WINNT/system32/inetsrv/inetmgr.exe).
4. Select the Default web site and create a virtual directory (right click, select New/Virtual directory). For this
example we will call the directory msapps.
5. In the Alias field enter msapps and click Next.
6. Enter the path to the root of your application (eg. “c:/msapps”) and click Next.
7. Set the directory permissions and click Finish.
8. Select the msapps virtual directory previously created and open the directory property sheets (by right clicking
and selecting properties) and then click on the Virtual directory tab.
9. Click on the Configuration button and then click the App Mapping tab.
10. Click Add and in the Executable box type: path/to/php4/php.exe %s %s. You MUST have the %s %s on the
end, PHP will not function properly if you fail to do this. In the Extension box, type the file name extension to
be associated with your PHP scripts. Usual extensions needed to be associated are phtml and php. You must
repeat this step for each extension.
11. Create a temp directory in Explorer to store MapServer created GIFs.
Note: This directory is specified in the IMAGEPATH parameter of the WEB Object in the Mapfile. For this
example we will call the temp directory ms_tmp (eg. C:/tmp/ms_tmp).
12. Open the Internet Service Manager again.
13. Select the Default web site and create a virtual directory called ms_tmp (right click, select New/Virtual direc-
tory). Set the path to the ms_tmp directory (eg. C:/tmp/ms_tmp) . The directory permissions should at least be
set to Read/Write Access.

4.3.3 FAQ / Common Problems

Questions Regarding Documentation

Q Is there any documentation available?

A The main reference document is the PHP MapScript reference, which describes all of the current
classes, properties and methods associated with the PHP/MapScript module.
To get a more complete description of each class and the meaning of their member variables, see the
MapScript reference and the MapFile reference.
The MapServer Wiki also has PHP/MapScript build and installation notes and some php code snip-
pets.

Q Where can I find sample scripts?

4.3. PHP MapScript Installation 49

MapServer Documentation, Release 6.0

A Some examples are included in directory mapserver/mapscript/php3/examples/ in the MapServer

source distribution. A good one to get started is test_draw_map.phtml: it’s a very simple script
that just draws a map, legend and scalebar in an HTML page.
A good intermediate example is the PHP MapScript By Example guide (note that this document was
created for an earlier MapServer version but the code might be still useful).
The next example is the GMap demo. You can download the whole source and data files from the
MapTools.org download page.

Questions About Installation

Q How can I tell that the module is properly installed on my server?

A Create a file called phpinfo.phtml with the following contents:
<?php dl("php_mapscript.so");
phpinfo();
?>

Make sure you replace the php_mapscript.so with the name under which you installed it, it could be
php_mapscript_46.so on Unix, or php_mapscript_46.dll on Windows
You can then try the second test page mapserver/mapscript/php3/examples/test_draw_map.phtml.
This page simply opens a MapServer .map file and inserts its map, legend, and scalebar in an HTML
page. Modify the page to access one of your own MapServer .map files, and if you get the expected
result, then everything is probably working fine.

Q I try to display my .phtml or .php page in my browser but the page is shown as it would it
Notepad.
A The problem is that your PHP installation does not recognize “.phtml” as a PHP file extension. As-
suming you’re using PHP4 under Apache then you need to add the following line with the other
PHP-related AddType lines in the httpd.conf:
AddType application/x-httpd-php .phtml

For a more detailed explanation, see the Example Steps of a Full Windows Installation section of
this document.

Q I installed the PROJ.4, GDAL, or one of the support libraries on my system, it is recognized by
MapServer’s “configure” as a system lib but at runtime I get an error: “libproj.so.0: No such
file or directory”.
A You are probably running a RedHat Linux system if this happened to you. This happens because the
libraries install themselves under /usr/local/lib but this directory is not part of the runtime library
path by default on your system.
(I’m still surprised that “configure” picked proj.4 as a system lib since it’s not in the system’s lib
path...probably something magic in autoconf that we’ll have to look into)
There are a couple of possible solutions:
1. Add a “setenv LD_LIBRARY_PATH” to your httpd.conf to contain that directory
2. Edit /etc/ld.so.conf to add /usr/local/lib, and then run “/sbin/ldconfig”. This will permanently
add /usr/local/lib to your system’s runtime lib path.

50 Chapter 4. Installation
 MapServer Documentation, Release 6.0

3. Configure MapServer with the following options:

--with-proj=/usr/local --enable-runpath

and the /usr/local/lib directory will be hardcoded in the exe and .so files
I (Daniel Morissette) personally prefer option #2 because it is permanent and applies to everything
running on your system.

Q Does PHP/MapScript have to be setup as a CGI? If so, why?

A Yes, please see the PHP/MapScript CGI page in the MapServer Wiki for details.

Q I have compiled PHP as a CGI and when PHP tries to load the php_mapscript.so, I get an
“undefined symbol: _register_list_destructors” error. What’s wrong?
A Your PHP CGI executable is probably not linked to support loading shared libraries. The MapServer
configure script must have given you a message about a flag to add to the PHP Makefile to enable
shared libs.
Edit the main PHP Makefile and add “-rdynamic” to the LDFLAGS at the top of the Makefile, then
relink your PHP executable.
Note: The actual parameter to add to LDFLAGS may vary depending on the system you’re running
on. On Linux it is “-rdynamic”, and on *BSD it is “-export-dynamic”.

Q What are the best combinations of MapScript and PHP versions?

A The best combinations are:
• MapScript 4.10 with PHP 5.2.1 and up
• MapScript 4.10 with PHP 4.4.6 and up

Q I am dynamically loading gd.so and php_mapscript.so and running into problems, why?
A The source of the problems could be a mismatch of GD versions. The PHP GD module compiles its
own version of libgd, and if the GD library is loaded before the mapscript library, mapscript will
use the php-specific version. Wherever possible you should use a gd.so built with the same GD as
PHPMapScript. A workaround is to load the php_mapscript module before the GD module.

4.4 .NET MapScript Compilation

Author Tamas Szekeres

Contact szekerest at gmail.com
Revision $Revision: 8386 $
Date $Date: 2009-01-04 14:03:29 -0800 (Sun, 04 Jan 2009) $

4.4. .NET MapScript Compilation 51

MapServer Documentation, Release 6.0

4.4.1 Compilation

Before compiling C# MapScript you should compile MapServer with the options for your requirements. For more
information about the compilation of MapServer please see Win32 Compilation and Installation Guide. It is highly
recommended to minimize the library dependency of your application, so when compiling MapServer enable only the
features really needed. To compile the C# binding SWIG 1.3.31 or later is required.

Warning: This document may refer to older library versions. You may want to try to use more recent library
versions for your build.

Win32 compilation targeting the MS.NET framework 1.1

You should compile MapServer, MapScript and all of the subsequent libraries using Visual Studio 2003. Download
and uncompress the latest SWIGWIN package that contains the precompiled swig.exe Open the Visual Studio .NET
2003 Command Prompt and step into the /mapscript/csharp directory. Edit makefile.vc and set the SWIG variable to
the location of your swig.exe
Use
nmake -f makefile.vc

to compile mapscript.dll and mapscript_csharp.dll.

Win32 compilation targeting the MS.NET framework 2.0

You should compile MapServer, MapScript and all of the subsequent libraries using Visual Studio 2005. Download
and uncompress the latest SWIGWIN package that contains the precompiled swig.exe Open the Visual Studio 2005
Command Prompt and step into the /mapscript/csharp directory Edit makefile.vc and set the SWIG variable to the
location of your swig.exe.
Use
nmake -f makefile.vc

to compile mapscript.dll and mapscript_csharp.dll.

Win32 compilation targeting the MONO framework

Before the compilation you should download and install the recent mono Win32 setup package (eg. mono-1.1.13.2-
gtksharp-2.8.1-win32-1.exe) Edit makefile.vc and set the CSC variable to the location of your mcs.exe. Alternatively
you can define
MONO = YES

in your nmake.opt file.

You should use the same compiler for compiling MapScript as the compiler has been used for the MapServer compi-
lation. To compile MapScript open the Command Prompt supplied with your compiler and use
nmake -f makefile.vc

to compile mapscript.dll and mapscript_csharp.dll.

52 Chapter 4. Installation
 MapServer Documentation, Release 6.0

Alternative compilation methods on Windows

Beginning from MapServer 4.8.3 you can invoke the C# compilation from the MapServer directory by uncommenting
DOT_NET in nmake.opt
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# .NET/C# MapScript
# ----------------------------------------------------------------------
# .NET will of course only work with MSVC 7.0 and 7.1. Also note that
# you will definitely want USE_THREAD defined.
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#DOT_NET = YES

and invoking the compilation by

nmake -f makefile.vc csharp

You can also use

nmake -f makefile.vc install

for making the compilation an copying the targets into a common output directory.

Testing the compilation

For testing the compilation and the runtime environment you can use
nmake -f makefile.vc test

within the csharp directory for starting the sample applications compiled previously. Before making the test the
location of the corresponding libraries should be included in the system PATH.

Linux compilation targeting the MONO framework

Before the compilation you should download and install the recent mono Linux package. Some distributions have pre-
compiled binaries to install, but for using the latest version you might want to compile and install it from the source.
Download and uncompress the latest SWIG release. You should probably compile it from the source if pre-compiled
binaries are not available for your platform.
Before compiling MapScript, MapServer should be configured and compiled. Beginning from MapServer 4.8.2 during
configuration the mapscript/csharp/Makefile will be created according to the configuration options. Edit this file and
set the SWIG and CSC for the corresponding executable pathes if the files could not be accessed by default. To compile
at a console step into the /mapscript/csharp directory use.
make

to compile libmapscript.so and mapscript_csharp.dll.

For testing the compilation and the runtime environment you can use
make test

for starting the sample applications compiled previously.

OSX compilation targeting the MONO framework

Beginning from 4.10.0 the csharp/Makefile supports the OSX builds. Before making the build the recent MONO
package should be installed on the system.

4.4. .NET MapScript Compilation 53

MapServer Documentation, Release 6.0

Before compiling MapScript, MapServer should be configured and compiled. Beginning from MapServer 4.8.2 during
configuration the mapscript/csharp/Makefile will be created according to the configuration options. Edit this file and
set the SWIG and CSC for the corresponding executable pathes if the files could not be accessed by default. To compile
at a console step into the /mapscript/csharp directory use
make

to compile libmapscript.dylib and mapscript_csharp.dll.

For testing the compilation and the runtime environment you can use
make test

for starting the sample applications compiled previously.

To run the applications mapscript_csharp.dll.config is needed along with the mapscript_csharp.dll file. This file is
created during the make process

4.4.2 Installation

The files required for your application should be manually installed. It is highly recommended to copy the files into
the same folder as the executable resides.

4.4.3 Known issues

Visual Studio 2005 requires a manifest file to load the CRT native assembly wrapper

If you have compiled MapServer for using the CRT libraries and you are using the MS.NET framework 2.0 as the
execution runtime you should supply a proper manifest file along with your executable, like:
<?xml version="1.0" encoding="utf-8"?>
<assembly xsi:schemaLocation="urn:schemas-microsoft-com:asm.v1 assembly.adaptive.xsd" manifestVersion
<assemblyIdentity name="drawmap.exe" version="1.0.0.0" type="win32" />
<dependency>
<dependentAssembly asmv2:dependencyType="install" asmv2:codebase="Microsoft.VC80.CRT.manifest" asmv2:
<assemblyIdentity name="Microsoft.VC80.CRT" version="8.0.50608.0" publicKeyToken="1fc8b3b9a1e18e3b" p
<hash xmlns="urn:schemas-microsoft-com:asm.v2">
<dsig:Transforms>
<dsig:Transform Algorithm="urn:schemas-microsoft-com:HashTransforms.Identity" />
</dsig:Transforms>
<dsig:DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1" />
<dsig:DigestValue>UMOlhUBGeKRrrg9DaaPNgyhRjyM=</dsig:DigestValue>
</hash>
</dependentAssembly>
</dependency>
</assembly>

This will inform the CLR that your exe depends on the CRT and the proper assembly wrapper is to be used. If you are
using the IDE the manifest file could be pregenerated by adding a reference to Microsoft.VC80.CRT.manifest within
the /Microsoft Visual Studio 8/VC/redist/x86/Microsoft.VC80.CRT directory.

Manifests for the dll-s must be embedded as a resource

According to the windows makefile the MapScript compilation target (mapscript.dll) is linked with the /MD option. In
this case the VS2005 linker will generate a manifest file containing the unmanaged assembly dependency. The sample
contents of the manifest file are:

54 Chapter 4. Installation
 MapServer Documentation, Release 6.0

<?xml version=’1.0’ encoding=’UTF-8’ standalone=’yes’?>

<assembly xmlns=’urn:schemas-microsoft-com:asm.v1’ manifestVersion=’1.0’>
<dependency>
<dependentAssembly>
<assemblyIdentity type=’win32’ name=’Microsoft.VC80.CRT’ version=’8.0.50608.0’ processorArchitecture=
</dependentAssembly>
</dependency>
</assembly>

Like previously mentioned if you are creating a windows application the common language runtime will search for a
manifest file for the application. The name of the manifest file should be the same as the executable append and end
with the .manifest extension. However if the host process is not controlled by you (like web mapping applications using
aspnet_wp.exe as the host process) you will not be certain if the host process (.exe) will have a manifest containing a
reference to the CRT wrapper. In this case you may have to embed the manifest into the dll as a resource using the mt
tool like:
mt /manifest mapscript.dll.manifest /outputresource:mapscript.dll;#2

the common language runtime will search for the embedded resource and load the CRT assembly properly.
Normally it is enough to load the CRT with the root dll (mapscript.dll), but it is not harmful embedding the manifest
into the dependent libraries as well.

Issue with regex and Visual Studio 2005

When compiling with Microsoft Visual Studio 2005 variable name collision may occur between regex.c and crtdefs.h.
For more details see:
http://trac.osgeo.org/mapserver/ticket/1651

C# MapScript library name mapping with MONO

Using the MapScript interface created by the SWIG interface generator the communication between the C# wrapper
classes (mapscript_csharp.dll) and the C code (mapscript.dll) takes place using platform invoke like:
[DllImport("mapscript", EntryPoint="CSharp_new_mapObj")]
public static extern IntPtr new_mapObj(string jarg1);

The DllImport declaration contains the library name, however to transform the library name into a file name is platform
dependent. On Windows the library name is simply appended with the .dll extension (mapscript.dll). On the Unix
systems the library file name normally starts with the lib prefix and appended with the .so extension (libmapscript.so).
Mapping of the library name may be manually controlled using a dll.config file. This simply maps the library file
the DllImport is looking for to its unix equivalent. The file normally contains the following information (map-
script_csharp.dll.config):
<configuration>
<dllmap dll="mapscript" target="libmapscript.so" />
</configuration>

and with the OSX builds:

<configuration>
<dllmap dll="mapscript" target="libmapscript.dylib" />
</configuration>

4.4. .NET MapScript Compilation 55

MapServer Documentation, Release 6.0

The file should be placed along with the corresponding mapscript_csharp.dll file, and created by default during the
make process. For more information see:
http://trac.osgeo.org/mapserver/ticket/1596 http://www.mono-project.com/Interop_with_Native_Libraries

Localization issues with MONO/Linux

According to http://trac.osgeo.org/mapserver/ticket/1762 MapServer may not operate equally well on different locale
settings. Especially when the decimal separator is other than “.” inside the locale of the process may cause parse errors
when the mapfile contains float numbers. Since the MONO process takes over the locale settings of the environment
it is worth considering to set the default locale to “C” of the host process, like:
LC_ALL=C mono ./drawmap.exe ../../tests/test.map test_csharp.png

4.4.4 Most frequent errors

This chapter will summarize the most frequent problems the user can run into. The issues were collected mainly from
the -users list and the IRC.

Unable to load dll (MapScript)

You can get this problem on Windows and in most cases it can be dedicated to a missing or an unloadable shared
library. The error message talks about mapscript.dll but surely one or more of the dll-s are missing that libmap.dll
depends on. So firstly you might want to check for the dependencies of your libmap.dll in your application directory.
You can use the Visual Studio Dependency Walker to accomplish this task. You can also use a file monitoring tool
(like SysInternal’s filemon) to detect the dll-s that could not be loaded. I propose to store all of the dll-s required by
your application in the application folder. If you can run the drawmap C# sample application with your mapfile your
compilation might be correct and all of the dlls are available.
You may find that the MapScript C# interface behaves differently for the desktop and the ASP.NET applications.
Although you can run the drawmap sample correctly you may encounter the dll loading problem with the ASP.NET
applications. When creating an ASP.NET project your application folder will be ‘Inetpubwwwroot[YourApp]bin’
by default. The host process of the application will aspnet_wp.exe or w3wp.exe depending on your system. The
application will run under a different security context than the interactive user (under the context of the ASPNET
user by default). When placing the dll-s outside of your application directory you should consider that the PATH
environment variable may differ between the interactive and the ASPNET user and/or you may not have enough
permission to access a dll outside of your application folder.

4.4.5 Bug reports

If you find a problem dedicated to the MapScript C# interface feel free to file a bug report to the Issue Tracker.

4.5 IIS Setup for MapServer

Author Debbie Paqurek

Last Updated 2005/12/12

56 Chapter 4. Installation
 MapServer Documentation, Release 6.0

Table of Contents
• IIS Setup for MapServer
– Base configuration
– Php.ini file
– Internet Services Manager
– Under the tree for your new website - add virtual directories for
– Test PHP
– Mapfiles for IIS
– Configuration files:

Some help on how to set up MapServer/Chameleon/PhpPgAdmin on Microsoft IIS (v5.0). Contains note on changes
to the php.ini file and necessary changes to the MapServer mapfiles. Please contribute or make changes as required.

4.5.1 Base configuration

• Windows 2000
• IIS 5.0
• MS4W 1.2.1
• Chameleon 2.2
• PHP 4.3.11
• MapServer 4.7
• PhpPgAdmin 3.5.4 (if using postgresql/postgis)
• Postgres 8.0.3 (if using postgresql/postgis)
• Postgis 1.0.3 (if using postgresql/postgis)
This setup assumes that MS4W was unzipped to form c:\ms4w\ directory.<br>

4.5.2 Php.ini file

• session.save_path (absolute path to your tmp directory)

• extension_dir (relative path to your php/extensions directory)
• cgi.force_redirect = 0</li>
• enable the pg_sql extension (php_pgsql.dll) (for Postgresql)

4.5.3 Internet Services Manager

Under your website tree, create a new website (e.g. msprojects). View the properties for the new website.
Web Site Tab
• set the IP address and under the Advanced tab put the complete Host Header name (e.g.msprojects.gc.ca).
Home Directory Tab
• content should come from: A directory located on this computer.
• Local Path: c:\ms4w\Apache\htdocs

4.5. IIS Setup for MapServer 57

MapServer Documentation, Release 6.0

• Read access + whatever else you need

• Execute Permissions: Scripts only
• Configuration button - App Mappings (Add extensions .php and .phtml, Executable is
c:\ms4w\Apache\cgi-bin\php.exe,select All verbs, Script Engine, and check that file exists<br>
Documents Tab
• Add index.phtml and index.html
• Directory Security Tab
– Anonymous access amd authentication control
– Select Anonymous access and the edit button should indicate the IUSR_account
Server Extensions Tab
• Enable authoring is selected and client scripting says Javascript

4.5.4 Under the tree for your new website - add virtual directories for

cgi-bin Under Properties, virtual directory tab Local Path should point to c:\ms4w\apache\cgi-bin. Select Read.
Execute Permissions should say “scripts and executables”
ms_tmp Under Properties, virtual directory tab Local Path should point to c:\ms4w\tmp\ms_tmp. Select Read, Write.
Execute Permissions should say “scripts only”. This is where temporary images are written to so in the File
system Security tab (use windows explorer), the c:\ms4w\tmp\ms_tmp directory should have permissions set
for the Internet Guest Account (Read and execute, Read, Write, List Folder Contents).
tmp Under Properties, virtual directory tab Local Path should point to c:\ms4w\tmp. Select Read, Write. Execute
Permissions should say “scripts only”. This is where chameleon writes sessions to so in the File system Secu-
rity tab (use windows explorer), the c:\ms4w\tmp directory should have permissions set for the Internet Guest
Accounnt (Read and execute, Read, Write, List Folder Contents).
chameleon Under Properties, virtual directory tab Local Path should point to C:\ms4w\apps\chameleon\htdocs. Select
Read. Execute Permissions should say “scripts only”. Under the Chameleon tree, you can add virtual directories
for admin (c:\ms4w\apps\chameleon\admin\htdocs), samples (c:\ms4w\apps\chameleon\samples\htdocs), cwc2
(c:\ms4w\apps\chameleon\cwc2\htdocs)
phppgadmin If using postgresql/postgis, under Properties, virtual directory tab Local Path should point to
C:\ms4w\Apache\htdocs\phpPgAdmin. Select Read, Write. Execute Permissions should say “scripts and ex-
ecutables”. Under Documents - add index.php.
Note: We had to unzip the phppgadmin package into this directory in order to get phppgadmin to show us the login
page at http://yourserver/phppgadmin/index.php. You might want additional security on this directory.
gmap Good for testing purposes. Remember to change your mapfiles as discussed in Mapfiles for IIS below. Under
Properties, virtual directory tab Local Path should point to C:\ms4w\apps\gmap\htdocs. Select Read. Execute
Permissions should say “scripts only”.

4.5.5 Test PHP

In a command line window, navigate to c:\ms4w\apache\cgi-bin and run php -i. This should return the out-
put that the phpinfo() function returns. I got an error about how it couldn’t find ntwdblib.dll. I found this in
c:\ms4w\apache\php\dlls and I copied it to the cgi-bin directory.

58 Chapter 4. Installation
 MapServer Documentation, Release 6.0

4.5.6 Mapfiles for IIS

• Add a config line to the MAP level of the mapfile

CONFIG PROJ_LIB "c:\ms4w\proj\nad\"

• change the IMAGEPATH to be an absolute path to your tmp/ms_tmp folder

IMAGEPATH "c:\ms4w\tmp\ms_tmp"

4.5.7 Configuration files:

For Chameleon
C:\ms4w\apps\chameleon\config\chameleon.xml
C:\ms4w\apps\chameleon\config\cwc2.xml

For phppgadmin: (if using postgresql/postgis)

C:\ms4w\apps\phpPgAdmin\conf\config.inc.php

4.6 Oracle Installation

Author Till Adams

Last Updated 2007/02/16

Table of Contents
• Oracle Installation
– Preface
– System Assumptions
– Compile MapServer
– Set Environment Variables

4.6.1 Preface

This document explains the whole configuration needed to get the connect between MapServer CGI and an Oracle
database server on a linux (Ubuntu) box. The aim of this document is just to put a lot of googled knowledge in ONE
place. Hopefully it will preserve many of people spending analog amount of time than I did!
This manual was written, because I spent several days googling around to get my UMN having access to an oracle
database. I’m NOT an oracle expert, so the aim of this document is just to put a lot of googled knowledge in ONE
place. Hopefully it will preserve many of people spending analog amount of time than I did! (Or: If you have the
choice: Try PostGIS ;-))
Before we start, some basic knowledge, I didn’t know before:
• MapServer can access oracle spatial as well as geodata from any oracle locator installation! Oracle locator
comes with every oracle instance, there is no need for an extra license.
• There is no need for further installation of any packages beside oracle/oracle OCI

4.6. Oracle Installation 59

MapServer Documentation, Release 6.0

4.6.2 System Assumptions

We assume that Oracle is already installed, there is a database and there is some geodata in the database. The following
paths should be known by the reader:
• ORACLE_HOME
• ORACLE_SID
• ORACLE_BASE
• LD_LIBRARY_PATH
We also assume that you have installed apache2 (our version was 2.0.49) and you are used to work with Linux/UNIX
systems. We also think you are able to handle the editor vi/vim.
We ensure that the Oracle user who later accesses the database has write-access to the oracle_home directory.
We also assume, that you already have setup the tnsnames.ora file. It should look like that:
MY_ORACLE =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = host)(PORT = 1521))
(CONNECT_DATA =
(SERVICE_NAME = your_name)
)
)

It is important that you know the NAME of the datasource, in this example this is “MY_ORACLE” and will be used
further on. Done that, you’re fine using User/Password@MY_ORACLE in your mapfile to connect to the oracle
database. But first we have to do some more stuff.

4.6.3 Compile MapServer

Compile as normal compilation and set this flag:

--with-oraclespatial=/path/to/oracle/home/</p>

If MapServer configure and make runs well, try

./mapserv -v

This should at least give this output:

INPUT=ORACLESPATIAL

If you got that, you’re fine from the MapServer point of view.

4.6.4 Set Environment Variables

It is important to set all environment variables correctly. There are one the one hand system-wide environment vari-
ables to be set, on the other hand there should be set some for the cgi-directory in your Apache configuration.

System Variables

On Ubuntu (and on many other systems) there is the file “/etc/profile” which sets environment variables for all users
on the system (you may also dedicate user-specific environment variables by editing the users “.profile” file in their
home directory, but usually the oracle database users are not users of the system with their own home)

60 Chapter 4. Installation
 MapServer Documentation, Release 6.0

Set the following variables:

$ cd /etc

$ echo export ORACLE_HOME=/path/to/oracle/home >> /etc/profile

# **(e.g. ORACLE_HOME=/app/oracle/ora10g)

$ echo export ORACLE_BASE=path/to/oracle >> /etc/profile

# **(e.g. ORACLE_HOME=/app/oracle)

$ echo export ORACLE_SID=MY_ORACLE >> /etc/profile

$ echo export LD_LIBRARY_PATH=path/to/oracle/home/lib >> /etc/profile

# **(e.g. ORACLE_HOME=/app/oracle/ora10g/lib)

The command comes silent, so there is no system output if you didn’t mistype anything!

Setting the Apache Environment

Sometimes it is confusing WHERE to set WHAT in the splitted apache2.conf-files. In the folder
“/etc/apache2/sites_available” you find your sites-file. If you did not do sth. Special e.g. installing virtual hosts,
the file is named “default”. In this file, the apache cgi-directory is defined. Our file looks like this:
ScriptAlias /cgi-bin/ /var/www/cgi-bin/
<Directory "/var/www/cgi-bin">
AllowOverride None
Options ExecCGI -MultiViews +SymLinksIfOwnerMatch
Order allow,deny
Allow from all
</Directory></p>

In this file, the local apache environment variables must be set. We did it within a location-block like this:
<Location "/cgi-bin/">
SetEnv ORACLE_HOME "/path/to/oracle/home"
</Location></p>

Where /cgi-bin/ in the opening location block refers to the script alias /cgi-bin/ and the TNS_ADMIN directory point
to the location of the tnsnames.ora file.
Then restart apache:
$ /etc/init.d/apache2 force-reload

Create mapfile

Before we start creating our mapfile ensure that you have a your access data (User/Password) and that you know the
Oracle SRID, which could be different from the proj-EPSG!
The data access parameters:
• CONNECTIONTYPE oraclespatial
• CONNECTION ‘user/password@MY_ORACLE‘

4.6. Oracle Installation 61

MapServer Documentation, Release 6.0

• DATA ‘GEOM FROM MY_LAYER USING SRID 82032’

[...]
Where:
• GEOM is the name of the geometry column
• MY_LAYER the name of the table
• 82032 is equivalent to the EPSG code 31468 (German projection system)

Testing & Error handling

So you are fine now. Load the mapfile in your application and try it. If everything goes well: Great, if not, possibly this
ugly error-emssage occurs (this one cmae by querying MapServer through the WMS interface as a GetMap-request):
<ServiceExceptionReport version="1.0.1">
<ServiceException>
msDrawMap(): Image handling error. Failed to draw layer named ’test1’.
msOracleSpatialLayerOpen(): OracleSpatial error. Cannot create OCI Handlers.
Connection failure. Check the connection string. Error: .
</ServiceException>
</ServiceExceptionReport>

This points us towards, that there might be a problem with the connection to the database. First of all, let’s check, if
the mapfile is all right. Therefore we use the MapServer utility program shp2img.
Let’s assume you are in the directory, where you compiled MapServer and run shp2img:
$ cd /var/src/mapserver_version/

$ shp2img -m /path/to/mapfile/mapfile.map -i png -o /path/to/output/output.png

The output of the command should look like this:

[Fri Feb 2 14:32:17 2007].522395 msDrawMap(): Layer 0 (test1), 0.074s
[Fri Feb 2 14:32:17 2007].522578 msDrawMap(): Drawing Label Cache, 0.000s
[Fri Feb 2 14:32:17 2007].522635 msDrawMap() total time: 0.075s

If not, this possibly points you towards any error in your mapfile or in the way to access the data directly. In this
case, take a look at Oracle Spatial. If there is a problem with your oracle connect, the same message as above
(MsDrawMap() ...) occurs. Check your mapfile syntax and/or the environment settings for Oracle.
For Debian/Ubuntu it’s worth also checking the file “/etc/environment” and test-wise to add the system variables
comparable to System Variables
If the output is OK, you may have a look at the generated image (output.png). Then your problem reduces to the access
of apache to oracle home directory. Carefully check your apache configuration. Please note, that the apache.config file
differs in several linux-distributions. For this paper we talk about Ubuntu, which should be the same as Debian.

62 Chapter 4. Installation
 CHAPTER 5

Mapfile

Author Steve Lime

Contact steve.lime at dnr.state.mn.us
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Author Jean-François Doyon
Contact jdoyon at ccrs.nrcan.gc.ca
The Mapfile is the heart of MapServer. It defines the relationships between objects, points MapServer to where data
are located and defines how things are to be drawn.
There are some important concepts that you must understand before you can reliably use mapfiles to configure
MapServer. First is the concept of a LAYER. A layer is the combination of data plus styling. Data, in the form of
attributes plus geometry, are given styling using CLASS and STYLE directives.
See Also:
An Introduction to MapServer for “An Introduction to the Mapfile”

5.1 Cartographical Symbol Construction with MapServer

Author Peter Freimuth

Contact pf at mapmedia.de
Author Arnulf Christl
Contact arnulf.christl at wheregroup.com
Author Håvard Tveite
Contact havard.tveite at umb.no
Revision $Revision: 11679 $
Date $Date: 2011-05-12 03:14:36 -0700 (Thu, 12 May 2011) $

63
MapServer Documentation, Release 6.0

Table of Contents
• Cartographical Symbol Construction with MapServer
– Abstract
– Introduction
* Multiple Rendering and Overlay
* Symbol Scaling
* Mapserver and symbol specification
– Using Cartographical Symbols in MapServer
* Output formats
* Symbol units
* Scaling of Symbols
– Construction of Point Symbols
* Symbols of TYPE VECTOR and ELLIPSE
* Symbols of TYPE truetype
* Symbols of TYPE pixmap
* Symbol definitions for the figure that demonstrates point symbols
* Combining symbols
– Construction of Line Symbols
* Overlaying lines
* Use of the PATTERN and GAP parameters
* Use of the OFFSET parameter
* Asymmetrical line styling with point symbols
– Area Symbols
* Hatch fill
* Polygon fills with symbols of TYPE pixmap
* Polygon fills with symbols of TYPE vector
– Examples (Mapserver 4)
* Basic Symbols
* Complex Symbols
* Mapfile Changes - line styling
– Tricks
* Changing the center of a point symbol
– Current Problems / Open Issues
* GAP - PATTERN incompatibility
– Summing up

5.1.1 Abstract

This Document refers to the syntax of MAP and symbol files for MapServer 6. The first version of the document
was based on the results of a project carried out at the University of Hannover, Institute of Landscape and Nature
Conservation. It was initiated by Mr. Dipl. Ing. Roland Hachmann. Parts have been taken from a study carried
through by Karsten Hoffmann, student of Geography and Cartography at the FU Berlin. In the context of a hands-on
training in the company GraS GmbH Mr. Hoffman mainly dealed with the development of symbols. (Download study
report in German) His degree dissertation will also concern this subject.
The document has been heavily revised for Mapserver 6.

64 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

5.1.2 Introduction

A map is an abstract representation that makes use of point, line and area symbols. Bertin (1974) created a clear and
logical symbol scheme in which symbols can be varied referring to graphical variables. The following graphical vari-
ables can be used within MapServer: FORM, SIZE, PATTERN, COLOR and LIGHTNESS. Point and area symbols
as well as text fonts (ttf) can additionally be displayed with a frame which we call OUTLINE.
The following figure shows the theoretical structure of cartographical symbols, which are also used in MapServer:

Figure 5.1: Structure of Cartographical Symbols‘

Multiple Rendering and Overlay

Say you want to display a highway with a black border line, two yellow lanes and a red center lane. This calls for a
combination of signatures.
Complex cartographical effects can be achieved by rendering the same vector data with different symbols, sizes and
colours on top of each other. This can be done using separate LAYERs. This could, however, have performance effects
for the application, as every rendering process of the same geometry will take up additional processor time. The
preferred solution is to use multiple STYLEs to create complex symbols by overlay.
To create the highway symbol mentioned above with a total width of 9 units, the lowest STYLE (in drawing order) will
be a broad black line with a width of 9 units. The second level STYLE will be a yellow line with a width of 7 units,
and the topmost STYLE will be a red line with a width of 1 unit. That way each yellow coloured lane will have a width
of (7-1)/2 = 3 units.

5.1. Cartographical Symbol Construction with MapServer 65

MapServer Documentation, Release 6.0

Combining symbols can be a solution for many kinds of cartographical questions. A combination of different geometry
types is also possible. A polygon data set can be rendered as lines to frame the polygons with a line signature. It can
also be rendred as polygons with a symbol filling the polygon. When the polygon fill is rendered on top of the lines,
the inner part of the underlying outline is covered by the fill symbol of the polygon. What is observed here is a clipping
effect tha in will result in an asymmetric symbol for the boundary line. To present the outline without clipping, just
reorder the LAYERs or STYLEs and put the outline presentation on top of the fill.
Yet another way to construct advanced line signatures for framed polygons is to tamper with the original geometries
by buffering or clipping the original geometry such that the new objects lie inside the original polygons or grow over
the borders. PostGIS can help achieve a lot of effects.
The OPACITY parameter of LAYER and STYLE can be used to achieve transparency (making lower symbols “shine”
through upper symbols).

Symbol Scaling

There are two basically different ways to handle the display size of symbols and cartographical elements in a map at
different scales. The size of cartographical elements is either set in screen pixel or in real world units.
• If the size is set in real world units (for example meters), it will shrink and grow according to the scale at which
the map is displayed.
• When sizes are given in screen pixels, symbols look the same at all scales.
MapServer implements the “screen pixels” size type for displaying cartographical elements. “Real world units”, as
described above, can be achieved using the SYMBOLSCALEDENOM parameter of the layer.

Mapserver and symbol specification

In a MapServer application, symbol parameters are organised in the map file as follows:
• Each LAYER has a TYPE parameter that defines the type of geometry (point, line or polygon). The symbols are
rendered at points, along lines or over areas accordingly.
• Basic symbols are defined in SYMBOL elements, using the parameters TYPE, POINTS, IMAGE, FILLED, and
more (SYMBOL elements can be collected in separate symbol files for reuse).
• Colour, lightness, size and outline are defined inside the STYLE sections of a CLASS section using the parameters
COLOR, SIZE, WIDTH and OUTLINECOLOR.
• Patterns for styling lines and polygons are defined in STYLE sections using GAP and PATTERN.
• Several basic elements can be combined to achieve a complex signature using several STYLEs inside one CLASS.
The following example shows the interaction of some of these elements and explains the configuration in the LAYER
and the SYMBOL sections necessary for rendering a cartographical point symbol (a red square with a 1 pixel wide
black outline and a smaller blue circle inside):

Figure 5.2: The generated overlay symbol

66 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Table 5.1: Commented Layer and Symbol sections.

LAYER section of the map file SYMBOL (from a separate symbol file or in-line in
the map file)
LAYER # Start of the layer definition
SYMBOL # Start of symbol definition
NAME "mytest" # The name of the layer NAME "square" # Name of the symbol (referenced in
TYPE POINT # Point geometries TYPE vector # The type of the symbol
STATUS DEFAULT # This layer will alwaysPOINTS
be drawn # Start of the geometry defining th
DATA t e s t # Refers to the dataset
0 0 test.shp
CLASS # Start of a Class definition
0 1
STYLE # Start of the first Style1 1
SYMBOL "square" # Symbol to be used1 (reference)
0
SIZE 16 # Size of the symbol
0 0in pixels
COLOR 255 0 0 # Colour (RGB) - END
red # end of POINTS
OUTLINECOLOR 0 0 0 # Outline colour FILLED
(RGB) -true
black # This area symbol should be filled
END # end of STYLE END # end of SYMBOL
STYLE # Start of the second Style
SYMBOL "circle" # Symbol to be SYMBOL
used (reference)# Start of symbol definition
SIZE 10 # Size of the symbol
NAME in pixels #
"circle" Name of the symbol (referenced in
COLOR 0 0 255 # Colour (RGB) - TYPE
blue ellipse # The type of the symbol
END # end of STYLE FILLED true # Start of the geometry defining th
END # end of CLASS POINTS
END # end of LAYER 1 1
END # end of POINTS
FILLED true # This area symbol should be filled
END # end of SYMBOL

5.1.3 Using Cartographical Symbols in MapServer

Vectors, truetype fonts and raster images are basic graphical elements that are defined by the TYPE parameter in the
STYLE element. This section explains how thezs elements can be combined to create complex cartographical symbols,
and it describes some other important aspects of map rendering in Mapserver.

Output formats

MapServer support raster output formats (e.g. PNG, JPEG and GIF) and vector output formats (e.g. PDF, SWG). The
raster formats (except for GIF) use anti-aliasing. See OUTPUTFORMAT for more.

Symbol units

The units used for specifying dimensions is defined in the SIZEUNITS parameter of the LAYE‘R. ‘SIZEUNITS can be
pixels, feet, inches, kilometers, meters, miles, dd and nauticalmiles. The default is pixels.
The MAP element’s RESOLUTION and DEFRESOLUTION parameters will determine the actual scale of the resulting
map and hence the size of the symbols on the map. DEFRESOLUTION is by default 72 dpi (dots per inch). If
RESOLUTION is specified to 144 (and DEFRESOLUTION is 72), all dimensions specified in the map file will be
multiplied by 144/72 = 2. This can be used to produce higher resolution images.
Dimensions can be specified using decimals.

5.1. Cartographical Symbol Construction with MapServer 67

MapServer Documentation, Release 6.0

Scaling of Symbols

The SYMBOLSCALEDENOM parameter in the LAYER section specifies the scale at which the symbol or text label
is displayed in exactly the dimensions defined in the STYLEs (for instance using SIZE and WIDTH). Observe that all
the parameters concerned with the symbol dimensions (SIZE, WIDTH, ...) are tightly connected to the SYMBOL-
SCALEDENOM parameter. The MAXSIZE and MINSIZE parameters inside the STYLE block limit the scaling of
symbols to the maximum and minimum size specified here (but does not affect the size calculations).
When the scale changes, the elements (defined in STYLEs) of a composite cartographical symbol may change their
positions relative to each other, especially at very small map scales (large scale denominators). They can also slightly
change their shape when they are displayed as tiny little images.
It is not possibile to define the display intervals with MINSCALEDENOM and MAXSCALEDENOM in the STYLE-
section, so this kind of tuning has to be solved at the LAYER level. To do this, create several layers with the same
geometries for different scale levels.
Always observe that cartographical symbols depend a lot on the scale! So be careful with the interaction of content,
symbols and scale when creating projects. All three parameters heavily interact and have to be coordinated to produce
a good map.

5.1.4 Construction of Point Symbols

In the figure below, point symbols of TYPE truetype, pixmap, ellipse and vector are demonstrated. The precise position
of the point for which the symbol is rendered is shown with a small red dot. A small blue dot is used to show an offset
position.

Figure 5.3: Basic point symbol TYPEs, showing effects of size, offset, angle and outlinecolor

68 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Symbols of TYPE VECTOR and ELLIPSE

The TYPE vector defines the shape of a symbol by setting X and Y values in a local two dimensional coordinate
system with X values increasing to the right and Y values increasing downwards.
The coordinates defining the symbol is listed in the POINTS element, which is explicitly ended using END. The
maximum number of points can be increased by changing the parameter MS_MAXVECTORPOINTS in the file map-
symbols.h before compilation. The current default is 100. Note that by setting the end point equal to the start point you
obtain a closed shape (polygon). A vector symbol can consist of several elements (lines, polygons). The coordinates
-99 -99 are used to separate the elements.
When creating symbols of TYPE vector you should observe some style guidelines. Avoid downtilted lines in area
symbols, as they will lead to heavy aliasing effects. Furthermore you should not go below a useful minimum size,
which is relevant for all types of symbols. Keep in mind that for pixel images, every symbol of TYPE vector has to be
rendered using pixels (pixmap symbols may be used directly for drawing).
Note: Current behaviour is for the bounding box of a vector symbol to always have 0,0 (in the SYMBOL coordinate
system) as its upper left corner. This can be used to generate asymmetrical symbols.
To create ellipses (and circles), use the TYPE ellipse. The shape of the ellipse is set in the POINTS elements (X - size
in the horizontal direction, Y - size in the vertical direction). To create a circle, let X and Y have the same value.

Symbols of TYPE truetype

You can use symbols from truetype fonts. The symbol settings are defined in the SYMBOL element. Specify the
character or the ASCII number of the character to be used in the CHARACTER parameter. The FONT parameter is
used to specify the font to be used (the alias name from the font file - often “fonts.list”). The FONTSET parameter of
the MAP element must be set for fonts to work.
??With the parameter ANTIALIAS you define whether to apply antialiasing to the symbols or characters. It is recom-
mended to do this especially with more complex symbols and and whenever they don’t fit well into the raster matrix
or show a visible pixel structure.??
In the STYLE section of the LAYER object it is possible to define colours for truetype symbols (as with signatures of
the TYPE vector). You can specify both fill colour and outline colour.
To find out the character number of a symbol use one of the following options:
• Use the software FontMap (Shareware, with free trial version for download, thanks Till!)
• Use the MS Windows truetype map
• Trial and Error :-)
Please note that the numbering of the so-called “symbol fonts” starts at 61440! So if you want to use character &#84,
you have to use 61440 + 84 = &#61524. (ain’t that a pain!!)
You can also place truetype characters and strings on the map using LABEL. The you can control the placing of the text
by using the POSITION parameter [ul|uc|ur|cl|cc|cr|ll|lc|lr], that specifies the position relative to the geometric origin
of the geometry.

Symbols of TYPE pixmap

Symbols of the TYPE pixmap are simply small raster images. The file name of the raster image is specified in the
IMAGE parameter of the SYMBOL element. MapServer supports the raster formats GIF and PNG for pixmaps.
Observe the colour depth of the images and avoid using 24 bit PNG symbols displayed in 8 bit mode as this may cause
unexpected colour leaps.

5.1. Cartographical Symbol Construction with MapServer 69

MapServer Documentation, Release 6.0

When using raster images, the colour cannot be modified in the SYMBOL element subsequently.
You can specify a colour with the TRANPARENT parameter which will not be displayed - i.e. it will be transparent.
As a result, all underlying objects and colours are visible.
Note: The TRANSPARENT parameter for pixmaps only works for the GD renderer (GIF rendering). The AGG
renderer supports natively transparent images, but does not support the TRANSPARENT parameter.
The SIZE parameter defines the height of pixmap symbols when rendered. The pixel structure will show when the
SIZE grows too large. If you are using symbol scaling (SYMBOLSCCALEDENOM is set) and want to prevent this
from happening, you should set the MAXSIZE parameter.
Pixmap symbols can be rotated using the ANGLE parameter.

Symbol definitions for the figure that demonstrates point symbols

This code was used to produce the symbols in the point symbol figure.
First, the symbol definitions:
SYMBOL
NAME "o-flag-trans"
TYPE pixmap
IMAGE "o-flag-trans.png"
END # SYMBOL

SYMBOL
NAME "circlef"
TYPE ellipse
FILLED true
POINTS
10 10
END # POINTS
END # SYMBOL

SYMBOL
NAME "P"
TYPE truetype
FONT "arial"
CHARACTER "P"
END # SYMBOL

SYMBOL
NAME "v-line"
TYPE vector
POINTS
0 0
5 10
10 0
END
END

SYMBOL
NAME "v-poly"
TYPE vector
FILLED false
POINTS
0 0
3.5 8

70 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

7 0
5.2 0
3.5 4
1.8 0
0 0
END
END

Then, the layers and styles used for producing the polygon V symbols in the point symbol figure:
LAYER # Vector v - polygon
STATUS DEFAULT
TYPE POINT
FEATURE
POINTS
10 30
END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
COLOR 0 0 0
END # STYLE
STYLE
SYMBOL "circlef"
COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

LAYER # Vector v - polygon, size

STATUS DEFAULT
TYPE POINT
FEATURE
POINTS
20 30
END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
COLOR 0 0 0
SIZE 30
END # STYLE
STYLE
SYMBOL "circlef"
COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

LAYER # Vector v - polygon, size, angle

STATUS DEFAULT
TYPE POINT
FEATURE
POINTS
30 30

5.1. Cartographical Symbol Construction with MapServer 71

MapServer Documentation, Release 6.0

END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
COLOR 0 0 0
SIZE 30
ANGLE 60
END # STYLE
STYLE
SYMBOL "circlef"
COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

LAYER # Vector v - polygon, size, offset

STATUS DEFAULT
TYPE POINT
FEATURE
POINTS
40 30
END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
COLOR 0 0 0
SIZE 30
OFFSET 0 15
END # STYLE
STYLE
SYMBOL "circlef"
COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

LAYER # Vector v - polygon, size, angle, offset

STATUS DEFAULT
TYPE POINT
FEATURE
POINTS
50 30
END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
COLOR 0 0 0
SIZE 30
ANGLE 60
OFFSET 0 15
END # STYLE
STYLE
SYMBOL "circlef"

72 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

LAYER # Vector v - polygon, size outline

STATUS DEFAULT
TYPE POINT
FEATURE
POINTS
60 30
END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
COLOR 0 0 0
SIZE 30
OUTLINECOLOR 0 255 0
END # STYLE
STYLE
SYMBOL "circlef"
COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

LAYER # Vector v - polygon, size, outline, width

STATUS DEFAULT
TYPE POINT
FEATURE
POINTS
70 30
END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
COLOR 0 0 0
SIZE 30
OUTLINECOLOR 0 255 0
WIDTH 4
END # STYLE
STYLE
SYMBOL "circlef"
COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

LAYER # Vector v - polygon, size, outline, no color

STATUS DEFAULT
TYPE POINT
FEATURE
POINTS

5.1. Cartographical Symbol Construction with MapServer 73

MapServer Documentation, Release 6.0

80 30
END # Points
END # Feature
CLASS
STYLE
SYMBOL "v-poly"
SIZE 30
OUTLINECOLOR 0 255 0
END # STYLE
STYLE
SYMBOL "circlef"
COLOR 255 0 0
SIZE 4
END # STYLE
END # CLASS
END # LAYER

Combining symbols

The following figure shows how to combine several basic symbols to create a complex point symbol. The combination
is achieved by adding several STYLEs within one layer. Each STYLE element references one SYMBOL element. All
the basic symbols are centered and overlayed when rendered.
Notice that the SIZE parameter in the STYLE element refers to the height of the symbol (extent in the Y direction).
A standing rectangle will thus display with a smaller area than a lying rectangle, although both have the same SIZE
parameter and the same maximum Y value in the SYMBOL element. When combining several basic point symbols on
top of each other, they will not always be centered correctly due to the integer mathematics required when rendering
raster images. It is recommended not to combine elements with even and odd numbered SIZE parameters, as this tends
to produce larger irregularities.

5.1.5 Construction of Line Symbols

For displaying line geometries you specify the width of the lines using the WIDTH parameter and the colour using the
COLOR parameter. If no colour is specified, the line will not be rendered. If no width is specified, a thin line (one unit
(pixel) wide) will result.

Overlaying lines

When combining several styles / symbols on a line, they will be positioned on the baseline which is defined by the
geometry of the object. In most cases MapServer correctly centers symbols. The combination of a line displayed in 16
units width and overlayed with a 10 unit width line results in a line symbol with a 3 unit border. If the cartographical
symbol is to contain a centered line with a width of 1 pixel, then the widths should be reconfigured, for example to 11
and 17 units. As a rule of thumb don’t combine even numbered and odd numbered widths.

Use of the PATTERN and GAP parameters

The PATTERN and GAP parameters can be used to produce styled lines in MapServer.
To create line patterns, use the PATTERN parameter of the STYLE. Here you define dashes by specifying the length
of the first dash, followed by the length of the first gap, then the length of the second dash, followed by the second gap,
and so on. This pattern will be repeated as many times as that pattern will fit into the line. LINECAP can be used to
control how the ends of the dashes are rendered. LINEJOIN can be used to control how sharp bends are rendered. In
the left column of the figure, you will find three examples where PATTERN has been used. Number 2 from below uses

74 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Figure 5.4: Construction of Point Symbols

5.1. Cartographical Symbol Construction with MapServer 75

MapServer Documentation, Release 6.0

LINECAP butt, number 3 from below uses LINECAP round (and LINEJOIN miter) and number 4 from below uses
LINECAP butt (and is overlayed over a wider, dark grey line). To produce dots, use 0 for dash length with LINECAP
‘round’.
Styled lines can be specified using GAP and a symbol for styling. In the figure, you will find four examples where
GAP has been used (in the right column). At the bottom a SYMBOL of TYPE ellipse has been used, then a SYMBOL
of TYPE vector, then a SYMBOL of TYPE font and at the top a SYMBOL of TYPE pixmap.
Note: It is currently not possible to specify an offset (start gap) when creating asymmetrical patterns.
The following figure shows how to use styles to define different kinds of line symbols.

Figure 5.5: Construction of Line Symbols

Note: For the line at the bottom of the right column, the red dots should have coincided with the black dots, but the
GAP length does not work as expected.
Below you will find the SYMBOLs and STYLEs that were used to produce the line symbols in “Construction of Line
Symbols”. The LAYERs are ordered from bottom to top of the figure.
Styles and symbols for lines
SYMBOL
NAME "circlef"
TYPE ellipse
FILLED true
POINTS
1 1
END # POINTS

76 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

END # SYMBOL

SYMBOL
NAME "P"
TYPE truetype
FONT "arial"
CHARACTER "P"
END # SYMBOL

SYMBOL
NAME "vertline"
TYPE vector
FILLED true
POINTS
0 5
0 10
1.4 10
1.4 5
0 5
END # POINTS
END # SYMBOL

SYMBOL
NAME "o-flag-trans"
TYPE pixmap
IMAGE "o-flag-trans.png"
END # SYMBOL

######## Left column ###############

LAYER # Simple line

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
5 5
25 10
45 10
35 5
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 6.5
END # STYLE
END # CLASS
END # LAYER

LAYER # Dashed line

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
5 15
25 20
45 20
35 15

5.1. Cartographical Symbol Construction with MapServer 77

MapServer Documentation, Release 6.0

END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 5.0
LINECAP butt
PATTERN 40 5 END
END # STYLE
END # CLASS
END # LAYER

LAYER # Dashed line, varying

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
5 25
25 30
45 30
35 25
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 5.0
LINECAP round #[butt|round|square|triangle]
LINEJOIN miter #[round|miter|bevel]
LINEJOINMAXSIZE 3
PATTERN 40 17 0 17 0 17 0 17 END
END # STYLE
END # CLASS
END # LAYER

LAYER # Line dash overlay

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
5 35
25 40
45 40
35 35
END # Points
END # Feature
CLASS
STYLE
COLOR 102 102 102
WIDTH 4.0
END # STYLE
STYLE
COLOR 255 255 255
WIDTH 2.0
LINECAP BUTT
PATTERN 8 12 END
END # STYLE
END # CLASS

78 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

END # LAYER

LAYER # Line overlay - 2

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
5 45
25 50
45 50
35 45
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 16.0
END # STYLE
STYLE
COLOR 209 66 0
WIDTH 10.0
END # STYLE
END # CLASS
END # LAYER

LAYER # Line overlay - 3

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
5 55
25 60
45 60
35 55
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 17.0
END # STYLE
STYLE
COLOR 209 66 0
WIDTH 11.0
END # STYLE
STYLE
COLOR 0 0 0
WIDTH 1.0
END # STYLE
END # CLASS
END # LAYER

######## right column ############

LAYER # Line - ellipse overlay

STATUS DEFAULT
TYPE LINE
FEATURE

5.1. Cartographical Symbol Construction with MapServer 79

MapServer Documentation, Release 6.0

POINTS
50 5
70 10
90 10
80 5
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 3.6
END # STYLE
STYLE
COLOR 0 0 0
SYMBOL "circlef"
SIZE 10
GAP 42
END # STYLE
STYLE
COLOR 255 0 0
SYMBOL "circlef"
SIZE 3
GAP 42
END # STYLE
END # CLASS
END # LAYER

LAYER # Line - symbol overlay

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
50 15
70 20
90 20
80 15
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 2.8
END # STYLE
STYLE
COLOR 0 0 0
SYMBOL "vertline"
SIZE 20.0
ANGLE 30
GAP -50
#OFFSET -5 -99
END # STYLE
END # CLASS
END # LAYER

LAYER # Line - font overlay

STATUS DEFAULT
TYPE LINE
FEATURE

80 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

POINTS
50 25
70 30
90 30
80 25
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 6
END # STYLE
STYLE
COLOR 102 0 0
SYMBOL "P"
SIZE 20
#SIZE 70
GAP -30
#OUTLINECOLOR 0 255 0
#WIDTH 4
#OFFSET -10 -99
END # STYLE
END # CLASS
END # LAYER

LAYER # Line - pixmap overlay

STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
50 35
70 40
90 40
80 35
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 6
END # STYLE
STYLE
COLOR 102 0 0
SYMBOL "o-flag-trans"
SIZE 20
#SIZE 70
GAP -30
#OUTLINECOLOR 0 255 0
#WIDTH 4
#OFFSET -10 -99
END # STYLE
END # CLASS
END # LAYER

LAYER # Line - pixmap overlay

STATUS DEFAULT
TYPE LINE
FEATURE

5.1. Cartographical Symbol Construction with MapServer 81

MapServer Documentation, Release 6.0

POINTS
50 45
70 50
90 50
80 45
END # Points
END # Feature
CLASS
STYLE
COLOR 0 0 0
WIDTH 6
END # STYLE
STYLE
COLOR 102 0 0
SYMBOL "o-flag-trans"
SIZE 20
GAP -30
OFFSET -10 -99
END # STYLE
END # CLASS
END # LAYER

By default, all lines (and patterns) will be drawn with rounded ends (extending the lines slightly beyond their ends).
This effect gets more obvious the larger the width of line is. It is possible to alter this behaviour using the LINECAP
parameter of the STYLE. LINECAP butt will give butt ends (stops the line exactly at the end), with no extension of
the line. LINECAP square will give square ends, with an extension of the line. LINECAP round is the default.

LINEJOIN

The different values for the parameter LINEJOIN have the following visual effects. Default is round. miter will follow
line borders until they intersect and fill the resulting area. none will render each segment using linecap butt. The figure
below illustrates the different linejoins.

Figure 5.6: Different kinds of linejoins

LINEJOINMAXSIZE

Specify the maximum length of m (only relevant for LINEJOIN type MITER). The value is a multiplication factor
(default 3). The actual max length is calculated as follows:
m - calculated join size
d - line width
m_max = d * LINEJOINMAXSIZE

If m > m_max, then the connection length will be set to m_max.

82 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Figure 5.7: Miter linejoin

Use of the OFFSET parameter

In STYLE, an OFFSET parameter can be set to shift symbols in the X and Y direction. The displacement is not
influenced by the direction of the line geometry. Therefore the point symbols used for styling are all shifted in the
same direction, independent of the direction of the line (as defined in style number 2 in the map file example below -
red line in the map image). A positive X value shifts to the right. A positive Y value shifts downwards.
To generate lines that are shifted relative to the original lines, -99 has to be used for the Y value of the OFFSET. Then
the X value defines the distance to the line from the original geometry (perpendicular to the line). A positive X value
will shift to the right (when viewed in the direction of the line), a negative X value will shift to the left.
The example below shows how OFFSET works with the use of -99 (blue and green lines) and without the use of -99
(red line). The thin black line shows the location of the line geometry.

5.1. Cartographical Symbol Construction with MapServer 83

MapServer Documentation, Release 6.0

Table 5.2: Line offset

Map file excerpt Resulting map image
LAYER #
STATUS DEFAULT
TYPE LINE
FEATURE
POINTS
20 20
280 160
280 20
160 20
160 60
END # Points
END # Feature
CLASS
STYLE # no offset
COLOR 0 0 0 # black
WIDTH 1
END # STYLE
STYLE # simple offset left and down
COLOR 255 0 0 # red
WIDTH 2
OFFSET -8 12
END # STYLE
STYLE # left offset rel. to line direction
COLOR 0 0 255 # blue
WIDTH 5
OFFSET -16 -99
END # STYLE
STYLE # right offset rel. to line direction
COLOR 0 255 0 # green
WIDTH 5
OFFSET 16 -99
END # STYLE
END # CLASS
END # LAYER

Asymmetrical line styling with point symbols

Line number 2 and 5 from below in the right column of the “Construction of Line Symbols” figure is an example of
asymmetrical line styling using a point symbol. This can be achieved either by using an OFFSET (with a Y value of
-99), or by using an asymmetrical point symbol of TYPE vector, as described in the tricks section below. Line number
2 from below is produced using an asymmetrical point symbol - this is best method for placing symbols on lines. Line
number 5 from below is produced using STYLE OFFSET. As can be seen, the symbols are here the rendered on the
offset line, meaning that at sharp bends, some symbols will be placed away from the line.

5.1.6 Area Symbols

Areas (polygons) can be filled with symbols to create for instance hatches and graticules.

84 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

The symbols are by default used as tiles, and rendered (without spacing) one after the other in the x and y direction,
filling the whole polygon.
If the SIZE parameter is used in the STYLE, the symbols will be scaled to the specified height.
The GAP parameter of the STYLE can be used to increase the spacing of the symbols.
The AGG renderer uses anti-aliasing by default, so edge effects around the symbols can occure.

Hatch fill

Simple line hatches (e.g. horizontal, vertical and diagonal) can be created by filling the polygon with a hatch symbol.
The SIZE parameter in the STYLE that uses a SYMBOL of type hatch specifies the distance from center to center
between the lines (the default is 1). The WIDTH parameter specifies the width of the lines in the hatch pattern (default
is 1). The ANGLE parameter specifies the direction of the lines (default is 0 - horizontal lines).
The figure demonstrates the use of SIZE (bottom left), WIDTH (bottom left), ANGLE (top right) and overlay (top
right) of hatches.

Figure 5.9: Hatch examples

The table shows excerpts of the map file that was used to produce the figure.

5.1. Cartographical Symbol Construction with MapServer 85

MapServer Documentation, Release 6.0

CLASS section SYMBOL definition

LAYER # hatch SYMBOL
... NAME "hatchsymbol"
CLASS TYPE hatch
STYLE END
SYMBOL "hatchsymbol"
COLOR 0 0 0
SIZE 10
END # STYLE
END # CLASS
END # LAYER

LAYER # hatch with angle

...
CLASS
STYLE
SYMBOL "hatchsymbol"
COLOR 0 0 0
SIZE 10
ANGLE 45
END # STYLE
END # CLASS
END # LAYER

LAYER # hatch with wide lines

...
CLASS
STYLE
SYMBOL "hatchsymbol"
COLOR 0 0 0
SIZE 10
WIDTH 5
END # STYLE
END # CLASS
END # LAYER

LAYER # cross hatch

...
CLASS
STYLE
SYMBOL "hatchsymbol"
COLOR 255 153 0
SIZE 10
WIDTH 4
END # STYLE
STYLE
SYMBOL "hatchsymbol"
COLOR 0 0 255
SIZE 20
ANGLE 90
END # STYLE
END # CLASS
END # LAYER

86 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Polygon fills with symbols of TYPE pixmap

Polygons can be filled with pixmaps.

Note: If the STYLE SIZE parameter is different from the image height of the pixmap, there will be rendering artefacts
around the pixmaps (visible as a grid with the “background” colour). This is due to the anti-aliasing done by the AGG
renderer.
Pixmap symbols can be rotated using the ANGLE parameter, but for polygon fills, this produces strange effects, and is
not recommended.
To create complex area symbols, e.g. with defined distances between single characters or hatches with broad lines,
pixmap fill is probably the best option. Depending on the desired pattern you have to generate the raster image with
high precision using a graphical editor. The figure below is an example of how to obtain a regular allocation of symbols
with defined spacing.

Figure 5.10: Raster image for a regular symbol fill**

You can use other shapes than circles. B defines the width and H the height of the raster image. For a regular
arrangment of symbols in a 45 degree angle B = H. For symbols, which are regularly arranged in parallel and without
offset between each other one centered symbol with same x and y distances to the imageborder is enough.
The following figure shows an example of how you can design a pixmap to produce a hatch with wide lines.
To create a 45 degree hatch use:
B = H and x = y

Note: When using the MapServer legend observe that each raster pixmap is displayed only once in the original size
in the middle of the legend box.
The example below shows some pixmap symbols which can be used as area symbols with transparency. The raster
images were created using FreeHand, finished with Photoshop and exported to PNG with special attention regarding
the colour palette. Observe that you have to specify a COLOR in the STYLE section of the CLASS to display raster
pixmaps although it has no influence on the output and colour of the image.
Construction of a horizontally arranged area symbol

5.1. Cartographical Symbol Construction with MapServer 87

MapServer Documentation, Release 6.0

Figure 5.11: Raster image for a hatched fill

CLASS section SYMBOL definition

CLASS SYMBOL
STYLE NAME "in_the_star"
COLOR 255 255 0 TYPE PIXMAP
END IMAGE "stern.png"
STYLE TRANSPARENT 8
SYMBOL "in_the_star" END
COLOR 0 0 0
OUTLINECOLOR 0 0 0
END
END

Figure 5.12: Polygon fill - regular grid pattern

Construction of a diagonally arranged area symbol

88 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

CLASS section SYMBOL definition

CLASS SYMBOL
STYLE NAME "in_point1"
SYMBOL "in_point1" TYPE PIXMAP
COLOR 0 0 0 IMAGE "flaeche1_1.png"
OUTLINECOLOR 0 0 0 TRANSPARENT 13
END END
END

Figure 5.13: Polygon fill - diagonal pattern

Construction of a hatch
CLASS section SYMBOL definition
CLASS SYMBOL
STYLE NAME "in_hatch"
COLOR 255 255 0 TYPE PIXMAP
END IMAGE "schraffur.png"
STYLE TRANSPARENT 2
SYMBOL "in_hatch" END
COLOR 0 0 0
OUTLINECOLOR 0 0 0
END
END

Figure 5.14: Polygon fill - hatch

Polygon fills with symbols of TYPE vector

Polygons can be filled with symbols of TYPE vector. As for the other symbol fills, the pattern will be generated by
using the specified symbol for the tiles. The bounding box of the symbol is used when tiling.
The upper right corner of the bounding box of a symbols of TYPE vector is always (0, 0). The lower left corner of the
bounding is determined by the maximum x and y values of the symbol definition.
Creating vector symbols for polygon fills is done in much the same way as for pixmap symbols. Precision is necessary
to get nice symmetrical symbols.

5.1. Cartographical Symbol Construction with MapServer 89

MapServer Documentation, Release 6.0

Both polygon (FILL on) and line (FILL off) vector symbols can be used. For line symbols, the WIDTH parameter of
the STYLE‘will give the line width and the ‘SIZE parameter will specify the height of the symbol.
STYLE ANGLE can be used for polygon fills, but will only rotate each individual symbol, not the pattern as a whole.
To generate rotated patterns is therefore quite demanding.
Below you will find some examples of vector symbols used for polygon fills. The polygon fill is accompanied by the
vector symbol used for the fill. The centre of the vector symbol is indicated with a red dot.

Figure 5.15: Polygon fills - vector

5.1.7 Examples (Mapserver 4)

The examples in this section were made for MapServer 4. Some of them may not work with later versions, but they
contain a lot of useful symbol definitions and are therefore provided as reference.
The symbols were created with MAP files and SYMBOL files, which can be download_old_symbols. If you want to
use these MAP files please note, that your MapServer must at least be able to handle 50 symbols. Otherwise you will
get an error while loading the SYMBOL files.

90 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Basic Symbols

5.1. Cartographical Symbol Construction with MapServer 91

MapServer Documentation, Release 6.0

92 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Complex Symbols

5.1. Cartographical Symbol Construction with MapServer 93

MapServer Documentation, Release 6.0

94 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Mapfile Changes - line styling

All line styling must be specified in layer STYLE. The following keywords have been removed from SYMBOL:
STYLE PATTERN GAP LINECAP LINEJOIN LINEJOINSIZE
The following keywords have been added to STYLE:
PATTERN GAP LINECAP LINEJOIN LINEJOINSIZE
The SYMBOL STYLE keyword has been renamed to PATTERN. The cartoline SYMBOL TYPE has been removed.

5.1.8 Tricks

Changing the center of a point symbol

Mapserver does all transformations (offset, rotation) with respect to the symbol center point. The center point is
calculated from the symbol’s bounding box. In some cases it can be useful to change the center point of a symbol.
Currently there is no way of explicitly specifying the center point of a SYMBOL. A mechanism for this (a new keyword
in SYMBOL that specifies the symbol center point) has been suggested in RFC45, but has not been implemented so
far.

5.1. Cartographical Symbol Construction with MapServer 95

MapServer Documentation, Release 6.0

When determining the position of the symbol center point, the lower x and y values of the bounding box is always set
to 0.
Here are some examples of what can be achieved by taking advantage of this for point symbols and decorated lines.
There are three examples in the illustration, and each example shows the result with and without the offset trick. At
the top arrows are added to lines using GEOMTRANSFORM start / end. In the middle, tags are added to lines using
GAP and ANGLE. At the bottom, a point symbol is shifted and rotated. The red dots represent the center points, and
the blue dots the offsets.

Figure 5.16: Shifting trick

Below you will find three tables that contain the SYMBOLS and the STYLE mechanisms that were used to genereate
shifted symbols in the illustration.

96 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Table 5.3: Symbol tricks - shift - arrows

SYMBOLs LAYER STYLEs
SYMBOL LAYER # Line
NAME "arrow-offset-end" STATUS DEFAULT
TYPE vector TYPE LINE
FILLED true FEATURE
POINTS POINTS
-5 0.4 20 65
-2 0.4 40 70
-2 0 60 70
0 0.8 70 65
-2 1.6 END # Points
-2 1.2 END # Feature
-5 1.2 CLASS
-5 0.4 STYLE
END # POINTS COLOR 0 0 0
END # SYMBOL WIDTH 15
SYMBOL LINECAP butt
NAME "arrow-offset-start" END # STYLE
TYPE vector STYLE
FILLED true G E O M TRANSFORM "start"
POINTS COLOR 0 255 0
5 0.4 SYMBOL "arrow-offset-start"
8 0.4 SIZE 15.0
8 0 ANGLE AUTO
10 0.8 END # STYLE
8 1.6 STYLE
8 1.2 G E O M TRANSFORM "start"
5 1.2 COLOR 255 0 0
5 0.4 SYMBOL "circlef"
END # POINTS SIZE 3
END # SYMBOL END # STYLE
STYLE
G E O M TRANSFORM "end"
COLOR 0 255 0
SYMBOL "arrow-offset-end"
SIZE 15.0
ANGLE AUTO
END # STYLE
STYLE
G E O M TRANSFORM "end"
COLOR 255 0 0
SYMBOL "circlef"
SIZE 3
END # STYLE
END # CLASS
END # LAYER

5.1. Cartographical Symbol Construction with MapServer 97

MapServer Documentation, Release 6.0

Table 5.4: Symbol tricks - shift - asymmetrical tags

SYMBOLs LAYER STYLEs
SYMBOL LAYER # Line - symbol overlay
NAME "fence-tag" STATUS DEFAULT
TYPE vector TYPE LINE
POINTS FEATURE
0 5 POINTS
0 10 20 50
END # POINTS 40 55
END # SYMBOL 60 55
70 50
SYMBOL END # Points
NAME "vert-line" END # Feature
TYPE vector CLASS
POINTS STYLE
0 0 COLOR 0 0 0
0 10 WIDTH 4
END # POINTS END # STYLE
END # SYMBOL STYLE
COLOR 0 0 0
SYMBOL "fence-tag"
SIZE 40.0
WIDTH 3
ANGLE 30
GAP -50
END # STYLE
STYLE
COLOR 255 0 0
SYMBOL "circlef"
SIZE 1
GAP -50
END # STYLE
END # CLASS
END # LAYER

98 Chapter 5. Mapfile
 MapServer Documentation, Release 6.0

Table 5.5: Symbol tricks - shift - asymmetrical symbols. Unshifted symbol v-line, shifted symbol v-line-offs
SYMBOLs
SYMBOL
NAME "v-line"
TYPE vector
POINTS
0 0
5 10
10 0
END # POINTS
END # SYMBOL

SYMBOL
NAME "v-line-offs"
TYPE vector
POINTS
0 10
5 20
10 10
END # POINTS
END # SYMBOL

5.1.9 Current Problems / Open Issues

GAP - PATTERN incompatibility

Creating advanced line symbols involving dashed lines are difficult due to the incompatibility of the dashed line
mechanisms (PATTERN) and the symbol on line placement mechanisms (GAP). A solution could be to allow GAP
be a list instead of a single number (perhaps renaming to GAPs or DISTANCES), but it would also be necessary to
introduce a new parameter to specify the distance to the first symbol on the line (‘STARTGAP‘?).
GAP does not support two dimensions, so the same gap will have to be used for for the x and the y directions. The
introduction of new parameters - GAPX and GAPY could be a solution to this.

5.1.10 Summing up

We hope that this document will help you to present your data in a cartographically nice manner with MapServer and
explains some basics and possibilities of the concept of MapServer as well as some weaknesses. It would be great to
put together a cartographical symbols library for the profit of everyone. This especially concerns truetype fonts, which
have been developed for some projects and contain some typical signatures for cartographical needs.
You can also view the discussion paper for the improvement of the MapServer Graphic-Kernel (German only).

5.2 CLASS

BACKGROUNDCOLOR [r] [g] [b] Color to use for non-transparent symbols.

COLOR [r] [g] [b] Color to use for drawing features.

5.2. CLASS 99
MapServer Documentation, Release 6.0

DEBUG [on|off] Enables debugging of the class object. Verbose output is generated and sent to the standard error
output (STDERR) or the MapServer logfile if one is set using the LOG parameter in the WEB object.
See Also:
MS RFC 28: Redesign of LOG/DEBUG output mechanisms
EXPRESSION [string] Four types of expressions are now supported to define class membership. String compar-
isons, regular expressions, simple logical expressions, and string functions. If no expression is given, then all
features are said to belong to this class.
• String comparisons are case sensitive and are the fastest to evaluate. No special delimiters are neces-
sary although string must be quoted if they contain special characters. (As a matter of good habit, it is
recommended you quote all strings).
• Regular expressions function just like previous versions of MapServer. However, you must now delimit a
regular expression using /regex/. No quotes should be used.
• Logical expressions allow you to build fairly complex tests based on one or more attributes and there-
fore are only available with shapefiles. Logical expressions are delimited by parentheses “(expres-
sion)”. Attribute names are delimited by square brackets “[ATTRIBUTE]”. These names are case
sensitive and must match the items in the shapefile. For example: EXPRESSION ([POPULATION]
> 50000 AND ‘[LANGUAGE]’ eq ‘FRENCH’) ... The following logical operators are supported:
=,>,<,<=,>=,=,or,and,lt,gt,ge,le,eq,ne. As you might expect this level of complexity is slower to process.
• One string function exists: length(). This obviously computes the length of a string. An example follows:
EXPRESSION (length(’[NAME_E]’) < 8)

String comparisons and regular expressions work from the classitem defined at the layer level. You may
mix expression types within the different classes of a layer.
GROUP [string] Allows for grouping of classes. It is only used when a CLASSGROUP at the LAYER level is set.
If the CLASSGROUP parameter is set, only classes that have the same group name would be considered at
rendering time. An example of a layer with grouped classes might contain:
LAYER
...
CLASSGROUP "group1"
...
CLASS
NAME "name1"
GROUP "group1"
...
END
CLASS
NAME "name2"
GROUP "group2"
...
END
CLASS
NAME "name3"
GROUP "group1"
...
END
...
END # layer

KEYIMAGE [filename] Full filename of the legend image for the CLASS. This image is used when building a legend
(or requesting a legend icon via MapScript or the CGI application).

100 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

LABEL Signals the start of a LABEL object.

MAXSCALEDENOM [double] Minimum scale at which this CLASS is drawn. Scale is given as the denominator
of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Implemented in MapServer
5.0, to replace the deprecated MAXSCALE parameter.
See Also:
Map Scale
MAXSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is MAXSCALEDENOM in-
stead. The deprecated MAXSCALE is the minimum scale at which this CLASS is drawn. Scale is given as the
denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Deprecated
since version 5.0.
MAXSIZE [integer] Maximum size in pixels to draw a symbol. Default is 50.
MINSCALEDENOM [double] Maximum scale at which this CLASS is drawn. Scale is given as the denominator of
the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Implemented in MapServer
5.0, to replace the deprecated MINSCALE parameter.
See Also:
Map Scale
MINSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is MINSCALEDENOM in-
stead. The deprecated MINSCALE is the maximum scale at which this CLASS is drawn. Scale is given as the
denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Deprecated
since version 5.0.
MINSIZE [integer] Minimum size in pixels to draw a symbol. Default is 0.
NAME [string] Name to use in legends for this class. If not set class won’t show up in legend.
OUTLINECOLOR [r] [g] [b] Color to use for outlining polygons and certain marker symbols. Line symbols do not
support outline colors.
SIZE [integer] Height, in pixels, of the symbol/pattern to be used. Only useful with scalable symbols. For vector
(and ellipse) symbol types the default size is based on the range of Y values in the POINTS defining the symbol.
For pixmaps, the default is the vertical size of the image. Default size is 1 for TTF symbols.
STATUS [on|off] Sets the current display status of the class. Default turns the class on.
STYLE Signals the start of a STYLE object. A class can contain multiple styles.
SYMBOL [integer|string|filename] The symbol name or number to use for all features if attribute tables are not used.
The number is the index of the symbol in the symbol file, starting at 1, the 5th symbol in the file is therefore
symbol number 5. You can also give your symbols names using the NAME keyword in the symbol definition
file, and use those to refer to them. Default is 0, which results in a single pixel, single width line, or solid
polygon fill, depending on layer type.
You can also specify a gif or png filename. The path is relative to the location of the mapfile.
TEMPLATE [filename] Template file or URL to use in presenting query results to the user. See Templating for more
info.
TEXT [string] Static text to label features in this class with. This overrides values obtained from the LABELITEM.
The string may be given as an expression delimited using the ()’s. This allows you to concatenate multiple
attributes into a single label. For example: ([FIRSTNAME],[LASTNAME]).
You can also “stack” 2 symbols to achieve interesting effects. You define the second symbol, which effectively sits
“on top” of the symbol normally defined above. See Cartographical Symbol Construction with MapServer for more
information.

5.2. CLASS 101

MapServer Documentation, Release 6.0

The following parameters allow you to define the symbol, and they are equivalent to their non-overlay counterparts:
• OVERLAYBACKGROUNDCOLOR
• OVERLAYCOLOR
• OVERLAYOUTLINECOLOR
• OVERLAYSIZE
• OVERLAYMINSIZE
• OVERLAYMAXSIZE
• OVERLAYSYMBOL

5.3 CLUSTER

5.3.1 Description

Since version 6.0, MapServer has the ability to combine multiple features from a point layer into single (aggregated)
features based on their relative positions. Only POINT layers are supported. This feature was added through MS RFC
69: Support for clustering of features in point layers.

5.3.2 Supported Layer Types

POINT

5.3.3 Mapfile Parameters

MAXDISTANCE [double] Specifies the distance of the search region (rectangle or ellipse) in pixel positions.
REGION [string] Defines the search region around a feature in which the neighbouring features are negotiated. Can
be ‘rectangle’ or ‘ellipse’.
GROUP [string] This expression evaluates to a string and only the features that have the same group value are ne-
gotiated. This parameter can be omitted. The evaluated group value is available in the ‘Cluster:Group’ feature
attribute.
FILTER [string] We can define the FILTER expression filter some of the features from the final output. This expres-
sion evaluates to a boolean value and if this value is false the corresponding shape is filtered out. This expression
is evaluated after the the feature negotiation is completed, therefore the ‘Cluster:FeatureCount’ parameter can
also be used, which provides the option to filter the shapes having too many or to few neighbors within the
search region.

5.3.4 Mapfile Snippet

LAYER
NAME "my-cluster"
TYPE POINT
...
C L U S T E R
M A X D I S T A N C E 20 # in pixels
R E G I ON ellipse # can be rectangle or ellipse

102 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

GROUP (expression) # we can define an expression to create separate groups for each value
FILTER (expression) # we can define a logical expression to specify the grouping condition
END
LABELITEM "Cluster:FeatureCount"
CLASS
...
LABEL
...
END
END
...
END

5.3.5 Feature attributes

The clustered layer itself provides the following aggregated attributes:

1. Cluster:FeatureCount - count of the features in the clustered shape
2. Cluster:Group - The group value of the cluster (to which the group expression is evaluated)
These attributes (in addition to the attributes provided by the original data source) can be used to configure the labels
of the features and can also be used in expressions. The ITEMS processing option can be used to specify a subset of
the attributes from the original layer in the query operations according to the user’s preference.
We can use simple aggregate functions (Min, Max, Sum, Count) to specify how the clustered attribute should be
calculated from the original attributes. The aggregate function should be specified as a prefix separated by ‘:’ in the
attibute definition, like: [Max:itemname]. If we don’t specify aggregate functions for the source layer attributes, then
the actual value of the cluster attribute will be non-deterministic if the cluster contains multiple shapes with different
values. The Count aggregate function in fact provides the same value as Cluster:FeatureCount.

5.3.6 Example: Clustering Railway Stations

The following example uses a point datasource, in this case in KML format, to display clusters of railway stations.
Two classes are used: one to style and label the cluster, and one to style and label the single railway station.
Note: Since we can’t declare 2 labelitems, for the single railway class we use the TEXT parameter to label the station.

Mapfile Layer

####################
# Lightrail Stations
####################
SYMBOL
NAME "lightrail"
TYPE PIXMAP
IMAGE "../etc/lightrail.png"
END
LAYER
NAME "lightrail"
GROUP "default"
STATUS DEFAULT
TYPE POINT
CONNECTIONTYPE OGR
CONNECTION "lightrail-stations.kml"

5.3. CLUSTER 103

MapServer Documentation, Release 6.0

DATA "lightrail-stations"
LABELITEM "Cluster:FeatureCount"
CLASSITEM "Cluster:FeatureCount"
###########################
# Define the cluster object
###########################
C L U S T E R
M A X D I S T A N C E 50
R E G I ON "ellipse"
END
################################
# Class1: For the cluster symbol
################################
CLASS
NAME "Clustered Lightrail Stations"
EXPRESSION ("[Cluster:FeatureCount]" != "1")
STYLE
SIZE 30
SYMBOL "citycircle"
COLOR 255 0 0
END
LABEL
FONT s c b
TYPE TRUETYPE
SIZE 8
COLOR 255 255 255
ALIGN CENTER
PRIORITY 10
BUFFER 1
PARTIALS TRUE
POSITION cc
END
END
################################
# Class2: For the single station
################################
CLASS
NAME "Lightrail Stations"
EXPRESSION "1"
STYLE
SIZE 30
SYMBOL "lightrail"
END
TEXT "[Name]"
LABEL
FONT s c b
TYPE TRUETYPE
SIZE 8
COLOR 0 0 0
OUTLINECOLOR 255 255 255
ALIGN CENTER
PRIORITY 9
BUFFER 1
PARTIALS FALSE
POSITION ur
END
END
# the following is used for a query

104 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

TOLERANCE 50
UNITS PIXELS
HEADER "../htdocs/templates/cluster_header.html"
FOOTER "../htdocs/templates/cluster_footer.html"
TEMPLATE "../htdocs/templates/cluster_query.html"
END

Map Image

5.4 Display of International Characters in MapServer

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-08-10

5.4. Display of International Characters in MapServer 105

MapServer Documentation, Release 6.0

Table of Contents
• Display of International Characters in MapServer
– Credit
– Related Links
– Requirements
– How to Enable in Your Mapfile
* Step 1: Verify ICONV Support and MapServer Version
* Step 2: Verify That Your Files’ Encoding is Supported by ICONV
* Step 3: Add ENCODING Parameter to your LABEL Object
* Step 4: Test with the shp2img utility
– Example Using PHP MapScript
– Notes

5.4.1 Credit

The following functionality was added to MapServer 4.4.0 as a part of a project sponsored by the Information-
technology Promotion Agency (IPA), in Japan. Project members included: Venkatesh Raghavan, Masumoto Shinji,
Nonogaki Susumu, Nemoto Tatsuya, Hirai Naoki (Osaka City University, Japan), Mario Basa, Hagiwara Akira, Niwa
Makoto, Mori Toru (Orkney Inc., Japan), and Hattori Norihiro (E-Solution Service, Inc., Japan).

5.4.2 Related Links

• MapServer ticket:858

5.4.3 Requirements

• MapServer >= 4.4.0

• MapServer compiled with the libiconv library

5.4.4 How to Enable in Your Mapfile

The mapfile LABEL object’s parameter named ENCODING can be used to convert strings from its original encoding
system into one that can be understood by the True Type Fonts. The ENCODING parameter accepts the encoding
name as its parameter.
This uses GNU’s libiconv ( http://www.gnu.org/software/libiconv/ ) so theoretically, every string with an encoding
system supported by libiconv can be displayed as labels in MapServer as long as it has a matching font-set.

Step 1: Verify ICONV Support and MapServer Version

Execute ‘’mapserv -v’ at the commandline, and verify that your MapServer version >= 4.4.0 and it includes ‘’SUP-
PORTS=ICONV’‘, such as:
> mapserv -v

MapServer version 5.6.5 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=PDF OUTPUT=SWF OUTPUT=
SUPPORTS=PROJ SUPPORTS=AGG SUPPORTS=FREETYPE SUPPORTS=ICONV SUPPORTS=FRIBIDI SUPPORTS=WMS_SERVER
SUPPORTS=WMS_CLIENT SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER SUPPORTS=SOS_SERVER

106 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

SUPPORTS=FASTCGI SUPPORTS=THREADS SUPPORTS=GEOS SUPPORTS=RGBA_PNG SUPPORTS=TILECACHE INPUT=JPEG

INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE

Step 2: Verify That Your Files’ Encoding is Supported by ICONV

Since MapServer uses the libiconv library to handle encodings, you can check the list of supported encodings here:
http://www.gnu.org/software/libiconv/
Unix users can also use the iconv -l command on a system with libiconv installed to get the complete list of supported
encodings on that specific system.

Step 3: Add ENCODING Parameter to your LABEL Object

Now you can simply add the ENCODING parameter to your mapfile LAYER object, such as:
MAP
...
LAYER
...
CLASS
...
LABEL
...
ENCODING "SHIFT_JIS"
END
END
END
END

One of the benefits of having an “ENCODING” parameter within the LABEL object is that different LAYERS with
different encoding systems can be combined together and display labels within a single map. For example, labels
from a Layer using Shapefile as it source which contains attributes in SHIFT-JIS can be combined with a Layer from
a PostGIS database server with EUC-JP attributes. A sample Mapfile can look like this:
LAYER
NAME "chimei"
DATA c h i m e i
STATUS DEFAULT
TYPE POINT
LABELITEM "NAMAE"
CLASS
NAME "CHIMEI"
STYLE
COLOR 10 100 100
END
LABEL
TYPE TRUETYPE
FONT k o c h i - g o t h i c
COLOR 220 20 20
SIZE 10
POSITION CL
PARTIALS FALSE
BUFFER 0
ENCODING S J I S
END
END

5.4. Display of International Characters in MapServer 107

MapServer Documentation, Release 6.0

END

LAYER
NAME "chimeipg"
CONNECTION "user=username password=password dbname=gis host=localhost port=5432"
CONNECTIONTYPE postgis
DATA "the_geom from chimei"
STATUS DEFAULT
TYPE POINT
LABELITEM "NAMAE"
CLASS
NAME "CHIMEI PG"
STYLE
COLOR 10 100 100
END
LABEL
TYPE TRUETYPE
FONT k o c h i - m i n c h o
COLOR 20 220 20
SIZE 10
POSITION CL
PARTIALS FALSE
BUFFER 0
ENCODING E UC- J P
END
END
END

Step 4: Test with the shp2img utility

• see shp2img commandline utility

5.4.5 Example Using PHP MapScript

For PHP Mapscript, the Encoding parameter is included in the LabelObj Class, so that the encoding parameter of a
layer can be modified such as:
// Loading the php_mapscript library
dl("php_mapscript.so");

// Loading the map file

$map = ms_newMapObj("example.map");

// get the desired layer

$layer = $map->getLayerByName("chimei");

// get the layer’s class object

$class = $layer->getClass(0);

// get the class object’s label object

$clabel= $class->label;

// get encoding parameter

$encode_str = $clabel->encoding;
print "Encoding = ".$encode_str."\n";

108 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

// set encoding parameter

$clabel->set("encoding","UTF-8");

5.4.6 Notes

During initial implementation, this functionality was tested using the different Japanese encoding systems: Shift-JIS,
EUC-JP, UTF-8, as well as Thai’s TIS-620 encoding system.

5.5 Expressions

Author Dirk Tilger

Contact dirk at MIRIUP.DE
Author Umberto Nicoletti
Contact umberto.nicoletti at gmail.com
Revision $Revision: 8295 $
Date $Date: 2008-12-26 21:08:04 -0800 (Fri, 26 Dec 2008) $
Last Updated 2007/07/09

Contents
• Expressions
– Introduction
– Expression Types
– “MapServer expressions”

5.5.1 Introduction

As of version 4.6.1, expressions are used in two places. They’re used to filter layers for specific records in the dataset
and they’re used in CLASS EXPRESSIONs to specify to which items this CLASS does apply to.

5.5.2 Expression Types

Expression are used to match attribute values with certain logical checks. There are three different types of expressions
you can use with MapServer:
• String comparisons: A single attribute is compared with a string value.
• Regular expressions: A single attribute is matched with a regular expression.
• Logical “MapServer expressions”: One or more attributes are compared using logical expressions.

5.5. Expressions 109

MapServer Documentation, Release 6.0

String comparison

String comparison means as the name suggests that attribute values are checked if they are equal to some value. String
comparison are the simplest form of MapServer expressions and the fastest option. To use a string comparison for
filtering a LAYER, both FILTERITEM and FILTER must be set. FILTERITEM is set to the attribute name. FILTER is
set to the value for comparison. The same rule applies to CLASSITEM and EXPRESSION in the CLASS object.
Example for a simple string comparison filter
FILTER "2005"
FILTERITEM "year"

would match all records that have the attribute “year” set to “2005”. The rendered map would appear as if the dataset
would only contain those items that have the “year” set to “2005”.
Similarly, a classification for the items matched above would be done by setting the CLASSITEM in the layer and the
EXPRESSION in the class
LAYER
NAME "example"
CLASSITEM "year"
...
CLASS
NAME "year-2005"
EXPRESSION "2005"
...
END
END

For a reason explained later on the values for both CLASSITEM and FILTERITEM should start neither with an ‘/’ nor
with a ‘(‘ character.

Regular expression comparison

Regular expressions are a standard text pattern matching mechanism from the UNIX world. The functionality of regu-
lar expression matching is provided by the operating system on UNIX systems and therefore slightly operating system
dependent. However their minimum set of features are those defined by the POSIX standard. The documentation of
the particular regular expression library is usually in the “regex” manual page (“man regex”).
Regular expression with MapServer work similarly to string comparison, but allow more complex operation. They are
slower than pure string comparisons, but might be still faster than logical expression. As with the string comparison
use regular expressions, a FILTERITEM or a CLASSITEM has to defined, respectively.
A regular expression typically consists of characters with special meanings and characters that are interpreted as they
are. Alphanumeric characters (A-Z, a-z and 0-9) are taken as they are. Characters with special means are:
• . will match a single character
• [ and ] are used for grouping. For example [A-Z] would match the characters A,B,C,...,X,Y,Z.
• {, }, and * are used to specify how often something should match.
• ^ matches the beginning, $ matches the end of the value.
• The backslash \ is used to take away the special meaning. For example $ would match the dollar sign.
The following LAYER configuration would have all records rendered on the map that have “hotel” in the attribute
named “placename”

110 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

LAYER
NAME "regexp-example"
FILTERITEM "placename"
FILTER /hotel/
...
END

Note: The regular expression is case-sensitive, thus records having “Hotel” in them would not have matched.
Example: Match records that have a value for the current century (as of 2005 ;) in the attribute “year”
FILTERITEM "year"
FILTER /^20[0-9][0-9]/

Example: Match all the records that are either purely numerical or empty
FILTER /^[0-9]*$/

Note: If you experience frequently segmentation faults when working with MapServer and regular expressions, it
might be that your current working environment is linked against more than one regular expression library. This can
happen when MapServer is linked with components that bring their own copy, like the Apache httpd or PHP. In these
cases the author has made best experiences with making all those components using the regular expression library of
the operating system (i.e. the one in libc). That involved editing the build files of some of the components, however.

5.5.3 “MapServer expressions”

MapServer expressions are the most complex and depending how they are written can become quite slow. They can
match any of the attributes and thus allow filtering and classification depending on more than one attribute. Besides
pure logical operations there are also expressions allow also certain arithmetic, string- and time operations.
To be able to use a MapServer expression for a FILTER or EXPRESSION value, the expression has to be finally of a
logical value.

Logical expressions

Syntactically, a logical expression is everything encapsulated in round brackets. Logical expressions take logical
values as their input and return logical values. A logical expression is either ‘true’ or ‘false’.
• ( ( ... ) AND ( ... ) ) ( ( ... ) && ( ... ) ) ... will become true when both of the two logical expressions in the
innermost brackets are true.
• ( ( ... ) OR ( ... ) ) ( ( ... ) || ( ... ) ) ... will become true when at least one of the two logical expressions in the
innermost brackets is true.
• NOT ( ... ) ! ( ... ) ... will become true, when the logical expression in the brackets becomes false.

String operations that result in a logical value

Syntactically, a sting is something encapsulated in double-quotes.

• ( “String1” eq “String2” ) ( “String1” == “String2” ) ( “String1” = “String2” ) ... will become true when both
strings are equal. This operation is identical to the MapServer string comparison described earlier.
• ( “String1” != “String2” ) ( “String1” ne “String2” ) ... will become true when both strings are not equal.
• ( “String1” < “String2” ) ( “String1” lt “String2” ) ... will become true when “String1” is lexicographically
smaller than “String2”

5.5. Expressions 111

MapServer Documentation, Release 6.0

• ( “String1” > “String2” ) ( “String1” gt “String2” ) ... will become true when “String1” is lexicographically
larger than “String2”.
• ( “String1” <= “String2” ) ( “String1” le “String2” ) ... will become true when “String1” is not lexicographically
larger than “String2”
• ( “String1” >= “String2” ) ( “String1” ge “String2” ) ... will become true when “String1” is not lexicographically
smaller than “String2”.
• ( “String1” IN “token1,token2,...,tokenN” ) ... will become true when “String1” is in equal one of the given
tokens.
Note:
The separator for those tokens is the comma. That means that you must not add unnecessary white space
to the list and that you cannot compare to tokens that have a comma in it.
• ( “String1” =~ /regexp/ ) ... will become true, when “String1” matches the regular expression “regexp”. This
operation is identical to the regular expression matching described earlier.

String operations that return string values

There is only one operation for strings that returns a string value:
• “String1” + “String2 ... will return “String1String2”, thus the two string concatenated to each other.

Arithmetic expressions returning logical values

The basic element for the arithmetic operation is the number. There are some purely arithmetic operations that are
returning numbers as their value. They will be covered in the next section.
• ( n1 eq n2 ) ( n1 == n2 ) ( n1 = n2 ) ... will become true when both numbers are equal.
• ( n1 != n2 ) ( n1 ne n2 ) ... will become true when both numbers are not equal.
• ( n1 < n2 ) ( n1 lt n2 ) ... will become true when n1 is smaller than n2
• ( n1 > n2 ) ( n1 gt n2 ) ... will become true when n1 is larger than n2.
• ( n1 <= n2 ) ( n1 le n2 ) ... will become true when n1 is smaller or equal n2
• ( n1 >= n2 ) ( n1 ge n2 ) ... will become true when n1 is larger or equal n2.
• ( n1 IN “number1,number2,...,numberN” ) ... will become true when n1 is equal to one of the given numbers.

Arithmetic expression returning a number

As stated in the previous section, MapServer can do purely numerical operations with numbers.
• n1 + n2 ... will become the sum of n1 and n2
• n1 - n2 ... will become n2 subtracted from n1
• n1 * n2 ... will become n1 multiplicated with n2
• n1 / n2> ... will become n1 divided by n2
• -n1 ... will become n1 with negated sign
• n1 ^ n2 ... will become n1 by a power of n2
• length ( “String1” ) ... will become the number of characters of “String1”

112 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

Note: When the numerical operations above are used like logical operations, the following rule applies: values equal
to zero will be taken as ‘false’ and everything else will be ‘true’. That means the expression
:: ( 6 + 5 ) ...
would evaluate as true, but
( 5 - 5 )
...

would evaluate as false.

Temporal expressions

MapServer uses an internal time type to do comparison. To convert a keys value into this time type it will check the
list below from the top down if the specified time matches and if so, it will do the conversion.
• YYYY-MM-DDTHH:MM:SSZ (‘Z’ and ‘T’ being the characters itself)</i>
• YYYY-MM-DDTHH:MM:SS (‘T’ being the character itself)</i>
• YYYY-MM-DD HH:MM:SS
• YYYY-MM-DDTHH:MM (‘T’ being the character itself)</i>
• YYYY-MM-DD HH:MM
• YYYY-MM-DDTHH (‘T’ being the character itself)</i>
• YYYY-MM-DD HH
• YYYY-MM-DD
• YYYY-MM
• YYYY
• THH:MM:SSZ (‘Z’ and ‘T’ being the characters itself)</i>
• THH:MM:SS
For temporal values obtained this way, the following operations are supported:
• ( n1 eq n2 ) ( n1 == n2 ) ( n1 = n2 ) ... will become true when both times are equal.
• ( t1 != t2 ) ( t1 ne t2 ) ... will become true when both times are not equal.
• ( t1 < t2 ) ( t1 lt t2 ) ... will become true when t1 is earlier than t2
• ( t1 > t2 ) ( t1 gt t2 ) ... will become true when t1 is later than t2.
• ( t1 <= t2 ) ( t1 le t2 ) ... will become true when t1 is earlier or same t2
• ( n1 >= n2 ) ( n1 ge n2 ) ... will become true when t1 is later or same t2.

How the attributes are referenced

To make a meaningful use of the expressions above, we need to get the attribute values into the expressions. That
is done by enclosing the attribute key into square brackets, like this: [KEY]. Then before the expression is evaluated
every occurrence of “[KEY]” will be replaced by the value for attribute “KEY”.
Example: how a simple string comparison would be evaluated. The filter is set to:

5.5. Expressions 113

MapServer Documentation, Release 6.0

FILTER ( "[BUILDING_NAME]" == "National Museum" )

There is a attribute “BUILDING_NAME” and its value is “National Government Building”. Thus the expression
actually evaluated is...
"National Government Building" == "National Museum" )

...and as such should be false.

Some layers do not really come with metadata. For raster layers for example special attributes have been defined that
can be used for classification:
• [PIXEL] ... will become the pixel value as number
• [RED], [GREEN], [BLUE] ... will become the color value for the red, green and blue component in the pixel
value, respectively.

Quotes escaping in strings

Note: Quotes escaping is not supported in MapServer versions lower than 5.0.
Starting with MapServer 5.0, if your dataset contains double-quotes, you can use a C-like escape sequence in the
expression string. For example if your key “NAME” has the value ‘National “hero” statue’ you could write the
FILTER expression as follows:
FILTER ( "[NAME]" == "National \"hero\" statue" )
...

to escape a single quote use the following sequence instead:

FILTER ( "[NAME]" == "National \’hero\’ statue" )

5.6 FEATURE

POINTS A set of xy pairs terminated with an END, for example:

POINTS 1 1 50 50 1 50 1 1 END

Note: POLYGON/POLYLINE layers POINTS must start and end with the same point (i.e. close the feature).
ITEMS Comma separated list of the feature attributes:
ITEMS "value1;value2;value3"

Note: Specifying the same number of items is recommended for each features of the same layer. The item names
should be specified as a PROCESSING option of the layer.
TEXT [string] String to use for labeling this feature.
WKT [string] A geometry expressed in OpenGIS Well Known Text geometry format. This feature is only supported
if MapServer is built with OGR or GEOS support.
WKT "POLYGON((500 500, 3500 500, 3500 2500, 500 2500, 500 500))"
WKT "POINT(2000 2500)"

Note: Inline features should be defined as their own layers in the mapfile. If another CONNECTIONTYPE is
specified in the same layer, MapServer will always use the inline features to draw the layer and ignore the other
CONNECTIONTYPEs.

114 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

5.7 FONTSET

Author Kari Guerts

Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Revision $Revision: 8295 $
Date $Date: 2008-12-26 21:08:04 -0800 (Fri, 26 Dec 2008) $
Last Updated 2008/10/08

Contents
• FONTSET
– Format

5.7.1 Format

The format is very simple. Each line contains 2 items: an alias and the name/path of the font separated by white space.
The alias is simply the name you refer to the font as in your Mapfile (eg. times-bold). The name is the actual name of
the TrueType file. If not full path then it is interpreted as relative to the location of the fontset. Here’s the fontset I use
(the font.list file and all .ttf files are stored in the same sub-directory).
Note: Aliases are case sensitive. Excellent reference information about the TrueType format and online font resources
is available from the FreeType.
arial arial.ttf
arial-bold arialbd.ttf
arial-italic ariali.ttf
arial-bold-italic arialbi.ttf
arial_black ariblk.ttf
comic_sans comic.ttf
comic_sans-bold comicbd.ttf
courier cour.ttf
courier-bold courbd.ttf
courier-italic couri.ttf
courier-bold-italic courbi.ttf
georgia georgia.ttf
georgia-bold georgiab.ttf
georgia-italic georgiai.ttf
georgia-bold-italic georgiaz.ttf
impact impact.ttf
monotype.com monotype.ttf
recreation_symbols recreate.ttf
times times.ttf
times-bold timesbd.ttf
times-italic timesi.ttf
times-bold-italic timesbi.ttf
trebuchet_ms trebuc.ttf
trebuchet_ms-bold trebucbd.ttf
trebuchet_ms-italic trebucit.ttf
trebuchet_ms-bold-italic trebucbi.ttf
verdana verdana.ttf
verdana-bold verdanab.ttf

5.7. FONTSET 115

MapServer Documentation, Release 6.0

verdana-italic verdanai.ttf
verdana-bold-italic verdanaz.ttf

5.8 GRID

LABELFORMAT [DD|DDMM|DDMMSS|C format string] Format of the label. “DD” for degrees, “DDMM” for
degrees minutes, and “DDMMSS” for degrees, minutes, seconds. A C-style formatting string is also allowed,
such as “%g°” to show decimal degrees with a degree symbol. The default is decimal display of whatever SRS
you’re rendering the GRID with.
MINARCS [double] The minimum number of arcs to draw. Increase this parameter to get more lines. Optional.
MAXARCS [double] The maximum number of arcs to draw. Decrease this parameter to get fewer lines. Optional.
MININTERVAL [double] The minimum number of intervals to try to use. The distance between the grid lines, in
the units of the grid’s coordinate system. Optional.
MAXINTERVAL [double] The maximum number of intervals to try to use. The distance between the grid lines, in
the units of the grid’s coordinate system. Optional.
MINSUBDIVIDE [double] The minimum number of segments to use when rendering an arc. If the lines should be
very curved, use this to smooth the lines by adding more segments. Optional.
MAXSUBDIVIDE [double] The maximum number of segments to use when rendering an arc. If the graticule should
be very straight, use this to minimize the number of points for faster rendering. Optional, default 256.
The following is an example of a GRID object in use:
LAYER
NAME "grid"
METADATA
"DESCRIPTION" "Grid"
END
TYPE LINE
STATUS ON
CLASS
NAME "Graticule"
COLOR 0 0 0
LABEL
COLOR 255 0 0
FONT "fritqat"
TYPE truetype
SIZE 8
POSITION AUTO
PARTIALS FALSE
BUFFER 5
OUTLINECOLOR 255 255 255
END
END
PROJECTION
"init=epsg:4326"
END
GRID
LABELFORMAT "DDMM"
# LABELFORMAT ’%g°’ # dec degrees with symbol
MAXARCS 10
MAXINTERVAL 10
MAXSUBDIVIDE 2

116 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

# LABELFORMAT ’%7.0f m’ # nice if a projected SRS used

# MININTERVAL 20000
# MAXSUBDIVIDE 2
END
END # Layer

5.9 INCLUDE

When this directive is encountered parsing switches to the included file immediately. As a result the included file can
be comprised of any valid mapfile syntax. For example:
INCLUDE ’myLayer.map’

Performance does not seem to be seriously impacted with limited use, however in high performance instances you
may want to use includes in a pre-processing step to build a production mapfile. The C pre-processor can also be used
(albeit with a different syntax) and is far more powerful.

5.9.1 Notes

• Supported in versions 4.10 and higher.

• The name of the file to be included MUST be quoted (single or double quotes).
• Includes may be nested, up to 5 deep.
• File location can be given as a full path to the file, or (in MapServer >= 4.10.1) as a path relative to the mapfile.
• Debugging can be problematic because:
1. the file an error occurs in does not get output to the user
2. the line number counter is not reset for each file. Here is one possible error that is thrown when the include
file cannot be found:
msyylex(): Unable to access file. Error opening included file "parks_include.map"

5.9.2 Example
MAP
NAME "include_mapfile"
EXTENT 0 0 500 500
SIZE 250 250

INCLUDE "test_include_symbols.map"
INCLUDE "test_include_layer.map"
END

where test_include_symbols.map contains:

SYMBOL
NAME ’square’
TYPE VECTOR
FILLED TRUE
POINTS 0 0 0 1 1 1 1 0 0 0 END
END

5.9. INCLUDE 117

MapServer Documentation, Release 6.0

and test_include_layer.map contains:

LAYER
TYPE POINT
STATUS DEFAULT
FEATURE
POINTS 10 10 40 20 300 300 400 10 10 400 END
END
CLASS
NAME ’Church’
COLOR 0 0 0
SYMBOL ’square’
SIZE 7
STYLE
SYMBOL "square"
SIZE 5
COLOR 255 255 255
END
STYLE
SYMBOL "square"
SIZE 3
COLOR 0 0 255
END
END
END

5.10 JOIN

5.10.1 Description

Joins are defined within a LAYER object. It is important to understand that JOINs are ONLY available once a query
has been processed. You cannot use joins to affect the look of a map. The primary purpose is to enable lookup tables
for coded data (e.g. 1 => Forest) but there are other possible uses.

5.10.2 Supported Formats

• DBF/XBase files
• CSV (comma delimited text file)
• PostgreSQL and PostGIS tables
• MySQL tables

5.10.3 Mapfile Parameters:

CONNECTION [string] Parameters required for the join table’s database connection (not required for DBF or CSV
joins). The following is an example connection for PostgreSQL:
CONNECTION "host=127.0.0.1 port=5432 user=postgres password=postgres dbname=somename"
CONNECTIONTYPE OGR

The following is an example connection for MySQL:

118 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

CONNECTION "MYSQL:mydbname,user=root,password=mysql,port=3306"
CONNECTIONTYPE OGR

CONNECTIONTYPE [string] Type of connection (not required for DBF or CSV joins). For PostgreSQL and
MySQL connections use ‘OGR’.
FOOTER [filename] Template to use after a layer’s set of results have been sent. In other words, this header HTML
will be displayed after the contents of the TEMPLATE HTML.
FROM [item] Join item in the dataset. This is case sensitive.
HEADER [filename] Template to use before a layer’s set of results have been sent. In other words, this header HTML
will be displayed before the contents of the TEMPLATE HTML.
NAME [string] Unique name for this join. Required.
TABLE [filename|tablename] For file-based joins this is the name of XBase or comma delimited file (relative to
the location of the mapfile) to join TO. For PostgreSQL and MySQL support this is the name of the Post-
greSQL/MySQL table to join TO.
TEMPLATE [filename] Template to use with one-to-many joins. The template is processed once for each record
and can only contain substitutions for items in the joined table. Refer to the column in the joined table in your
template like [joinname_columnname], where joinname is the NAME specified for the JOIN object.
TO [item] Join item in the table to be joined. This is case sensitive.
TYPE [ONE-TO-ONE|ONE-TO-MANY] The type of join. Default is one-to-one.

5.10.4 Example 1: Join from SHP file to DBF file

Mapfile Layer

LAYER
NAME "prov_bound"
TYPE POLYGON
STATUS DEFAULT
DATA "prov.shp"
CLASS
NAME "Province"
STYLE
OUTLINECOLOR 120 120 120
COLOR 255 255 0
END
END
TEMPLATE "../htdocs/cgi-query-templates/prov.html"
HEADER "../htdocs/cgi-query-templates/prov-header.html"
FOOTER "../htdocs/cgi-query-templates/footer.html"
JOIN
NAME "test"
TABLE "../data/lookup.dbf"
FROM "ID"
TO "IDENT"
TYPE ONE-TO-ONE
END
END # layer

5.10. JOIN 119

MapServer Documentation, Release 6.0

Ogrinfo

>ogrinfo lookup.dbf lookup -summary

INFO: Open of ‘lookup.dbf’
using driver ‘ESRI Shapefile’ successful.

Layer name: lookup

Geometry: None
Feature Count: 12
Layer SRS WKT:
(unknown)
IDENT: Integer (2.0)
VAL: Integer (2.0)

>ogrinfo prov.shp prov -summary

INFO: Open of ‘prov.shp’
using driver ‘ESRI Shapefile’ successful.

Layer name: prov

Geometry: Polygon
Feature Count: 12
Extent: (-2340603.750000, -719746.062500) - (3009430.500000, 3836605.250000)
Layer SRS WKT:
(unknown)
NAME: String (30.0)
ID: Integer (2.0)

Template

<tr bgcolor="#EFEFEF"><td align="left">[NAME]</td><td align="left">[test_VAL]</td></tr>

5.10.5 Example 2: Join from SHP file to PostgreSQL table

Mapfile Layer

LAYER
NAME "prov_bound"
TYPE POLYGON
STATUS DEFAULT
DATA "prov.shp"
CLASS
NAME "Province"
STYLE
OUTLINECOLOR 120 120 120
COLOR 255 255 0
END
END
TOLERANCE 20
TEMPLATE "../htdocs/cgi-query-templates/prov.html"
HEADER "../htdocs/cgi-query-templates/prov-header.html"
FOOTER "../htdocs/cgi-query-templates/footer.html"
JOIN
NAME "test"
CONNECTION "host=127.0.0.1 port=5432 user=postgres password=postgres dbname=join"
CONNECTIONTYPE ogr

120 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

TABLE "lookup"
FROM "ID"
TO "ident"
TYPE ONE-TO-ONE
END
END # layer

Ogrinfo

>ogrinfo -ro PG:"host=127.0.0.1 port=5432 user=postgres password=postgre dbname=join" lookup -summary

INFO: Open of ‘PG:host=127.0.0.1 port=5432 user=postgres password=postgres dbname=join’
using driver ‘PostgreSQL’ successful.

Layer name: lookup

Geometry: Unknown (any)
Feature Count: 12
Layer SRS WKT:
(unknown)
ident: Integer (0.0)
val: Integer (0.0)

Template

<tr bgcolor="#EFEFEF"><td align="left">[NAME]</td><td align="left">[test_val]</td></tr>

Note: When testing with MapServer 4.10.0 on Windows this postgresql join caused a mapserv.exe crash. However
when testing this with a MapServer build > 4.10.0 the crash did not occur.

5.10.6 Example 3: Join from SHP file to CSV file

Mapfile Layer

LAYER
NAME "prov_bound"
TYPE POLYGON
STATUS DEFAULT
DATA "prov.shp"
CLASS
NAME "Province"
STYLE
OUTLINECOLOR 120 120 120
COLOR 255 255 0
END
END
TOLERANCE 20
TEMPLATE "../htdocs/cgi-query-templates/prov.html"
HEADER "../htdocs/cgi-query-templates/prov-header.html"
FOOTER "../htdocs/cgi-query-templates/footer.html"
JOIN
NAME "test"
TABLE "../data/lookup.csv"
FROM "ID"
TO "IDENT"

5.10. JOIN 121

MapServer Documentation, Release 6.0

TYPE ONE-TO-ONE
END
END # layer

CSV File Structure

"IDENT","VAL"
1,12
2,11
3,10
4,9
5,8
6,7
7,6
8,5
9,4
10,3
11,2
12,1

Ogrinfo

>ogrinfo lookup.csv lookup -summary

INFO: Open of ‘lookup.csv’
using driver ‘CSV’ successful.

Layer name: lookup

Geometry: None
Feature Count: 12
Layer SRS WKT:
(unknown)
IDENT: String (0.0)
VAL: String (0.0)

Template

<tr bgcolor="#EFEFEF"><td align="left">[NAME]</td><td align="left">[test_VAL]</td></tr>

5.10.7 Example 4: Join from SHP file to MySQL table

Mapfile Layer

LAYER
NAME "prov_bound"
TYPE POLYGON
STATUS DEFAULT
DATA "prov.shp"
CLASS
NAME "Province"
STYLE
OUTLINECOLOR 120 120 120
COLOR 255 255 0

122 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

END
END
TOLERANCE 20
TEMPLATE "../htdocs/cgi-query-templates/prov.html"
HEADER "../htdocs/cgi-query-templates/prov-header.html"
FOOTER "../htdocs/cgi-query-templates/footer.html"
JOIN
NAME "test"
CONNECTION "MYSQL:test,user=root,password=mysql,port=3306"
CONNECTIONTYPE ogr
TABLE "lookup"
FROM "ID"
TO "ident"
TYPE ONE-TO-ONE
END
END # layer

Ogrinfo

>ogrinfo MYSQL:test,user=root,password=mysql,port=3306 lookup -summary

INFO: Open of ‘MYSQL:test,user=root,password=mysql,port=3306’
using driver ‘MySQL’ successful.

Layer name: lookup

Geometry: None
Feature Count: 48
Layer SRS WKT:
(unknown)
FID Column = ident
val: Integer (0.0)

Template

<tr bgcolor="#EFEFEF"><td align="left">[NAME]</td><td align="left">[test_val]</td></tr>

Note: When testing with MapServer 5.6.5 and MySQL 5.1.49 on Windows this MySQL join caused a mapserv.exe
crash.

5.11 LABEL

ALIGN [left|center|right] Since version 5.4 . Specifies text alignment for multiline labels (see WRAP) Note that
the alignment algorithm is far from precise, so don’t expect fabulous results (especially for right alignment) if
you’re not using a fixed width font.
ANGLE [double|auto|follow|attribute]
• Angle, given in degrees, to draw the label.
• AUTO allows MapServer to compute the angle. Valid for LINE layers only.
• FOLLOW was introduced in version 4.10 and tells MapServer to compute a curved label for appropriate
linear features (see MS RFC 11: Support for Curved Labels for specifics).

5.11. LABEL 123

MapServer Documentation, Release 6.0

• [Attribute] was introduced in version 5.0, to specify the item name in the attribute table to use for an-
gle values. The hard brackets [] are required. For example, if your shapefile’s DBF has a field named
“MYANGLE” that holds angle values for each record, your LABEL object might contain:
LABEL
COLOR 150 150 150
OUTLINECOLOR 255 255 255
FONT "sans"
TYPE truetype
SIZE 6
ANGLE [ M Y ANGLE]
POSITION AUTO
PARTIALS FALSE
END

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
ANTIALIAS [true|false] Should text be antialiased? Note that this requires more available colors, decreases drawing
performance, and results in slightly larger output images. Only useful for GD (gif) rendering. Default is false.
Has no effect for the other renderers (where anti-aliasing can not be turned off).
BACKGROUNDCOLOR [r] [g] [b] Color to draw a background rectangle (i.e. billboard). Off by default. Depre-
cated since version 6.0: Use STYLE object and GEOMTRANSFORM labelpoly.
BACKGROUNDSHADOWCOLOR [r] [g] [b] Color to draw a background rectangle (i.e. billboard) shadow. Off
by default. Deprecated since version 6.0: Use STYLE object and GEOMTRANSFORM labelpoly.
BACKGROUNDSHADOWSIZE [x][y] How far should the background rectangle be offset? Default is 1. Depre-
cated since version 6.0: Use STYLE object and GEOMTRANSFORM labelpoly.
BUFFER [integer] Padding, in pixels, around labels. Useful for maintaining spacing around text to enhance read-
ability. Available only for cached labels. Default is 0.
COLOR [r] [g] [b] | [attribute]
• Color to draw text with.
• [Attribute] was introduced in version 5.0, to specify the item name in the attribute table to use for color
values. The hard brackets [] are required. For example, if your shapefile’s DBF has a field named “MY-
COLOR” that holds color values for each record, your LABEL object might contain:
LABEL
COLOR [ M Y COLOR]
OUTLINECOLOR 255 255 255
FONT "sans"
TYPE truetype
SIZE 6
POSITION AUTO
PARTIALS FALSE
END

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
ENCODING [string] Supported encoding format to be used for labels. If the format is not supported, the label will
not be drawn. Requires the iconv library (present on most systems). The library is always detected if present on
the system, but if not the label will not be drawn.
Required for displaying international characters in MapServer. More information can be found in the Label
Encoding document.
FONT [name|attribute]
• Font alias (as defined in the FONTSET) to use for labeling.

124 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

• [Attribute] was introduced in version 5.6 to specfify the font alias.

FORCE [true|false] Forces labels for a particular class on, regardless of collisions. Available only for cached labels.
Default is false.
MAXLENGTH [integer] Introduced in mapserver 5.4, this keyword interacts with the WRAP keyword so that line
breaks only occur after the defined number of characters

Table 5.6: Interaction with WRAP keyword

maxlength = 0 maxlength > 0 maxlength < 0
wrap = always wrap at the newline at the first WRAP character hard wrap (always break at
‘char’ WRAP character after MAXLENGTH characters exactly MAXLENGTH
characters)
no no processing skip label if it contains more than hard wrap (always break at
wrap MAXLENGTH characters exactly MAXLENGTH
characters)
The associated RFC document for this feature is MS RFC 40: Support Label Text Transformations.
MAXOVERLAPANGLE [double] Angle threshold to use in filtering out ANGLE FOLLOW labels in which char-
acters overlap (floating point value in degrees). This filtering will be enabled by default starting with MapServer
6.0. The default MAXOVERLAPANGLE value will be 22.5 degrees, which also matches the default in
GeoServer. Users will be free to tune the value up or down depending on the type of data they are dealing
with and their tolerance to bad overlap in labels. As per RFC 60, if MAXOVERLAPANGLE is set to 0, then we
fall back on pre-6.0 behavior which was to use maxoverlapangle = 0.4*MS_PI (40% of 180 degrees = 72degree).
The associated RFC document for this feature is MS RFC 60: Labeling enhancement: ability to skip ANGLE
FOLLOW labels with too much character overlap.
MAXSIZE [double] Maximum font size to use when scaling text (pixels). Default is 256. Starting from version 5.4,
the value can also be a fractional value (and not only integer).
MINDISTANCE [integer] Minimum distance between duplicate labels. Given in pixels.
MINFEATURESIZE [integer|auto] Minimum size a feature must be to be labeled. Given in pixels. For line data
the overall length of the displayed line is used, for polygons features the smallest dimension of the bounding box
is used. “Auto” keyword tells MapServer to only label features that are larger than their corresponding label.
Available for cached labels only.
MINSIZE [double] Minimum font size to use when scaling text (pixels). Default is 4. Starting from version 5.4, the
value can also be a fractional value (and not only integer).
OFFSET [x][y] Offset values for labels, relative to the lower left hand corner of the label and the label point. Given
in pixels. In the case of rotated text specify the values as if all labels are horizontal and any rotation will be
compensated for.
OUTLINECOLOR [r] [g] [b] | [attribute]
• Color to draw a one pixel outline around the characters in the text.
• [attribute] was introduced in version 5.0, to specify the item name in the attribute table to use for color
values. The hard brackets [] are required. For example, if your shapefile’s DBF has a field named “MY-
OUTCOLOR” that holds color values for each record, your LABEL object might contain:
LABEL
COLOR 150 150 150
OUTLINECOLOR [ M Y O U T COLOR]
FONT "sans"
TYPE truetype
SIZE 6
POSITION AUTO

5.11. LABEL 125

MapServer Documentation, Release 6.0

PARTIALS FALSE
END

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
OUTLINEWIDTH [integer] Width of the outline if OUTLINECOLOR has been set. Defaults to 1. Currently only
the AGG renderer supports values greater than 1, and renders these as a ‘halo’ effect: recommended values are
3 or 5.
PARTIALS [true|false] Can text run off the edge of the map? Default is true.
POSITION [ul|uc|ur|cl|cc|cr|ll|lc|lr|auto] Position of the label relative to the labeling point (layers only). First letter
is “Y” position, second letter is “X” position. “Auto” tells MapServer to calculate a label position that will not
interfere with other labels. With points, MapServer selects from the 8 outer positions (i.e. excluding cc). With
polygons, MapServer selects from cc (added in MapServer 5.4), uc, lc, cl and cr as possible positions. With
lines, it only uses lc or uc, until it finds a position that doesn’t collide with labels that have already been drawn.
If all positions cause a conflict, then the label is not drawn (Unless the label’s FORCE a parameter is set to
“true”). “Auto” placement is only available with cached labels.
PRIORITY [integer]|[item_name]|[attribute] The priority parameter (added in v5.0) takes an integer value between
1 (lowest) and 10 (highest). The default value is 1. It is also possible to bind the priority to an attribute
(item_name) using square brackets around the [item_name]. e.g. “PRIORITY [someattribute]”
Labels are stored in the label cache and rendered in order of priority, with the highest priority levels rendered
first. Specifying an out of range PRIORITY value inside a map file will result in a parsing error. An out of range
value set via MapScript or coming from a shape attribute will be clamped to the min/max values at rendering
time. There is no expected impact on performance for using label priorities.
[Attribute] was introduced in version 5.6.
REPEATDISTANCE [integer]
• Introduced in version 5.6. The label will be repeated on every line of a multiline shape and will be repeated
multiple times along a given line at an interval of REPEATDISTANCE pixels.
The associated RFC document for this feature is MS RFC 57: Labeling enhancements: ability to repeat
labels along a line/multiline.
SHADOWCOLOR [r] [g] [b] Color of drop shadow. An label with the same text will be rendered in this color before
the main label is drawn, resulting in a shadow effect on the the label characters. The offset of the renderered
shadow is set with SHADOWSIZE.
SHADOWSIZE [x][y]|[attribute][attribute]|[x][attribute]|[attribute][y] Shadow offset in pixels, see SHADOW-
COLOR.
[Attribute] was introduced in version 6.0, and can be used like:
SHADOWSIZE 2 2
SHADOWSIZE [ s h a d o w s i z e X ] 2
SHADOWSIZE 2 [ s h a d o w s i z e Y ]
SHADOWSIZE [shadowsize] [shadowsize]

SIZE [double]|[tiny|small|medium|large|giant]|[attribute]
• Text size. Use “integer” to give the size in pixels of your TrueType font based label, or any of the other 5
listed keywords to bitmap fonts.
• Starting from version 5.4, the value can also be a fractional value (and not only integer).
• [Attribute] was introduced in version 5.0, to specify the item name in the attribute table to use for size val-
ues. The hard brackets [] are required. For example, if your shapefile’s DBF has a field named “MYSIZE”
that holds size values for each record, your LABEL object might contain:

126 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

LABEL
COLOR 150 150 150
OUTLINECOLOR 255 255 255
FONT "sans"
TYPE truetype
SIZE [ M Y SIZE]
POSITION AUTO
PARTIALS FALSE
END

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
STYLE The start of a STYLE object.
Label specific mechanisms of the STYLE object are the GEOMTRANSFORM options:
GEOMTRANSFORM [labelpnt|labelpoly] Creates geometries that can be used for styling the label.
• labelpnt generates a point with its center at the middle of the bounding rectangle of the text.
• labelpoly generates the bounding rectangle for the text, with 1 pixel of padding added in all directions.
The resulting geometries can be styled using the mechanisms available in the STYLE object.
New in version 6.0.
TYPE [bitmap|truetype] Type of font to use. Generally bitmap fonts are faster to draw then TrueType fonts. How-
ever, TrueType fonts are scalable and available in a variety of faces. Be sure to set the FONT parameter if you
select TrueType.
Note that bitmap fonts are only supported with the AGG and GD renderers.
WRAP [character] Character that represents an end-of-line condition in label text, thus resulting in a multi-line label.
Interacts with MAXLENGTH for conditional line wrapping after a given number of characters

5.12 LAYER

CLASS Signals the start of a CLASS object.

Inside a layer, only a single class will be used for the rendering of a feature. Each feature is tested against each
class in the order in which they are defined in the mapfile. The first class that matches the its min/max scale
constraints and its EXPRESSION check for the current feature will be used for rendering.
CLASSITEM [attribute] Item name in attribute table to use for class lookups.
CLASSGROUP [string] Specify the class’s group that would be considered at rendering time. The CLASS object’s
GROUP parameter must be used in combination with CLASSGROUP.
CLUSTER Signals the start of a CLUSTER object.
The CLUSTER configuration option provides to combine multiple features from the layer into single (aggre-
gated) features based on their relative positions. Supported only for POINT layers.
See Also:
MS RFC 69: Support for clustering of features in point layers
CONNECTION [string] Database connection string to retrieve remote data.
An SDE connection string consists of a hostname, instance name, database name, username and password
separated by commas.

5.12. LAYER 127

MapServer Documentation, Release 6.0

A PostGIS connection string is basically a regular PostgreSQL connection string, it takes the form of
“user=nobody password=****** dbname=dbname host=localhost port=5432”
An Oracle connection string: user/pass[@db]
See Also:
See Vector Data for specific connection information for various data sources.
CONNECTIONTYPE [local|sde|ogr|postgis|oraclespatial|wms|wfs|plugin|union] Type of connection. Default is
local. See additional documentation for any other type.
See Also:
See Vector Data for specific connection information for various data sources. See Union Layer for combining
layers, added in MapServer 6.0
Note: mygis is another connectiontype, but it is deprecated; please see the MySQL section of the Vector Data
document for connection details.
DATA [filename]|[sde parameters][postgis table/column][oracle table/column] Full filename of the spatial data to
process. No file extension is necessary for shapefiles. Can be specified relative to the SHAPEPATH option from
the Map Object.
If this is an SDE layer, the parameter should include the name of the layer as well as the geometry column, i.e.
“mylayer,shape,myversion”.
If this is a PostGIS layer, the parameter should be in the form of “<columnname> from <tablename>”, where
“columnname” is the name of the column containing the geometry objects and “tablename” is the name of the
table from which the geometry data will be read.
For Oracle, use “shape FROM table” or “shape FROM (SELECT statement)” or even more complex Oracle
compliant queries! Note that there are important performance impacts when using spatial subqueries however.
Try using MapServer’s FILTER whenever possible instead. You can also see the SQL submitted by forcing an
error, for instance by submitting a DATA parameter you know won’t work, using for example a bad column
name.
See Also:
See Vector Data for specific connection information for various data sources.
DEBUG [off|on|0|1|2|3|4|5] Enables debugging of a layer in the current map.
Debugging with MapServer versions >= 5.0:
Verbose output is generated and sent to the standard error output (STDERR) or the MapServer errorfile if one
is set using the “MS_ERRORFILE” environment variable. You can set the environment variable by using the
CONFIG parameter at the MAP level of the mapfile, such as:
CONFIG "MS_ERRORFILE" "/ms4w/tmp/ms_error.txt"

You can also set the environment variable in Apache by adding the following to your httpd.conf:
SetEnv MS_ERRORFILE "/ms4w/tmp/ms_error.txt"

Once the environment variable is set, the DEBUG mapfile parameter can be used to control the level of debug-
ging output. Here is a description of the possible DEBUG values:
• DEBUG O or OFF - only msSetError() calls are logged to MS_ERRORFILE. No msDebug() output at
all. This is the default and corresponds to the original behavior of MS_ERRORFILE in MapServer 4.x
• DEBUG 1 or ON - includes all output from DEBUG 0 plus msDebug() warnings about common pitfalls,
failed assertions or non-fatal error situations (e.g. missing or invalid values for some parameters, missing
shapefiles in tileindex, timeout error from remote WMS/WFS servers, etc.)

128 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

• DEBUG 2 - includes all output from DEBUG 1 plus notices and timing information useful for tuning
mapfiles and applications
• DEBUG 3 - all of DEBUG 2 plus some debug output useful in troubleshooting problems such as WMS
connection URLs being called, database connection calls, etc. This is the recommended level for debug-
ging mapfiles.
• DEBUG 4 - DEBUG 3 plus even more details...
• DEBUG 5 - DEBUG 4 plus any msDebug() output that might be more useful to the developers than to the
users.
You can also set the debug level by using the “MS_DEBUGLEVEL” environment variable.
The DEBUG setting can also be specified for the entire map, by setting the DEBUG parameter in the MAP
object.
For more details on this debugging mechanism, please see MS RFC 28: Redesign of LOG/DEBUG output
mechanisms.
Debugging with MapServer versions < 5:
Verbose output is generated and sent to the standard error output (STDERR) or the MapServer logfile if one is
set using the LOG parameter in the WEB object. Apache users will see timing details for drawing in Apache’s
error_log file. Requires MapServer to be built with the DEBUG=MSDEBUG option (–with-debug configure
option).
DUMP [true|false] Switch to allow MapServer to return data in GML format. Useful when used with WMS GetFea-
tureInfo operations. “false” by default.
EXTENT [minx] [miny] [maxx] [maxy] The spatial extent of the data. In most cases you will not need to specify
this, but it can be used to avoid the speed cost of having MapServer compute the extents of the data. An
application can also possibly use this value to override the extents of the map.
FEATURE Signals the start of a FEATURE object.
FILTER [string] This parameter allows for data specific attribute filtering that is done at the same time spatial filter-
ing is done, but before any CLASS expressions are evaluated. For OGR and shapefiles the string is simply a
mapserver regular expression. For spatial databases the string is a SQL WHERE clause that is valid with respect
to the underlying database.
For example: FILTER ([type]=’road’ and [size]<2)
FILTERITEM [attribute] Item to use with simple FILTER expressions. OGR and shapefiles only.
FOOTER [filename] Template to use after a layer’s set of results have been sent. Multiresult query modes only.
GRID Signals the start of a GRID object.
GROUP [name] Name of a group that this layer belongs to. The group name can then be reference as a regular layer
name in the template files, allowing to do things like turning on and off a group of layers at once.
HEADER [filename] Template to use before a layer’s set of results have been sent. Multiresult query modes only.
JOIN Signals the start of a JOIN object.
LABELANGLEITEM [attribute] (As of MapServer 5.0 this parameter is no longer available. Please see the LABEL
object’s ANGLE parameter) For MapServer versions < 5.0, this is the item name in attribute table to use for
class annotation angles. Values should be in degrees. Deprecated since version 5.0.
LABELCACHE [on|off] Specifies whether labels should be drawn as the features for this layer are drawn, or whether
they should be cached and drawn after all layers have been drawn. Default is on. Label overlap removal, auto
placement etc... are only available when the label cache is active.
LABELITEM [attribute] Item name in attribute table to use for class annotation (i.e. labeling).

5.12. LAYER 129

MapServer Documentation, Release 6.0

LABELMAXSCALEDENOM [double] Minimum scale at which this LAYER is labeled. Scale is given as the
denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Implemented
in MapServer 5.0, to replace the deprecated LABELMAXSCALE parameter.
See Also:
Map Scale
LABELMAXSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is LABELMAXS-
CALEDENOM instead. The deprecated LABELMAXSCALE is the minimum scale at which this LAYER is
labeled. Scale is given as the denominator of the actual scale fraction, for example for a map at a scale of
1:24,000 use 24000. Deprecated since version 5.0.
LABELMINSCALEDENOM [double] Maximum scale at which this LAYER is labeled. Scale is given as the de-
nominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Implemented in
MapServer 5.0, to replace the deprecated LABELMINSCALE parameter.
See Also:
Map Scale
LABELMINSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is LABELMIN-
SCALEDENOM instead. The deprecated LABELMINSCALE is the maximum scale at which this LAYER
is labeled. Scale is given as the denominator of the actual scale fraction, for example for a map at a scale of
1:24,000 use 24000. Deprecated since version 5.0.
LABELREQUIRES [expression] Sets context for labeling this layer, for example:
LABELREQUIRES "![orthoquads]"

means that this layer would NOT be labeled if a layer named “orthoquads” is on. The expression consists of
a boolean expression based on the status of other layers, each [layer name] substring is replaced by a 0 or a 1
depending on that layer’s STATUS and then evaluated as normal. Logical operators AND and OR can be used.
LABELSIZEITEM [attribute] (As of MapServer 5.0 this parameter is no longer available. Please see the LABEL
object’s SIZE parameter) For MapServer versions < 5.0, this is the item name in attribute table to use for class
annotation sizes. Values should be in pixels. Deprecated since version 5.0.
MAXFEATURES [integer] Specifies the number of features that should be drawn for this layer in the CURRENT
window. Has some interesting uses with annotation and with sorted data (i.e. lakes by area).
MAXGEOWIDTH [double] Maximum width, in the map’s geographic units, at which this LAYER is drawn. If
MAXSCALEDENOM is also specified then MAXSCALEDENOM will be used instead. (added in MapServer
5.4.0)
The width of a map in geographic units can be found by calculating the following from the extents:
[maxx] - [minx]

MAXSCALEDENOM [double] Minimum scale at which this LAYER is drawn. Scale is given as the denominator
of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Implemented in MapServer
5.0, to replace the deprecated MAXSCALE parameter.
See Also:
Map Scale
MAXSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is MAXSCALEDENOM in-
stead. The deprecated MAXSCALE is the minimum scale at which this LAYER is drawn. Scale is given as the
denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Deprecated
since version 5.0.

130 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

METADATA This keyword allows for arbitrary data to be stored as name value pairs. This is used with OGC WMS
to define things such as layer title. It can also allow more flexibility in creating templates, as anything you put
in here will be accessible via template tags.
Example:
METADATA
title "My layer title"
author "Me!"
END

MINGEOWIDTH [double] Minimum width, in the map’s geographic units, at which this LAYER is drawn. If
MINSCALEDENOM is also specified then MINSCALEDENOM will be used instead. (added in MapServer
5.4.0)
The width of a map in geographic units can be found by calculating the following from the extents:
[maxx] - [minx]

MINSCALEDENOM [double] Maximum scale at which this LAYER is drawn. Scale is given as the denominator
of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Implemented in MapServer
5.0, to replace the deprecated MINSCALE parameter.
See Also:
Map Scale
MINSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is MINSCALEDENOM in-
stead. The deprecated MINSCALE is the maximum scale at which this LAYER is drawn. Scale is given as the
denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Deprecated
since version 5.0.
NAME [string] Short name for this layer. Limit is 20 characters. This name is the link between the mapfile and web
interfaces that refer to this name. They must be identical. The name should be unique, unless one layer replaces
another at different scales. Use the GROUP option to associate layers with each other.
OFFSITE [r] [g] [b] Sets the color index to treat as transparent for raster layers.
OPACITY [integer|alpha] Sets the opacity level (or the inability to see through the layer) of all classed pixels for
a given layer. The value can either be an integer in the range (0-100) or the named symbol “ALPHA”. A
value of 100 is opaque and 0 is fully transparent. Implemented in MapServer 5.0, to replace the deprecated
TRANSPARENCY parameter.
The “ALPHA” symbol directs the MapServer rendering code to honor the indexed or alpha transparency of
pixmap symbols used to style a layer. This is only needed in the case of RGB output formats, and should be
used only when necessary as it is expensive to render transparent pixmap symbols onto an RGB map image.
PLUGIN [filename] Additional library to load by MapServer, for this layer. This is commonly used to load specific
support for SDE and Microsoft SQL Server layers, such as:
CONNECTIONTYPE PLUGIN
CONNECTION "hostname,port:xxx,database,username,password"
PLUGIN "C:/ms4w/Apache/specialplugins/msplugin_sde_92.dll"
DATA "layername,geometrycolumn,SDE.DEFAULT"

POSTLABELCACHE [true|false] Tells MapServer to render this layer after all labels in the cache have been drawn.
Useful for adding neatlines and similar elements. Default is false.
PROCESSING [string] Passes a processing directive to be used with this layer. The supported processing directives
vary by layer type, and the underlying driver that processes them.
• Attributes Directive - The ITEMS processing option allows to specify the name of attributes for inline
layers or specify the subset of the attributes to be used by the layer, such as:

5.12. LAYER 131

MapServer Documentation, Release 6.0

PROCESSING "ITEMS=itemname1,itemname2,itemname3"

• Connection Pooling Directive - This is where you can enable connection pooling for certain layer layer
types. Connection pooling will allow MapServer to share the handle to an open database or layer con-
nection throughout a single map draw process. Additionally, if you have FastCGI enabled, the connection
handle will stay open indefinitely, or according to the options specified in the FastCGI configuration.
Oracle Spatial, ArcSDE, OGR and PostGIS/PostgreSQL currently support this approach.
PROCESSING "CLOSE_CONNECTION=DEFER"

• OGR Styles Directive - This directive can be used for obtaining label styles through MapScript. For more
information see the MapServer’s OGR document.
PROCESSING "GETSHAPE_STYLE_ITEMS=all"

• Raster Directives - All raster processing options are described in Raster Data. Here we see the SCALE
and BANDs directives used to autoscale raster data and alter the band mapping.
PROCESSING "SCALE=AUTO"
PROCESSING "BANDS=3,2,1"

PROJECTION Signals the start of a PROJECTION object.

REQUIRES [expression] Sets context for displaying this layer (see LABELREQUIRES).
SIZEUNITS [pixels|feet|inches|kilometers|meters|miles|nauticalmiles] Sets the unit of CLASS object SIZE values
(default is pixels). Useful for simulating buffering. Nauticalmiles was added in MapServer 5.6.
STATUS [on|off|default] Sets the current status of the layer. Often modified by MapServer itself. Default turns the
layer on permanently.
Note: In CGI mode, layers with STATUS DEFAULT cannot be turned off using normal mechanisms. It is
recommended to set layers to STATUS DEFAULT while debugging a problem, but set them back to ON/OFF in
normal use.
Note: For WMS, layers in the server mapfile with STATUS DEFAULT are always sent to the client.
STYLEITEM [attribute] Item to use for feature specific styling. The style information may be represented by a
separate attribute (style string) attached to the feature. MapServer supports the following style string represen-
tations:
• MapServer STYLE definition - The style string can be represented as a MapServer STYLE block accord-
ing to the following example:
STYLE BACKGROUNDCOLOR 128 0 0 COLOR 0 0 208 END

• MapServer CLASS definition - By specifying the entire CLASS instead of a single style allows to use
further options (like setting expressions, label attributes, multiple styles) on a per feature basis.
• OGR Style String - MapServer support rendering the OGR style string format according to the OGR -
Feature Style Specification documentation. Currently only a few data sources support storing the styles
along with the features (like MapInfo, AutoCAD DXF, Microstation DGN), however those styles can
easily be transferred to many other data sources as a separate attribute by using the ogr2ogr command line
tool as follows:

o g r 2ogr - s q l "select *, OGR_STYLE from srclayer" "dstlayer" "srclayer"

We can specify the value: “AUTO” in order to have the driver to retrieve the style information. Currently
only the OGR driver supports the “AUTO” setting,

132 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

SYMBOLSCALEDENOM [double] The scale at which symbols and/or text appear full size. This allows for dy-
namic scaling of objects based on the scale of the map. If not set then this layer will always appear at the
same size. Scaling only takes place within the limits of MINSIZE and MAXSIZE as described above. Scale is
given as the denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000.
Implemented in MapServer 5.0, to replace the deprecated SYMBOLSCALE parameter.
See Also:
Map Scale
SYMBOLSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is SYMBOLSCALEDE-
NOM instead. The deprecated SYMBOLSCALE is the scale at which symbols and/or text appear full size. This
allows for dynamic scaling of objects based on the scale of the map. If not set then this layer will always appear
at the same size. Scaling only takes place within the limits of MINSIZE and MAXSIZE as described above.
Scale is given as the denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use
24000. Deprecated since version 5.0.
TEMPLATE [file|url] Used as a global alternative to CLASS TEMPLATE.
TILEINDEX [filename|layername] Name of the tileindex file or layer. A tileindex is similar to an ArcInfo library
index. The tileindex contains polygon features for each tile. The item that contains the location of the tiled data
is given using the TILEITEM parameter. When a file is used as the tileindex for shapefile or raster layers, the
tileindex should be a shapefile. For CONNECTIONTYPE OGR layers, any OGR supported datasource can be
a tileindex. Normally the location should contain the path to the tile file relative to the shapepath, not relative to
the tileindex itself. If the DATA parameter contains a value then it is added to the end of the location. When a
tileindex layer is used, it works similarly to directly referring to a file, but any supported feature source can be
used (ie. postgres, oracle).
Note: All files in the tileindex should have the same coordinate system, and for vector files the same set of
attributes in the same order.
TILEITEM [attribute] Item that contains the location of an individual tile, default is “location”.
TOLERANCE [double] Sensitivity for point based queries (i.e. via mouse and/or map coordinates). Given in TOL-
ERANCEUNITS. If the layer is a POINT or a LINE, the default is 3. For all other layer types, the default is 0.
To restrict polygon searches so that the point must occur in the polygon set the tolerance to zero.
TOLERANCEUNITS [pixels|feet|inches|kilometers|meters|miles|nauticalmiles|dd] Units of the TOLERANCE
value. Default is pixels. Nauticalmiles was added in MapServer 5.6.
TRANSPARENCY [integer|alpha] - deprecated Since MapServer 5.0 the proper parameter to use is OPACITY.
The deprecated TRANSPARENCY parameter sets the transparency level of all classed pixels for a given layer.
The value can either be an integer in the range (0-100) or the named symbol “ALPHA”. Although this parameter
is named “transparency”, the integer values actually parameterize layer opacity. A value of 100 is opaque and 0
is fully transparent.
The “ALPHA” symbol directs the MapServer rendering code to honor the indexed or alpha transparency of
pixmap symbols used to style a layer. This is only needed in the case of RGB output formats, and should be
used only when necessary as it is expensive to render transparent pixmap symbols onto an RGB map image.
Deprecated since version 5.0.
See Also:
OPACITY
TRANSFORM [true|false ul|uc|ur|lc|cc|lr|ll|lc|lr] Tells MapServer whether or not a particular layer needs to be
transformed from some coordinate system to image coordinates. Default is true. This allows you to create
shapefiles in image/graphics coordinates and therefore have features that will always be displayed in the same
location on every map. Ideal for placing logos or text in maps. Remember that the graphics coordinate system
has an origin in the upper left hand corner of the image, contrary to most map coordinate systems.

5.12. LAYER 133

MapServer Documentation, Release 6.0

Version 4.10 introduces the ability to define features with coordinates given in pixels (or percentages, see
UNITS), most often inline features, relative to something other than the UL corner of an image. That is what
‘TRANSFORM FALSE’ means. By setting an alternative origin it allows you to anchor something like a copy-
right statement to another portion of the image in a way that is independent of image size.
TYPE [point|line|polygon|circle|annotation|raster|query|chart] Specifies how the data should be drawn. Need not
be the same as the shapefile type. For example, a polygon shapefile may be drawn as a point layer, but a point
shapefile may not be drawn as a polygon layer. Common sense rules. Annotation means that a label point
will be calculated for the features, but the feature itself will not be drawn although a marker symbol can be
optionally drawn. this allows for advanced labeling like numbered highway shields. Points are labeled at that
point. Polygons are labeled first using a centroid, and if that doesn’t fall in the polygon a scanline approach
is used to guarantee the label falls within the feature. Lines are labeled at the middle of the longest arc in the
visible portion of the line. Query only means the layer can be queried but not drawn.
In order to differentiate between POLYGONs and POLYLINEs (which do not exist as a type), simply respec-
tively use or omit the COLOR keyword when classifying. If you use it, it’s a polygon with a fill color, otherwise
it’s a polyline with only an OUTLINECOLOR.
A circle must be defined by a a minimum bounding rectangle. That is, two points that define the smallest square
that can contain it. These two points are the two opposite corners of said box.
The following is an example using inline points to draw a circle:
LAYER
NAME ’inline_circles’
TYPE CIRCLE
STATUS ON
FEATURE
POINTS
74.01 -53.8
110.7 -22.16
END
END
CLASS
STYLE
COLOR 0 0 255
END
END
END

See Also:
For CHART layers, see the Dynamic Charting HowTo.
UNITS [feet|inches|kilometers|meters|miles|nauticalmiles|dd|pixels|percentages] Units of the layer. Percentages
was added in MapServer 4.10 and is mostly geared for inline features. Nauticalmiles was added in MapServer
5.6.

5.13 LEGEND

The size of the legend image is NOT known prior to creation so be careful not to hard-code width and height in the
<IMG> tag in the template file.
IMAGECOLOR [r] [g] [b] Color to initialize the legend with (i.e. the background).
INTERLACE [on|off] Default is [on]. This keyword is now deprecated in favor of using the FORMATOPTION
“INTERLACE=ON” line in the OUTPUTFORMAT declaration. Deprecated since version 4.6.
LABEL Signals the start of a LABEL object

134 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

OUTLINECOLOR [r] [g] [b] Color to use for outlining symbol key boxes.
POSITION [ul|uc|ur|ll|lc|lr] Where to place an embedded legend in the map. Default is lr.
KEYSIZE [x][y] Size of symbol key boxes in pixels. Default is 20 by 10.
KEYSPACING [x][y] Spacing between symbol key boxes ([y]) and labels ([x]) in pixels. Default is 5 by 5.
POSTLABELCACHE [true|false] Tells MapServer to render this legend after all labels in the cache have been
drawn. Useful for adding neatlines and similar elements. Default is false.
STATUS [on|off|embed] Is the legend image to be created.
TEMPLATE [filename] HTML legend template file.
See Also:
HTML Legends with MapServer
TRANSPARENT [on|off] Should the background color for the legend be transparent. This flag is now deprecated in
favor of declaring transparency within OUTPUTFORMAT declarations. Default is off. Deprecated since version
4.6.

5.14 MAP

ANGLE [double] Angle, given in degrees, to rotate the map. Default is 0. The rendered map will rotate in a clockwise
direction. The following are important notes:
• Requires a PROJECTION object specified at the MAP level and for each LAYER object (even if all layers
are in the same projection).
• Requires MapScript (SWIG, PHP MapScript). Does not work with CGI mode.
• If using the LABEL object’s ANGLE or the LAYER object’s LABELANGLEITEM parameters as well,
these parameters are relative to the map’s orientation (i.e. they are computed after the MAP object’s
ANGLE). For example, if you have specified an ANGLE for the map of 45, and then have a layer LA-
BELANGLEITEM value of 45, the resulting label will not appear rotated (because the resulting map is
rotated clockwise 45 degrees and the label is rotated counter-clockwise 45 degrees).
• More information can be found on the MapRotation Wiki Page.
CONFIG [key] [value] This can be used to specify several values at run-time, for both MapServer and GDAL/OGR
libraries. Developers: values will be passed on to CPLSetConfigOption(). Details on GDAL/OGR options
are found in their associated driver documentation pages (GDAL/OGR). The following options are available
specifically for MapServer:
CGI_CONTEXT_URL [value] The CONFIG parameter can be used to enable loading a map context from a
URL. See the Map Context HowTo for more info.
MS_ENCRYPTION_KEY [filename] The CONFIG parameter can be used to specify an encryption key that
is used with MapServer’s msencypt utility.
MS_ERRORFILE [filename] The CONFIG parameter can be used to write MapServer errors to a file (as of
MapServer 5.0). With MapServer 5.x, a full path (absolute reference) is required, including the filename.
Starting with MapServer 6.0, a filename with relative path can be passed via this CONFIG directive, in
which case the filename is relative to the mapfile location. Note that setting MS_ERRORFILE via an
environment variable always requires an absolute path since there would be no mapfile to make the path
relative to. For more on this see the DEBUG parameter below.
MS_NONSQUARE [yes|no] The CONFIG parameter can be used to allow non-square WMS requests.

5.14. MAP 135

MapServer Documentation, Release 6.0

ON_MISSING_DATA [FAIL|LOG|IGNORE] The CONFIG parameter can be used to tell MapServer how
to handle missing data in tile indexes (as of MapServer 5.3-dev, r8015). Previous MapServer versions
required a compile-time switch (“IGNORE_MISSING_DATA”), but this is no longer required.
FAIL This will cause MapServer to throw an error and exit (to crash, in other words) on a missing file in
a tile index. This is the default.

CONFIG "ON_MISSING_DATA" "FAIL"

LOG This will cause MapServer to log the error message for a missing file in a tile index, and continue
with the map creation. Note: DEBUG parameter and CONFIG “MS_ERRORFILE” need to be set for
logging to occur, so please see the DEBUG parameter below for more information.

CONFIG "ON_MISSING_DATA" "LOG"

IGNORE This will cause MapServer to not report or log any errors for missing files, and map creation
will occur normally.

CONFIG "ON_MISSING_DATA" "IGNORE"

PROJ_LIB [path] The CONFIG parameter can be used to define the location of your EPSG files for the Proj.4
library. Setting the [key] to PROJ_LIB and the [value] to the location of your EPSG files will force
PROJ.4 to use this value. Using CONFIG allows you to avoid setting environment variables to point to
your PROJ_LIB directory. Here are some examples:
1. Unix

CONFIG "PROJ_LIB" "/usr/local/share/proj/"

2. Windows

CONFIG "PROJ_LIB" "C:/somedir/proj/nad/"

DATAPATTERN [regular expression] This defines a regular expression to be applied to requests to change DATA
parameters via URL requests (i.e. map_layername_data=...). If a pattern doesn’t exist then web users can’t
monkey with support files via URLs. This allows you to isolate one application from another if you desire, with
the default operation being very conservative. See also TEMPLATEPATTERN.
DEBUG [off|on|0|1|2|3|4|5] Enables debugging of all of the layers in the current map.
Debugging with MapServer versions >= 5.0:
Verbose output is generated and sent to the standard error output (STDERR) or the MapServer errorfile if one
is set using the “MS_ERRORFILE” environment variable. You can set the environment variable by using the
CONFIG parameter
at the MAP level of the mapfile, such as:
CONFIG "MS_ERRORFILE" "/ms4w/tmp/ms_error.txt"

You can also set the environment variable in Apache by adding the following to your httpd.conf:
SetEnv MS_ERRORFILE "/ms4w/tmp/ms_error.txt"

Once the environment variable is set, the DEBUG mapfile parameter can be used to control the level of debug-
ging output. Here is a description of the possible DEBUG values:
• DEBUG O or OFF - only msSetError() calls are logged to MS_ERRORFILE. No msDebug() output at
all. This is the default and corresponds to the original behavior of MS_ERRORFILE in MapServer 4.x

136 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

• DEBUG 1 or ON - includes all output from DEBUG 0 plus msDebug() warnings about common pitfalls,
failed assertions or non-fatal error situations (e.g. missing or invalid values for some parameters, missing
shapefiles in tileindex, timeout error from remote WMS/WFS servers, etc.)
• DEBUG 2 - includes all output from DEBUG 1 plus notices and timing information useful for tuning
mapfiles and applications
• DEBUG 3 - all of DEBUG 2 plus some debug output useful in troubleshooting problems such as WMS
connection URLs being called, database connection calls, etc. This is the recommended level for debug-
ging mapfiles.
• DEBUG 4 - DEBUG 3 plus even more details...
• DEBUG 5 - DEBUG 4 plus any msDebug() output that might be more useful to the developers than to the
users.
You can also set the debug level by using the “MS_DEBUGLEVEL” environment variable.
The DEBUG setting can also be specified for a layer, by setting the DEBUG parameter in the LAYER object.
For more details on this debugging mechanism, please see the Debugging MapServer document.
Debugging with MapServer versions < 5:
Verbose output is generated and sent to the standard error output (STDERR) or the MapServer logfile if one is
set using the LOG parameter in the WEB object. Apache users will see timing details for drawing in Apache’s
error_log file. Requires MapServer to be built with the DEBUG=MSDEBUG option (–with-debug configure
option).
EXTENT [minx] [miny] [maxx] [maxy] The spatial extent of the map to be created. In most cases you will need to
specify this, although MapServer can sometimes (expensively) calculate one if it is not specified.
FONTSET [filename] Filename of fontset file to use. Can be a path relative to the mapfile, or a full path.
IMAGECOLOR [r] [g] [b] Color to initialize the map with (i.e. background color). When transparency is enabled
(TRANSPARENT ON) for the typical case of 8-bit pseudocolored map generation, this color will be marked as
transparent in the output file palette. Any other map components drawn in this color will also be transparent, so
for map generation with transparency it is best to use an otherwise unused color as the background color.
IMAGEQUALITY [int] Deprecated Use FORMATOPTION “QUALITY=n” in the OUTPUTFORMAT declaration
to specify compression quality for JPEG output. Deprecated since version 4.6.
IMAGETYPE [gif|png|jpeg|wbmp|gtiff|swf|userdefined] Output format to generate. See details in the OUTPUT-
FORMAT section for available formats. The name here must match the ‘NAME’ of a user defined or internally
generated OUTPUTFORMAT section.
INTERLACE [on|off] Deprecated Use FORMATOPTION “INTERLACE=ON” in the OUTPUTFORMAT declara-
tion to specify if the output images should be interlaced. Deprecated since version 4.6.
LAYER Signals the start of a LAYER object.
LEGEND Signals the start of a LEGEND object.
MAXSIZE [integer] Sets the maximum size of the map image. This will override the default value. For example,
setting this to 2048 means that you can have up to 2048 pixels in both dimensions (i.e. max of 2048x2048).
NAME [name] Prefix attached to map, scalebar and legend GIF filenames created using this mapfile. It should be
kept short.
PROJECTION Signals the start of a PROJECTION object.
QUERYMAP Signals the start of a QUERYMAP object.
REFERENCE Signals the start of a REFERENCE MAP object.

5.14. MAP 137

MapServer Documentation, Release 6.0

RESOLUTION [int] Sets the pixels per inch for output, only affects scale computations and nothing else, default is
72.
SCALEDENOM [double] Computed scale of the map. Set most often by the application. Scale is given as the
denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Implemented
in MapServer 5.0, to replace the deprecated SCALE parameter.
See Also:
Map Scale
SCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is SCALEDENOM instead. The
deprecated SCALE is the computed scale of the map. Set most often by the application. Scale is given as the
denominator of the actual scale fraction, for example for a map at a scale of 1:24,000 use 24000. Deprecated
since version 5.0.
SCALEBAR Signals the start of a SCALEBAR object.
SHAPEPATH [filename] Path to the directory holding the shapefiles or tiles. There can be further subdirectories
under SHAPEPATH.
SIZE [x][y] Size in pixels of the output image (i.e. the map).
STATUS [on|off] Is the map active? Sometimes you may wish to turn this off to use only the reference map or scale
bar.
SYMBOLSET [filename] Filename of the symbolset to use. Can be a path relative to the mapfile, or a full path.
SYMBOL Signals the start of a SYMBOL object.
TEMPLATEPATTERN [regular expression] This defines a regular expression to be applied to requests to change
TEMPLATE parameters via URL requests (i.e. map_layername_template=...). If a pattern doesn’t exist then
web users can’t monkey with support files via URLs. This allows you to isolate one application from another if
you desire, with the default operation being very conservative. See also DATAPATTERN.
TRANSPARENT [on|off] Use FORMATOPTION “TRANSPARENT=ON” in the OUTPUTFORMAT declaration
to specify if the output images should be transparent. Deprecated since version 4.6.
UNITS [feet|inches|kilometers|meters|miles|nauticalmiles|dd] Units of the map coordinates. Used for scalebar and
scale computations. Nauticalmiles was added in MapServer 5.6.
WEB Signals the start of a WEB object.

5.15 OUTPUTFORMAT

A map file may have zero, one or more OUTPUTFORMAT object declarations, defining available output formats
supported including formats like PNG, GIF, JPEG, GeoTIFF, SVG, PDF and KML.
If OUTPUTFORMAT sections declarations are not found in the map file, the following implicit declarations will be
made. Only those for which support is compiled in will actually be available. The GeoTIFF depends on building with
GDAL support, and the PDF and SVG depend on building with cairo support.
OUTPUTFORMAT
NAME "png"
DRIVER AGG/PNG
MIMETYPE "image/png"
IMAGEMODE RGB
EXTENSION "png"
END
OUTPUTFORMAT
NAME gif

138 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

DRIVER "GD/GIF"
MIMETYPE "image/gif"
IMAGEMODE PC256
EXTENSION "gif"
END
OUTPUTFORMAT
NAME p n g 8
DRIVER "AGG/PNG8"
MIMETYPE "image/png; mode=8bit"
IMAGEMODE RGB
EXTENSION "png"
FORMATOPTION "QUANTIZE_FORCE=on"
FORMATOPTION "QUANTIZE_COLORS=256"
END
OUTPUTFORMAT
NAME jpeg
DRIVER "AGG/JPEG"
MIMETYPE "image/jpeg"
IMAGEMODE RGB
EXTENSION "jpg"
END
OUTPUTFORMAT
NAME s v g
DRIVER "CAIRO/SVG"
MIMETYPE "image/svg+xml"
IMAGEMODE RGB
EXTENSION "svg"
END
OUTPUTFORMAT
NAME pdf
DRIVER "CAIRO/PDF"
MIMETYPE "application/x-pdf"
IMAGEMODE RGB
EXTENSION "pdf"
END
OUTPUTFORMAT
NAME GTiff
DRIVER "GDAL/GTiff"
MIMETYPE "image/tiff"
IMAGEMODE RGB
EXTENSION "tif"
END
OUTPUTFORMAT
NAME k m l
DRIVER "KML"
MIMETYPE "application/vnd.google-earth.kml.xml"
IMAGEMODE RGB
EXTENSION "kml"
END
OUTPUTFORMAT
NAME k m z
DRIVER "KMZ"
MIMETYPE "application/vnd.google-earth.kmz"
IMAGEMODE RGB
EXTENSION "kmz"
END
OUTPUTFORMAT
NAME "cairopng"

5.15. OUTPUTFORMAT 139

MapServer Documentation, Release 6.0

DRIVER C A I R O /PNG
MIMETYPE "image/png"
IMAGEMODE RGB
EXTENSION "png"
END

NAME [name] The name to use use in the IMAGETYPE keyword of the map file to select this output format.
This name is also used in metadata describing wxs formats allowed, and can be used (sometimes along with
mimetype) to select the output format via keywords in OGC requests. (optional)
DRIVER [name] The name of the driver to use to generate this output format. Some driver names include the defini-
tion of the format if the driver supports multiple formats. For AGG, the possbile driver names are “AGG/PNG”
and “AGG/JPEG”. For GD the possible driver names are “GD/Gif” and “GD/PNG”. For output through OGR the
OGR driver name is appended, such as “OGR/Mapinfo File”. For output through GDAL the GDAL shortname
for the format is appended, such as “GDAL/GTiff”. Note that PNG, JPEG and GIF output can be generated with
either GDAL or GD (GD is generally more efficient). TEMPLATE should be used for template based output.
(mandatory)
IMAGEMODE [PC256/RGB/RGBA/INT16/FLOAT32/FEATURE] Selects the imaging mode in which the output
is generated. Does matter for non-raster formats like Flash. Not all formats support all combinations. For
instance GD supports only PC256. (optional)
• PC256: Produced a pseudocolored result with up to 256 colors in the palette (legacy MapServer mode).
Only supported for GD/GIF and GD/PNG.
• RGB: Render in 24bit Red/Green/Blue mode. Supports all colors but does not support transparency.
• RGBA: Render in 32bit Red/Green/Blue/Alpha mode. Supports all colors, and alpha based transparency.
All features are rendered against an initially transparent background.
• BYTE: Render raw 8bit pixel values (no presentation). Only works for RASTER layers (through GDAL)
and WMS layers currently.
• INT16: Render raw 16bit signed pixel values (no presentation). Only works for RASTER layers (through
GDAL) and WMS layers currently.
• FLOAT32: Render raw 32bit floating point pixel values (no presentation). Only works for RASTER layers
(through GDAL) and WMS layers currently.
• FEATURE: Output is a non-image result, such as features written via templates or OGR.
MIMETYPE [type] Provide the mime type to be used when returning results over the web. (optional)
EXTENSION [type] Provide the extension to use when creating files of this type. (optional)
TRANSPARENT [ON/OFF] Indicates whether transparency should be enabled for this format. Note that trans-
parency does not work for IMAGEMODE RGB output. Not all formats support transparency (optional). When
transparency is enabled for the typical case of 8-bit pseudocolored map generation, the IMAGECOLOR color
will be marked as transparent in the output file palette. Any other map components drawn in this color will
also be transparent, so for map generation with transparency it is best to use an otherwise unused color as the
background color.
FORMATOPTION [option] Provides a driver or format specific option. Zero or more FORMATOPTION statement
may be present within a OUTPUTFORMAT declaration. (optional)
• AGG/JPEG: The “QUALITY=n” option may be used to set the quality of jpeg produced (value from
0-100).
• GD/PNG: The “INTERLACE=[ON/OFF]” option may be used to turn interlacing on or off.
• GD/GIF: The “INTERLACE=[ON/OFF]” option may be used to turn interlacing on or off.

140 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

• GDAL/GTiff: Supports the TILED=YES, BLOCKXSIZE=n, BLOCKYSIZE=n, INTER-

LEAVE=[PIXEL/BAND] and COMPRESS=[NONE,PACKBITS,JPEG,LZW,DEFLATE] format specific
options.
• GDAL/*: All FORMATOPTIONs are passed onto the GDAL create function. Options supported by GDAL
are described in the detailed documentation for each GDAL format
• GDAL/*: NULLVALUE=n is used in raw image modes (IMAGEMODE BYTE/INT16/FLOAT) to pre-
initialize the raster and an attempt is made to record this in the resulting file as the nodata value. This is
automatically set in WCS mode if rangeset_nullvalue is set.
• OGR/*: See OGR Output document for details of OGR format options.
• AGG/PNG: COMPRESSION=n is used to determine the ZLIB compression applied to the png creation.
n is expected to be an integer value from 0 to 9, with 0 meaning no compression (not recommended),
1 meaning fastest compression, and 9 meaning best compression. The compression levels come at a
cost (be it in terms of cpu processing or file size, chose the setting that suits you most). The default is
COMPRESSION=6.
• AGG/PNG supports quantizing from 24/32 bits to 8bits, in order to reduce the final image size (and there-
fore save bandwidth) (see also http://trac.osgeo.org/mapserver/ticket/2436#comment:4 for strategies when
applying these options):
– “QUANTIZE_FORCE=on” used to reduce an RGB or RGBA image into an 8bit (or less) paletted
images. The colors used in the palette are selected to best fit the actual colors in the RGB or RGBA
image.
– “QUANTIZE_COLORS=256” used to specify the number of colors to be used when applying quan-
tization. Maximum value is 256. Specifying anything between 17 and 255 is probably a waste of
quality as each pixel is still encoded with a full byte. Specifying a value under 16 will produce tiny
images, but severly degraded.
– “PALETTE_FORCE=on” is used to reduce image depth with a predefined palette. This option is
incompatible with the previous quantization options.
– “PALETTE=/path/to/palette.txt” is used to define the absolute path where palette colors can be found.
This file must contain 256 entries of r,g,b triplets for RGB imagemodes, or r,g,b,a quadruplets for
RGBA imagemodes. The expected format is one triplet (or quadruplet) per line, each value separated
by commas, and each triplet/quadruplet on a single line. If you want to use transparency with a palette,
it is important to have these two colors in the palette file: 0,0,0,0 and 255,255,255,255.
Note: 0,0,0,0 is important if you have fully transparent areas. 255,255,255,255 is opaque white. The
important colors to have in your palette really depend on your actual map, although 0,0,0,0 , 0,0,0,255
, and 255,255,255,255 are very likely to show up most of the time.

5.16 PROJECTION

To set up projections you must define two projection objects: one for the output image (in the MAP object) and one for
each layer (in the LAYER objects) to be projected. MapServer relies on the Proj.4 library for projections. Projection
objects therefore consist of a series of PROJ.4 keywords, which are either specified within the object directly or
referred to in an EPSG file. An EPSG file is a lookup file containing projection parameters, and is part of the PROJ.4
library.
The following two examples both define the same projection (UTM zone 15, NAD83), but use 2 different methods:
Example 1: Inline Projection Parameters

5.16. PROJECTION 141

MapServer Documentation, Release 6.0

PROJECTION
"proj=utm"
"ellps=GRS80"
"datum=NAD83"
"zone=15"
"units=m"
"north"
"no_defs"
END

Note: For a list of all of the possible PROJ.4 projection parameters, see the PROJ.4 parameters page.
Example 2: EPSG Projection Use
PROJECTION
"init=epsg:26915"
END

Note: This refers to an EPSG lookup file that contains a ‘26915’ code with the full projection parameters. “epsg” in
this instance is case-sensitive because it is referring to a file name. If your file system is case-sensitive, this must be
lower case, or MapServer (Proj.4 actually) will complain about not being able to find this file.
Note: See http://spatialreference.org/ref/epsg/26915 for more information on this coordinate system.
The next two examples both display how to possibly define unprojected lat/longs (“geographic”):
Example 3: Inline Projection Parameters
PROJECTION
"proj=latlong"
"ellps=WGS84"
"datum=WGS84"
END

Example 4: epsg Projection Use

PROJECTION
"init=epsg:4326"
END

5.16.1 Important Notes

• If all of your data in the mapfile is in the same projection, you DO NOT have to specify any projection objects.
MapServer will assume that all of the data is in the same projection.
• Think of the MAP-level projection object as your output projection. The EXTENT and UNITS values at the
MAP-level must be in the output projection units. Also, if you have layers in other projections (other than the
MAP-level projection) then you must define PROJECTION objects for those layers, to tell MapServer what
projections they are in.
• If you specify a MAP-level projection, and then only one other LAYER projection object, MapServer will
assume that all of the other layers are in the specified MAP-level projection.
• Always refer to the EPSG file in lowercase, because it is a lowercase filename and on Linux/Unix systems this
parameter is case sensitive.

5.16.2 For More Information

• If you get projection errors, refer to the Errors to check if your exact error has been discussed.

142 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

• Search the MapServer-users email list archives, odds are that someone has faced your exact issue before.
• See the PROJ.4 user guides for complete descriptions of supported projections and coordinate systems.
• Refer to the Cartographical Map Projections page for background information on projections.

5.17 QUERYMAP

COLOR [r] [g] [b] Color in which features are highlighted. Default is yellow.
SIZE [x][y] Size of the map in pixels. Defaults to the size defined in the map object.
STATUS [on|off] Is the query map to be drawn?
STYLE [normal|hilite|selected] Sets how selected features are to be handled. Layers not queried are drawn as usual.
• Normal: Draws all features according to the settings for that layer.
• Hilite: Draws selected features using COLOR. Non-selected features are drawn normally.
• Selected: draws only the selected features normally.

5.18 REFERENCE

Three types of reference maps are supported. The most common would be one showing the extent of a map in an
interactive interface. It is also possible to request reference maps as part of a query. Point queries will generate an
image with a marker (see below) placed at the query point. Region based queries will depict the extent of the area of
interest. Finally, feature based queries will display the selection feature(s) used.
COLOR [r] [g] [b] Color in which the reference box is drawn. Set any component to -1 for no fill. Default is red.
EXTENT [minx][miny][maxx][maxy] The spatial extent of the base reference image.
IMAGE [filename] Full filename of the base reference image. Must be a GIF image.
MARKER [integer|string] Defines a symbol (from the symbol file) to use when the box becomes too small (see
MINBOXSIZE and MAXBOXSIZE below). Uses a crosshair by default.
MARKERSIZE [integer] Defines the size of the symbol to use instead of a box (see MARKER above).
MINBOXSIZE [integer] If box is smaller than MINBOXSIZE (use box width or height) then use the symbol defined
by MARKER and MARKERSIZE.
MAXBOXSIZE [integer] If box is greater than MAXBOXSIZE (use box width or height) then draw nothing (Often
the whole map gets covered when zoomed way out and it’s perfectly obvious where you are).
OUTLINECOLOR [r] [g] [b] Color to use for outlining the reference box. Set any component to -1 for no outline.
SIZE [x][y] Size, in pixels, of the base reference image.
STATUS [on|off] Is the reference map to be created? Default it off.

5.19 SCALEBAR

Scalebars currently do not make use of TrueType fonts. The size of the scalebar image is NOT known prior to
rendering, so be careful not to hard-code width and height in the <IMG> tag in the template file. Future versions will
make the image size available.

5.17. QUERYMAP 143

MapServer Documentation, Release 6.0

ALIGN [left|center|right] Defines how the scalebar is aligned within the scalebar image. Default is center. Available
in versions 5.2 and higher. New in version 5.2.
BACKGROUNDCOLOR [r] [g] [b] Color to use for scalebar background, not the image background.
COLOR [r] [g] [b] Color to use for drawing all features if attribute tables are not used.
IMAGECOLOR [r] [g] [b] Color to initialize the scalebar with (i.e. background).
INTERLACE [true|false] Should output images be interlaced? Default is [on]. This keyword is now deprecated in
favour of using the FORMATOPTION “INTERLACE=ON” line in the OUTPUTFORMAT declaration. Depre-
cated since version 4.6.
INTERVALS [integer] Number of intervals to break the scalebar into. Default is 4.
LABEL Signals the start of a LABEL object
OUTLINECOLOR [r] [g] [b] Color to use for outlining individual intervals. Set any component to -1 for no outline
which is the default.
POSITION [ul|uc|ur|ll|lc|lr] Where to place an embedded scalebar in the image. Default is lr.
POSTLABELCACHE [true|false] For use with embedded scalebars only. Tells the MapServer to embed the scale-
bar after all labels in the cache have been drawn. Default is false.
SIZE [x][y] Size in pixels of the scalebar. Labeling is not taken into account.
STATUS [on|off|embed] Is the scalebar image to be created, and if so should it be embedded into the image? Default
is off. (Please note that embedding scalebars require that you define a markerset. In essence the scalebar
becomes a custom marker that is handled just like any other annotation.)
STYLE [integer] Chooses the scalebar style. Valid styles are 0 and 1.
TRANSPARENT [on|off] Should the background color for the scalebar be transparent. This flag is now deprecated
in favor of declaring transparency within OUTPUTFORMAT declarations. Default is off. Deprecated since
version 4.6.
UNITS [feet|inches|kilometers|meters|miles|nauticalmiles] Output scalebar units, default is miles. Used in con-
junction with the map’s units to develop the actual graphic. Note that decimal degrees are not valid scalebar
units. Nauticalmiles was added in MapServer 5.6.

5.20 STYLE

Style holds parameters for symbolization and styling. Multiple styles may be applied within a CLASS or LABEL.
This object appeared in 4.0 and the intention is to separate logic from looks. The final intent is to have named styles
(Not yet supported) that will be re-usable through the mapfile. This is the way of defining the appearance of an object
(a CLASS or a LABEL).
ANGLE [double|attribute|AUTO] Angle, given in degrees, to rotate the symbol (counter clockwise). Default is 0
(no rotation). If you have an attribute that specifies angles in a clockwise direction (compass direction), you
have to adjust the angle attribute values before they reach Mapserver (360-ANGLE), as it is not possible to use
a mathematical expression for ANGLE.
• For points, it specifies the rotation of the symbol around its center.
• For decorated lines, the behaviour depends on the value of the GAP element.
– For negative GAP values it specifies the rotation of the decoration symbol relative to the direction of
the line. An angle of 0 means that the symbol’s x-axis is oriented along the direction of the line.

144 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

– For non-negativ (or absent) GAP values it specifies the rotation of the decoration symbol around its
center. An angle of 0 means that the symbol is not rotated.
• For polygons, it specifies the angle of the lines in a HATCH symbol (0 - horizontal lines), or it specifies
the rotation of the symbol used to generate the pattern in a polygon fill (it does not specify the rotation of
the fill as a whole). For its use with hatched lines, see Example #7 in the symbology examples.
• [attribute] was introduced in version 5.0, to specify the attribute to use for angle values. The hard brackets
[] are required. For example, if your data source has an attribute named “MYROTATE” that holds angle
values for each feature, your STYLE object for hatched lines might contain:
STYLE
SYMBOL ’hatch-test’
COLOR 255 0 0
ANGLE [ M Y R O T A T E ]
SIZE 4.0
WIDTH 3.0
END

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
• The AUTO keyword was added in version 5.4, and currently only applies when coupled with the GEOM-
TRANSFORM keyword.
ANGLEITEM [string] ANGLE[attribute] must now be used instead. Deprecated since version 5.0.
ANTIALIAS [true|false] Should TrueType fonts be antialiased. Only useful for GD (gif) rendering. Default is false.
Has no effect for the other renderers (where anti-aliasing can not be turned off).
BACKGROUNDCOLOR [r] [g] [b] Color to use for non-transparent symbols.
COLOR [r] [g] [b] | [attribute]
• Color to use for drawing features. r, g and b shall be integers [0..255]. To specify green, the following is used:
COLOR 0 255 0

• [attribute] was introduced in version 5.0, to specify the attribute to use for color values. The hard brackets []
are required. For example, if your data set has an attribute named “MYPAINT” that holds color values for each
record, use: object for might contain:

COLOR [ M Y P A I N T ]

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
GAP [double] GAP specifies the distance between SYMBOLs (center to center) for decorated lines and polygon fills
in layer SIZEUNITS. For polygon fills, GAP specifies the distance between SYMBOLs in both the X and the Y
direction. For lines, the centers of the SYMBOLs are placed on the line. As of MapServer 5.0 this also applies
to PixMap symbols.
When scaling of symbols is in effect (SYMBOLSCALEDENOM is specified for the LAYER), GAP specifies the
distance in layer SIZEUNITS at the map scale 1:SYMBOLSCALEDENOM.
• For lines, the first symbol will be placed GAP/2 from the start of the line.
• For lines, a negative GAP value will cause the symbols’ X axis to be aligned relative to the tangent of the
line.
• For lines, a positive GAP value aligns the symbols’ X axis relative to the X axis of the output device.
• For lines, a GAP of 0 (the default value) will cause the symbols to be rendered edge to edge
• For polygons, a missing GAP or a GAP of 0 will cause the symbols to be rendered edge to edge.

5.20. STYLE 145

MapServer Documentation, Release 6.0

Symbols can be rotated using ANGLE. New in version 6.0: moved from SYMBOL
Note: The behaviour of GAP has not been stable over time. It has specified the amount of space between
the symbols, and also something in between the amount of space between the symbols and the center to center
distance. The goal is to have GAP specify the center to center distance, but in version 6.0 it is the amount of
space between the symbols that is specified.
GEOMTRANSFORM [bbox|end|labelpnt|labelpoly|start|vertices] Used to indicate that the current feature will be
transformed before the actual style is applied. Introduced in version 5.4.
• bbox: produces the bounding box of the geometry.
• end: produces the last point of the geometry. When used with ANGLE AUTO, it can for instance be used
to render an arrowhead.
• labelpnt: used for LABEL styles. It produces a point that is placed at the label’s reference point.
• labelpoly: used for LABEL styles. Produces a polygon that covers the label plus a 1 pixel padding.
• start: produces the first point of the current geometry. When used with ANGLE AUTO, it can for instance
be used to render arrow tail on line segments.
• vertices: produces all the intermediate vertices (points) of the underlying geometry (the start and end are
excluded). When used with ANGLE AUTO, the marker is oriented by the half angle formed by the two
adjacent line segments.
LINECAP [butt|round|square] Sets the line cap type for lines. Default is round. See Cartographical Symbol Con-
struction with MapServer for explanation and examples. New in version 6.0: moved from SYMBOL
LINEJOIN [round|miter|bevel] Sets the line join type for lines. Default is round. See Cartographical Symbol
Construction with MapServer for explanation and examples. New in version 6.0: moved from SYMBOL
LINEJOINMAXSIZE [int]
Sets the max length of the miter LINEJOIN type. The value represents a coefficient which multiplies a
current symbol size. Default is 3. See Cartographical Symbol Construction with MapServer for explana-
tion and examples. New in version 6.0: moved from SYMBOL
MAXSIZE [double] Maximum size in layer SIZEUNITS to draw a symbol. Default is 500. Starting from version
5.4, the value can also be a decimal value (and not only integer).
MAXWIDTH [double] Maximum width in layer SIZEUNITS to draw the line work. Default is 32. Starting from
version 5.4, the value can also be a decimal value (and not only integer).
MINSIZE [double] Minimum size in layer SIZEUNITS to draw a symbol. Default is 0. Starting from version 5.4,
the value can also be a decimal value (and not only integer).
MINWIDTH [double] Minimum width in layer SIZEUNITS to draw the line work. Default is 0. Starting from
version 5.4, the value can also be a decimal value (and not only integer).
OFFSET [x][y] Geometry offset values in layer SIZEUNITS.
When scaling of symbols is in effect (SYMBOLSCALEDENOM is specified for the LAYER), OFFSET gives
offset values in layer SIZEUNITS at the map scale 1:SYMBOLSCALEDENOM.
An OFFSET of 20 40 will shift the geometry 20 SIZEUNITS to the left and 40 SIZEUNITS down before render-
ing.
For lines, an OFFSET of n -99 will produce a line geometry that is shifted n SIZEUNITS perpendicular to the
original line geometry. A positive n shifts the line to the right when seen along the direction of the line. A
negative n shifts the line to the left when seen along the direction of the line.
OPACITY [integer|attribute] Opacity to draw the current style (applies to 5.2+, AGG Rendering Specifics only, does
not apply to pixmap symbols)

146 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

• [attribute] was introduced in version 5.6, to specify the attribute to use for opacity values.
OUTLINECOLOR [r] [g] [b] | [attribute]
• Color to use for outlining polygons and certain marker symbols (ellipse, vector and truetype). Has no effect for
lines. The width of the outline can be specified using WIDTH. If no WIDTH is specified, an outline of one pixel
will be drawn. r, g and b shall be integers [0..255]. To specify green, the following is used:
OUTLINECOLOR 0 255 0
WIDTH 3.0

• [attribute] was introduced in version 5.0, to specify the attribute to use for color values. The hard brackets []
are required. For example, if your data set has an attribute named “MYPAINT” that holds color values for each
record, use: object for might contain:

OUTLINECOLOR [ M Y P A I N T ]

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
PATTERN [double on] [double off] [double on] [double off] ... END
Currently used to defines a dash pattern for line work (lines, polygon outlines, ...). The numbers (doubles)
specify the lengths of the dashes and gaps of the dash pattern in layer SIZEUNITS.
When scaling of symbols is in effect (SYMBOLSCALEDENOM is specified for the LAYER), the numbers
specify the lengths of the dashes and gaps in layer SIZEUNITS at the map scale 1:SYMBOLSCALEDE-
NOM.
To specify a dashed line that is 5 units wide, with dash lengths of 5 units and gaps of 5 units, the following
style can be used:
STYLE
COLOR 0 0 0
WIDTH 5.0
LINECAP BUTT
PATTERN 5.0 5.0 END
END

New in version 6.0: moved from SYMBOL

SIZE [double|attribute]
Height, in layer SIZEUNITS, of the symbol/pattern to be used. Only useful with scalable symbols (SYM-
BOL TYPE ellipse, hatch, pixmap, truetype, vector). Default is 1.0.
When scaling of symbols is in effect (SYMBOLSCALEDENOM is specified for the LAYER), SIZE gives
the height, in layer SIZEUNITS, of the symbol/pattern to be used at the map scale 1:SYMBOLSCALEDE-
NOM.
• For symbols of TYPE hatch, the SIZE is the center to center distance between the lines. For its use
with hatched lines, see Example#8 in the symbology examples.
• [attribute] was introduced in version 5.0, to specify the attribute to use for size values. The hard
brackets [] are required. For example, if your data set has an attribute named “MYHIGHT” that
holds size values for each feature, your STYLE object for hatched lines might contain:
STYLE
SYMBOL ’hatch-test’
COLOR 255 0 0
ANGLE 45
SIZE [ M Y H I G H T ]

5.20. STYLE 147

MapServer Documentation, Release 6.0

WIDTH 3.0
END

The associated RFC document for this feature is MS RFC 19: Style & Label attribute binding.
• Starting from version 5.4, the value can also be a decimal value (and not only integer).
SIZEITEM [string] SIZE [attribute] must now be used instead. Deprecated since version 5.0.
SYMBOL [integer|string|filename|attribute] The symbol name or number to use for all features if attribute tables
are not used. The number is the index of the symbol in the symbol file, starting at 1, the 5th symbol in the file
is therefore symbol number 5. You can also give your symbols names using the NAME keyword in the symbol
definition file, and use those to refer to them. Default is 0, which results in a single pixel, single width line, or
solid polygon fill, depending on layer type.
You can also specify a GIF or PNG filename. The path is relative to the location of the mapfile.
[attribute] was introduced in version 5.6, to use an attribute to specify the symbol name. The hard brackets []
are required.
WIDTH [double|attribute]
WIDTH refers to the thickness of line work drawn, in layer SIZEUNITS. Default is 1.0.
When scaling of symbols is in effect (SYMBOLSCALEDENOM is specified for the LAYER), WIDTH
refers to the thickness of the line work in layer SIZEUNITS at the map scale 1:SYMBOLSCALEDENOM.
• If used with SYMBOL and OUTLINECOLOR, WIDTH specifies the width of the symbol outlines.
This applies to SYMBOL TYPE vector (polygons), ellipse and truetype.
• For lines, WIDTH specifies the width of the line.
• For polygons, if used with OUTLINECOLOR, WIDTH specifies the thickness of the polygon outline.
• For a symbol of SYMBOL TYPE hatch, WIDTH specifies the thickness of the hatched lines. For its
use with hatched lines, see Example #7 in the symbology examples.
• [attribute] was added in version 5.2 to specify the attribute to use for the width value. The hard
brackets [] are required.
• Starting from version 5.4, the value can also be a decimal value (and not only integer).

5.21 SYMBOL

• Symbol definitions can be included within the main map file or, more commonly, in a separate file. Symbol
definitions in a separate file are designated using the SYMBOLSET keyword, as part of the MAP object. This
recommended setup is ideal for re-using symbol definitions across multiple MapServer applications.
• There are 3 main types of symbols in MapServer: Markers, Lines and Shadesets.
• Symbol 0 is always the degenerate case for a particular class of symbol. For points, symbol 0 is a single pixel,
for shading (i.e. filled polygons) symbol 0 is a solid fill, and for lines, symbol 0 is a single pixel wide line.
• Symbol definitions contain no color information, colors are set within STYLE objects.
• For MapServer versions < 5 there is a maximum of 64 symbols per file. This can be changed by editing
mapsymbol.h and changing the value of MS_MAXSYMBOLS at the top of the file. As of MapServer 5.0 there
is no symbol limit.
• More information can be found in the Construction of Cartographic Symbols document.

148 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

ANTIALIAS [true|false] Should TrueType fonts be antialiased. Only useful for GD (gif) rendering. Default is false.
Has no effect for the other renderers (where anti-aliasing can not be turned off).
CHARACTER [char] Character used to reference a particular TrueType font character. You’ll need to figure out the
mapping from the keyboard character to font character.
FILLED [true|false] Sets the symbol to be filled with a user defined color (See the STYLE object). For marker
symbols, if OUTLINECOLOR was specified then the symbol is outlined with it.
FONT [string] Name of TrueType font to use as defined in the FONTSET.
GAP [int] This keyword has been moved to STYLE in version 6.0. Deprecated since version 6.0.
IMAGE [string] Image (GIF or PNG) to use as a marker or brush for type PIXMAP symbols.
NAME [string] Alias for the symbol. To be used in CLASS STYLE objects.
LINECAP [butt|round|square|triangle] This keyword has been moved to STYLE in version 6.0. Deprecated since
version 6.0.
LINEJOIN [round|miter|bevel] This keyword has been moved to STYLE in version 6.0. Deprecated since version
6.0.
LINEJOINMAXSIZE [int] This keyword has been moved to STYLE in version 6.0. Deprecated since version 6.0.
PATTERN [num on] [num off] [num on] ... END This keyword has been moved to STYLE in version 6.0. Depre-
cated since version 6.0.
POINTS [x y] [x y] ... END
Signifies the start of a sequence of points that make up a symbol of TYPE vector or that define the x and y
radius of a symbol of TYPE ellipse. The end of this section is signified with the keyword END. The x and
y values can be given using decimal numbers. The maximum x and y values define the bounding box of
the symbol. The size (actually height) of a symbol is defined in the STYLE. You can create non-contiguous
paths by inserting “-99 -99” at the appropriate places.
x values increase to the right, y values increase downwards.
For symbols of TYPE ellipse, a single point is specified that defines the x and y radius of the ellipse.
Circles are created when x and y are equal.
Note: If a STYLE using this symbol doesn’t contain an explicit size, then the default symbol size will
be based on the range of “y” values in the point coordinates. e.g. if the y coordinates of the points in the
symbol range from 0 to 5, then the default size for this symbol will be assumed to be 5.
STYLE [num on] [num off] [num on] ... END Renamed to PATTERN in MapServer 5.0. Deprecated since version
5.0.
TRANSPARENT [color index] Sets a transparent color for the input image for pixmap symbols, or determines
whether all shade symbols should have a transparent background. For shade symbols it may be desirable to
have background features “show through” a transparent hatching pattern, creating a more complex map. By
default a symbol’s background is the same as the parent image (i.e. color 0). This is user configurable.
TYPE [ellipse|hatch|pixmap|simple|truetype|vector]
• ellipse: radius values in the x and y directions define an ellipse.
• hatch: produces hatched lines throughout the (polygon) shape.
• pixmap: a user supplied image will be used as the symbol.
• simple: default symbol type (1 pixel point, 1 pixel line, solid fill).
• truetype: TrueType font to use as defined in the MAP FONTSET.
• vector: a vector drawing is used to define the shape of the symbol.

5.21. SYMBOL 149

MapServer Documentation, Release 6.0

5.22 Symbology Examples

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Author Håvard Tveite
Contact havard.tveite at umb.no
Date $Date: 2011-05-11 02:01:34 -0700 (Wed, 11 May 2011) $
Revision $Revision: 11672 $
Last Updated 2011/05/11

Table of Contents
• Symbology Examples
– Example 1. Dashed Line
– Example 2. TrueType font marker symbol
– Example 3. Vector triangle marker symbol
– Example 4. Non-contiguous vector marker symbol (Cross)
– Example 5. Circle vector symbol
– Example 6. Downward diagonal fill
– Example 7. Using the Symbol Type HATCH (new in 4.6)
– Example 8. Styled lines using GAP

5.22.1 Example 1. Dashed Line

This example creates a dashed line that is 5 SIZEUNITS wide, with 10 SIZEUNITS on, 5 off, 5 on, 10 off ...
LAYER
...
CLASS
...
STYLE
COLOR 0 0 0
WIDTH 5
LINECAP butt
PATTERN 10 5 5 10 END
END
END
END

5.22.2 Example 2. TrueType font marker symbol

This example symbol is a star, used to represent the national capital, hence the name. The font name in defined in
the FONTSET file. The code number “114” varies, you can use MS Windows’ character map to figure it out, or
guestimate.
SYMBOL
NAME "natcap"
TYPE TRUETYPE
FONT "geo"

150 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

FILLED true
ANTIALIAS true # only necessary for GD rendering
CHARACTER "&#114;"
END

5.22.3 Example 3. Vector triangle marker symbol

This example is fairly straight forward. Note that to have 3 sides you need 4 points, hence the first and last points are
identical. The triangle is not filled.
SYMBOL
NAME "triangle"
TYPE vector
POINTS
0 4
2 0
4 4
0 4
END
END

5.22.4 Example 4. Non-contiguous vector marker symbol (Cross)

This example draws a cross, that is 2 lines (vectors) that are not connected end-to-end (Like the triangle in the previous
example). The negative values separate the two.
SYMBOL
NAME "cross"
TYPE vector
POINTS
2.0 0.0
2.0 4.0
-99 -99
0.0 2.0
4.0 2.0
END
END

5.22.5 Example 5. Circle vector symbol

This example creates a simple filled circle. Using non-equal values for the point will give you an actual ellipse.
SYMBOL
NAME "circle"
TYPE ellipse
FILLED true
POINTS
1 1
END
END

5.22.6 Example 6. Downward diagonal fill

This example creates a symbol that can be used to create a downward diagonal fill for polygons.

5.22. Symbology Examples 151

MapServer Documentation, Release 6.0

SYMBOL
NAME "downwarddiagonalfill"
TYPE vector
TRANSPARENT 0
POINTS
0 1
1 0
END
END

5.22.7 Example 7. Using the Symbol Type HATCH (new in 4.6)

As of MapServer 4.6, you can use the symbol type HATCH to produce hatched lines. The following will display
hatched lines at a 45 degree angle, 10 SIZEUNITS apart (center to center), and 3 SIZEUNITS wide.
Symbol definition:
SYMBOL
NAME ’hatch-test’
TYPE HATCH
END

Layer definition:
LAYER
...
CLASS
...
STYLE
SYMBOL ’hatch-test’
COLOR 255 0 0
ANGLE 45
SIZE 10
WIDTH 3
END
END
END

Other parameters available for HATCH are: MINSIZE, MAXSIZE, MINWIDTH, and MAXWIDTH.

5.22.8 Example 8. Styled lines using GAP

This example shows how to style lines with symbols.

A 5 SIZEUNITS wide black line is decorated with ellipses that are 15 SIZEUNITS long (and 7.5 SIZEUNITS‘wide).
The ellipses are placed 30 ‘SIZEUNITS apart, and the negative GAP value ensures that the ellipses are oriented relative
to the direction of the line. The ellipses are rotated 30 degrees counter clock-wise from their position along the line.
Symbol definition:
SYMBOL
NAME "ellipse2"
TYPE ellipse
FILLED true
POINTS
1 2
END
END

152 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

Layer definition:
LAYER
...
CLASS
...
STYLE
WIDTH 5
COLOR 0 0 0
END
STYLE
SYMBOL ’ellipse2’
COLOR 0 0 0
ANGLE 30
SIZE 15
GAP -30
END
END
END

5.23 Templating

Author Frank Koormann

Contact frank.koormann at intevation.de
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Revision $Revision: 9819 $
Date $Date: 2010-02-20 12:18:22 -0800 (Sat, 20 Feb 2010) $
Last Updated 2010/01/25

Table of Contents
• Templating
– Introduction
– Format
– Example Template

5.23.1 Introduction

Templates are used:

• to define the look of a MapServer CGI application interface and
• to present the results of a query.
They guide the presentation of results, either a query or a map, to the user. Templates are almost always HTML files
although they can also be a URL (e.g.. http://www.somewhere.com/[ATTRIBUTE]/info.html). URL templates can
only be used with simple QUERY or ITEMQUERY results so many substitutions defined below are not available for
them. Simple pan/zoom interfaces use a single template file while complicated queries often require many templates.
Templates often use JavaScript to enhance the basic interface.

5.23. Templating 153

MapServer Documentation, Release 6.0

Notes

• Templates must contain the magic string ‘mapserver template’ in the first line of the template. Often this takes
the form of an HTML, javascript or XML comment. This line is not written to the client. The magic string is
not case sensitive.
• All CGI parameters can be referenced in template substitutions, MapServer specific parameters as well as user
defined ones. In principle parameters are handed through by the MapServer 1:1. This feature is essential for
implementing MapServer applications.
The reference below only lists special template substitution strings which are needed to obtain information
modified by the MapServer, e.g. a new scale, query results, etc.
• Template substitution strings are case sensitive.
• Attribute item substitutions must be the same case as the item names in the dbase file.
• ArcView and ArcInfo generally produce dbase files with item names that are all uppercase. Appropriate URL
encoding (i.e. ‘ ‘ to ‘+’) is applied when templates are URLs.
• Some substitutions are also available in escaped form (i.e. URL encoded).
As an example this is needed when generating links within a template. This might pass the current mapextent to a new
MapServer call. [mapext] is substituted by a space delimited set of lower left and upper right coordinates. This would
break the URL. [mapext_esc] is substituted by a proper encoded set.

5.23.2 Format

Templates are simply HTML files or URL strings that contains special characters that are replaced by mapserv each
time the template is processed. The simple substitution allows information such as active layers or the spatial extent
to be passed from the user to mapserv and back again. Most often the new values are dumped into form variables that
will be passed on again. The list of special characters and form variables is given below. HTML templates can include
just about anything including JavaScript and Java calls.
In HTML files, the attribute values can be inside quotes(“”). Writing attribute values inside quotes allows you to set
special characters in value that you couldn’t use normaly (ie: ],=,” and space). To write a single quote in a attribute
value, just use two quotes (“”).

General

[version] The MapServer version number.

[id] Unique session id. The id can be passed in via a form but is more commonly generated by the software. In that
case the id is a concatenation of UNIX time (or NT equivalent) and the process id. Unless you’re getting more
requests in a second than the system has process ids the id can be considered unique. ;->
[host] Hostname of the web server.
[port] Port the web server is listening to.
[post or get variable name], [post or get variable name_esc] The contents of any variables passed to the
MapServer, whether they were used or not, can be echoed this way. One use might be to have the user set
a map title or north arrow style in an interactive map composer. The system doesn’t care about the values, but
they might be real important in creating the final output, e.g. if you specified a CGI parameter like myvalue=....
you can access this in the template file with [myvalue].
Also available as escaped version.

154 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

[web_meta data key],[web_meta data key_esc] Web object meta data access (e.g [web_projection]
Also available as escaped version.
[errmsg], [errmsg_esc] Current error stack output. Various error messages are delimited by semi-colons.
Also available as escaped version.

File Reference

[img] Path (relative to document root) of the new image, just the image name if IMAGE_URL is not set in the mapfile.
In a map interface template, [img] is substituted with the path to the map image. In a query results template, it
is substituted with the path to the querymap image (if a QUERYMAP object is defined in the Mapfile).
[ref] Path (relative to document root) of the new reference image.
[legend] Path (relative to document root) of new legend image rendered by the MapServer.
Since version 3.5.1 a new HTML Legend template is provided by MapServer. If a template is defined in the
Mapfile the [legend] string is replaced by the processed legend as. See the HTML Legends with MapServer for
details.
[scalebar] Path (relative to document root) of new scalebar image.
[queryfile] Path to the query file (if savequery was set as a CGI Parameter).
[map] Path to the map file (if savemap was set as a CGI Parameter).

Image Geometry

[center] Computed image center in pixels. Useful for setting imgxy form variable when map sizes change.
[center_x], [center_y] Computed image center X or Y coordinate in pixels.
[mapsize], [mapsize_esc] Current image size in cols and rows (separated by spaces).
Also available as escaped version.
[mapwidth], [mapheight] Current image width or height.
[scaledenom] Current image scale. The exact value is not appropriate for user information but essential for some
applications. The value can be rounded e.g. using JavaScript or server side post processing.
[scale] - deprecated Since MapServer 5.0 the proper parameter to use is [scaledenom] instead. The deprecated [scale]
is the current image scale. The exact value is not appropriate for user information but essential for some appli-
cations. The value can be rounded e.g. using JavaScript or server side post processing.
[cellsize] Size of an pixel in the current image in map units. Useful for distance measurement tools in user interfaces.

Map Geometry

[mapx], [mapy] X and Y coordinate of mouse click.

[mapext], [mapext_esc] Full mapextent (separated by spaces).
Also available as escaped version. (mapext_esc is deprecated in MapServer 5.2. You should use the “escape=”
argument instead)
The default template [mapext] returns coordinates in the format of: mixx miny maxx maxy
Available arguments:

5.23. Templating 155

MapServer Documentation, Release 6.0

• escape= Escape the coordinates returned. Default is “none”. Valid values are:
– “url” Use URL escape codes to encode the coordinates returned.
– “none” Do not escape.
• expand= Expand the bounds of the extents by a specific value. Specified in map coordinates. For example,
[mapext] might return:

123456 123456 567890 567890

and [mapext expand=1000] would therefore return:

122456 122456 568890 568890

• format= Format of the coordinates. Default is “$minx $miny $maxx $maxy”. For example, to add
commas to the coordinates you would use:

[mapext format="$minx,$miny,$maxx,$maxy"]

• precision= The number of decimal places to output for coordinates (default is 0).
[minx], [miny], [maxx], [maxy] Minimum / maximum X or Y coordinate of new map extent.
[dx], [dy] The differences of minimum / maximum X or Y coordinate of new map extent.
Useful for creating cachable extents (i.e. 0 0 dx dy) with legends and scalebars
[rawext], [rawext_esc] Raw mapextent, that is the extent before fitting to a window size (separated by spaces). In
cases where input came from imgbox (via Java or whatever) rawext refers to imgbox coordinates transformed
to map units. Useful for spatial query building.
Also available as escaped version. (rawext_esc is deprecated in MapServer 5.2. You should use the “escape=”
argument instead)
The default template [rawext] returns coordinates in the format of: mixx miny maxx maxy
Available arguments:
• escape= Escape the coordinates returned. Default is “none”. Valid values are:
– “url” Use URL escape codes to encode the coordinates returned.
– “none” Do not escape.
• expand= Expand the bounds of the extents by a specific value. Specified in map coordinates. For example,
[rawext] might return:

123456 123456 567890 567890

and [rawext expand=1000] would therefore return:

122456 122456 568890 568890

• format= Format of the coordinates. Default is “$minx $miny $maxx $maxy”. For example, to add
commas to the coordinates you would use:

[rawext format="$minx,$miny,$maxx,$maxy"]

• precision= The number of decimal places to output for coordinates (default is 0).

156 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

[rawminx], [rawminy], [rawmaxx], [rawmaxy] Minimum / maximum X or Y coordinate of a raw map/search ex-
tent.
The following substitutions are only available if the MapServer was compiled with PROJ support and a PROJECTION
is defined in the Mapfile.
[maplon], [maplat] Longitude / latitude value of mouse click. Available only when projection enabled.
[mapext_latlon], [mapext_latlon_esc] Full mapextent (separated by spaces). Available only when projection en-
abled.
Also available as escaped version. (mapext_latlon_esc is deprecated in MapServer 5.2. You should use the
“escape=” argument instead)
The default template [mapext_latlon] returns coordinates in the format of: mixx miny maxx maxy
Available arguments:
• escape= Escape the coordinates returned. Default is “none”. Valid values are:
– “url” Use URL escape codes to encode the coordinates returned.
– “none” Do not escape.
• expand= Expand the bounds of the extents by a specific value. Specified in map coordinates. For example,
[mapext_latlon] might return:

123456 123456 567890 567890

and [mapext_latlon expand=1000] would therefore return:

122456 122456 568890 568890

• format= Format of the coordinates. Default is “$minx $miny $maxx $maxy”. For example, to add
commas to the coordinates you would use:

[mapext_latlon format="$minx,$miny,$maxx,$maxy"]

• precision= The number of decimal places to output for coordinates (default is 0).
[minlon], [minlat], [maxlon] [maxlat] Minimum / maximum longitude or latitude value of mapextent. Available
only when projection enabled.
[refext], [refext_esc] Reference map extent (separated by spaces).
This template has been added with version 4.6 on behalf of an enhancement request. See the thread in the
MapServer ticket#1102 for potential use cases.
Also available as escaped version. (refext_esc is deprecated in MapServer 5.2. You should use the “escape=”
argument instead)
The default template [refext] returns coordinates in the format of: mixx miny maxx maxy
Available arguments:
• escape= Escape the coordinates returned. Default is “none”. Valid values are:
– “url” Use URL escape codes to encode the coordinates returned.
– “none” Do not escape.
• expand=
Expand the bounds of the extents by a specific value. Specified in map coordinates. For example,
[refext] might return:

5.23. Templating 157

MapServer Documentation, Release 6.0

123456 123456 567890 567890

and [refext expand=1000] would therefore return:

122456 122456 568890 568890

• format= Format of the coordinates. Default is “$minx $miny $maxx $maxy”. For example, to add
commas to the coordinates you would use:

[refwext format="$minx,$miny,$maxx,$maxy"]

• precision= The number of decimal places to output for coordinates (default is 0).

Layer

[layers] | [layers_esc] All active layers space delimited. Used for a “POST” request.
Also available as escaped version.
[toggle_layers] | [toggle_layers_esc] List of all layers that can be toggled, i.e. all layers defined in the Mapfile which
status is currently not default.
Also available as escaped version.
[layername_check | select] Used for making layers persistent across a map creation session. String is replaced with
the keyword “checked”, “selected” or “” if layername is on. Layername is the name of a layer as it appears in
the Mapfile. Does not work for default layers.
[layername_meta data key] Layer meta data access (e.g. [streets_build] the underscore is essential).

Zoom

[zoom_minzoom to maxzoom_check|select] Used for making the zoom factor persistent. Zoom values can range
from -25 to 25 by default. The string is replaced with the HTML keyword “checked”, “selected” or “” depending
on the current zoom value.
E.g. if the zoom is 12, a [zoom_12_select] is replaced with “selected”, while a [zoom_13_select] in the same
HTML template file is not.
[zoomdir_-1|0|1_check|select] Used for making the zoom direction persistent. Use check with a radio control or
select with a selection list. See the demo for an example. The string is replaced with the HTML keyword
“checked”, “selected” or “” depending on the current value of zoomdir.

Query

The following substitutions are only available when the template is processed as a result of a query.
[shpext], [shpext_esc] Extent of current shape plus a 5 percent buffer. Available only when processing query results.
The default template [shpext] returns coordinates in the format of: mixx miny maxx maxy
Available arguments:
• escape= Escape the coordinates returned. Default is “none”. Valid values are:
– “url” Use URL escape codes to encode the coordinates returned.
– “none” Do not escape.

158 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

• expand= Expand the bounds of the extents by a specific value. Specified in map coordinates. For example,
[shpext] might return:

123456 123456 567890 567890

and [shpext expand=1000] would therefore return:

122456 122456 568890 568890

• format= Format of the coordinates. Default is “$minx $miny $maxx $maxy”. For example, to add
commas to the coordinates you would use:

[shpext format="$minx,$miny,$maxx,$maxy"]

• precision= The number of decimal places to output for coordinates (default is 0).
[shpminx], [shpminy], [shpmaxx], [shpmaxy] Minimum / maximum X or Y coordinate of shape extent. Available
only when processing query results.
[shpmid] Middle of the extent of current shape. Available only when processing query results.
[shpmidx], [shpmidy] X or Y coordinate of middle of the extent of the current shape. Available only when processing
query results.
[shpidx] Index value of the current shape. Available only when processing query results.
[shpclass] Classindex value of the current shape. Available only when processing query results.
[shpxy formatting options] The list of shape coordinates, with list formatting options, especially useful for SVG.
The default template [shpxy] returns a comma separated list of space delimited of coordinates (i.e. x1 y1, x2 y2,
x3 y3).
Available only when processing query results.
Available attributes (h = header, f=footer, s=separator):
• cs= Coordinate separator, default is comma (“,”).
• xh=, xf= Characters to put before and after the x coordinates (defaults are xh=”” and xf=” “).
• yh= yf= Characters to put before and after the y coordinates (defaults are no characters).
• ph=, pf=, ps= Characters to put before and after and separators between feature parts (e.g. holes, defaults
are no characters).
• sh=, sf= Characters to put before and after a feature (defaults are no
characters)
• precision= The number of decimal places to output for coordinates (default is 0).
• proj= The output projection definition for the coordinates (default is none), a special value of “image”
will convert to image coordinates.
As a simple example:
[shpxy xh="(" yf=")"] will result in: (x1 y1),(x2 y2),(x3 y3)

[tileindex] Index value of the current tile. If no tiles used for the current shape this is replaced by “-1”. Available only
when processing query results.

5.23. Templating 159

MapServer Documentation, Release 6.0

[item formatting options] An attribute table “item”, with list formatting options. The “name” attribute is required.
Available only when processing query results.
Available attributes:
• name = The name of an attribute, case insenstive. (required)
• precision = The number of decimal places to use for numeric data. Use of this will force display as a
number and will lead to unpredicable results with non-numeric data.
• pattern = Regular expression to compare the value of an item against. The tag is output only if there is a
match.
• uc = Set this attribute to “true” to convert the attribute value to upper case.
• lc = Set this attribute to “true” to convert the attribute value to lower case.
• commify = Set this attribute to “true” to add commas to a numeric value. Again, only useful with numeric
data.
• escape = Default escaping is for HTML, but you can escape for inclusion in a URL (=url), or not escape
at all (=none).
• format = A format string used to output the attribute value. The token “$value” is used to place the value
in a more complex presentation. Default is to output only the value.
• nullformat = String to output if the attribute value is NULL, empty or doesn’t match the pattern (if de-
fined). If not set and any of these conditions occur the item tag is replaced with an empty string.
As a simple example:
[item name="area" precision="2" commify="2" format="Area is $value"]

[attribute name],[attrribute name_esc],[attribute item name_raw] Attribute name from the data table of a queried
layer. Only attributes for the active query layers are accessible. Case must be the same as what is stored in the
data file. ArcView, for example, uses all caps for shapefile field names. Available only when processing query
results.
By default the attributes are encoded especially for HTML representation. In addition the escaped version (for
use in URLs) as well as the raw data is available.
[Join name_attribute name],[Join name_attribute name_esc], [Join name_attribute name_raw] One-to-one
joins: First the join name (as specified in the Mapfile has to be given, second the tables fields can be accessed
similar to the layers attribute data. Available only when processing query results.
By default the attributes are encoded especially for HTML representation. In addition the escaped version (for
use in URLs) as well as the raw data is available.
[join_Join name] One-to-many joins: The more complex variant. If the join type is multiple (one-to-many) the
template is replaced by the set of header, template file and footer specified in the Mapfile.
[metadata_meta data key], [metadata_meta data key_esc] Queried layer meta data access (e.g [meta-
data_projection]
Also available as escaped version.
For query modes that allow for multiple result sets, the following string substitutions are available. For FEATURESE-
LECT and FEATURENSELECT modes the totals a re adjusted so as not to include the selection layer. The selection
layer results ARE available for display to the user.
[nr] Total number of results. Useful in web header and footers. Available only when processing query results.
[nl] Number of layers returning results. Useful in web header and footers. Available only when processing query
results.

160 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

[nlr] Total number of results within the current layer. Useful in web header and footers. Available only when pro-
cessing query results.
[rn] Result number within all layers. Starts at 1. Useful in web header and footers. Available only when processing
query results.
[lrn] Result number within the current layer. Starts at 1. Useful in query templates. Available only when processing
query results.
[cl] Current layer name. Useful in layer headers and footers. Available only when processing query results.

5.23.3 Example Template

A small example to give an idea how to work with templates. Note that it covers MapServer specific templates (e.g.
the [map], [mapext]) and user defined templates (e.g. [htmlroot] or [program]) used to store application settings.
1 <!-- MapServer Template -->
2 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
3 "http://www.w3.org/TR/html4/transitional.dtd">
4 <html>
5 <head>
6 <title>MapServer Template Sample</title>
7 </head>
8

9 <body>
10 MapServer Template Sample<br>
11

12 <!-- The central form the application is based on. -->

13 <form method="GET" action="[program]">
14

15 <!-- CGI MapServer applications are server stateless in principle,

16 all information must be "stored" in the client. This includes
17 some basic settings as below.
18 The example is based on the pan and zoom test suite:
19 http://maps.dnr.state.mn.us/mapserver_demos/tests36/ -->
20 <input type="hidden" name="map" value="[map]">
21 <input type="hidden" name="imgext" value="[mapext]">
22 <input type="hidden" name="imgxy" value="149.5 199.5">
23 <input type="hidden" name="program" value="[program]">
24 <input type="hidden" name="htmlroot" value="[htmlroot]">
25 <input type="hidden" name="map_web" value="[map_web]">
26

27 <!-- A table for minimal page formatting. -->

28 <table border=0 cellpadding=5>
29 <tr>
30 <!-- First column: Map and scale bar -->
31 <td align=center>
32 <!-- The map -->
33 <input type="image" name="img" src="[img]"
34 style="border:0;width:300;height:400">
35 <br>
36 <!-- The scale bar-->
37 <img src="[scalebar]" alt="Scale Bar">
38 </td>
39

40 <!-- Second column: Zoom direction, Legend and Reference -->

41 <td valign=top>
42 <!-- Zoom direction -->

5.23. Templating 161

 MapServer Documentation, Release 6.0

43 <b>Map Controls</b><br>
44 Set your zoom option:<br>
45 <select name="zoom" size="1">
46 <option value="2" [ z o o m _ 2 _ s e l e c t ] > Zoom in 2 times
47 <option value="1" [ z o o m _ 1 _ s e l e c t ] > Recenter Map
48 <option value="-2" [ z o o m _ - 2 _ s e l e c t ] > Zoom out 2 times
49 </select>
50 <br>
51

52 <!-- Legend -->

53 <b>Legend</b><br>
54 <img src="[legend]" alt="Legend"><br><br><br><br>
55

56 <!-- Reference map -->

57 <input type="image" name="ref" src="[ref]"
58 style="border:0;width:150;height:150">
59 </td>
60 </tr>
61 </table>
62

63 </form>
64

65 </body>
66 </html>

5.24 Union Layer

Author Tamas Szekeres

Contact szekerest at gmail.com
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Last Updated 2011-04-11

Table of Contents
• Union Layer
– Description
– Requirements
– Mapfile Configuration
– Feature attributes
– Classes and Styles
– Projections
– Examples
* Mapfile Example
* PHP MapScript Example

162 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

5.24.1 Description

Since version 6.0, MapServer has the ability to display features from multiple layers (called ‘source layers’) in a single
mapfile layer. This feature was added through MS RFC 68: Support for combining features from multiple layers.

5.24.2 Requirements

This is a native MapServer option that doesn’t use any external libraries to support it.

5.24.3 Mapfile Configuration

• The CONNECTIONTYPE parameter must be set to UNION.

• The CONNECTION parameter must contain a comma separated list of the source layer names.
• All of the source layers and the union layer must be the same TYPE (e.g. all must be TYPE POINT, or all TYPE
POLYGON etc.)
Note: You may wish to disable the visibility (change their STATUS) of the source layers to avoid displaying the
features twice.
For example:
LAYER
NAME "union-layer"
TYPE POINT
STATUS DEFAULT
CONNECTIONTYPE U N I ON
CONNECTION "layer1,layer2,layer3" # reference to the source layers
PROCESSING "ITEMS=itemname1,itemname2,itemname3"
...
END
LAYER
NAME "layer1"
TYPE POINT
STATUS OFF
CONNECTIONTYPE OGR
CONNECTION ...
...
END
LAYER
NAME "layer2"
TYPE POINT
STATUS OFF
CONNECTIONTYPE S H A P E
CONNECTION ...
...
END
LAYER
NAME "layer3"
TYPE POINT
STATUS OFF
CONNECTIONTYPE S H A P E
CONNECTION ...
...
END

5.24. Union Layer 163

MapServer Documentation, Release 6.0

5.24.4 Feature attributes

In the LAYER definition you may refer to any attributes supported by each of the source layers. In addition to the
source layer attributes the union layer provides the following additional attributes:
1. Combine:SourceLayerName - The name of the source layer the feature belongs to
2. Combine:SourceLayerGroup - The group of the source layer the feature belongs to
During the selection / feature query operations only the ‘Combine:SourceLayerName’ and ‘Com-
bine:SourceLayerGroup’ attributes are provided by default. The set of the provided attributes can manually be
overridden (and further attributes can be exposed) by using the ITEMS processing option (refer to the example above).

5.24.5 Classes and Styles

We can define the symbology and labelling for the union layers in the same way as for any other layer by specifying the
classes and styles. In addition the STYLEITEM AUTO option is also supported for the union layer, which provides to
display the features as specified at the source layers. The source layers may also use the STYLEITEM AUTO setting
if the underlying data source provides that.

5.24.6 Projections

For speed, it is recommended to always use the same projection for the union layer and source layers. However
MapServer will reproject the source layers to the union layer if requested. (for more information on projections in
MapServer refer to PROJECTION)

5.24.7 Examples

Mapfile Example

The follow example contains 3 source layers in different formats, and one layer (yellow) in a different projection.
The union layer uses the STYLEITEM “AUTO” parameter to draw the styles from the source layers. (in this case
MapServer will reproject the yellow features, in EPSG:4326, for the union layer, which is in EPSG:3978).

164 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

MAP
...
PROJECTION
"init=epsg:3978"
END
...
LAYER
NAME ’unioned’
TYPE POLYGON
STATUS DEFAULT
CONNECTIONTYPE U N I ON
CONNECTION "red,green,yellow"
STYLEITEM "AUTO"
# Define an empty class that will be filled at runtime from the color and
# styles read from each source layer.
CLASS
END
PROJECTION
"init=epsg:3978"
END
END

LAYER
NAME ’red’
TYPE POLYGON
STATUS OFF
DATA ’nb.shp’
CLASS
NAME ’red’
STYLE

5.24. Union Layer 165

MapServer Documentation, Release 6.0

OUTLINECOLOR 0 0 0
COLOR 255 85 0
END
END
END

LAYER
NAME ’green’
TYPE POLYGON
STATUS OFF
CONNECTIONTYPE OGR
CONNECTION ’ns.mif’
CLASS
NAME ’green’
STYLE
OUTLINECOLOR 0 0 0
COLOR 90 218 71
END
END
END

LAYER
NAME ’yellow’
TYPE POLYGON
STATUS OFF
CONNECTIONTYPE OGR
CONNECTION ’pei.gml’
CLASS
NAME ’yellow’
STYLE
OUTLINECOLOR 0 0 0
COLOR 255 255 0
END
END
PROJECTION
"init=epsg:4326"
END
END

END # Map

PHP MapScript Example

<?php

// open map
$oMap = ms_newMapObj( "D:/ms4w/apps/osm/map/osm.map" );

// create union layer

$oLayer = ms_newLayerObj($oMap);
$oLayer->set("name", "unioned");
$oLayer->set("type", MS_LAYER_POLYGON);
$oLayer->set("status", MS_ON);
$oLayer->setConnectionType(MS_UNION);
$oLayer->set("connection", "red,green,yellow");
$oLayer->set("styleitem", "AUTO");
$oLayer->setProjection("init=epsg:3978");

166 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

// create empty class

$oClass = ms_newClassObj($oLayer);
...

?>

5.25 Variable Substitution

Syntax: ‘%’ + variable name + ‘%’

Example 1. Connecting securely to a Spatial Database
You want to map some senstitive data held in a PostGIS database. The username and password to be used for the
database connection are held in 2 cookies previously set by a seperate authentication mechanism, “uid” and “passwd”.
CONNECTION "user=%uid% password=%passwd% dbname=postgis"

Example 2. Handling temporary files

You have a user based discovery application that generates shapefiles and stores them in a user’s home directory on
the server. The “username” comes from a cookie, the “filename” comes from a request parameter.
DATA "/home/%username%/tempshp/%filename%"

This feature is only available in the CGI version of MapServer through a mapfile pre-processor. If you are using
MapScript, you will have to code the substitution logic into your application yourself (By writing your own pre-
processor).

5.26 WEB

BROWSEFORMAT [mime-type] Format of the interface output, using MapServer CGI. (added to MapServer 4.8.0)
The default value is “text/html”. Example:
BROWSEFORMAT "image/svg+xml"

EMPTY [url] URL to forward users to if a query fails. If not defined the value for ERROR is used.
ERROR [url] URL to forward users to if an error occurs. Ugly old MapServer error messages will appear if this is
not defined
FOOTER [filename] Template to use AFTER anything else is sent. Multiresult query modes only.
HEADER [filename] Template to use BEFORE everything else has been sent. Multiresult query modes only.
IMAGEPATH [path] Path to the temporary directory fro writing temporary files and images. Must be writable by
the user the web server is running as. Must end with a / or depending on your platform.
IMAGEURL [path] Base URL for IMAGEPATH. This is the URL that will take the web browser to IMAGEPATH
to get the images.
LEGENDFORMAT [mime-type] Format of the legend output, using MapServer CGI. (added to MapServer 4.8.0)
The default value is “text/html”. Example:
LEGENDFORMAT "image/svg+xml"

LOG [filename] Since MapServer 5.0 the recommeded parameters to use for debugging are the MAP object’s CON-
FIG and DEBUG parameters instead (see the Debugging MapServer document).

5.25. Variable Substitution 167

MapServer Documentation, Release 6.0

File to log MapServer activity in. Must be writable by the user the web server is running as. Deprecated since
version 5.0.
MAXSCALEDENOM [double] Minimum scale at which this interface is valid. When a user requests a map at
a smaller scale, MapServer automatically returns the map at this scale. This effectively prevents user from
zooming too far out. Scale is given as the denominator of the actual scale fraction, for example for a map at a
scale of 1:24,000 use 24000. Implemented in MapServer 5.0, to replace the deprecated MAXSCALE parameter.
Deprecated since version 5.0.
MAXSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is MAXSCALEDENOM in-
stead. The deprecated MAXSCALE is the minimum scale at which this interface is valid. When a user requests
a map at a smaller scale, MapServer automatically returns the map at this scale. This effectively prevents user
from zooming too far out. Scale is given as the denominator of the actual scale fraction, for example for a map
at a scale of 1:24,000 use 24000.
See Also:
Map scale
MAXTEMPLATE [file|url] Template to be used if below the minimum scale for the app (the denominator of the
requested scale is larger than MAXSCALEDENOM), useful for nesting apps.
METADATA This keyword allows for arbitrary data to be stored as name value pairs. This is used with OGC WMS
to define things such as layer title. It can also allow more flexibility in creating templates, as anything you put
in here will be accessible via template tags. Example:
METADATA
title "My layer title"
author "Me!"
END

MINSCALEDENOM [double] Maximum scale at which this interface is valid. When a user reqests a map at a larger
scale, MapServer automatically returns the map at this scale. This effectively prevents the user from zooming
in too far. Scale is given as the denominator of the actual scale fraction, for example for a map at a scale of
1:24,000 use 24000. Implemented in MapServer 5.0, to replace the deprecated MINSCALE parameter.
See Also:
Map scale
MINSCALE [double] - deprecated Since MapServer 5.0 the proper parameter to use is MINSCALEDENOM in-
stead. The deprecated MINSCALE is the maximum scale at which this interface is valid. When a user reqests a
map at a larger scale, MapServer automatically returns the map at this scale. This effectively prevents the user
from zooming in too far. Scale is given as the denominator of the actual scale fraction, for example for a map at
a scale of 1:24,000 use 24000. Deprecated since version 5.0.
MINTEMPLATE Template to be used if above the maximum scale for the app (the denominator of the requested
scale is smaller than MINSCALEDENOM), useful for nesting apps.
QUERYFORMAT [mime-type] Format of the query output. (added to MapServer 4.8.0) This works for
mode=query (using query templates in CGI mode), but not for mode=browse. The default value is “text/html”.
Example:
QUERYFORMAT "image/svg+xml"

TEMPLATE [filename|url] Template file or URL to use in presenting the results to the user in an interactive mode
(i.e. map generates map and so on ... )

168 Chapter 5. Mapfile

 MapServer Documentation, Release 6.0

5.27 Notes

• The Mapfile is NOT case-sensitive.

• Strings containing non-alphanumeric characters or a MapServer keyword MUST be quoted. It is recommended
to put ALL strings in double-quotes.
• For MapServer versions < 5, there was a default maximum of 200 layers per mapfile (there is no layer limit with
MapServer >= 5). This can be changed by editing the map.h file to change the value of MS_MAXLAYERS to
the desired number and recompiling. Here are other important default limits when using a MapServer version <
5:
– MAXCLASSES 250 (set in map.h)
– MAXSTYLES 5 (set in map.h)
– MAXSYMBOLS 64 (set in mapsymbol.h)
MapServer versions >= 5 have no limits for classes, styles, symbols, or layers.
• File paths may be given as absolute paths, or as paths relative to the location of the mapfile. In addition, data
files may be specified relative to the SHAPEPATH.
• The mapfile has a hierarchical structure, with the MAP object being the “root”. All other objects fall under this
one.
• Comments are designated with a #.
• Attributes are named using the following syntax: [ATTRIBUTENAME] ... Note that the name of the attribute
included between the square brackets IS CASE SENSITIVE. Generally ESRI generated shapefiles have their
attributes (.dbf column names) all in upper-case for instance, and for PostGIS, ALWAYS use lower-case.
• MapServer Regular Expressions are used through the operating system’s C Library. For information on how to
use and write Regular Expressions on your system, you should read the documentation provided with your C
Library. On Linux, this is GLibC, and you can read “man 7 regex” ... This man page is also available on most
UNIX’s. Since these RegEx’s are POSIX compliant, they should be the same on Windows as well, so windows
users can try searching the web for “man 7 regex” since man pages are available all over the web.

5.27. Notes 169

MapServer Documentation, Release 6.0

170 Chapter 5. Mapfile

 CHAPTER 6

MapScript

Release 6.0
Date May 12, 2011

6.1 Introduction

This is language agnostic documentation for the MapScript interface to MapServer generated by SWIG. This document
is intended for developers and to serve as a reference for writers of more extensive, language specific documentation
located at Mapfile

6.1.1 Appendices

Language-specific extensions are described in the following appendices

Python Appendix

6.1.2 Documentation Elements

Classes will be documented in alphabetical order in the manner outlined below. Attributes and methods will be
formatted as definition lists with the attribute or method as item, the type or return type as classifier, and a concise
description. To make the document as agnostic as possible, we refer to the following types: int, float, and string. There
are yet no mapscript methods that return arrays or sequences or accept array or sequence arguments.
We will use the SWIG term immutable to indicate that an attribute’s value is read-only.

6.1.3 fooObj

A paragraph or two about class fooObj.

fooObj Attributes

attribute [type [access]] Concise description of the attribute.

171
MapServer Documentation, Release 6.0

Attribute name are completely lower case. Multiple words are packed together like outlinecolor.
Note that because of the way that mapscript is generated many confusing, meaningless, and even dangerous attributes
are creeping into objects. See outputFormatObj.refcount for example. Until we get a grip on the structure members
we are exposing to SWIG this problem will continue to grow.

fooObj Methods

method(type mandatory_parameter [, type optional_parameter=default]) [type] Description of the method in-

cluding elaboration on the method arguments, the method’s actions, and returned values. Optional parameters
and their default values are enclosed in brackets.
Class method names are camel case with a leading lower case character like getExpressionString.

6.1.4 Additional Documentation

There’s no point in duplicating the MapServer Mapfile Reference, which remains the primary reference for mapscript
class attributes.

6.2 SWIG MapScript API Reference

Author Sean Gillies

Author Steve Lime
Contact steve.lime at dnr.state.mn.us
Author Frank Warmerdam
Contact warmerdam at pobox.com
Author Umberto Nicoletti
Contact umberto.nicoletti at gmail.com
Author Tamas Szekeres
Contact szekerest at gmail.com
Author Daniel Morissette
Contact dmorisette at mapgears.com
Revision $Revision: 11476 $
Date $Date: 2011-04-06 04:25:40 -0700 (Wed, 06 Apr 2011) $

172 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Contents
• SWIG MapScript API Reference
– Introduction
* Appendices
* Documentation Elements
* fooObj
* Additional Documentation
– MapScript Functions
– MapScript Classes
* classObj
* colorObj
* errorObj
* fontSetObj
* hashTableObj
* imageObj
* intarray
* labelCacheMemberObj
* labelCacheObj
* labelObj
* layerObj
* legendObj
* lineObj
* mapObj
* markerCacheMemberObj
* outputFormatObj
* OWSRequest
* pointObj
* projectionObj
* rectObj
* referenceMapObj
* resultCacheMemberObj
* resultCacheObj
* scalebarObj
* shapefileObj
* shapeObj
* styleObj
* symbolObj
* symbolSetObj
* webObj

6.2.1 Introduction

This is language agnostic documentation for the mapscript interface to MapServer generated by SWIG. This document
is intended for developers and to serve as a reference for writers of more extensive, language specific documentation
in DocBook format for the MDP.

Appendices

Language-specific extensions are described in the following appendices

Python MapScript Appendix

6.2. SWIG MapScript API Reference 173

MapServer Documentation, Release 6.0

Documentation Elements

Classes will be documented in alphabetical order in the manner outlined below. Attributes and methods will be
formatted as definition lists with the attribute or method as item, the type or return type as classifier, and a concise
description. To make the document as agnostic as possible, we refer to the following types: int, float, and string. There
are yet no mapscript methods that return arrays or sequences or accept array or sequence arguments.
We will use the SWIG term immutable to indicate that an attribute’s value is read-only.

fooObj

A paragraph or two about class fooObj.

fooObj Attributes

attribute [type [access]] Concise description of the attribute.

Attribute name are completely lower case. Multiple words are packed together like outlinecolor.
Note that because of the way that mapscript is generated many confusing, meaningless, and even dangerous attributes
are creeping into objects. See outputFormatObj.refcount for example. Until we get a grip on the structure members
we are exposing to SWIG this problem will continue to grow.

fooObj Methods

method(type mandatory_parameter [, type optional_parameter=default]) [type] Description of the method in-

cluding elaboration on the method arguments, the method’s actions, and returned values. Optional parameters
and their default values are enclosed in brackets.
Class method names are camel case with a leading lower case character like getExpressionString.

Additional Documentation

There’s no point in duplicating the MapServer Mapfile Reference, which remains the primary reference for mapscript
class attributes.

6.2.2 MapScript Functions

msCleanup() [void] msCleanup() attempts to recover all dynamically allocated resources allocated by MapServer
code and dependent libraries. It it used primarily for final cleanup in scripts that need to do memory leak testing
to get rid of “noise” one-time allocations. It should not normally be used by production code.
msGetVersion() [string] Returns a string containing MapServer version information, and details on what optional
components are built in. The same report as produced by “mapserv -v”.
msGetVersionInt() [int] Returns the MapServer version number (x.y.z) as an integer (x*10000 + y*100 + z). (New
in v5.0) e.g. V5.4.3 would return 50403.
msResetErrorList() [void] Clears the current error stack.
msIO_installStdoutToBuffer() [void] Installs a mapserver IO handler directing future stdout output to a memory
buffer.

174 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

msIO_installStdinFromBuffer() [void] Installs a mapserver IO handler directing future stdin reading (ie. post re-
quest capture) to come from a buffer.
msIO_resetHandlers() [void] Resets the default stdin and stdout handlers in place of “buffer” based handlers.
msIO_getStdoutBufferString() [string] Fetch the current stdout buffer contents as a string. This method does not
clear the buffer.
msIO_getStdoutBufferBytes() [binary data] Fetch the current stdout buffer contents as a binary buffer. The exact
form of this buffer will vary by mapscript language (eg. string in Python, byte[] array in Java and C#, unhandled
in perl)
msIO_stripStdoutBufferContentType() [string] Strip the Content-type header off the stdout buffer if it has one, and
if a content type is found it is return (otherwise NULL/None/etc).
msIO_stripStdoutBufferContentHeaders(): void Strip all Content-* headers off the stdout buffer if it has ones.

6.2.3 MapScript Classes

classObj

An instance of classObj is associated with with one instance of layerObj.

+-------+ 0..* 1 +-------+
| Class | <--------> | Layer |
+-------+ +-------+

The other important associations for classObj are with styleObj, labelObj, and hashTableObj.
+-------+ 1 0..* +-------+
| Class | ---------> | Style |
+-------+ +-------+

+-------+ 1 0..1 +-------+

| Class | ---------> | Label |
+-------+ +-------+

+-------+ 1 1 +-----------+
| Class | ---------> | HashTable |
+-------+ | -- |
| metadata |
+-----------+

Multiple class styles are now supported in 4.1. See the styleObj section for details on use of multiple class styles.

classObj Attributes

debug [int] MS_TRUE or MS_FALSE

keyimage [string] TODO Not sure what this attribute is for
label [labelObj immutable] Definition of class labeling
layer [layerObj immutable] Reference to the parent layer
maxscale [float] The maximum scale at which class is drawn
metadata [hashTableObj immutable] class metadata hash table.
minscale [float] The minimum scale at which class is drawn

6.2. SWIG MapScript API Reference 175

MapServer Documentation, Release 6.0

name [string] Unique within a layer

numstyles [int] Number of styles for class. In the future, probably the 4.4 release, this attribute will be made im-
mutable.
status [int] MS_ON or MS_OFF. Draw features of this class or do not.
template [string] Template for queries
title [string] Text used for legend labeling
type [int] The layer type of its parent layer

classObj Methods

new classObj( [ layerObj parent_layer=NULL ] ) [classObj] Create a new child classObj instance at the tail (high-
est index) of the class array of the parent_layer. A class can be created outside the context of a parent layer by
omitting the single constructor argument.
clone( ) [classObj] Return an independent copy of the class without a parent layer.
createLegendIcon( mapObj map, layerObj layer, int width, int height ) [imageObj] Draw and return a new leg-
end icon.
drawLegendIcon( mapObj map, layerObj layer, int width, int height, imageObj image, int dstx, int dsty ) [int]
Draw the legend icon onto image at dstx, dsty. Returns MS_SUCCESS or MS_FAILURE.
getExpressionString() [string] Return a string representation of the expression enclosed in the quote characters ap-
propriate to the expression type.
getFirstMetaDataKey() [string] Returns the first key in the metadata hash table. With getNextMetaDataKey(), pro-
vides an opaque iterator over keys.
getMetaData( string key ) [string] Return the value of the classObj metadata at key.
getNextMetaDataKey( string lastkey ) [string] Returns the next key in the metadata hash table or NULL if lastkey
is the last valid key. If lastkey is NULL, returns the first key of the metadata hash table.
Note: getFirstMetaDataKey(), getMetaData(), and getNextMetaDataKey() are deprecated and will be removed in a
future version. Replaced by direct metadata access, see hashTableObj.
getStyle( int index ) [styleObj] Return a reference to the styleObj at index in the styles array.
See the styleObj section for more details on multiple class styles.
getTextString() [string] Return a string representation of the text enclosed in the quote characters appropriate to the
text expression type (logical or simple string).
insertStyle( styleObj style [, int index=-1 ] ) [int] Insert a copy of style into the styles array at index index. Default
is -1, or the end of the array. Returns the index at which the style was inserted.
moveStyleDown( int index ) [int] Swap the styleObj at index with the styleObj index + 1.
moveStyleUp( int index ) [int] Swap the styleObj at index with the styleObj index - 1.
removeStyle( int index ) [styleObj] Remove the styleObj at index from the styles array and return a copy.
setExpression( string expression ) [int] Set expression string where expression is a MapServer regular, logical or
string expression. Returns MS_SUCCESS or MS_FAILUIRE.
setMetaData( string key, string value ) [int] Insert value into the classObj metadata at key. Returns MS_SUCCESS
or MS_FAILURE.

176 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Note: setMetaData() is deprecated and will be removed in a future version. Replaced by direct metadata access, see
hashTableObj.
setText( string text ) [int] Set text string where text is a MapServer text expression. Returns MS_SUCCESS or
MS_FAILUIRE.
Note: Older versions of MapScript (pre-4.8) featured the an undocumented setText() method that required a layerObj
be passed as the first argument. That argument was completely bogus and has been removed.

colorObj

Since the 4.0 release, MapServer colors are instances of colorObj. A colorObj may be a lone object or an attribute of
other objects and have no other associations.

colorObj Attributes

blue [int] Blue component of color in range [0-255]

green [int] Green component of color in range [0-255]
red [int] Red component of color in range [0-255]
pen [int] Don’t mess with this unless you know what you are doing!
Note: Because of the issue with pen, setting colors by individual components is unreliable. Best practice is to use
setRGB(), setHex(), or assign to a new instance of colorObj().

colorObj Methods

new colorObj( [ int red=0, int green=0, int blue=0, int pens=-4 ] ) [colorObj] Create a new instance. The color ar-
guments are optional.
setRGB( int red, int green, int blue ) [int] Set all three RGB components. Returns MS_SUCCESS or
MS_FAILURE.
setHex( string hexcolor ) [int] Set the color to values specified in case-independent hexadecimal notation. Calling
setHex(‘#ffffff’) assigns values of 255 to each color component. Returns MS_SUCCESS or MS_FAILURE.
toHex() [string] Complement to setHex, returning a hexadecimal representation of the color components.

errorObj

This class allows inspection of the MapServer error stack. Only needed for the Perl module as the other language
modules expose the error stack through exceptions.

errorObj Attributes

code [int] MapServer error code such as MS_IMGERR (1).

message [string] Context-dependent error message.
routine [string] MapServer function in which the error was set.

6.2. SWIG MapScript API Reference 177

MapServer Documentation, Release 6.0

errorObj Methods

next [errorObj] Returns the next error in the stack or NULL if the end has been reached.

fontSetObj

A fontSetObj is always a ‘fontset’ attribute of a mapObj.

fontSetObj Attributes

filename [string immutable] Path to the fontset file on disk.

fonts [hashTableObj immutable] Mapping of fonts.
numfonts [int immutable] Number of fonts in set.

fontSetObj Methods

None

hashTableObj

A hashTableObj is a very simple mapping of case-insensitive string keys to single string values. Map, Layer, and Class
metadata have always been hash hables and now these are exposed directly. This is a limited hash that can contain no
more than 41 values.

hashTableObj Attributes

numitems [int immutable] Number of hash items.

hashTableObj Methods

clear( ) [void] Empties the table of all items.

get( string key [, string default=NULL ] ) [string] Returns the value of the item by its key, or default if the key does
not exist.
nextKey( [string key=NULL] ) [string] Returns the name of the next key or NULL if there is no valid next key. If
the input key is NULL, returns the first key.
remove( string key ) [int] Removes the hash item by its key. Returns MS_SUCCESS or MS_FAILURE.
set( string key, string value ) [int] Sets a hash item. Returns MS_SUCCESS or MS_FAILURE.

imageObj

An image object is a wrapper for GD and GDAL images.

178 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

imageObj Attributes

format [outputFormatObj immutable] Image format.

height [int immutable] Image height in pixels.
imagepath [string immutable] If image is drawn by mapObj.draw(), this is the mapObj’s web.imagepath.
imageurl [string immutable] If image is drawn by mapObj.draw(), this is the mapObj’s web.imageurl.
renderer [int] MS_RENDER_WITH_GD, MS_RENDER_WITH_SWF, MS_RENDER_WITH_RAWDATA,
MS_RENDER_WITH_PDF, or MS_RENDER_WITH_IMAGEMAP. Don’t mess with this!
size [int immutable] To access this attribute use the getSize method.
Note: the getSize method is inefficient as it does a call to getBytes and then computes the size of the byte array. The
bytearray is then immediately discarded. In most cases it is more efficient to call getBytes directly.
width [int immutable] Image width in pixels.

imageObj Methods

new imageObj( int width, int height [, outputFormatObj format=NULL [, string filename=NULL ] ] )
[imageObj] Create new instance of imageObj. If filename is specified, an imageObj is created from the
file and any specified width, height, and format parameters will be overridden by values of the image in
filename. Otherwise, if format is specified an imageObj is created using that format. See the format attribute
above for details. If filename is not specified, then width and height should be specified.
getBytes() [binary data] Returns the image contents as a binary buffer. The exact form of this buffer will vary by
mapscript language (eg. string in Python, byte[] array in Java and C#, unhandled in perl)
getSize() [int] Resturns the size of the binary buffer representing the image buffer.
Note: the getSize method is inefficient as it does a call to getBytes and then computes the size of the byte array. The
byte array is then immediately discarded. In most cases it is more efficient to call getBytes directly.
save( string filename [, mapObj parent_map=NULL ] ) [int] Save image to filename. The optional parent_map
parameter must be specified if saving GeoTIFF images.
write( [ FILE file=NULL ] ) [int] Write image data to an open file descriptor or, by default, to stdout. Returns
MS_SUCCESS or MS_FAILURE.
Note: This method is current enabled for Python and C# only. C# supports writing onto a Stream object. User-
contributed typemaps are needed for Perl, Ruby, and Java.
Note: The free() method of imageObj has been deprecated. In MapServer revisions 4+ all instances of imageObj will
be properly disposed of by the interpreter’s garabage collector. If the application can’t wait for garabage collection,
then the instance can simply be deleted or undef’d.

intarray

An intarray is a utility class generated by SWIG useful for manipulating map layer draw-
ing order. See mapObj::getLayersDrawingOrder for discussion of mapscript use and see
http://www.swig.org/Doc1.3/Library.html#Library_nn5 for a complete reference.

intarray Attributes

None

6.2. SWIG MapScript API Reference 179

MapServer Documentation, Release 6.0

intarray Methods

new intarray( int numitems ) [intarray] Returns a new instance of the specified length.

labelCacheMemberObj

An individual feature label. The labelCacheMemberObj class is associated with labelCacheObj.

+------------------+ 0..* 1 +------------+
| LabelCacheMember | <--------- | LabelCache |
+------------------+ +------------+

labelCacheMemberObj Attributes

classindex [int immutable] Index of the class of the labeled feature.

featuresize [float immutable] TODO
label [labelObj immutable] Copied from the class of the labeled feature.
layerindex [int immutable] The index of the layer of the labeled feature.
numstyles [int immutable] Number of styles as for the class of the labeled feature.
point [pointObj immutable] Label point.
poly [shapeObj immutable] Label bounding box.
shapeindex [int immutable] Index within shapefile of the labeled feature.
status [int immutable] Has the label been drawn or not?
styles [styleObj immutable] TODO this should be protected from SWIG.
text [string immutable] Label text.
tileindex [int immutable] Tileindex of the layer of the labeled feature.

labelCacheMemberObj Methods

None.
Note: No real scripting control over labeling currently, but there may be some interesting new possibilities if users
have control over labeling text, position, and status.

labelCacheObj

Set of a map’s cached labels. Has no other existence other than as a ‘labelcache’ attribute of a mapObj. Associated
with labelCacheMemberObj and markerCacheMemberObj.
+------------+ 1 0..* +-------------------+
| LabelCache | ---------> | LabelCacheMember |
+------------+ + ----------------- +
| MarkerCacheMember |
+-------------------+

180 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

labelCacheObj Attributes

cachesize [int immutable] TODO

markercachesize [int immutable] TODO
numlabels [int immutable] Number of label members.
nummarkers [int immutable] Number of marker members.

labelCacheObj Methods

freeCache( ) [void] Free the labelcache.

labelObj

A labelObj is associated with a classObj, a scalebarObj, or a legendObj.

+-------+ 0..1 1 +----------+
| Label | <--------- | Class |
+-------+ | -------- |
| Scalebar |
| -------- |
| Legend |
+----------+

labelObj Attributes

angle [float] TODO

antialias [int] MS_TRUE or MS_FALSE
autoangle [int] MS_TRUE or MS_FALSE
autofollow [int] MS_TRUE or MS_FALSE. Tells mapserver to compute a curved label for appropriate linear features
(see MS RFC 11: Support for Curved Labels for specifics).
autominfeaturesize: int MS_TRUE or MS_FALSE
backgroundcolor [colorObj] Color of background rectangle or billboard.
backgroundshadowcolor [colorObj] Color of background rectangle or billboard shadow.
backgroundshadowsizex [int] Horizontal offset of drop shadow in pixels.
backgroundshadowsizey [int] Vertical offset of drop shadow in pixels.
buffer [int] Maybe this should’ve been named ‘padding’ since that’s what it is: padding in pixels around a label.
color [colorObj] Foreground color.
encoding [string] Supported encoding format to be used for labels. If the format is not supported, the label will not
be drawn. Requires the iconv library (present on most systems). The library is always detected if present on the
system, but if not the label will not be drawn. Required for displaying international characters in MapServer.
More information can be found at: http://www.foss4g.org/FOSS4G/MAPSERVER/mpsnf-i18n-en.html.
font [string] Name of TrueType font.
force [int] MS_TRUE or MS_FALSE.

6.2. SWIG MapScript API Reference 181

MapServer Documentation, Release 6.0

maxsize [int] Maximum height in pixels for scaled labels. See symbolscale attribute of layerObj.
mindistance [int] Minimum distance in pixels between duplicate labels.
minfeaturesize [int] Features of this size of greater will be labeled.
minsize [int] Minimum height in pixels.
offsetx [int] Horizontal offset of label.
offsety [int] Vertical offset of label.
outlinecolor [colorObj] Color of one point outline.
partials [int] MS_TRUE (default) or MS_FALSE. Whether or not labels can flow past the map edges.
position [int] MS_UL, MS_UC, MS_UR, MS_CL, MS_CC, MS_CR, MS_LL, MS_LC, MS_LR, or MS_AUTO.
shadowcolor [colorObj] Color of drop shadow.
shadowsizex [int] Horizontal offset of drop shadow in pixels.
shadowsizey [int] Vertical offset of drop shadow in pixels.
size [int] Annotation height in pixels.
type [int] MS_BITMAP or MS_TRUETYPE.
wrap [string] Character on which legend text will be broken to make multi-line legends.

labelObj Methods

None

layerObj

A layerObj is associated with mapObj. In the most recent revision, an intance of layerObj can exist outside of a
mapObj.
+-------+ 0..* 0..1 +-----+
| Layer | <--------> | Map |
+-------+ +-----+

The other important association for layerObj is with classObj

+-------+ 1 0..* +-------+
| Layer | <--------> | Class |
+-------+ +-------+

and hashTableObj
+-------+ 1 1 +-----------+
| Layer | ---------> | HashTable |
+-------+ | -- |
| metadata |
+-----------+

182 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

layerObj Attributes

bandsitem [string] The attribute from the index file used to select the source raster band(s) to be used. Normally
NULL for default bands processing.
classitem [string] The attribute used to classify layer data.
connection [string] Layer connection or DSN.
connectiontype [int] See MS_CONNECTION_TYPE in mapserver.h for possible values. When setting the connec-
tion type setConnectionType() should be used in order to initialize the layer vtable properly.
data [string] Layer data definition, values depend upon connectiontype.
debug [int] Enable debugging of layer. MS_ON or MS_OFF (default).
dump [int] Switch to allow mapserver to return data in GML format. MS_TRUE or MS_FALSE. Default is
MS_FALSE.
extent [rectObj] optional limiting extent for layer features.
filteritem [string] Attribute defining filter.
footer [string] TODO
group [string] Name of a group of layers.
header [string] TODO
index [int immutable] Index of layer within parent map’s layers array.
labelangleitem [string] Attribute defining label angle.
labelcache [int] MS_ON or MS_OFF. Default is MS_ON.
labelitem [string] Attribute defining feature label text.
labelmaxscale [float] Maximum scale at which layer will be labeled.
labelminscale [float] Minimum scale at which layer will be labeled.
labelrequires [string] Logical expression.
labelsizeitem [string] Attribute defining label size.
map [mapObj immutable] Reference to parent map.
maxfeatures [int] Maximum number of layer features that will be drawn. For shapefile data this means the first N
features where N = maxfeatures.
maxscale [float] Maximum scale at which layer will be drawn.
metadata [hashTableObj immutable] Layer metadata.
minscale [float] Minimum scale at which layer will be drawn.
name [string] Unique identifier for layer.
numclasses [int immutable] Number of layer classes.
numitems [int immutable] Number of layer feature attributes (items).
numjoins [int immutable] Number of layer joins.
numprocessing [int immutable] Number of raster processing directives.
offsite [colorObj] transparent pixel value for raster layers.
postlabelcache [int] MS_TRUE or MS_FALSE. Default is MS_FALSE.

6.2. SWIG MapScript API Reference 183

MapServer Documentation, Release 6.0

requires [string] Logical expression.

sizeunits [int] Units of class size values. MS_INCHES, MS_FEET, MS_MILES, MS_NAUTICALMILES,
MS_METERS, MS_KILOMETERS, MS_DD or MS_PIXELS
status [int] MS_ON, MS_OFF, or MS_DEFAULT.
styleitem [string] Attribute defining styles.
symbolscale [float] Scale at which symbols are default size.
template [string] Template file. Note that for historical reasons, the query attribute must be non-NULL for a layer to
be queryable.
tileindex [string] Layer index file for tiling support.
tileitem [string] Attribute defining tile paths.
tolerance [float] Search buffer for point and line queries.
toleranceunits [int] MS_INCHES, MS_FEET, MS_MILES, MS_NAUTICALMILES, MS_METERS,
MS_KILOMETERS, MS_DD or MS_PIXELS
transform [int] Whether or not layer data is to be transformed to image units. MS_TRUE or MS_FALSE. Default is
MS_TRUE. Case of MS_FALSE is for data that are in image coordinates such as annotation points.
transparency [int] Layer opacity percentage in range [0, 100]. The special value of MS_GD_ALPHA (1000) indi-
cates that the alpha transparency of pixmap symbols should be honored, and should be used only for layers that
use RGBA pixmap symbols.
type [int] See MS_LAYER_TYPE in mapserver.h.
units [int] Units of the layer. See MS_UNITS in mapserver.h.

layerObj Methods

new layerObj( [ mapObj parent_map=NULL ] ) [layerObj] Create a new layerObj in parent_map. The layer index
of the new layerObj will be equal to the parent_map numlayers - 1. The parent_map arg is now optional and
Layers can exist outside of a Map.
addFeature( shapeObj shape ) [int] Add a new inline feature on a layer. Returns -1 on error. TODO: Is this similar
to inline features in a mapfile? Does it work for any kind of layer or connection type?
addProcessing( string directive ) [void] Adds a new processing directive line to a layer, similar to the PROCESSING
directive in a map file. Processing directives supported are specific to the layer type and underlying renderer.
applySLD( string sld, string stylelayer ) [int] Apply the SLD document to the layer object. The matching between
the sld document and the layer will be done using the layer’s name. If a namedlayer argument is passed (argu-
ment is optional), the NamedLayer in the sld that matchs it will be used to style the layer. See SLD HOWTO
for more information on the SLD support.
applySLDURL( string sld, string stylelayer ) [int] Apply the SLD document pointed by the URL to the layer object.
The matching between the sld document and the layer will be done using the layer’s name. If a namedlayer
argument is passed (argument is optional), the NamedLayer in the sld that matchs it will be used to style the
layer. See SLD HOWTO for more information on the SLD support.
clearProcessing() [int] Clears the layer’s raster processing directives. Returns the subsequent number of directives,
which will equal MS_SUCCESS if the directives have been cleared.
clone() [layerObj] Return an independent copy of the layer with no parent map.
close() [void] Close the underlying layer.

184 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Note: demote() is removed in MapServer 4.4

draw( mapObj map, imageObj image ) [int] Renders this layer into the target image, adding labels to the cache if
required. Returns MS_SUCCESS or MS_FAILURE. TODO: Does the map need to be the map on which the
layer is defined? I suspect so.
drawQuery( mapObj map, imageObj image ) : Draw query map for a single layer into the target image. Returns
MS_SUCCESS or MS_FAILURE.
executeWFSGetFeature( layer ) [string] Executes a GetFeature request on a WFS layer and returns the name of the
temporary GML file created. Returns an empty string on error.
generateSLD() [void] Returns an SLD XML string based on all the classes found in the layers.
getClass( int i ) [classObj] Fetch the requested class object. Returns NULL if the class index is out of the legal range.
The numclasses field contains the number of classes available, and the first class is index 0.
getExtent() [rectObj] Fetches the extents of the data in the layer. This normally requires a full read pass through the
features of the layer and does not work for raster layers.
getFeature( int shapeindex [, int tileindex=-1 ] ) [shapeObj] Return the layer feature at shapeindex and tileindex.
getFilterString() [string] Returns the current filter expression.
getFirstMetaDataKey() [string] Returns the first key in the metadata hash table. With getNextMetaDataKey(), pro-
vides an opaque iterator over keys.
getItem( int i ) [string] Returns the requested item. Items are attribute fields, and this method returns the item name
(field name). The numitems field contains the number of items available, and the first item is index zero.
getMetaData( string key ) [string] Return the value at key from the metadata hash table.
getNextMetaDataKey( string lastkey ) [string] Returns the next key in the metadata hash table or NULL if lastkey
is the last valid key. If lastkey is NULL, returns the first key of the metadata hash table.
Note: getFirstMetaDataKey(), getMetaData(), and getNextMetaDataKey() are deprecated and will be removed in a
future version. Replaced by direct metadata access, see hashTableObj.
getNumFeatures() [int] Returns the number of inline features in a layer. TODO: is this really only online features or
will it return the number of non-inline features on a regular layer?
getNumResults() [int] Returns the number of entries in the query result cache for this layer.
getProcessing( int index) [string] Return the raster processing directive at index.
getProjection( ) [string] Returns the PROJ.4 definition of the layer’s projection.
getResult( int i ) [resultCacheMemberObj] Fetches the requested query result cache entry, or NULL if the index is
outside the range of available results. This method would normally only be used after issuing a query operation.
Note: getNumResults() and getResult() are deprecated in MapServer 4.4. Users should instead use the new querying
API described in querying-HOWTO.txt. layerObj::getResults() is the entry point for the new API.
getResults() [resultCacheObj] Returns a reference to layer’s result cache. Should be NULL prior to any query, or
after a failed query or query with no results.
getShape( shapeObj shape, int tileindex, int shapeindex ) [int] Get a shape from layer data.
Note: getShape() is deprecated. Users should adopt getFeature() for new applications.
getWMSFeatureInfoURL( mapObj map, int click_x, int click_y, int feature_count, string info_format )
[string] Return a WMS GetFeatureInfo URL (works only for WMS layers) clickX, clickY is the location of
to query in pixel coordinates with (0,0) at the top left of the image. featureCount is the number of results to
return. infoFormat is the format the format in which the result should be requested. Depends on remote server’s

6.2. SWIG MapScript API Reference 185

MapServer Documentation, Release 6.0

capabilities. MapServer WMS servers support only “MIME” (and should support “GML.1” soon). Returns “”
and outputs a warning if layer is not a WMS layer or if it is not queriable.
insertClass( classObj class [, int index=-1 ] ) [int] Insert a copy of the class into the layer at the requested index.
Default index of -1 means insertion at the end of the array of classes. Returns the index at which the class was
inserted.
isVisible( ) [int] Returns MS_TRUE or MS_FALSE after considering the layer status, minscale, and maxscale within
the context of the parent map.
moveClassDown( int class ) [int] The class specified by the class index will be moved up into the array of layers.
Returns MS_SUCCESS or MS_FAILURE. ex. moveClassDown(1) will have the effect of moving class 1 down
to postion 2, and the class at position 2 will be moved to position 1.
moveClassUp( int class ) [int] The class specified by the class index will be moved up into the array of layers. Re-
turns MS_SUCCESS or MS_FAILURE. ex. moveClassUp(1) will have the effect of moving class 1 up to
postion 0, and the class at position 0 will be moved to position 1.
nextShape( ) [shapeObj] Called after msWhichShapes has been called to actually retrieve shapes within a given area
returns a shape object or MS_FALSE
example of usage :
mapObj map = new mapObj("d:/msapps/gmap-ms40/htdocs/gmap75.map");
layerObj layer = map.getLayerByName(’road’);
int status = layer.open();
status = layer.whichShapes(map.extent);
shapeObj shape;
while ((shape = layer.nextShape()) != null)
{
...
}
layer.close();

open() [void] Opens the underlying layer. This is required before operations like getFeature() will work, but is not
required before a draw or query call.
Note: promote() is eliminated in MapServer 4.4.
queryByAttributes( mapObj map, string qitem, string qstring, int mode ) [int] Query layer for shapes that inter-
sect current map extents. qitem is the item (attribute) on which the query is performed, and qstring is the expres-
sion to match. The query is performed on all the shapes that are part of a CLASS that contains a TEMPLATE
value or that match any class in a layer that contains a LAYER TEMPLATE value.
Note that the layer’s FILTER/FILTERITEM are ignored by this function. Mode is MS_SINGLE or
MS_MULTIPLE depending on number of results you want. Returns MS_SUCCESS if shapes were found
or MS_FAILURE if nothing was found or if some other error happened.
queryByFeatures( mapObj map, int slayer ) [int] Perform a query set based on a previous set of results from an-
other layer. At present the results MUST be based on a polygon layer. Returns MS_SUCCESS if shapes were
found or MS_FAILURE if nothing was found or if some other error happened
queryByIndex( mapObj map, int shapeindex, int tileindex [, int bAddToQuery=MS_FALSE ]) [int] Pop a
query result member into the layer’s result cache. By default clobbers existing cache. Returns MS_SUCCESS
or MS_FAILURE.
queryByPoint( mapObj map, pointObj point, int mode, float buffer ) [int] Query layer at point location specified
in georeferenced map coordinates (i.e. not pixels). The query is performed on all the shapes that are part of a
CLASS that contains a TEMPLATE value or that match any class in a layer that contains a LAYER TEMPLATE
value. Mode is MS_SINGLE or MS_MULTIPLE depending on number of results you want. Passing buffer <=0
defaults to tolerances set in the map file (in pixels) but you can use a constant buffer (specified in ground units)

186 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

instead. Returns MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or if some other
error happened.
queryByRect( mapObj map, rectObj rect ) [int] Query layer using a rectangle specified in georeferenced map co-
ordinates (i.e. not pixels). The query is performed on all the shapes that are part of a CLASS that contains
a TEMPLATE value or that match any class in a layer that contains a LAYER TEMPLATE value. Returns
MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or if some other error happened.
queryByShape( mapObj map, shapeObj shape ) [int] Query layer based on a single shape, the shape has to be a
polygon at this point. Returns MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or
if some other error happened
removeClass( int index ) [classObj] Removes the class indicated and returns a copy, or NULL in the case of a failure.
Note that subsequent classes will be renumbered by this operation. The numclasses field contains the number
of classes available.
removeMetaData( string key ) [int] Delete the metadata hash at key. Returns MS_SUCCESS or MS_FAILURE.
Note: removeMetaData() is deprecated and will be removed in a future version. Replaced by direct metadata access,
see hashTableObj.
setConnectionType(int connectiontype, string library_str) [int] Changes the connectiontype of the layer and recre-
ates the vtable according to the new connection type. This method should be used instead of setting the con-
nectiontype parameter directly. In case when the layer.connectiontype = MS_PLUGIN the library_str parameter
should also be specified so as to select the library to load by mapserver. For the other connection types this
parameter is not used.
setExtent( float minx, float miny, float maxx, float maxy ) [int] Sets the extent of a layer. Returns MS_SUCCESS
or MS_FAILURE.
setFilter( string filter ) [int] Sets a filter expression similarly to the FILTER expression in a map file. Returns
MS_SUCCESS on success or MS_FAILURE if the expression fails to parse.
setMetaData( string key, string value ) [int] Assign value to the metadata hash at key. Return MS_SUCCESS or
MS_FAILURE.
Note: setMetaData() is deprecated and will be removed in a future version. Replaced by direct metadata access, see
hashTableObj.
setProcessingKey( string key, string value ) [void] Adds or replaces a processing directive of the form “key=value”.
Unlike the addProcessing() call, this will replace an existing processing directive for the given key value. Pro-
cessing directives supported are specific to the layer type and underlying renderer.
setProjection( string proj4 ) [int] Set the layer projection using a PROJ.4 format projection definition (ie.
“+proj=utm +zone=11 +datum=WGS84” or “init=EPSG:26911”). Returns MS_SUCCESS or MS_FAILURE.
setWKTProjection( string wkt ) [int] Set the layer projection using OpenGIS Well Known Text format. Returns
MS_SUCCESS or MS_FAILURE.
int whichShapes( rectObj rect ) [int] Performs a spatial, and optionally an attribute based feature search. The func-
tion basically prepares things so that candidate features can be accessed by query or drawing functions (eg using
nextShape function). Returns MS_SUCCESS or MS_FAILURE.
resultsGetShape(int shapeindex [, int tileindex = -1]) [shapeObj] Retrieve shapeObj from a layer’s resultset by in-
dex. Tileindex is optional and is used only for tiled shapefiles, Simply omit or pass tileindex = -1 for other data
sources. Added in MapServer 5.6.0 due to the one-pass query implementation.

legendObj

legendObj is associated with mapObj

6.2. SWIG MapScript API Reference 187

MapServer Documentation, Release 6.0

+--------+ 0..1 1 +-----+

| Legend | <--------> | Map |
+--------+ +-----+

and with labelObj.

+--------+ 1 1 +-------+
| Legend | ---------> | Label |
+--------+ +-------+

legendObj Attributes

height [int] Legend height.

imagecolor [colorObj] Legend background color.
keysizex [int] Width in pixels of legend keys.
keysizey [int] Pixels.
keyspacingx [int] Horizontal padding around keys in pixels.
keyspacingy [int] Vertical padding.
label [labelObj immutable] legend label.
map [mapObj immutable] Reference to parent mapObj.
outlinecolor [colorObj] key outline color.
position [int] MS_UL, MS_UC, MS_UR, MS_LL, MS_LC, or MS_LR.
postlabelcache [int] MS_TRUE or MS_FALSE.
status [int] MS_ON, MS_OFF, or MS_EMBED.
template [string] Path to template file.
width [int] Label width.

legendObj Methods

None

lineObj

A lineObj is composed of one or more pointObj instances.

+------+ 0..1 1..* +-------+
| Line | ---------> | Point |
+------+ +-------+

lineObj Attributes

numpoints [int immutable] Number of points in the line.

188 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

lineObj Methods

add(pointObj point) [int] Add point to the line. Returns MS_SUCCESS or MS_FAILURE.
get(int index) [pointObj] Return reference to point at index.
project(projectionObj proj_in, projectionObj proj_out) [int] Transform line in place from proj_in to proj_out. Re-
turns MS_SUCCESS or MS_FAILURE.
set(int index, pointObj point) [int] Set the point at index to point. Returns MS_SUCCESS or MS_FAILURE.

mapObj

A mapObj is primarily associated with instances of layerObj.

+-----+ 0..1 0..* +-------+
| Map | <--------> | Layer |
+-----+ +-------+

Secondary associations are with legendObj, scalebarObj, referenceMapObj,

+-----+ 1 0..1 +--------------+
| Map | ---------> | Legend |
+-----+ | ------------ |
| Scalebar |
| ------------ |
| ReferenceMap |
+--------------+

outputFormatObj.
+-----+ 1 1..* +--------------+
| Map | ---------> | OutputFormat |
+-----+ +------------- +

mapObj Attributes

cellsize [float] Pixel size in map units.

configoptions [hashObj immutable] A hash table of configuration options from CONFIG keywords in the .map. Di-
rect access to config options is discouraged. Use the setConfigOption() and getConfigOption() methods instead.
datapattern [string] TODO not sure this is meaningful for mapscript.
debug [int] MS_TRUE or MS_FALSE.
extent [rectObj] Map’s spatial extent.
fontset [fontSetObj immutable] The map’s defined fonts.
height [int] Map’s output image height in pixels.
Note: direct setting of height is deprecated in MapServer version 4.4. Users should set width and height simultane-
ously using setSize().
imagecolor [colorObj] Initial map background color.
imagequality [int] JPEG image quality.
Note: map imagequality is deprecated in MapServer 4.4 and should instead be managed through map outputformats.

6.2. SWIG MapScript API Reference 189

MapServer Documentation, Release 6.0

imagetype [string immutable] Name of the current output format.

interlace [int] Output image interlacing.
Note: map interlace is deprecated in MapServer 4.4 and should instead be managed through map outputformats.
lablecache [labelCacheObj immutable] Map’s labelcache.
legend [legendObj immutable] Reference to map’s legend.
mappath [string] Filesystem path of the map’s mapfile.
maxsize [int] TODO ?
name [string] Unique identifier.
numlayers [int immutable] Number of map layers.
numoutputformats [int] Number of output formats.
outputformat [outputFormatObj] The currently selected output format.
Note: Map outputformat should not be modified directly. Use the selectOutputFormat() method to select named
formats.
outputformatlist [outputFormatObj[]] Array of the available output formats.
Note: Currently only available for C#. A proper typemaps should be implemented for the other languages.
querymap [queryMapObj immutable] TODO should this be exposed to mapscript?
reference [referenceMapObj immutable] Reference to reference map.
resolution [float] Nominal DPI resolution. Default is 72.
scale [float] The nominal map scale. A value of 25000 means 1:25000 scale.
scalebar [scalebarObj immutable] Reference to the scale bar.
shapepath [string] Base filesystem path to layer data.
status [int] MS_OFF, MS_ON, or MS_DEFAULT.
symbolset [symbolSetObj immutable] The map’s set of symbols.
templatepattern [string] TODO not sure this is meaningful for mapscript.
transparent [int] MS_TRUE or MS_FALSE.
Note: map transparent is deprecated in MapServer 4.4 and should instead be managed through map outputformats.
units [int] MS_DD, MS_METERS, etc.
web [webObj immutable] Reference to map’s web definitions.
width [int] Map’s output image width in pixels.
Note: direct setting of width is deprecated in MapServer version 4.4. Users should set width and height simultane-
ously using setSize().

mapObj Methods

new mapObj( [ string filename=” ] ) [mapObj] Create a new instance of mapObj. Note that the filename is now
optional.
appendOutputFormat( outputFormatObj format ) [int] Attach format to the map’s output format list. Returns the
updated number of output formats.

190 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

applyConfigOptions( ) [void] Apply the defined configuration options set by setConfigOption().

applySLD( string sldxml ) [int] Parse the SLD XML string sldxml and apply to map layers. Returns MS_SUCCESS
or MS_FAILURE.
applySLDURL( string sldurl ) [int] Fetch SLD XML from the URL sldurl and apply to map layers. Returns
MS_SUCCESS or MS_FAILURE.
clone( ) [mapObj] Returns a independent copy of the map, less any caches.
Note: In the Java module this method is named ‘cloneMap’.
draw( ) [imageObj] Draw the map, processing layers according to their defined order and status. Return an imageObj.
drawLabelCache( imageObj image ) [int] Draw map’s label cache on image. Returns MS_SUCCESS or
MS_FAILURE.
drawLegend( ) [imageObj] Draw map legend, returning an imageObj.
drawQuery( ) [imageObj] Draw query map, returning an imageObj.
drawReferenceMap( ) [imageObj] Draw reference map, returning an imageObj.
drawScalebar( ) [imageObj] Draw scale bar, returning an imageObj.
embedLegend( imageObj image ) [int] Embed map’s legend in image. Returns MS_SUCCESS or MS_FAILURE.
embedScalebar( imageObj image ) [int] Embed map’s scalebar in image. Returns MS_SUCCESS or
MS_FAILURE.
freeQuery( [ int qlayer=-1 ] ) [void] Clear layer query result caches. Default is -1, or all layers.
generateSLD( ) [string] Return SLD XML as a string for map layers.
getConfigOption( string key ) [string] Fetches the value of the requested configuration key if set. Returns NULL if
the key is not set.
getFirstMetaDataKey( ) [string] Returns the first key in the web.metadata hash table. With getNextMetaDataKey( ),
provides an opaque iterator over keys.
getLayer( int index ) [layerObj] Returns a reference to the layer at index.
getLayerByName( string name ) [layerObj] Returns a reference to the named layer.
getLayersDrawingOrder( ) [int*] Returns an array of layer indexes in drawing order.
Note: Unless the proper typemap is implemented for the module’s language a user is more likely to get back an
unuseable SWIG pointer to the integer array.
getMetaData( string key ) [string] Return the value at key from the web.metadata hash table.
getNextMetaDataKey( string lastkey ) [string] Returns the next key in the web.metadata hash table or NULL if
lastkey is the last valid key. If lastkey is NULL, returns the first key of the metadata hash table.
getNumSymbols( ) [int] Return the number of symbols in map.
getOutputFormatByName( string imagetype ) [outputFormatObj] Return the output format corresponding to
driver name imagetype or to format name imagetype. This works exactly the same as the IMAGETYPE di-
rective in a mapfile, is case insensitive and allows an output format to be found either by driver (like ‘GD/PNG’)
or name (like ‘PNG24’).
getProjection( ) [string] Returns the PROJ.4 definition of the map’s projection.
getSymbolByName( string name ) [int] Return the index of the named symbol in the map’s symbolset.

6.2. SWIG MapScript API Reference 191

MapServer Documentation, Release 6.0

Note: This method is poorly named and too indirect. It is preferrable to use the getSymbolByName method of
symbolSetObj, which really does return a symbolObj reference, or use the index method of symbolSetObj to get a
symbol’s index number.
insertLayer( layerObj layer [, int nIndex=-1 ] ) [int] Insert a copy of layer into the Map at index nIndex. The de-
fault value of nIndex is -1, which means the last possible index. Returns the index of the new Layer, or -1 in the
case of a failure.
loadMapContext( string filename [, int useUniqueNames=MS_FALSE ] ) [int] Load an OGC map context file to
define extents and layers of a map.
loadOWSParameters( OWSRequest request [, string version=‘1.1.1’ ] ) [int] Load OWS request parameters
(BBOX, LAYERS, &c.) into map. Returns MS_SUCCESS or MS_FAILURE.
loadQuery( string filename ) [int] Load a saved query. Returns MS_SUCCESS or MS_FAILURE.
moveLayerDown( int layerindex ) [int] Move the layer at layerindex down in the drawing order array, meaning that
it is drawn later. Returns MS_SUCCESS or MS_FAILURE.
moveLayerUp( int layerindex ) [int] Move the layer at layerindex up in the drawing order array, meaning that it is
drawn earlier. Returns MS_SUCCESS or MS_FAILURE.
nextLabel( ) [labelCacheMemberObj] Return the next label from the map’s labelcache, allowing iteration over labels.
Note: nextLabel() is deprecated and will be removed in a future version. Replaced by getLabel().
getLabel( int labelindex ) [labelCacheMemberObj] Return label at specified index from the map’s labelcache.
OWSDispatch( OWSRequest req ) [int] Processes and executes the passed OpenGIS Web Services request on the
map. Returns MS_DONE (2) if there is no valid OWS request in the req object, MS_SUCCESS (0) if an OWS
request was successfully processed and MS_FAILURE (1) if an OWS request was not successfully processed.
OWS requests include WMS, WFS, WCS and SOS requests supported by MapServer. Results of a dispatched
request are written to stdout and can be captured using the msIO services (ie. msIO_installStdoutToBuffer() and
msIO_getStdoutBufferString())
prepareImage( ) [imageObj] Returns an imageObj initialized to map extents and outputformat.
prepareQuery( ) [void] TODO this function only calculates the scale or am I missing something?
processLegendTemplate( string names[], string values[], int numitems ) [string] Process MapServer legend tem-
plate and return HTML.
processQueryTemplate( string names[], string values[], int numitems ) [string] Process MapServer query tem-
plate and return HTML.
processTemplate( int generateimages, string names[], string values[], int numitems ) [string] Process MapServer
template and return HTML.
Note: None of the three template processing methods will be useable unless the proper typemaps are implemented in
the module for the target language. Currently the typemaps are not implemented.
queryByFeatures( int layerindex ) [int] Query map layers, result sets contain features that intersect or are con-
tained within the features in the result set of the MS_LAYER_POLYGON type layer at layerindex. Returns
MS_SUCCESS or MS_FAILURE.
queryByPoint( pointObj point, int mode, float buffer ) [int] Query map layers, result sets contain one or more
features, depending on mode, that intersect point within a tolerance buffer. Returns MS_SUCCESS or
MS_FAILURE.
queryByRect( rectObj rect ) [int] Query map layers, result sets contain features that intersect or are contained within
rect. Returns MS_SUCCESS or MS_FAILURE.
queryByShape( shapeObj shape ) [int] Query map layers, result sets contain features that intersect or are contained
within shape. Returns MS_SUCCESS or MS_FAILURE.

192 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

removeLayer( int index ) [int] Remove the layer at index.

removeMetaData( string key ) [int] Delete the web.metadata hash at key. Returns MS_SUCCESS or
MS_FAILURE.
removeOutputFormat( string name ) [int] Removes the format named name from the map’s output format list. Re-
turns MS_SUCCESS or MS_FAILURE.
save( string filename ) [int] Save map to disk as a new map file. Returns MS_SUCCESS or MS_FAILURE.
saveMapContext( string filename ) [int] Save map definition to disk as OGC-compliant XML. Returns
MS_SUCCESS or MS_FAILURE.
saveQuery( string filename ) [int] Save query to disk. Returns MS_SUCCESS or MS_FAILURE.
saveQueryAsGML( string filename ) [int] Save query to disk. Returns MS_SUCCESS or MS_FAILURE.
selectOutputFormat( string imagetype ) [void] Set the map’s active output format to the internal format named
imagetype. Built-in formats are “PNG”, “PNG24”, “JPEG”, “GIF”, “GTIFF”.
setConfigOption( string key, string value ) [void] Set the indicated key configuration option to the indicated value.
Equivalent to including a CONFIG keyword in a map file.
setExtent( float minx, float miny, float maxx, float maxy ) [int] Set the map extent, returns MS_SUCCESS or
MS_FAILURE.
offsetExtent( float x, float y) [int] Offset the map extent based on the given distances in map coordinates, returns
MS_SUCCESS or MS_FAILURE.
scaleExtent( float zoomfactor, float minscaledenom, float maxscaledenom) [int] Scale the map extent using the
zoomfactor and ensure the extent within the minscaledenom and maxscaledenom domain. If minscalede-
nom and/or maxscaledenom is 0 then the parameter is not taken into account. returns MS_SUCCESS or
MS_FAILURE.
setCenter( pointObj center ) [int] Set the map center to the given map point, returns MS_SUCCESS or
MS_FAILURE.
setFontSet( string filename ) [int] Load fonts defined in filename into map fontset. The existing fontset is cleared.
Returns MS_SUCCESS or MS_FAILURE.
setImageType( string name ) [void] Sets map outputformat to the named format.
Note: setImageType() remains in the module but it’s use is deprecated in favor of selectOutputFormat().
setLayersDrawingOrder( int layerindexes[]) [int] Set map layer drawing order.
Note: Unless the proper typemap is implemented for the module’s language users will not be able to pass arrays or
lists to this method and it will be unusable.
setMetaData( string key, string value ) [int] Assign value to the web.metadata hash at key. Return MS_SUCCESS
or MS_FAILURE.
setOutputFormat( outputFormatObj format ) [void] Sets map outputformat.
setProjection( string proj4 ) [int] Set map projection from PROJ.4 definition string proj4.
setRotation( float rotation_angle ) [int] Set map rotation angle. The map view rectangle (specified in EXTENTS)
will be rotated by the indicated angle in the counter- clockwise direction. Note that this implies the rendered
map will be rotated by the angle in the clockwise direction. Returns MS_SUCCESS or MS_FAILURE.
setSize( int width, int height ) [int] Set map’s image width and height together and carry out the necessary subse-
quent geotransform computation. Returns MS_SUCCESS or MS_FAILURE.
setSymbolSet( string filename ) [int] Load symbols defined in filename into map symbolset. The existing symbolset
is cleared. Returns MS_SUCCESS or MS_FAILURE.

6.2. SWIG MapScript API Reference 193

MapServer Documentation, Release 6.0

setWKTProjection( string wkt ) [int] Sets map projection from OGC definition wkt.
zoomPoint( int zoomfactor, pointObj imgpoint, int width, int height, rectObj extent, rectObj maxextent )
[int] Zoom by zoomfactor to imgpoint in pixel units within the image of height and width dimensions and
georeferenced extent. Zooming can be constrained to a maximum maxextent. Returns MS_SUCCESS or
MS_FAILURE.
zoomRectangle( rectObj imgrect, int width, int height, rectObj extent, rectObj maxextent ) : int Zoom to a pixel
coordinate rectangle in the image of width and height dimensions and georeferencing extent. Zooming can be
constrained to a maximum maxextent. The imgrect rectangle contains the coordinates of the LL and UR coordi-
nates in pixel: the maxy in the rect object should be < miny value. Returns MS_SUCCESS or MS_FAILURE.
------- UR (values in the rect object : maxx, maxy)
| |
| |
| |
------
LL (values in the rectobject minx, miny)

zoomScale( float scale, pointObj imgpoint, int width, int height, rectObj extent, rectObj maxextent) [int] Like
the previous methods, but zooms to the point at a specified scale.

markerCacheMemberObj

An individual marker. The markerCacheMemberObj class is associated with labelCacheObj.

+------------------+ 0..* 1 +------------+
| MarkerCacheMember | <--------- | LabelCache |
+------------------+ +------------+

markerCacheMemberObj Attributes

id [int immutable] Id of the marker.

poly [shapeObj immutable] Marker bounding box.

markerCacheMemberObj Methods

None.

outputFormatObj

An outputFormatObj is associated with a mapObj

+--------------+ 1..* 1 +-----+
| OutputFormat | <--------- | Map |
+--------------+ +-----+

and can also be an attribute of an imageObj.

194 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

outputFormatObj Attributes

bands [int] The number of bands in the raster. Only used for the “raw” modes, MS_IMAGEMODE_BYTE,
MS_IMAGEMODE_INT16, and MS_IMAGEMODE_FLOAT32. Normally set via the BAND_COUNT for-
matoption ... this field should be considered read-only.
driver [string] A string such as ‘GD/PNG’ or ‘GDAL/GTiff’.
extension [string] Format file extension such as ‘png’.
imagemode [int] MS_IMAGEMODE_PC256, MS_IMAGEMODE_RGB, MS_IMAGEMODE_RGBA,
MS_IMAGEMODE_INT16, MS_IMAGEMODE_FLOAT32, MS_IMAGEMODE_BYTE, or
MS_IMAGEMODE_NULL.
mimetype [string] Format mimetype such as ‘image/png’.
name [string] A unique identifier.
renderer [int] MS_RENDER_WITH_GD, MS_RENDER_WITH_SWF, MS_RENDER_WITH_RAWDATA,
MS_RENDER_WITH_PDF, or MS_RENDER_WITH_IMAGEMAP. Normally set internally based on the
driver and some other setting in the constructor.
transparent [int] MS_ON or MS_OFF.

outputFormatObj Methods

new outputFormatObj( string driver [, string name=driver ] ) [outputFormatObj] Create new instance. If name is
not provided, the value of driver is used as a name.
getOption( string key [, string value=”” ] ) [string] Return the format option at key or value if key is not a valid hash
index.
setExtension( string extension ) [void] Set file extension for output format such as ‘png’ or ‘jpg’. Method could
probably be deprecated since the extension attribute is mutable.
setMimetype( string mimetype ) [void] Set mimetype for output format such as ‘image/png’ or ‘image/jpeg’.
Method could probably be deprecated since the mimetype attribute is mutable.
setOption( string key, string value ) [void] Set the format option at key to value. Format options are mostly driver
specific.
validate() [int] Checks some internal consistency issues, and returns MS_TRUE if things are OK and MS_FALSE if
there are problems. Some problems are fixed up internally. May produce debug output if issues encountered.

OWSRequest

Not associated with other mapscript classes. Serves as a message intermediary between an application and
MapServer’s OWS capabilities. Using it permits creation of lightweight WMS services:
wms_map = mapscript.mapObj(’wms.map’)
wms_request = mapscript.OWSRequest()

# Convert application request parameters (req.args)

for param, value in req.args.items():
wms_request.setParam(param, value)

# Map loads parameters from OWSRequest, adjusting its SRS, extents,

# active layers accordingly
wms_map.loadWMSRequest(’1.1.0’, wms_request)

6.2. SWIG MapScript API Reference 195

MapServer Documentation, Release 6.0

# Render the Map

img = wms_map.draw()

OWSRequest Attributes

NumParams [int immutable] Number of request parameters. Eventually should be changed to numparams lowercase
like other attributes.
postrequest [string] TODO
type [int] MS_GET_REQUEST or MS_POST_REQUEST.

OWSRequest Methods

new OWSRequest( ) [OWSRequest] Create a new instance.

setParameter( string name, string value ) [void] Set a request parameter. For example
request.setParameter(’REQUEST’, ’GetMap’)
request.setParameter(’BBOX’, ’-107.0,40.0,-106.0,41.0’)

Note: MapServer’s OWSRequest supports only single valued parameters.

getName( int index ) [string] Return the name of the parameter at index in the request’s array of parameter names.
getValue( int index ) [string] Return the value of the parameter at index in the request’s array of parameter values.
getValueByName( string name) [string] Return the value associated with the parameter name.
loadParams() [int] Initializes the OWSRequest object from the cgi environment variables REQUEST_METHOD,
QUERY_STRING and HTTP_COOKIE. Returns the number of name/value pairs collected. Warning: most
errors will result in a process exit!
loadParamsFromURL( string url ) [int] Initializes the OWSRequest object from the provided URL which is treated
like a QUERY_STRING. Note that REQUEST_METHOD=GET and no post data is assumed in this case. This
method was added in MapServer 6.0.

pointObj

A pointObj instance may be associated with a lineObj.

+-------+ 1..* 0..1 +------+
| Point | <--------- | Line |
+-------+ +------+

pointObj Attributes

m [float] Measure. Meaningful only for measured shapefiles. Given value -2e38 if not otherwise assigned to indicate
“nodata”.
x [float] Easting
y [float] Northing
z [float] Elevation

196 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

pointObj Methods

new pointObj( [ float x=0.0, float y=0.0, float z=0.0, float m=-2e38 ] ) [pointObj] Create new instance. Easting,
northing, and measure arguments are optional.
distanceToPoint( pointObj point ) [float] Returns the distance to point.
distanceToSegment( pointObj point1, pointObj point2 ) [float] Returns the minimum distance to a hypothetical
line segment connecting point1 and point2.
distanceToShape( shapeObj shape ) [float] Returns the minimum distance to shape.
draw( mapObj map, layerObj layer, imageObj image, int classindex, string text ) [int] Draw the point using the
styles defined by the classindex class of layer and labeled with string text. Returns MS_SUCCESS or
MS_FAILURE.
project( projectionObj proj_in, projectionObj proj_out ) [int] Reproject point from proj_in to proj_out. Transfor-
mation is done in place. Returns MS_SUCCESS or MS_FAILURE.
setXY( float x, float y [, float m=2e-38 ] ) [int] Set spatial coordinate and, optionally, measure values simultaneously.
The measure will be set only if the value of m is greater than the ESRI measure no-data value of 1e-38. Returns
MS_SUCCESS or MS_FAILURE.
setXYZ( float x, float y, float z [, float m=-2e38 ] ) [int] Set spatial coordinate and, optionally, measure values si-
multaneously. The measure will be set only if the value of m is greater than the ESRI measure no-data value of
-1e38. Returns MS_SUCCESS or MS_FAILURE.
setXYZM( float x, float y, float z, float m ) [int] Set spatial coordinate and, optionally, measure values simultane-
ously. The measure will be set only if the value of m is greater than the ESRI measure no-data value of -1e38.
Returns MS_SUCCESS or MS_FAILURE.
toString() [string] Return a string formatted like
{ ’x’: %f , ’y’: %f, ’z’: %f }

with the coordinate values substituted appropriately. Python users can get the same effect via the pointObj
__str__ method
>>> p = mapscript.pointObj(1, 1)
>>> str(p)
{ ’x’: 1.000000 , ’y’: 1.000000, ’z’: 1.000000 }

toShape() [shapeObj] Convience method to quickly turn a point into a shapeObj.

projectionObj

This class is not really fully implemented yet. MapServer’s Maps and Layers have Projection attributes, and these
are C projectionObj structures, but are not directly exposed by the mapscript module. Currently we have to do some
round-a-bout logic like this
point.project(projectionObj(mapobj.getProjection(),
projectionObj(layer.getProjection())

to project a point from map to layer reference system.

projectionObj Attributes

numargs [int immutable] Number of PROJ.4 arguments.

6.2. SWIG MapScript API Reference 197

MapServer Documentation, Release 6.0

projectionObj Methods

new projectionObj( string proj4 ) [projectionObj] Create new instance of projectionObj. Input parameter proj4 is a
PROJ.4 definition string such as “init=EPSG:4269”.
getUnits() [int] Returns the units of a projection object. Returns -1 on error.

rectObj

A rectObj may be a lone object or an attribute of another object and has no other associations.

rectObj Attributes

maxx [float] Maximum easting

maxy [float] Maximum northing
minx [float] Minimum easting
miny [float] Minimum northing

rectObj Methods

new rectObj( [ float minx=-1.0, float miny=-1.0, float maxx=-1.0, float maxy=-1.0, int imageunits=MS_FALSE ] )
[rectObj] Create new instance. The four easting and northing arguments are optional and default to -1.0. Note
the new optional fifth argument which allows creation of rectangles in image (pixel/line) units which are also
tested for validity.
draw( mapObj map, layerObj layer, imageObj img, int classindex, string text ) [int] Draw rectangle into img us-
ing style defined by the classindex class of layer. The rectangle is labeled with the string text. Returns
MS_SUCCESS or MS_FAILURE.
getCenter() [pointObj] Return the center point of the rectagle.
project( projectionObj proj_in, projectionObj proj_out ) [int] Reproject rectangle from proj_in to proj_out.
Transformation is done in place. Returns MS_SUCCESS or MS_FAILURE.
toPolygon() [shapeObj] Convert to a polygon of five vertices.
toString() [string] Return a string formatted like
{ ’minx’: %f , ’miny’: %f , ’maxx’: %f , ’maxy’: %f }

with the bounding values substituted appropriately. Python users can get the same effect via the rectObj __str__
method
>>> r = mapscript.rectObj(0, 0, 1, 1)
>>> str(r)
{ ’minx’: 0 , ’miny’: 0 , ’maxx’: 1 , ’maxy’: 1 }

referenceMapObj

A referenceMapObj is associated with mapObj.

198 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

+--------------+ 0..1 1 +-----+

| ReferenceMap | <--------> | Map |
+--------------+ +-----+

referenceMapObj Attributes

color [colorObj] Color of reference box.

extent [rectObj] Spatial extent of reference in units of parent map.
height [int] Height of reference map in pixels.
image [string] Filename of reference map image.
map [mapObj immutable] Reference to parent mapObj.
marker [int] Index of a symbol in the map symbol set to use for marker.
markername [string] Name of a symbol.
markersize [int] Size of marker.
maxboxsize [int] Pixels.
minboxsize [int] Pixels.
outlinecolor [colorObj] Outline color of reference box.
status [int] MS_ON or MS_OFF.
width [int] In pixels.

referenceMapObj Methods

None

resultCacheMemberObj

Has no associations with other MapScript classes and has no methods. By using several indexes, a resultCacheMem-
berObj refers to a single layer feature.

resultCacheMemberObj Attributes

classindex [int immutable] The index of the layer class into which the feature has been classified.
shapeindex [int immutable] Index of the feature within the layer.
tileindex [int immutable] Meaningful for tiled layers only, index of the shapefile data tile.

resultCacheObj

See querying-HOWTO.txt for extra guidance in using the new 4.4 query API.

6.2. SWIG MapScript API Reference 199

MapServer Documentation, Release 6.0

resultCacheObj Attributes

bounds [rectObj immutable] Bounding box of query results.

numresults [int immutable] Length of result set.

resultCacheObj Methods

getResult( int i ) [resultCacheObj] Returns the result at index i, like layerObj::getResult, or NULL if index is outside
the range of results.

scalebarObj

A scalebarObj is associated with mapObj.

+----------+ 0..1 1 +-----+
| Scalebar | <--------- | Map |
+----------+ +-----+

and also with labelObj

+----------+ 1 1 +-------+
| Scalebar | ---------> | Label |
+----------+ +-------+

scalebarObj Attributes

backgroundcolor [colorObj] Scalebar background color.

color [colorObj] Scalebar foreground color.
imagecolor [colorObj] Background color of scalebar.
height [int] Pixels.
intervals [int] Number of intervals.
label [labelObj] Scalebar label.
outlinecolor [colorObj] Foreground outline color.
position [int] MS_UL, MS_UC, MS_UR, MS_LL, MS_LC, or MS_LR.
postlabelcache [int] MS_TRUE or MS_FALSE.
status [int] MS_ON, MS_OFF, or MS_EMBED.
style [int] 0 or 1.
units [int] See MS_UNITS in mapserver.h.
width [int] Pixels.

scalebarObj Methods

None

200 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

shapefileObj

shapefileObj Attributes

bounds [rectObj] Extent of shapes

numshapes [int] Number of shapes
type [int] See mapshape.h for values of type.

shapefileObj Methods

new shapefileObj( string filename [, int type=-1 ] ) [shapefileObj] Create a new instance. Omit the type argument
or use a value of -1 to open an existing shapefile.
add( shapeObj shape ) [int] Add shape to the shapefile. Returns MS_SUCCESS or MS_FAILURE.
get( int i, shapeObj shape ) [int] Get the shapefile feature from index i and store it in shape. Returns MS_SUCCESS
or MS_FAILURE.
getShape( int i ) [shapeObj] Returns the shapefile feature at index i. More effecient than get.
TODO

shapeObj

Each feature of a layer’s data is a shapeObj. Each part of the shape is a closed lineObj.
+-------+ 1 1..* +------+
| Shape | --------> | Line |
+-------+ +------+

shapeObj Attributes

bounds [rectObj] Bounding box of shape.

classindex [int] The class index for features of a classified layer.
index [int] Feature index within the layer.
numlines [int immutable] Number of parts.
numvalues [int immutable] Number of shape attributes.
text [string] Shape annotation.
tileindex [int] Index of tiled file for tileindexed layers.
type [int] MS_SHAPE_POINT, MS_SHAPE_LINE, MS_SHAPE_POLYGON, or MS_SHAPE_NULL.

shapeObj Methods

new shapeObj( int type ) [shapeObj] Return a new shapeObj of the specified type. See the type attribute above. No
attribute values created by default. initValues should be explicitly called to create the required number of values.
add( lineObj line ) [int] Add line (i.e. a part) to the shape. Returns MS_SUCCESS or MS_FAILURE.

6.2. SWIG MapScript API Reference 201

MapServer Documentation, Release 6.0

boundary() [shapeObj] Returns the boundary of the existing shape. Requires GEOS support. Returns NULL/undef
on failure.
buffer( int distance ) [shapeObj] Returns a new buffered shapeObj based on the supplied distance (given in the co-
ordinates of the existing shapeObj). Requires GEOS support. Returns NULL/undef on failure.
contains( pointObj point ) [int] Returns MS_TRUE if the point is inside the shape, MS_FALSE otherwise.
contains( shapeObj shape2 ) [int] Returns MS_TRUE if shape2 is entirely within the shape. Returns -1 on error and
MS_FALSE otherwise. Requires GEOS support.
convexHull() [shapeObj] Returns the convex hull of the existing shape. Requires GEOS support. Returns
NULL/undef on failure.
copy( shapeObj shape_copy ) [int] Copy the shape to shape_copy. Returns MS_SUCCESS or MS_FAILURE.
clone() [shapeObj] Return an independent copy of the shape.
crosses( shapeObj shape2 ) [int] Returns MS_TRUE if shape2 crosses the shape. Returns -1 on error and
MS_FALSE otherwise. Requires GEOS support.
difference( shapeObj shape ) [shapeObj] Returns the computed difference of the supplied and existing shape. Re-
quires GEOS support. Returns NULL/undef on failure.
disjoint( shapeObj shape2 ) [int] Returns MS_TRUE if shape2 and the shape are disjoint. Returns -1 on error and
MS_FALSE otherwise. Requires GEOS support.
distanceToPoint( pointObj point ) [float] Return distance to point.
distanceToShape( shapeObj shape ) [float] Return the minimum distance to shape.
draw( mapObj map, layerObj layer, imageObj img ) [int] Draws the individual shape using layer. Returns
MS_SUCCESS or MS_FAILURE.
equals( shapeObj shape2 ) [int] Returns MS_TRUE if the shape and shape2 are equal (geometry only). Returns -1
on error and MS_FALSE otherwise. Requires GEOS support.
fromWKT( char \*wkt ) [shapeObj] Returns a new shapeObj based on a well-known text representation of a geom-
etry. Requires GEOS support. Returns NULL/undef on failure.
get( int index ) [lineObj] Returns a reference to part at index. Reference is valid only during the life of the shapeObj.
getArea() [double] Returns the area of the shape (if applicable). Requires GEOS support.
getCentroid() [pointObj] Returns the centroid for the existing shape. Requires GEOS support. Returns NULL/undef
on failure.
getLength() [double] Returns the length (or perimeter) of a shape. Requires GEOS support.
getValue( int i ) [string] Return the shape attribute at index i.
initValues( int numvalues ) [void] Allocates memory for the requested number of values.
intersects( shapeObj shape ) [int] Returns MS_TRUE if the two shapes intersect, MS_FALSE otherwise. Note, does
not require GEOS support but will use GEOS functions if available.
intersection( shapeObj shape ) [shapeObj] Returns the computed intersection of the supplied and existing shape.
Requires GEOS support. Returns NULL/undef on failure.
overlaps( shapeObj shape2 ) [int] Returns MS_TRUE if shape2 overlaps shape. Returns -1 on error and MS_FALSE
otherwise. Requires GEOS support.
project( projectionObj proj_in, projectionObj proj_out ) [int] Reproject shape from proj_in to proj_out. Trans-
formation is done in place. Returns MS_SUCCESS or MS_FAILURE.

202 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

setBounds [void] Must be called to calculate new bounding box after new parts have been added.
TODO: should return int and set msSetError.
setValue( int i, string value ) [int] Set the shape value at index i to value.
symDifference( shapeObj shape ) [shapeObj] Returns the computed symmetric difference of the supplied and exist-
ing shape. Requires GEOS support. Returns NULL/undef on failure.
touches( shapeObj shape2 ) [int] Returns MS_TRUE if the shape and shape2 touch. Returns -1 on error and
MS_FALSE otherwise. Requires GEOS support.
toWKT() [string] Returns the well-known text representation of a shapeObj. Requires GEOS support. Returns
NULL/undef on failure.
Union( shapeObj shape ) [shapeObj] Returns the union of the existing and supplied shape. Shapes must be of the
same type. Requires GEOS support. Returns NULL/undef on failure.
within( shapeObj shape2 ) [int] Returns MS_TRUE if the shape is entirely within shape2. Returns -1 on error and
MS_FALSE otherwise. Requires GEOS support.

styleObj

An instance of styleObj is associated with one instance of classObj.

+-------+ 0..* 1 +-------+
| Style | <-------- | Class |
+-------+ +-------+

An instance of styleObj can exist outside of a classObj container and be explicitly inserted into the classObj for use in
mapping.
new_style = new styleObj()
the_class.insertStyle(new_style)

It is important to understand that insertStyle inserts a copy of the styleObj instance, not a reference to the instance
itself.
The older use case
new_style = new styleObj(the_class)

remains supported. These will be the only ways to access the styles of a class. Programmers should no longer directly
access the styles attribute.

styleObj Attributes

angle [double] Angle, given in degrees, to draw the line work. Default is 0. For symbols of Type HATCH, this is the
angle of the hatched lines.
angleitem [string] Attribute/field that stores the angle to be used in rendering. Angle is given in degrees with 0
meaning no rotation.
antialias [int] MS_TRUE or MS_FALSE. Should TrueType fonts be antialiased.
backgroundcolor [colorObj] Background pen color.
color [colorObj] Foreground or fill pen color.

6.2. SWIG MapScript API Reference 203

MapServer Documentation, Release 6.0

mincolor [colorObj] Attribute for Color Range Mapping (MS RFC 6: Color Range Mapping of Continuous Feature
Values). mincolor, minvalue, maxcolor, maxvalue define the range for mapping a continuous feature value to a
continuous range of colors when rendering the feature on the map.
minsize [int] Minimum pen or symbol width for scaling styles.
minvalue [double] Attribute for Color Range Mapping (MS RFC 6: Color Range Mapping of Continuous Feature
Values). mincolor, minvalue, maxcolor, maxvalue define the range for mapping a continuous feature value to a
continuous range of colors when rendering the feature on the map.
minwidth [int] Minimum width of the symbol.
maxcolor [colorObj] Attribute for Color Range Mapping (MS RFC 6: Color Range Mapping of Continuous Feature
Values). mincolor, minvalue, maxcolor, maxvalue define the range for mapping a continuous feature value to a
continuous range of colors when rendering the feature on the map.
maxsize [int] Maximum pen or symbol width for scaling.
maxvalue [double] Attribute for Color Range Mapping (MS RFC 6: Color Range Mapping of Continuous Feature
Values). mincolor, minvalue, maxcolor, maxvalue define the range for mapping a continuous feature value to a
continuous range of colors when rendering the feature on the map.
maxwidth [int] Maximum width of the symbol.
offsetx [int] Draw with pen or symbol offset from map data.
offsety [int] Draw with pen or symbol offset from map data.
outlinecolor [colorObj] Outline pen color.
rangeitem [string] Attribute/field that stores the values for the Color Range Mapping (MS RFC 6: Color Range
Mapping of Continuous Feature Values).
size [int] Pixel width of the style’s pen or symbol.
sizeitem [string] Attribute/field that stores the size to be used in rendering. Value is given in pixels.
symbol [int] The index within the map symbolset of the style’s symbol.
symbolname [string immutable] Name of the style’s symbol.
width [int] Width refers to the thickness of line work drawn, in pixels. Default is 1. For symbols of Type HATCH,
the with is how thick the hatched lines are.

styleObj Methods

new styleObj( [ classObj parent_class ] ) [styleObj] Returns new default style Obj instance. The parent_class is
optional.
clone [styleObj] Returns an independent copy of the style with no parent class.
setSymbolByName(mapObj map, string symbolname) [int] Setting the symbol of the styleObj given the reference
of the map object and the symbol name.

symbolObj

A symbolObj is associated with one symbolSetObj.

+--------+ 0..* 1 +-----------+
| Symbol | <-------- | SymbolSet |
+--------+ +-----------+

204 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

A styleObj will often refer to a symbolObj by name or index, but this is not really an object association, is it?

symbolObj Attributes

antialias [int] MS_TRUE or MS_FALSE.

character [string] For TrueType symbols.
filled [int] MS_TRUE or MS_FALSE.
font [string] For TrueType symbols.
gap [int] Moved to STYLE
imagepath [string] Path to pixmap file.
inmapfile [int] If set to TRUE, the symbol will be saved inside the mapfile. Added in MapServer 5.6.1
linecap [int] Moved to STYLE
linejoin [int] Moved to STYLE
linejoinmaxsize [float] Moved to STYLE
name [string] Symbol name
numpoints [int immutable] Number of points of a vector symbol.
position [int] No more available?
sizex [float] TODO what is this?
sizey [float] TODO what is this?
stylelength [int] Number of intervals
transparent [int] TODO what is this?
transparentcolor [int] TODO is this a derelict attribute?
type [int] MS_SYMBOL_SIMPLE, MS_SYMBOL_VECTOR, MS_SYMBOL_ELLIPSE,
MS_SYMBOL_PIXMAP, or MS_SYMBOL_TRUETYPE.

symbolObj Methods

new symbolObj( string symbolname [, string imagefile ] ) [symbolObj] Create new default symbol named name.
If imagefile is specified, then the symbol will be of type MS_SYMBOL_PIXMAP.
getImage() [imageObj] Returns a pixmap symbol’s imagery as an imageObj.
getPoints() [lineObj] Returns the symbol points as a lineObj.
setImage( imageObj image ) [int] Set a pixmap symbol’s imagery from image.
setPoints( lineObj line ) [int] Sets the symbol points from the points of line. Returns the updated number of points.
setStyle( int index, int value ) [int] Set the style at index to value. Returns MS_SUCCESS or MS_FAILURE.

6.2. SWIG MapScript API Reference 205

MapServer Documentation, Release 6.0

symbolSetObj

A symbolSetObj is an attribute of a mapObj and is associated with instances of symbolObj.

+-----------+ 1 0..* +--------+
| SymbolSet | --------> | Symbol |
+-----------+ +--------+

symbolSetObj Attributes

filename [string] Symbolset filename

numsymbols [int immutable] Number of symbols in the set.

symbolSetObj Methods

new symbolSetObj( [ string symbolfile ] ) [symbolSetObj] Create new instance. If symbolfile is specified, symbols
will be loaded from the file.
appendSymbol( symbolObj symbol ) [int] Add a copy of symbol to the symbolset and return its index.
getSymbol( int index ) [symbolObj] Returns a reference to the symbol at index.
getSymbolByName( string name ) [symbolObj] Returns a reference to the symbol named name.
index( string name ) [int] Return the index of the symbol named name or -1 in the case that no such symbol is found.
removeSymbol( int index ) [symbolObj] Remove the symbol at index and return a copy of the symbol.
save( string filename ) [int] Save symbol set to a file. Returns MS_SUCCESS or MS_FAILURE.

webObj

Has no other existence than as an attribute of a mapObj. Serves as a container for various run-time web application
definitions like temporary file paths, template paths, etc.

webObj Attributes

empty [string] TODO

error [string] TODO
extent [rectObj] Clipping extent.
footer [string] Path to footer document.
header [string] Path to header document.
imagepath [string] Filesystem path to temporary image location.
imageurl [string] URL to temporary image location.
log [string] TODO
map [mapObj immutable] Reference to parent mapObj.
maxscale [float] Maximum map scale.
maxtemplate [string] TODO

206 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

metadata [hashTableObj immutable] metadata hash table.

minscale [float] Minimum map scale.
mintemplate [string] TODO
queryformat [string] TODO
template [string] Path to template document.

webObj Methods

None.

6.3 PHP MapScript

Author Daniel Morissette

Contact dmorissette at mapgears.com
Author Yewondwossen Assefa
Contact yassefa at dmsolutions.ca
Author Alan Boudreault
Contact aboudreault at mapgears.com
Revision $Revision: 11476 $
Date $Date: 2011-04-06 04:25:40 -0700 (Wed, 06 Apr 2011) $
Note: If you are using MapServer 5.6 and older, please refer to the PHP MapScript 5.6 documen-
tation instead.
Note: If you are migrating your existing application that is based on MapServer 5.6 or older, to
MapServer 6.0 or beyond, please read the PHP MapScript Migration Guide for important changes.

6.3. PHP MapScript 207

MapServer Documentation, Release 6.0

Contents
• PHP MapScript
– Introduction
– Versions Supported
– How to Get More Information on PHP MapScript
– Important Note
– Constants
– Functions
– Classes
* classObj
* colorObj
* errorObj
* gridObj
* hashTableObj
* imageObj
* labelcacheMemberObj
* labelcacheObj
* labelObj
* layerObj
* legendObj
* lineObj
* mapObj
* outputformatObj
* OwsrequestObj
* pointObj
* projectionObj
* querymapObj
* rectObj
* referenceMapObj
* resultObj
* scalebarObj
* shapefileObj
* shapeObj
* styleObj
* symbolObj
* webObj
* clusterObj
• Memory Management

6.3.1 Introduction

This is a PHP module that makes MapServer’s MapScript functionalities available in a PHP Dynamically Loadable
Library. In simple terms, this module will allow you to use the powerful PHP scripting language to dynamically create
and modify map images in MapServer.

6.3.2 Versions Supported

PHP 5.2.0 or more recent is required; since MapServer 6.0, support for PHP 4, PHP 5.0 and PHP 5.1 have been
dropped. PHP MapScript was originally developed for PHP 3.0.14, and after MapServer 3.5 support for PHP 3 was
dropped.

208 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

The module has been tested and used on Linux, Solaris, *BSD, and Windows.

6.3.3 How to Get More Information on PHP MapScript

• For installation questions regarding the PHP MapScript module, see PHP MapScript Installation.
• The MapServer Wiki has information on this module, that was contributed by users.
• New PHP MapScript users should read the php_example document.
• The project’s home is the PHP/MapScript page on MapTools.org.
• Also, see the MapScript, and the Mapfile sections of this site.
• Refer to the main PHP site for their official documentation.

6.3.4 Important Note

• Constant names and class member variable names are case-sensitive in PHP.

6.3.5 Constants

The following MapServer constants are available:

Boolean values MS_TRUE, MS_FALSE, MS_ON, MS_OFF, MS_YES, MS_NO
Map units MS_INCHES, MS_FEET, MS_MILES, MS_METERS, MS_KILOMETERS, MS_DD, MS_PIXELS,
MS_NAUTICALMILES
Layer types MS_LAYER_POINT, MS_LAYER_LINE, MS_LAYER_POLYGON, MS_LAYER_RASTER,
MS_LAYER_ANNOTATION, MS_LAYER_QUERY, MS_LAYER_CIRCLE, MS_LAYER_TILEINDEX,
MS_LAYER_CHART
Layer/Legend/Scalebar/Class Status MS_ON, MS_OFF, MS_DEFAULT, MS_EMBED, MS_DELETE
Layer alpha transparency allows alpha transparent pixmaps to be used with RGB map images MS_GD_ALPHA
Font types MS_TRUETYPE, MS_BITMAP
Label positions MS_UL, MS_LR, MS_UR, MS_LL, MS_CR, MS_CL, MS_UC, MS_LC, MS_CC, MS_XY,
MS_AUTO, MS_AUTO2, MS_FOLLOW, MS_NONE
Bitmap font styles MS_TINY , MS_SMALL, MS_MEDIUM, MS_LARGE, MS_GIANT
Shape types MS_SHAPE_POINT, MS_SHAPE_LINE, MS_SHAPE_POLYGON, MS_SHAPE_NULL
Shapefile types MS_SHP_POINT, MS_SHP_ARC, MS_SHP_POLYGON, MS_SHP_MULTIPOINT
Query/join types MS_SINGLE, MS_MULTIPLE
Querymap styles MS_NORMAL, MS_HILITE, MS_SELECTED
Connection Types MS_INLINE, MS_SHAPEFILE, MS_TILED_SHAPEFILE, MS_SDE, MS_OGR,
MS_TILED_OGR, MS_POSTGIS, MS_WMS, MS_ORACLESPATIAL, MS_WFS, MS_GRATICULE,
MS_RASTER, MS_PLUGIN, MS_UNION
Error codes MS_NOERR, MS_IOERR, MS_MEMERR, MS_TYPEERR, MS_SYMERR, MS_REGEXERR,
MS_TTFERR, MS_DBFERR, MS_GDERR, MS_IDENTERR, MS_EOFERR, MS_PROJERR,
MS_MISCERR, MS_CGIERR, MS_WEBERR, MS_IMGERR, MS_HASHERR, MS_JOINERR,
MS_NOTFOUND, MS_SHPERR, MS_PARSEERR, MS_SDEERR, MS_OGRERR, MS_QUERYERR,

6.3. PHP MapScript 209

MapServer Documentation, Release 6.0

MS_WMSERR, MS_WMSCONNERR, MS_ORACLESPATIALERR, MS_WFSERR, MS_WFSCONNERR,

MS_MAPCONTEXTERR, MS_HTTPERR, MS_WCSERR
Symbol types MS_SYMBOL_SIMPLE, MS_SYMBOL_VECTOR, MS_SYMBOL_ELLIPSE,
MS_SYMBOL_PIXMAP, MS_SYMBOL_TRUETYPE
Image Mode types (outputFormatObj) MS_IMAGEMODE_PC256, MS_IMAGEMODE_RGB,
MS_IMAGEMODE_RGBA, MS_IMAGEMODE_INT16, MS_IMAGEMODE_FLOAT32,
MS_IMAGEMODE_BYTE, MS_IMAGEMODE_FEATURE, MS_IMAGEMODE_NULL
Style/Attribue binding MS_STYLE_BINDING_SIZE, MS_STYLE_BINDING_ANGLE,
MS_STYLE_BINDING_COLOR, MS_STYLE_BINDING_OUTLINECOLOR,
MS_STYLE_BINDING_SYMBOL
Label/Attribute binding MS_LABEL_BINDING_SIZE, MS_LABEL_BINDING_ANGLE,
MS_LABEL_BINDING_COLOR, MS_LABEL_BINDING_OUTLINECOLOR,
MS_LABEL_BINDING_FONT, MS_LABEL_BINDING_PRIORITY, MS_LABEL_BINDING_POSITION,
MS_LABEL_BINDING_SHADOWSIZEX, MS_LABEL_BINDING_SHADOWSIZEY
Alignment MS_ALIGN_LEFT, MS_ALIGN_CENTER, MS_ALIGN_RIGHT
OwsRequest MS_GET_REQUEST, MS_POST_REQUEST

6.3.6 Functions

string ms_GetVersion() Returns the MapServer version and options in a string. This string can be parsed to find out
which modules were compiled in, etc.
int ms_GetVersionInt() Returns the MapServer version number (x.y.z) as an integer (x*10000 + y*100 + z). (New
in v5.0) e.g. V5.4.3 would return 50403.
array ms_TokenizeMap(string map_file_name) Preparses a mapfile through the MapServer parser and return an ar-
ray with one item for each token from the mapfile. Strings, logical expressions, regex expressions and comments
are returned as individual tokens.
void ms_ioinstallstdouttobuffer() Installs a mapserver IO handler directing future stdout output to a memory buffer.
void ms_ioinstallstdinfrombuffer() Installs a mapserver IO handler directing future stdin reading (ie. post request
capture) to come from a buffer.
void ms_iogetstdoutbufferstring() Fetch the current stdout buffer contents as a string. This method does not clear
the buffer.
int ms_iogetStdoutBufferBytes() Writes the current buffer to stdout. The PHP header() function should be used to
set the documents’s content-type prior to calling the function. Returns the number of bytes written if output is
sent to stdout. See MapScript Wrappers for WxS Services for more info.
void ms_ioresethandlers() Resets the default stdin and stdout handlers in place of “buffer” based handlers.
void ms_iostripstdoutbuffercontenttype() Strip the Content-type header off the stdout buffer if it has one, and if a
content type is found it is return. Otherwise return false.

6.3.7 Classes

The following class objects are available through PHP MapScript.

210 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

classObj

Constructor

Class Objects can be returned by the layerObj class, or can be created using:
new classObj(layerObj layer [, classObj class])

or using the old constructor

classObj ms_newClassObj(layerObj layer [, classObj class])

The second argument class is optional. If given, the new class created will be a copy of this class.

Members

Type Name
string name
string title
int type
int status (MS_ON, MS_OFF or MS_DELETE)
double minscaledenom
double maxscaledenom
string template
labelObj label
int numstyles (read-only)
string keyimage
string group
hashTableObj metadata

Methods

int updateFromString(string snippet) Update a class from a string snippet. Returns

MS_SUCCESS/MS_FAILURE.
$oClass->updateFromString(’CLASS STYLE COLOR 255 0 255 END END’); /*set the color */

int set(string property_name, new_value) Set object property to a new value.

int setExpression(string expression) Set the expression string for the class object.
string getExpressionString() Returns the expression string for the class object.
int settext(string text) Set the text string for the class object.
string getTextString() Returns the text string for the class object.
int drawLegendIcon(int width, int height, imageObj im, int dstX, int dstY) Draw the legend icon on im object at
dstX, dstY. Returns MS_SUCCESS/MS_FAILURE.
imageObj createLegendIcon(int width, int height) Draw the legend icon and return a new imageObj.
styleObj getStyle(int index) Return the style object using an index. index >= 0 && index < class->numstyles.
int movestyleup(int index) The style specified by the style index will be moved up into the array of classes. Returns
MS_SUCCESS or MS_FAILURE. ex class->movestyleup(1) will have the effect of moving style 1 up to position
0, and the style at position 0 will be moved to position 1.

6.3. PHP MapScript 211

MapServer Documentation, Release 6.0

int movestyledown(int index) The style specified by the style index will be moved down into the array of classes.
Returns MS_SUCCESS or MS_FAILURE. ex class->movestyledown(0) will have the effect of moving style 0
up to position 1, and the style at position 1 will be moved to position 0.
int deletestyle(int index) Delete the style specified by the style index. If there are any style that follow the deleted
style, their index will decrease by 1.
Note: if you are using the numstyles parameter while using the deletestyle function on the class object you need to
refetch a new class object.
Example :
//class has 2 styles
$class = $oLayer->getclass(0);
$class->deletestyle(1);
echo $class->numstyles; : will echo 2

$class = $oLayer->getclass(0);
echo $class->numstyles; : will echo 1

int getMetaData(string name) Fetch class metadata entry by name. Returns “” if no entry matches the name. Note
that the search is case sensitive.
Note: getMetaData’s query is case sensitive.
int setMetaData(string name, string value) Set a metadata entry for the class. Returns
MS_SUCCESS/MS_FAILURE.
int removeMetaData(string name) Remove a metadata entry for the class. Returns MS_SUCCESS/MS_FAILURE.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

colorObj

Constructor

Instances of colorObj are always embedded inside other classes.

Members

Type Name
int red
int green
int blue

Methods

void setRGB(int red, int green, int blue) Set red, green, blue values.

212 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

errorObj

Instances of errorObj are created internally by MapServer as errors happen. Errors are managed as a chained list with
the first item being the most recent error. The head of the list can be fetched using ms_GetErrorObj(), and the list can
be cleared using ms_ResetErrorList()

Functions

errorObj ms_GetErrorObj() Returns a reference to the head of the list of errorObj.

void ms_ResetErrorList() Clear the current error list. Note that clearing the list invalidates any errorObj handles
obtained via the $error->next() method.

Members

Type Name
int code //See error code constants above
string routine
string message

Method

errorObj next() Returns the next errorObj in the list, or NULL if we reached the end of the list.
Example: This example draws a map and reports all errors generated during the draw() call, errors can potentially
come from multiple layers.
ms_ResetErrorList();
$img = $map->draw();
$error = ms_GetErrorObj();
while($error && $error->code != MS_NOERR)
{
printf("Error in %s: %s<br>\n", $error->routine, $error->message);
$error = $error->next();
}

gridObj

Constructor

The grid is always embedded inside a layer object defined as a grid (layer->connectiontype = MS_GRATICULE) (for
more docs : http://trac.osgeo.org/mapserver/wiki/MapServerGrid)
A layer can become a grid layer by adding a grid object to it using : ms_newGridObj(layerObj layer)
$oLayer = ms_newlayerobj($oMap);
$oLayer->set("name", "GRID");
ms_newgridobj($oLayer);
$oLayer->grid->set("labelformat", "DDMMSS");

6.3. PHP MapScript 213

MapServer Documentation, Release 6.0

Members

Type Name
double minsubdivide
double maxsubdivide
double minarcs
double maxacrs
double mininterval
double maxinterval
string labelformat

Methods

int set(string property_name, new_value) Set object property to a new value.

hashTableObj

Constructor

Instance of hashTableObj is always embedded inside the classObj, layerObj, mapObj and webObj. It is uses a read
only.
$hashTable = $oLayer->metadata;
$key = null;
while ($key = $hashTable->nextkey($key))
echo "Key: ".$key." value: ".$hashTable->get($key)."<br/>";

Methods

string get(string key) Fetch class metadata entry by name. Returns “” if no entry matches the name. Note that the
search is case sensitive.
int set(string key, string value) Set a metadata entry in the hashTable. Returns MS_SUCCESS/MS_FAILURE.
int remove(string key) Remove a metadata entry in the hashTable. Returns MS_SUCCESS/MS_FAILURE.
void clear() Clear all items in the hashTable (To NULL).
string nextkey(string previousKey) Return the next key or first key if previousKey = NULL. Return NULL if no
item is in the hashTable or end of hashTable is reached

imageObj

Constructor

Instances of imageObj are always created by the mapObj class methods.

214 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Members

Type Name
int width (read-only)
int height (read-only)
int resolution (read-only)
int resolutionfactor (read-only)
string imagepath
string imageurl

Methods

int saveImage([string filename, MapObj oMap]) Writes image object to specified filename. Passing no filename
or an empty filename sends output to stdout. In this case, the PHP header() function should be used to set the
document’s content-type prior to calling saveImage(). The output format is the one that is currently selected
in the map file. The second argument oMap is not manadatory. It is usful when saving to formats like GTIFF
that needs georeference informations contained in the map file. On success, it returns either MS_SUCCESS if
writing to an external file, or the number of bytes written if output is sent to stdout.
string saveWebImage() Writes image to temp directory. Returns image URL. The output format is the one that is
currently selected in the map file.
void pasteImage(imageObj srcImg, int transparentColorHex [[, int dstX, int dstY], int angle]) Copy srcImg on
top of the current imageObj. transparentColorHex is the color (in 0xrrggbb format) from srcImg that should
be considered transparent (i.e. those pixels won’t be copied). Pass -1 if you don’t want any transparent color.
If optional dstx,dsty are provided then it defines the position where the image should be copied (dstx,dsty =
top-left corner position). The optional angle is a value between 0 and 360 degrees to rotate the source image
counterclockwise. Note that if an angle is specified (even if its value is zero) then the dstx and dsty coordinates
specify the CENTER of the destination area. Note: this function works only with 8 bits GD images (PNG or
GIF).

labelcacheMemberObj

Accessible only through the mapObj (map->getLabel()).

Members

Type Name
int classindex (Read-Only)
int featuresize (Read-Only)
int layerindex (Read-Only)
int numstyles (Read-Only)
int shapeindex (Read-Only)
int status (Read-Only)
int markerid (Read-Only)
string text (Read-Only)
int tileindex (Read-Only)

6.3. PHP MapScript 215

MapServer Documentation, Release 6.0

Method

None

labelcacheObj

Accessible only through the mapObj (map->labelcache). This object is only used to give the possiblity to free the label
cache (map->labelcache->freeCache())

Method

boolean freeCache() Free the label cache. Always returns MS_SUCCESS. Ex : (map->labelcache->freeCache();

labelObj

Constructor

labelObj are always embedded inside other classes.

Members

Type Name
string font
int type
colorObj color
colorObj outlinecolor
int outlinewidth
colorObj shadowcolor
int shadowsizex
int shadowsizey
colorObj backgroundcolor
colorObj backgroundshadowcolor
int backgroundshadowsizex
int backgroundshadowsizey
int size
int minsize
int maxsize
int position
int offsetx
int offsety
double angle
int autoangle
int autofollow
int buffer
int antialias
int wrap
int minfeaturesize
int autominfeaturesize
Continued on next page

216 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Table 6.1 – continued from previous page

int mindistance
int repeatdistance
int partials
int force
string encoding
int align
int maxlength
int minlength
int priority

Methods

int updateFromString(string snippet) Update a label from a string snippet. Returns

MS_SUCCESS/MS_FAILURE.
int set(string property_name, new_value) Set object property to a new value.
int setBinding(const labelbinding, string value) Set the attribute binding for a specified label property.
Example:
$oLabel->setbinding(MS_LABEL_BINDING_COLOR, "FIELD_NAME_COLOR");

This would bind the color parameter with the data (ie will extract the value of the color from the field called
“FIELD_NAME_COLOR”
string getBinding(const labelbinding) Get the attribute binding for a specified label property. Returns NULL if there
is no binding for this property.
Example:
$oLabel->setbinding(MS_LABEL_BINDING_COLOR, "FIELD_NAME_COLOR");
echo $oLabel->getbinding(MS_LABEL_BINDING_COLOR); // FIELD_NAME_COLOR

int removeBinding(const labelbinding) Remove the attribute binding for a specfiled style property.
Example:
$oStyle->removebinding(MS_LABEL_BINDING_COLOR);

void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

layerObj

Constructor

Layer Objects can be returned by the mapObj class, or can be created using:
layerObj ms_newLayerObj(MapObj map [, layerObj layer])

A second optional argument can be given to ms_newLayerObj() to create the new layer as a copy of an existing layer.
If a layer is given as argument then all members of a this layer will be copied in the new layer created.

6.3. PHP MapScript 217

MapServer Documentation, Release 6.0

Members

218 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Type Name
int numclasses (read-only)
int index (read-only)
int status (MS_ON, MS_OFF, MS_DEFAULT or MS_DELETE)
int debug
string classitem
string classgroup
string name
string group
string data
int type
int dump
double tolerance
int toleranceunits
int sizeunits
double symbolscaledenom
double minscaledenom
double maxscaledenom
double labelminscaledenom
double labelmaxscaledenom
int maxfeatures
int startindex
colorObj offsite
int annotate
int transform
int labelcache
int postlabelcache
string labelitem
string tileitem
string tileindex
string header
string footer
string connection
int connectiontype (read-only, use setConnectionType() to set it)
string filteritem
string template
int opacity
string styleitem
gridObj grid //only available on a layer defined as grid (MS_GRATICULE)
int num_processing
string requires
string labelrequires
hashTableObj metadata
hashTableObj bindvals
projectionObj projection
clusterObj cluster

Methods

int updateFromString(string snippet) Update a layer from a string snippet. Returns

MS_SUCCESS/MS_FAILURE.

6.3. PHP MapScript 219

MapServer Documentation, Release 6.0

$oLayer->updateFromString(’LAYER NAME land_fn2 END’); /*modify the name */

$oLayer->updateFromString(’LAYER CLASS STYLE COLOR 255 255 0 END END END’); /*add a new class*/

int set(string property_name, new_value) Set object property to a new value.

int draw(imageObj image) Draw a single layer, add labels to cache if required. Returns MS_SUCCESS or
MS_FAILURE on error.
int drawQuery(imageObj image) Draw query map for a single layer.
classObj getClass(int classIndex) Returns a classObj from the layer given an index value (0=first class)
int getClassIndex(shape, scaledenom [, classgroup, numclasses]) Get the class index of a shape for a given scale.
Returns -1 if no class matches. classgroup is an array of class ids to check (Optionnal). numclasses is the
number of classes that the classgroup array contains. By default, all the layer classes will be checked.
int queryByPoint(pointObj point, int mode, double buffer) Query layer at point location specified in georefer-
enced map coordinates (i.e. not pixels). The query is performed on all the shapes that are part of a CLASS
that contains a TEMPLATE value or that match any class in a layer that contains a LAYER TEMPLATE value.
Mode is MS_SINGLE or MS_MULTIPLE depending on number of results you want. Passing buffer -1 defaults
to tolerances set in the map file (in pixels) but you can use a constant buffer (specified in ground units) instead.
Returns MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or if some other error
happened (note that the error message in case nothing was found can be avoided in PHP using the ‘@’ control
operator).
int queryByRect(rectObj rect) Query layer using a rectangle specified in georeferenced map coordinates (i.e. not
pixels). The query is performed on all the shapes that are part of a CLASS that contains a TEMPLATE value
or that match any class in a layer that contains a LAYER TEMPLATE value. Returns MS_SUCCESS if shapes
were found or MS_FAILURE if nothing was found or if some other error happened (note that the error message
in case nothing was found can be avoided in PHP using the ‘@’ control operator).
int queryByShape(shapeObj shape) Query layer based on a single shape, the shape has to be a polygon at this
point. Returns MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or if some other
error happened (note that the error message in case nothing was found can be avoided in PHP using the ‘@’
control operator).
int queryByFeatures(int slayer) Perform a query set based on a previous set of results from another layer. At present
the results MUST be based on a polygon layer. Returns MS_SUCCESS if shapes were found or MS_FAILURE
if nothing was found or if some other error happened (note that the error message in case nothing was found can
be avoided in PHP using the ‘@’ control operator).
int queryByAttributes(string qitem, string qstring, int mode) Query layer for shapes that intersect current map ex-
tents. qitem is the item (attribute) on which the query is performed, and qstring is the expression to match. The
query is performed on all the shapes that are part of a CLASS that contains a TEMPLATE value or that match
any class in a layer that contains a LAYER TEMPLATE value. Note that the layer’s FILTER/FILTERITEM are
ignored by this function. Mode is MS_SINGLE or MS_MULTIPLE depending on number of results you want.
Returns MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or if some other error
happened (note that the error message in case nothing was found can be avoided in PHP using the ‘@’ control
operator).
int setFilter(string expression) Set layer filter expression.
string getFilterString() Returns the expression for this layer or NULL on error.
string getProjection() Returns a string representation of the projection. Returns NULL on error or if no projection is
set.
int setProjection(string proj_params) Set layer projection and coordinate system. Parameters are given as a single
string of comma-delimited PROJ.4 parameters. Returns MS_SUCCESS or MS_FAILURE on error.

220 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

int setWKTProjection(string proj_params) Same as setProjection(), but takes an OGC WKT projection definition
string as input.
Note: setWKTProjection requires GDAL support
int getNumResults() Returns the number of results from this layer in the last query.
resultObj getResult(int index) Returns a resultObj by index from a layer object with index in the range 0 to
numresults-1. Returns a valid object or FALSE(0) if index is invalid.
int open() Open the layer for use with getShape(). Returns MS_SUCCESS/MS_FAILURE.
int whichshapes(rectobj) Performs a spatial, and optionally an attribute based feature search. The function basically
prepares things so that candidate features can be accessed by query or drawing functions (eg using nextshape
function). Returns MS_SUCCESS or MS_FAILURE.
shapeobj nextShape() Called after msWhichShapes has been called to actually retrieve shapes within a given area.
Returns a shape object or NULL on error.
$map = ms_newmapobj("d:/msapps/gmap-ms40/htdocs/gmap75.map");
$layer = $map->getLayerByName(’road’);
$status = $layer->open();
$status = $layer->whichShapes($map->extent);
while ($shape = $layer->nextShape())
{
echo $shape->index ."<br>\n";
}
$layer->close();

void close() Close layer previously opened with open().

rectObj getExtent() Returns the layer’s data extents or NULL on error. If the layer’s EXTENT member is set then
this value is used, otherwise this call opens/closes the layer to read the extents. This is quick on shapefiles, but
can be an expensive operation on some file formats or data sources. This function is safe to use on both opened
or closed layers: it is not necessary to call open()/close() before/after calling it.
int addFeature(shapeObj shape) Add a new feature in a layer. Returns MS_SUCCESS or MS_FAILURE on error.
int getMetaData(string name) Fetch layer metadata entry by name. Returns “” if no entry matches the name. Note
that the search is case sensitive.
Note: getMetaData’s query is case sensitive.
int setMetaData(string name, string value) Set a metadata entry for the layer. Returns
MS_SUCCESS/MS_FAILURE.
int removeMetaData(string name) Remove a metadata entry for the layer. Returns MS_SUCCESS/MS_FAILURE.
string getWMSFeatureInfoURL(int clickX, int clickY, int featureCount, string infoFormat) Returns a WMS
GetFeatureInfo URL (works only for WMS layers) clickX, clickY is the location of to query in pixel coor-
dinates with (0,0) at the top left of the image. featureCount is the number of results to return. infoFormat is the
format the format in which the result should be requested. Depends on remote server’s capabilities. MapServer
WMS servers support only “MIME” (and should support “GML.1” soon). Returns “” and outputs a warning if
layer is not a WMS layer or if it is not queriable.
array getItems() Returns an array containing the items. Must call open function first. If there are no items, it returns
an empty array.
int setProcessing(string) Add the string to the processing string list for the layer. The layer->num_processing is
incremented by 1. Returns MS_SUCCESS or MS_FAILURE on error.
$oLayer->setprocessing("SCALE_1=AUTO");
$oLayer->setprocessing("SCALE_2=AUTO");

6.3. PHP MapScript 221

MapServer Documentation, Release 6.0

array getProcessing() Returns an array containing the processing strings. If there are no processing strings, it returns
an empty array.
void clearProcessing() Clears all the processing strings.
string executeWFSGetfeature() Executes a GetFeature request on a WFS layer and returns the name of the tempo-
rary GML file created. Returns an empty string on error.
int applySLD(string sldxml, string namedlayer) Apply the SLD document to the layer object. The matching be-
tween the sld document and the layer will be done using the layer’s name. If a namedlayer argument is passed
(argument is optional), the NamedLayer in the sld that matchs it will be used to style the layer. See SLD HowTo
for more information on the SLD support.
int applySLDURL(string sldurl, string namedlayer) Apply the SLD document pointed by the URL to the layer ob-
ject. The matching between the sld document and the layer will be done using the layer’s name. If a namedlayer
argument is passed (argument is optional), the NamedLayer in the sld that matchs it will be used to style the
layer. See SLD HowTo for more information on the SLD support.
string generateSLD() Returns an SLD XML string based on all the classes found in the layers.
int moveclassup(int index) The class specified by the class index will be moved up into the array of layers. Re-
turns MS_SUCCESS or MS_FAILURE. ex layer->moveclassup(1) will have the effect of moving class 1 up to
position 0, and the class at position 0 will be moved to position 1.
int moveclassdown(int index) The class specified by the class index will be moved down into the array of layers.
Returns MS_SUCCESS or MS_FAILURE. ex layer->moveclassdown(0) will have the effect of moving class 0
up to position 1, and the class at position 1 will be moved to position 0.
classObj removeClass(int index) Removes the class indicated and returns a copy, or NULL in the case of a failure.
Note that subsequent classes will be renumbered by this operation. The numclasses field contains the number
of classes available.
boolean isVisible() Returns MS_TRUE/MS_FALSE depending on whether the layer is currently visible in the map
(i.e. turned on, in scale, etc.).
int setConnectionType(int connectiontype [,string plugin_library]) Changes the connectiontype of the layer and
recreates the vtable according to the new connection type. This method should be used instead of setting the
connectiontype parameter directly. In the case when the layer.connectiontype = MS_PLUGIN the plugin_library
parameter should also be specified so as to select the library to load by MapServer. For the other connection
types this parameter is not used.
shapeObj getShape(resultObj result]) If the resultObj passed has a valid resultindex, retrieve shapeObj from a
layer’s resultset. (You get it from the resultObj returned by getResult() for instance). Otherwise, it will do
a single query on the layer to fetch the shapeindex
$map = new mapObj("gmap75.map");
$l = $map->getLayerByName("popplace");
$l->queryByRect($map->extent);
for ($i=0; $i<$l->getNumResults();$i++){
$s = $l->getShape($l->getResult($i));
echo $s->getValue($l,"Name");
echo "\n";
}

array getGridIntersectionCoordinates() Returns an array containing the grid intersection coordinates. If there are
no coordinates, it returns an empty array.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

222 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

legendObj

Constructor

Instances of legendObj are always are always embedded inside the mapObj.

Members

Type Name
int height
int width
int keysizex
int keysizey
int keyspacingx
int keyspacingy
colorObj outlinecolor //Color of outline of box, -1 for no outline
int status //MS_ON, MS_OFF, MS_EMBED
int position //for embeded legends, MS_UL, MS_UC, ...
int postlabelcache //MS_TRUE, MS_FALSE
labelObj label
colorObj imagecolor
string template

Methods

int updateFromString(string snippet) Update a legend from a string snippet. Returns

MS_SUCCESS/MS_FAILURE.
int set(string property_name, new_value) Set object property to a new value.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

lineObj

Constructor

new lineObj()

or using the old constructor

LineObj ms_newLineObj()

Members

Type Name
int numpoints (read-only)

6.3. PHP MapScript 223

MapServer Documentation, Release 6.0

Methods

int add(pointObj point) Add a point to the end of line. Returns MS_SUCCESS/MS_FAILURE.
int addXY(double x, double y [, double m]) Add a point to the end of line. Returns MS_SUCCESS/MS_FAILURE.
Note: the 3rd parameter m is used for measured shape files only. It is not mandatory.
int addXYZ(double x, double y, double z [, double m]) Add a point to the end of line. Returns
MS_SUCCESS/MS_FAILURE.
Note: the 4th parameter m is used for measured shape files only. It is not mandatory.
PointObj point(int i) Returns a reference to point number i.
int project(projectionObj in, projectionObj out) Project the line from “in” projection (1st argument) to “out” pro-
jection (2nd argument). Returns MS_SUCCESS/MS_FAILURE.

mapObj

Constructor

new mapObj(string map_file_name [, string new_map_path])

or using the old constructors

mapObj ms_newMapObj(string map_file_name [, string new_map_path]) Returns a new object to deal with a
MapServer map file.
mapObj ms_newMapObjFromString(string map_file_string [, string new_map_path]) Construct a new
mapObj from a mapfile string. Returns a new object to deal with a MapServer map file.
Note: By default, the SYMBOLSET, FONTSET, and other paths in the mapfile are relative to the mapfile location.
If new_map_path is provided then this directory will be used as the base path for all the rewlative paths inside the
mapfile.

Members

Type Name
int numlayers (read-only)
string name
string mappath
int status
int debug
int width (see setSize())
int height (see setSize())
int maxsize
outputformatObj outputformat
double resolution (pixels per inch, defaults to 72)
double defresolution (pixels per inch, defaults to 72)
rectObj extent;
double cellsize
int units (map units type)
double scaledenom (read-only, set by drawMap())
Continued on next page

224 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Table 6.3 – continued from previous page

string shapepath
int keysizex
int keysizey
int keyspacingx
int keyspacingy
webObj web
referenceMapObj reference
colorObj imagecolor
scalebarObj scalebar
legendObj legend
querymapObj querymap
string symbolsetfilename (read-only, set by setSymbolSet())
string fontsetfilename (read-only, set by setFontSet())
labelcacheObj labelcache (no members. Used only to be able to free the the label cache (ex : map->labelcache->free())
projectionObj projection
hashTableObj metadata

Methods

int set(string property_name, new_value) Set map object property to new value.
int getsymbolbyname(string symbol_name) Returns the symbol index using the name.
symbol getsymbolobjectbyid(int symbolid) Returns the symbol object using a symbol id. Refer to the symbol object
reference section for more details.
void preparequery() Calculate the scale of the map and set map->scaledenom.
imageObj prepareImage() Return a blank image object.
imageObj draw() Render map and return an image object or NULL on error.
imageObj drawQuery() Render a query map and return an image object or NULL on error.
imageObj drawLegend() Render legend and return an image object.
imageObj drawReferenceMap() Render reference map and return an image object.
imageObj drawScaleBar() Render scale bar and return an image object.
int embedlegend(imageObj image) embeds a legend. Actually the legend is just added to the label cache so you
must invoke drawLabelCache() to actually do the rendering (unless postlabelcache is set in which case it is
drawn right away). Returns MS_SUCCESS or MS_FAILURE on error.
int embedScalebar(imageObj image) embeds a scalebar. Actually the scalebar is just added to the label cache so
you must invoke drawLabelCache() to actually do the rendering (unless postlabelcache is set in which case it is
drawn right away). Returns MS_SUCCESS or MS_FAILURE on error.
int drawLabelCache(imageObj image) Renders the labels for a map. Returns MS_SUCCESS or MS_FAILURE on
error.
labelcacheMemberObj getLabel(int index) Returns a labelcacheMemberObj from the map given an index value
(0=first label) Labelcache has to be enabled.
while ($oLabelCacheMember = $oMap->getLabel($i)) {
/* do something with the labelcachemember */
++$i;
}

6.3. PHP MapScript 225

MapServer Documentation, Release 6.0

layerObj getLayer(int index) Returns a layerObj from the map given an index value (0=first layer)
layerObj getLayerByName(string layer_name) Returns a layerObj from the map given a layer name. Returns
NULL if layer doesn’t exist.
colorObj getcolorbyindex(int iCloIndex) Returns a colorObj corresponding to the color index in the palette
void setExtent(double minx, double miny, double maxx, double maxy) Set the map extents using the georef ex-
tents passed in argument. Returns MS_SUCCESS or MS_FAILURE on error.
int setCenter(pointObj center) Set the map center to the given map point. Returns MS_SUCCESS or
MS_FAILURE.
int offsetExtent(double x, double y) Offset the map extent based on the given distances in map coordinates. Returns
MS_SUCCESS or MS_FAILURE.
int scaleExtent(double zoomfactor, double minscaledenom, double maxscaledenom) Scale the map extent using
the zoomfactor and ensure the extent within the minscaledenom and maxscaledenom domain. If minscale-
denom and/or maxscaledenom is 0 then the parameter is not taken into account. Returns MS_SUCCESS or
MS_FAILURE.
int setRotation(double rotation_angle) Set map rotation angle. The map view rectangle (specified in EXTENTS)
will be rotated by the indicated angle in the counter- clockwise direction. Note that this implies the rendered
map will be rotated by the angle in the clockwise direction. Returns MS_SUCCESS or MS_FAILURE.
int setSize(int width, int height) Set the map width and height. This method updates the internal geotransform and
other data structures required for map rotation so it should be used instead of setting the width and height
members directly. Returns MS_SUCCESS or MS_FAILURE.
int zoompoint(int nZoomFactor, pointObj oPixelPos, int nImageWidth, int nImageHeight, rectObj oGeorefExt)
Zoom to a given XY postion. Returns MS_SUCCESS or MS_FAILURE on error.
Parameters are :
• Zoom factor : positive values do zoom in, negative values zoom out. Factor of 1 will recenter.
• Pixel position (pointObj) : x, y coordinates of the click, with (0,0) at the top-left
• Width : width in pixel of the current image.
• Height : Height in pixel of the current image.
• Georef extent (rectObj) : current georef extents.
• MaxGeoref extent (rectObj) : (optional) maximum georef extents. If provided then it will be impos-
sible to zoom/pan outside of those extents.
int zoomrectangle(rectObj oPixelExt, int nImageWidth, int nImageHeight, rectObj oGeorefExt) Set the map
extents to a given extents. Returns MS_SUCCESS or MS_FAILURE on error.
Parameters are :
• oPixelExt (rect object) : Pixel Extents
• Width : width in pixel of the current image.
• Height : Height in pixel of the current image.
• Georef extent (rectObj) : current georef extents.
int zoomscale(double nScaleDenom, pointObj oPixelPos, int nImageWidth, int nImageHeight, rectObj oGeorefExt [, rectObj oM
Zoom in or out to a given XY position so that the map is displayed at specified scale. Returns MS_SUCCESS
or MS_FAILURE on error.
Parameters are :

226 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

• ScaleDenom : Scale denominator of the scale at which the map should be displayed.
• Pixel position (pointObj) : x, y coordinates of the click, with (0,0) at the top-left
• Width : width in pixel of the current image.
• Height : Height in pixel of the current image.
• Georef extent (rectObj) : current georef extents.
• MaxGeoref extent (rectObj) : (optional) maximum georef extents. If provided then it will be impos-
sible to zoom/pan outside of those extents.
int queryByPoint(pointObj point, int mode, double buffer) Query all selected layers in map at point location spec-
ified in georeferenced map coordinates (i.e. not pixels). The query is performed on all the shapes that are part of
a CLASS that contains a Templating value or that match any class in a layer that contains a LAYER TEMPLATE
value. Mode is MS_SINGLE or MS_MULTIPLE depending on number of results you want. Passing buffer -1
defaults to tolerances set in the map file (in pixels) but you can use a constant buffer (specified in ground units)
instead. Returns MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or if some other
error happened (note that the error message in case nothing was found can be avoided in PHP using the ‘@’
control operator).
int queryByRect(rectObj rect) Query all selected layers in map using a rectangle specified in georeferenced map
coordinates (i.e. not pixels). The query is performed on all the shapes that are part of a CLASS that con-
tains a Templating value or that match any class in a layer that contains a LAYER TEMPLATE value. Returns
MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found or if some other error happened
(note that the error message in case nothing was found can be avoided in PHP using the ‘@’ control operator).
int queryByShape(shapeObj shape) Query all selected layers in map based on a single shape, the shape has to be
a polygon at this point. Returns MS_SUCCESS if shapes were found or MS_FAILURE if nothing was found
or if some other error happened (note that the error message in case nothing was found can be avoided in PHP
using the ‘@’ control operator).
int queryByFeatures(int slayer) Perform a query based on a previous set of results from a layer. At present the
results MUST be based on a polygon layer. Returns MS_SUCCESS if shapes were found or MS_FAILURE if
nothing was found or if some other error happened (note that the error message in case nothing was found can
be avoided in PHP using the ‘@’ control operator).
int queryByIndex(layerindex, tileindex, shapeindex[, addtoquery]) Add a specific shape on a given layer to the
query result. If addtoquery (which is a non mandatory argument) is set to MS_TRUE, the shape will be added
to the existing query list. Default behavior is to free the existing query list and add only the new shape.
int savequery(filename) Save the current query in a file. Returns MS_SUCESS or MS_FAILURE. Can be used with
loadquery
int loadquery(filename) Loads a query from a file. Returns MS_SUCESS or MS_FAILURE. To be used with save-
query.
void freequery(layerindex) Frees the query result on a specified layer. If the layerindex is -1, all queries on layers
will be freed.
int save(string filename) Save current map object state to a file. Returns -1 on error. Use absolute path. If a relative
path is used, then it will be relative to the mapfile location.
string getProjection() Returns a string representation of the projection. Returns NULL on error or if no projection is
set.
int setProjection(string proj_params, boolean bSetUnitsAndExtents) Set map projection and coordinate system.
Returns MS_SUCCESS or MS_FAILURE on error.
Parameters are given as a single string of comma-delimited PROJ.4 parameters. The argument : bSetUnit-
sAndExtents is used to automatically update the map units and extents based on the new projection. Possible

6.3. PHP MapScript 227

MapServer Documentation, Release 6.0

values are MS_TRUE and MS_FALSE. By defualt it is set at MS_FALSE.

int setWKTProjection(string proj_params, boolean bSetUnitsAndExtents) Same as setProjection(), but takes an
OGC WKT projection definition string as input. Returns MS_SUCCESS or MS_FAILURE on error.
Note: setWKTProjection requires GDAL support
int getMetaData(string name) Fetch metadata entry by name (stored in the WEB object in the map file). Returns “”
if no entry matches the name.
Note: getMetaData’s query is case sensitive.
int setMetaData(string name, string value) Set a metadata entry for the map (stored in the WEB object in the map
file). Returns MS_SUCCESS/MS_FAILURE.
int removeMetaData(string name) Remove a metadata entry for the map (stored in the WEB object in the map file).
Returns MS_SUCCESS/MS_FAILURE.
array getLayersIndexByGroup(string groupname) Return an array containing all the layer’s indexes given a group
name. If there are no layers, it returns an empty array.
array getAllGroupNames() Return an array containing all the group names used in the layers. If there are no groups,
it returns an empty array.
array getAllLayerNames() Return an array containing all the layer names. If there are no layers, it returns an empty
array.
int moveLayerUp(int layerindex) Move layer up in the hierarcy of drawing. Returns MS_SUCCESS or
MS_FAILURE on error.
int moveLayerDown(int layerindex) Move layer down in the hierarcy of drawing. Returns MS_SUCCESS or
MS_FAILURE on error.
array getLayersDrawingOrder() Return an array containing layer’s index in the order which they are drawn. If
there are no layers, it returns an empty array.
boolean setLayersDrawingOrder(array layeryindex) Set the layer’s order array. The argument passed must be a
valid array with all the layer’s index. Returns MS_SUCCESS or MS_FAILURE on error.
string processTemplate(array params, boolean generateimages) Process the template file specified in the web ob-
ject and return the result in a buffer. The processing consists of opening the template file and replace all the tags
found in it. Only tags that have an equivalent element in the map object are replaced (ex [scaledenom]). The are
two exceptions to the previous statement :
• [img], [scalebar], [ref], [legend] would be replaced with the appropriate url if the parameter generateimages
is set to MS_TRUE. (Note : the images corresponding to the different objects are generated if the object is
set to MS_ON in the map file)
• the user can use the params parameter to specify tags and their values. For example if the user have a
specific tag call [my_tag] and would like it to be replaced by “value_of_my_tag” he would do
$tmparray["my_tag"] = "value_of_my_tag";
$map->processtemplate($tmparray, MS_FALSE);

string processquerytemplate(array params, boolean generateimages) Process query template files and return the
result in a buffer. Second argument generateimages is not mandatory. If not given it will be set to TRUE.
See Also:
processtemplate
string processlegendtemplate(array params) Process legend template files and return the result in a buffer.

228 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

See Also:
processtemplate
int setSymbolSet(string fileName) Load and set a symbol file dynamically.
int getNumSymbols() Return the number of symbols in map.
int setFontSet(string fileName) Load and set a new FONTSET.
int selectOutputFormat(string type) Selects the output format to be used in the map. Returns
MS_SUCCESS/MS_FAILURE.
Note: the type used should correspond to one of the output formats declared in the map file. The type argument
passed is compared with the mimetype parameter in the output format structure and then to the name parameter in the
structure.
int saveMapContext(string filename) Available only if WMS support is enabled. Save current map object state
in WMS Map Context format. Only WMS layers are saved in the WMS Map Context XML file. Returns
MS_SUCCESS/MS_FAILURE.
int loadMapContext(string filename [, boolean unique_layer_name]) Available only if WMS support is enabled.
Load a WMS Map Context XML file into the current mapObj. If the map already contains some layers then
the layers defined in the WMS Map context document are added to the current map. The 2nd argument
unique_layer_name is optional and if set to MS_TRUE layers created will have a unique name (unique pre-
fix added to the name). If set to MS_FALSE the layer name will be the the same name as in the context. The
default value is MS_FALSE. Returns MS_SUCCESS/MS_FAILURE.
int applySLD(string sldxml) Apply the SLD document to the map file. The matching between the sld document and
the map file will be done using the layer’s name. See SLD HowTo for more information on the SLD support.
int applySLDURL(string sldurl) Apply the SLD document pointed by the URL to the map file. The matching
between the sld document and the map file will be done using the layer’s name. See SLD HowTo for more
information on the SLD support.
string generateSLD() Returns an SLD XML string based on all the classes found in all the layers.
string getconfigoption(string key) Returns the config value associated with the key. Returns an empty sting if key
not found.
int setconfigoption(string key, string value) Sets a config parameter using the key and the value passed
int applyconfigoptions() Applies the config options set in the map file. For example setting the PROJ_LIB using
the setconfigoption only modifies the value in the map object. applyconfigoptions will actually change the
PROJ_LIB value that will be used when dealing with projection.
int loadowsparameters(owsrequest request, string version) Load OWS request parameters (BBOX, LAYERS,
&c.) into map. Returns MS_SUCCESS or MS_FAILURE. 2nd argument version is not mandatory. If not
given, the version will be set to 1.1.1
int owsdispatch(owsrequest request) Processes and executes the passed OpenGIS Web Services request on the map.
Returns MS_DONE (2) if there is no valid OWS request in the req object, MS_SUCCESS (0) if an OWS
request was successfully processed and MS_FAILURE (1) if an OWS request was not successfully processed.
OWS requests include WMS, WFS, WCS and SOS requests supported by MapServer. Results of a dispatched
request are written to stdout and can be captured using the msIO services (ie. ms_ioinstallstdouttobuffer() and
ms_iogetstdoutbufferstring())
int insertLayer( layerObj layer [, int nIndex=-1 ] ) Insert a copy of layer into the Map at index nIndex. The default
value of nIndex is -1, which means the last possible index. Returns the index of the new Layer, or -1 in the case
of a failure.
layerObj removeLayer( int nIndex ) Remove a layer from the mapObj. The argument is the index of the layer to be
removed. Returns the removed layerObj on success, else null.

6.3. PHP MapScript 229

MapServer Documentation, Release 6.0

void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

outputformatObj

Constructor

Instance of outputformatObj is always embedded inside the mapObj. It is uses a read only.
No constructor available (coming soon, see ticket 979)

Members

Type Name
string name
string mimetype
string driver
string extension
int renderer
int imagemode // MS_IMAGEMODE_* value.
int transparent

Methods

int set(string property_name, new_value) Set object property to a new value.

void setOption(string property_name, string new_value) Add or Modify the format option list. return true on suc-
cess.
$oMap->outputformat->setOption("OUTPUT_TYPE", "RASTER");

string getOption(string property_name) Returns the associated value for the format option property passed as ar-
gument. Returns an empty string if property not found.
int validate() Checks some internal consistency issues, Returns MS_SUCCESS or MS_FAILURE. Some problems
are fixed up internally. May produce debug output if issues encountered.

OwsrequestObj

Constructor

new OWSRequestObj()

or using the old constructor

request = ms_newOwsrequestObj();

Create a new ows request object.

230 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Members

Type Name
int numparams (read-only)
int type (read-only): MS_GET_REQUEST or MS_POST_REQUEST

Methods

int loadparams() Initializes the OWSRequest object from the cgi environment variables REQUEST_METHOD,
QUERY_STRING and HTTP_COOKIE. Returns the number of name/value pairs collected.
int setparameter(string name, string value) Set a request parameter. For example :
$request->setparameter(’REQUEST’, ’GetMap’);

string getname(int index) Return the name of the parameter at index in the request’s array of parameter names.
string getvalue(int index) Return the value of the parameter at index in the request’s array of parameter values.
string getvaluebyname(string name) Return the value associated with the parameter name.

pointObj

Constructor

new pointObj()

or using the old constructor

PointObj ms_newPointObj()

Members

Type Name
double x
double y
double z (used for 3d shape files. set to 0 for other types)
double m (used only for measured shape files. set to 0 for other types.)

Methods

int setXY(double x, double y [, double m]) Set X,Y coordinate values.

Note: the 3rd parameter m is used for measured shape files only. It is not mandatory.
int setXYZ(double x, double y , double z, [, double m]) Set X,Y,Z coordinate values.
Note: the 4th parameter m is used for measured shape files only. It is not mandatory.
int draw(mapObj map, layerObj layer, imageObj img, int class_index, string text) Draws the individual point
using layer. The class_index is used to classify the point based on the classes defined for the layer. The text
string is used to annotate the point. Returns MS_SUCCESS/MS_FAILURE.

6.3. PHP MapScript 231

MapServer Documentation, Release 6.0

double distanceToPoint(pointObj poPoint) Calculates distance between two points.

double distanceToLine(pointObject p1, pointObject p2) Calculates distance between a point ad a lined defined by
the two points passed in argument.
double distanceToShape(shapeObj shape) Calculates the minimum distance between a point and a shape.
int project(projectionObj in, projectionObj out) Project the point from “in” projection (1st argument) to “out” pro-
jection (2nd argument). Returns MS_SUCCESS/MS_FAILURE.

projectionObj

Constructor

new projectionObj(string projectionString)

or using the old constructor

ProjectionObj ms_newProjectionObj(string projectionString)

Creates a projection object based on the projection string passed as argument.

$projInObj = ms_newprojectionobj("proj=latlong")

will create a geographic projection class.

The following example will convert a lat/long point to an LCC projection:
$projInObj = ms_newprojectionobj("proj=latlong");
$projOutObj = ms_newprojectionobj("proj=lcc,ellps=GRS80,lat_0=49,".
"lon_0=-95,lat_1=49,lat_2=77");
$poPoint = ms_newpointobj();
$poPoint->setXY(-92.0, 62.0);
$poPoint->project($projInObj, $projOutObj);

Methods

int getUnits() Returns the units of a projection object. Returns -1 on error.

querymapObj

Constructor

Instances of querymapObj are always are always embedded inside the mapObj.

Members

Type Name
int width
int height
int style (MS_NORMAL, MS_HILITE, MS_SELECTED)
colorObj color

232 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Methods

int updateFromString(string snippet) Update a queryMap object from a string snippet. Returns
MS_SUCCESS/MS_FAILURE.
int set(string property_name, new_value) Set object property to a new value.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

rectObj

Constructor

rectObj are sometimes embedded inside other objects. New ones can also be created with:
new rectObj()

or using the old constructor

RectObj ms_newRectObj()

Note: the members (minx, miny,maxx,maxy) are initialized to -1;

Members:

Type Name
double minx
double miny
double maxx
double maxy

Methods

int set(string property_name, new_value) Set object property to a new value.

void setextent(double minx, double miny, double maxx, double maxy) Set the rectangle extents.
int draw(mapObj map, layerObj layer, imageObj img, int class_index, string text) Draws the individual rectan-
gle using layer. The class_index is used to classify the rectangle based on the classes defined for the layer. The
text string is used to annotate the rectangle. Returns MS_SUCCESS/MS_FAILURE.
double fit(int width, int height) Adjust extents of the rectangle to fit the width/height specified.
int project(projectionObj in, projectionObj out) Project the rectangle from “in” projection (1st argument) to “out”
projection (2nd argument). Returns MS_SUCCESS/MS_FAILURE.

referenceMapObj

Constructor

Instances of referenceMapObj are always embedded inside the mapObj.

6.3. PHP MapScript 233

MapServer Documentation, Release 6.0

Members

Type Name
string image
int width
int height
int status
int marker
string markername
int markersize
int maxboxsize
int minboxsize
rectObj extent
ColorObj color
ColorObj outlinecolor

Methods

int updateFromString(string snippet) Update a referenceMap object from a string snippet. Returns
MS_SUCCESS/MS_FAILURE.
int set(string property_name, new_value) Set object property to a new value.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

resultObj

Constructor

new resultObj(int shapeindex)

or using the layerObj‘s getResult() method.

Members

Type Name
int shapeindex (read-only)
int tileindex (read-only)
int classindex (read-only)
int resultindex (read-only)

Method

None

234 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

scalebarObj

Constructor

Instances of scalebarObj are always are always embedded inside the mapObj.

Members

Type Name
int height
int width
int style
int intervals
colorObj color
colorObj backgroundcolor
colorObj outlinecolor
int units
int status //MS_ON, MS_OFF, MS_EMBED
int position //for embeded scalebars, MS_UL, MS_UC, ...
int postlabelcache
labelObj label
colorObj imagecolor
int align

Methods

int updateFromString(string snippet) Update a scalebar from a string snippet. Returns

MS_SUCCESS/MS_FAILURE.
int set(string property_name, new_value) Set object property to a new value.
int setimagecolor(int red, int green, int blue) Sets the imagecolor propery (baclground) of the object. Returns
MS_SUCCESS or MS_FAILURE on error.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

shapefileObj

Constructor

new shapeFileObj(string filename, int type)

or using the old constructor

shapefileObj ms_newShapefileObj(string filename, int type)

Opens a shapefile and returns a new object to deal with it. Filename should be passed with no extension.
To create a new file (or overwrite an existing one), type should be one of MS_SHP_POINT, MS_SHP_ARC,
MS_SHP_POLYGON or MS_SHP_MULTIPOINT. Pass type as -1 to open an existing file for read-only access, and
type=-2 to open an existing file for update (append).

6.3. PHP MapScript 235

MapServer Documentation, Release 6.0

Members

Type Name
int numshapes (read-only)
int type (read-only)
string source (read-only)
rectObj bounds (read-only)

Methods

shapeObj getShape(int i) Retrieve shape by index.

shapeObj getPoint(int i) Retrieve point by index.
shapeObj getTransformed(mapObj map, int i) Retrieve shape by index.
rectObj getExtent(int i) Retrieve a shape’s bounding box by index.
int addShape(shapeObj shape) Appends a shape to an open shapefile.
int addPoint(pointObj point) Appends a point to an open shapefile.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.
NOTE: The shape file is closed (and changes committed) when the object is destroyed. You can explicitly close
and save the changes by calling $shapefile->free(); unset($shapefile), which will also free the php object.

shapeObj

Constructor

new shapeObj(int type)

or using the old constructor

ShapeObj ms_newShapeObj(int type)

‘type’ is one of MS_SHAPE_POINT, MS_SHAPE_LINE, MS_SHAPE_POLYGON or MS_SHAPE_NULL

ShapeObj ms_shapeObjFromWkt(string wkt)

Creates new shape object from WKT string.

236 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Members

Type Name
string text
int classindex
int type (read-only)
int numlines (read-only)
int index
int tileindex (read-only)
rectObj bounds (read-only)
int numvalues (read-only)
array values (read-only)
The values array is an associative array with the attribute values for this shape. It is set only on shapes obtained from
layer->getShape(). The key to the values in the array is the attribute name, e.g.
$population = $shape->values["Population"];

Methods

int set(string property_name, new_value) Set object property to a new value.

int add(lineObj line) Add a line (i.e. a part) to the shape.
LineObj line(int i) Returns a reference to line number i.
int draw(mapObj map, layerObj layer, imageObj img) Draws the individual shape using layer. Returns
MS_SUCCESS/MS_FAILURE.
boolean contains(pointObj point) Returns MS_TRUE if the point is inside the shape, MS_FALSE otherwise.
boolean intersects(shapeObj shape) Returns MS_TRUE if the two shapes intersect, MS_FALSE otherwise.
int project(projectionObj in, projectionObj out) Project the shape from “in” projection (1st argument) to “out”
projection (2nd argument). Returns MS_SUCCESS/MS_FAILURE.
pointObj getpointusingmeasure(double m) Apply only on Measured shape files. Given a measure m, retun the
corresponding XY location on the shapeobject.
pointObj getmeasureusingpoint(pointObject point) Apply only on Measured shape files. Given an XY Location,
find the nearest point on the shape object. Return a point object of this point with the m value set.
string getvalue(layerObj layer, string filedname) Returns the value for a given field name.
shapeobj buffer(width) Returns a new buffered shapeObj based on the supplied distance (given in the coordinates of
the existing shapeObj). Only available if php/mapscript is built with GEOS library.
shapeobj convexhull() Returns a shape object representing the convex hull of shape. Only available if php/mapscript
is built with GEOS library.
shapeobj boundary() Returns the boundary of the shape. Only available if php/mapscript is built with GEOS library.
int containsShape(shapeobj shape2) Returns true if shape2 passed as argument is entirely within the shape. Else
return false. Only available if php/mapscript is built with GEOS library.
shapeobj union(shapeobj shape) Returns a shape object representing the union of the shape object with the one
passed as parameter. Only available if php/mapscript is built with GEOS library
shapeobj intersection(shapeobj shape) Returns a shape object representing the intersection of the shape object with
the one passed as parameter. Only available if php/mapscript is built with GEOS library.

6.3. PHP MapScript 237

MapServer Documentation, Release 6.0

shapeobj difference(shapeobj shape) Returns a shape object representing the difference of the shape object with the
one passed as parameter. Only available if php/mapscript is built with GEOS library.
shapeobj symdifference(shapeobj shape) Returns the computed symmetric difference of the supplied and existing
shape. Only available if php/mapscript is built with GEOS library.
int overlaps(shapeobj shape) Returns true if the shape passed as argument overlaps the shape. Else returns false.
Only available if php/mapscript is built with GEOS library.
int within(shapeobj shape2) Returns true if the shape is entirely within the shape2 passed as argument. Else returns
false. Only available if php/mapscript is built with GEOS library.
int crosses(shapeobj shape) Returns true if the shape passed as argument crosses the shape. Else return false. Only
available if php/mapscript is built with GEOS library.
int touches(shapeobj shape) Returns true if the shape passed as argument touches the shape. Else return false. Only
available if php/mapscript is built with GEOS library.
int equals(shapeobj shape) Returns true if the shape passed as argument is equal to the shape (geometry only). Else
return false. Only available if php/mapscript is built with GEOS library.
int disjoint(shapeobj shape) Returns true if the shape passed as argument is disjoint to the shape. Else return false.
Only available if php/mapscript is built with GEOS library.
pointObj getCentroid() Returns a point object representing the centroid of the shape. Only available if php/mapscript
is built with GEOS library.
double getArea() Returns the area of the shape (if applicable). Only available if php/mapscript is built with GEOS
library.
double getLength() Returns the length (or perimeter) of the shape. Only available if php/mapscript is built with
GEOS library.
pointObj getLabelPoint() Returns a point object with coordinates suitable for labelling the shape.
string toWkt() Returns WKT representation of the shape’s geometry.
int setBounds() Updates the bounds property of the shape. Must be called to calculate new bounding box after new
parts have been added.
int simplify(double tolerance) Given a tolerance, returns a simplified shape object using underlying GEOS library
or NULL on error.
int topologySimplifyPreservingSimplify(double tolerance) Given a tolerance, returns a simplified shape object us-
ing underlying GEOS library or NULL on error.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

styleObj

Constructor

Instances of styleObj are always embedded inside the classObj.

new styleObj(classObj class [, styleObj style])

or using the old constructor

styleObj ms_newStyleObj(classObj class [, styleObj style])

The second argument ‘style’ is optional. If given, the new style created will be a copy of the style passed as argument.

238 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

Members

Type Name
int symbol
string symbolname
double size
double minsize
double maxsize
int offsetx
int offsety
int antialias
colorObj color
colorObj backgroundcolor
colorObj outlinecolor
double width
double minwidth
double maxwidth
double angle
double minvalue
double maxvalue
string rangeitem
int opacity (this parameter is only supported for the AGG driver)

Methods

int updateFromString(string snippet) Update a style from a string snippet. Returns

MS_SUCCESS/MS_FAILURE.
int set(string property_name, new_value) Set object property to a new value.
int setBinding(const stylebinding, string value) Set the attribute binding for a specfiled style property. Added in
MapServer 5.0.
$oStyle->setbinding(MS_STYLE_BINDING_COLOR, "FIELD_NAME_COLOR");

This would bind the color parameter with the data (ie will extract the value of the color from the field called
“FIELD_NAME_COLOR”
string getBinding(const stylebinding) Get the attribute binding for a specfiled style property. Returns NULL if there
is no binding for this property.
$oStyle->setbinding(MS_STYLE_BINDING_COLOR, "FIELD_NAME_COLOR");
echo $oStyle->getbinding(MS_STYLE_BINDING_COLOR); // FIELD_NAME_COLOR

int removeBinding(const stylebinding) Remove the attribute binding for a specfiled style property. Added in
MapServer 5.0.
$oStyle->removebinding(MS_STYLE_BINDING_COLOR);

void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

6.3. PHP MapScript 239

MapServer Documentation, Release 6.0

symbolObj

Constructor

new symbolObj(mapObj map, string symbolname)

or using the old constructor

int ms_newSymbolObj(mapObj map, string symbolname)

Creates a new symbol with default values in the symbolist.

Note: Using the new constructor, the symbol is automatically returned. The old constructor returns the id of the new
symbol.
If a symbol with the same name exists, it (or its id) will be returned. To get a symbol object using the old constructor,
you need to use a method on the map object:
$nId = ms_newSymbolObj($map, "symbol-test");
$oSymbol = $map->getSymbolObjectById($nId);

Members

Type Name
string name
type name //Please refer to symbol type constants
int inmapfile If set to TRUE, the symbol will be saved inside the mapfile.
double sizex
double sizey
int numpoints (Read-Only)
int filled
int patternlength (Read-Only)
string imagepath (Read-Only))
int transparent
int transparentcolor
string character
int antialias
string font
int position

Methods

int set(string property_name, new_value) Set object property to a new value.

int setpoints(array double) Set the points of the symbol. Note that the values passed if an array containing the x and
y values of the points. Returns MS_SUCCESS/MS_FAILURE. Example: array[0] = 1 : x value of the first point
array[1] = 0 : y values of the first point array[2] = 1 : x value of the 2nd point ....
int setpattern(array int) Set the pattern of the symbol (used for dash patterns). Returns
MS_SUCCESS/MS_FAILURE.
array getpointsarray() Returns an array containing the points of the symbol. Refer to setpoints to see how the array
should be interpreted. If there are no points, it returns an empty array.
array getpatternarray() Returns an array containing the pattern. If there is no pattern, it returns an empty array.

240 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

int setimagepath(char filename) Loads a pixmap symbol specified by the filename. The file should be of either Gif
or Png format.
Example of usage:
1. create a symbol to be used as a dash line
$nId = ms_newsymbolobj($gpoMap, "mydash");
$oSymbol = $gpoMap->getsymbolobjectbyid($nId);
$oSymbol->set("filled", MS_TRUE);
$oSymbol->set("sizex", 1);
$oSymbol->set("sizey", 1);
$oSymbol->set("inmapfile", MS_TRUE);

$aPoints[0] = 1;
$aPoints[1] = 1;
$oSymbol->setpoints($aPoints);

$aPattern[0] = 10;
$aPattern[1] = 5;
$aPattern[2] = 5;
$aPattern[3] = 10;
$oSymbol->setpattern($aPattern);

$style->set("symbolname", "mydash");

2. Create a TrueType symbol

$nId = ms_newSymbolObj($gpoMap, "ttfSymbol");
$oSymbol = $gpoMap->getSymbolObjectById($nId);
$oSymbol->set("type", MS_SYMBOL_TRUETYPE);
$oSymbol->set("filled", true);
$oSymbol->set("character", "&#68;");
$oSymbol->set("font", "ttfFontName");

void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

webObj

Constructor

Instances of webObj are always are always embedded inside the mapObj.

6.3. PHP MapScript 241

MapServer Documentation, Release 6.0

Members

Type Name
string log
string imagepath
string temppath
string template
string imageurl
string header
string footer
string empty (read-only)
string error (read-only)
string mintemplate
string maxtemplate
double minscaledenom
double maxscaledenom
rectObj extent (read-only)
string queryformat
string legendformat
string browseformat
hashTableObj metadata

Methods

int updateFromString(string snippet) Update a web object from a string snippet. Returns
MS_SUCCESS/MS_FAILURE.
int set(string property_name, new_value) Set object property to a new value.
void free() Free the object properties and break the internal references. Note that you have to unset the php variable
to free totally the resources.

clusterObj

Constructor

Instance of clusterObj is always embedded inside the layerObj.

Members

Type Name
double maxdistance
double buffer
string region

Methods

int setGroup(string expression) Set layer group expression.

string getGroupString() Returns the expression for this cluster group or NULL on error.

242 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

int setFilter(string expression) Set layer filter expression.

string getFilterString() Returns the expression for this cluster filter or NULL on error.

6.4 Memory Management

Normally, you should not have to worry about the memory management because php has a garbage collector and will
free resources for you. If you write only small scripts that don’t do a lot of processing, it’s not worth to care about
that. Everything will be freed at the end of the script.
However, it may be useful to free resources during the execution if the script executes many tasks. To do so, you’ll
have to call the free() method of the mapscript objects and unset the php variables. The purpose of the free methods is
to break the circular references between an object and its properties to allow the zend engine to free the resources.
Here’s an example of a script that doesn’t free things during the execution:
$map = new mapObj("mapfile.map");
$of = $map->outputformat;
echo $map->extent->minx." - ".$map->extent->miny." - ".$map->extent->maxx." - ".$map->extent->maxy."\
echo "Outputformat name: $of->name\n";
unset($of);
unset($map); // Even if we unset the php variables, resources wont be freed
// Resources will be only freed at the end of the script

and the same script that frees resources as soon as it can

$map = new mapObj("mapfile.map");
$of = $map->outputformat;
echo $map->extent->minx." - ".$map->extent->miny." - ".$map->extent->maxx." - ".$map->extent->maxy."\
echo "Outputformat name: $of->name\n";
unset($of);
$map->free(); // break the circular references
// at this place, the outputformat ($of) and the rect object ($map->extent) resources are freed
unset($map);
// the map object is immediately freed after the unset (before the end of the script)

6.5 Python MapScript Appendix

Author Sean Gillies

Revision $Revision: 8278 $
Date $Date: 2008-12-23 13:34:31 -0800 (Tue, 23 Dec 2008) $

Contents
• Python MapScript Appendix
– Introduction
– Classes
– Exception Handling

6.4. Memory Management 243

MapServer Documentation, Release 6.0

6.5.1 Introduction

The Python MapScript module contains some class extension methods that have not yet been implemented for other
languages.

6.5.2 Classes

References to sections below will be added here as the documentation grows.

imageObj

The Python Imaging Library, http://www.pythonware.com/products/pil/, is an indispensible tool for image manipula-
tion. The extensions to imageObj are all geared towards better integration of PIL in MapScript applications.

imageObj Methods

imageObj( PyObject arg1, PyObject arg2 [, PyObject arg3 ] ) [imageObj] Create a new instance which is either
empty or read from a Python file-like object that refers to a GD format image.
The constructor has 2 different modes. In the blank image mode, arg1 and arg2 should be the desired width and
height in pixels, and the optional arg3 should be either an instance of outputFormatObj or a GD driver name as
a shortcut to a format. In the image file mode, arg1 should be a filename or a Python file or file-like object. If
the file-like object does not have a “seek” attribute (such as a urllib resource handle), then a GD driver name
must be provided as arg2.
Here’s an example of creating a 320 pixel wide by 240 pixel high JPEG using the constructor’s blank image
mode:
image = mapscript.imageObj(320, 240, ’GD/JPEG’)

In image file mode, interesting values of arg1 to try are instances of StringIO:
s = StringIO()
pil_image.save(s) # Save an image manipulated with PIL
ms_image = imageObj(s)

Or the file-like object returned from urlopen

url = urllib.urlopen(’http://mapserver.gis.umn.edu/bugs/ant.jpg’)
ms_image = imageObj(url, ’GD/JPEG’)

write( [ PyObject file ] ) [void] Write image data to a Python file-like object. Default is stdout.

pointObj

pointObj Methods

__str__() [string] Return a string formatted like

{ ’x’: %f , ’y’: %f }

with the coordinate values substituted appropriately. Usage example:

244 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

>>> p = mapscript.pointObj(1, 1)
>>> str(p)
{ ’x’: 1.000000 , ’y’: 1.000000 }

Note that the return value can be conveniently eval’d into a Python dictionary:
>>> p_dict = eval(str(p))
>>> p_dict[’x’]
1.000000

rectObj

rectObj Methods

__contains__( pointObj point ) [boolean] Returns True if point is inside the rectangle, otherwise returns False.
>>> r = mapscript.rectObj(0, 0, 1, 1)
>>> p = mapscript.pointObj(2, 0) # outside
>>> p in r
False
>>> p not in r
True

__str__() [string] Return a string formatted like

{ ’minx’: %f , ’miny’: %f , ’maxx’: %f , ’maxy’: %f }

with the bounding values substituted appropriately. Usage example:

>>> r = mapscript.rectObj(0, 0, 1, 1)
>>> str(r)
{ ’minx’: 0.000000 , ’miny’: 0.000000 , ’maxx’: 1.000000 , ’maxy’: 1.000000 }

Note that the return value can be conveniently eval’d into a Python dictionary:
>>> r_dict = eval(str(r))
>>> r_dict[’minx’]
0.000000

6.5.3 Exception Handling

The Python MapScript module maps a few MapServer errors into Python exceptions. Attempting to load a non-existent
mapfile raises an ‘IOError’, for example
>>> import mapscript
>>> mapfile = ’/no/such/file.map’
>>> m = mapscript.mapObj(mapfile)
Traceback (most recent call last):
File "<stdin>", line 1, in ?
File "/usr/lib/python2.3/site-packages/mapscript.py", line 799, in __init__
newobj = _mapscript.new_mapObj(*args)
IOError: msLoadMap(): Unable to access file. (/no/such/file.map)
>>>

The message of the error is written by ‘msSetError’ and so is the same message that CGI mapserv users see in error
logs.

6.5. Python MapScript Appendix 245

MapServer Documentation, Release 6.0

6.6 Python MapScript Image Generation

Author Sean Gillies

Revision $Revision: 8295 $
Date $Date: 2008-12-26 21:08:04 -0800 (Fri, 26 Dec 2008) $
Last Updated 2008/07/15

Table of Contents
• Python MapScript Image Generation
– Introduction
– Imagery Overview
– The imageObj Class
– Image Output
– Images and Symbols

6.6.1 Introduction

The MapScript HOWTO docs are intended to complement the API reference with examples of usage for specific
subjects. All examples in this document refer to the mapfile and testing layers distributed with MapServer 4.2+ and
found under mapserver/tests.

Pseudocode

All examples will use a pseudocode that is consistent with the language independent API reference. Each line is a
statement. For object attributes and methods we use the dot, ‘.’, operator. Creation and deletion of objects will be
indicated by ‘new’ and ‘del’ keywords. Other than that, the pseudocode looks a lot like Python.

6.6.2 Imagery Overview

The most common use of MapServer and MapScript is to create map imagery using the built-in GD format drivers:
GD/GIF, GD/PNG, GD/PNG24, and GD/JPEG. This imagery might be saved to a file on disk or be streamed directly
to another device.

6.6.3 The imageObj Class

Imagery is represented in MapScript by the imageObj class. Please see the API Reference (MapScript.txt) for class
attribute and method details.

Creating imageObj from a mapObj

The mapObj class has two methods that return instances of imageObj: ‘draw’, and ‘prepareImage’. The first returns a
full-fledged map image just as one would obtain from the mapserv CGI program
test_map = MapScript.mapObj(’tests/test.map’)
map_image = test_map.draw()

246 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

A properly sized and formatted blank image, without any layers, symbols, or labels, will be generated by ‘prepareIm-
age’
blank_image = test_map.prepareImage()

Creating a new imageObj

The imageObj class constructor creates new instances without need of a map
format = MapScript.outputFormatObj(’GD/JPEG’)
image = MapScript.imageObj(300, 200, format) # 300 wide, 200 high JPEG

and can even initialize from a file on disk

# First three args are overriden by attributes of the disk image file
disk_image = MapScript.imageObj(-1, -1, NULL, ’tests/test.png’)

6.6.4 Image Output

Creating files on disk

Imagery is saved to disk by using the ‘save’ method. By accessing the ‘extension’ attribute of an image’s format, the
proper file extension can be used without making any assumptions
filename = ’test.’ + map_image.format.extension
map_image.save(filename)

If the image is using a GDAL/GTiff-based format, a GeoTIFF file can be created on disk by adding a mapObj as a
second optional argument to ‘save’
map_image.save(filename, test_map)

Direct Output

An image can be dumped to an open filehandle using the ‘write’ method. By default, the filehandle is ‘stdout’
# Send an image to a web browser
print "Content-type: " + map_image.format.mimetype + "\n\n"
map_image.write()

This method is not fully functional for all SWIG MapScript languages. See the API Reference (MapScript.txt) for
details. The ‘write’ method is new in 4.4.

6.6.5 Images and Symbols

The symbolObj::getImage() method will return an instance of imageObj for pixmap symbols
symbol = test_map.symbolset.getSymbolByName(’home-png’)
image = symbol.getImage()

There is a symmetric ‘setImage’ method which loads imagery into a symbol, allowing pixmap symbols to be created
dynamically

6.6. Python MapScript Image Generation 247

MapServer Documentation, Release 6.0

new_symbol = MapScript.symbolObj(’from_image’)
new_symbol.type = MapScript.MS_SYMBOL_PIXMAP
new_symbol.setImage(image)
index = test_map.symbolset.appendSymbol(new_symbol)

The get/setImage methods are new in MapServer 4.4.

6.7 Mapfile Manipulation

Author Sean Gillies

Revision $Revision: 8365 $
Date $Date: 2008-12-31 07:49:02 -0800 (Wed, 31 Dec 2008) $

Contents
• Mapfile Manipulation
– Introduction
– Mapfile Overview
– The mapObj Class
– Children of mapObj
– Metadata

6.7.1 Introduction

The MapScript HowTo docs are intended to complement the API reference with examples of usage for specific sub-
jects. All examples in this document refer to the mapfile and testing layers distributed with MapServer 4.2+ and found
under mapserver/tests.

Pseudocode

All examples will use a pseudocode that is consistent with the language independent API reference. Each line is a
statement. For object attributes and methods we use the dot, ‘.’, operator. Creation and deletion of objects will be
indicated by ‘new’ and ‘del’ keywords. Other than that, the pseudocode looks a lot like Python.

6.7.2 Mapfile Overview

By “Mapfile” here, I mean all the elements that can occur in (nearly) arbitrary numbers within a MapScript mapObj:
Layers, Classes, and Styles. MapServer 4.4 has greatly improved capability to manipulate these objects.

6.7.3 The mapObj Class

An instance of mapObj is a parent for zero to many layerObj children.

248 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

New instances

The mapfile path argument to the mapscript.mapObj constructor is now optional

empty_map = new mapscript.mapObj

generates a default mapObj with no layers. A mapObj is initialized from a mapfile on disk in the usual manner:
test_map = new mapscript.mapObj(’tests/test.map’)

Cloning

An independent copy, less result and label caches, of a mapObj can be produced by the new mapObj.clone() method:
clone_map = test_map.clone()

Note: the Java MapScript module implements a “cloneMap” method to avoid conflict with the clone method of Java’s
Object class.

Saving

A mapObj can be saved to disk using the save method:

clone_map.save(’clone.map’)

Frankly, the msSaveMap() function which is the foundation for mapObj::save is incomplete. Your mileage may vary.

6.7.4 Children of mapObj

There is a common parent/child object API for Layers, Classes, and Styles in MapServer 4.4.

Referencing a Child

References to Layer, Class, and Style children are obtained by “getChild”-like methods of their parent:
layer_i = test_map.getLayer(i)
class_ij = layer_i.getClass(j)
style_ijk = class_ij.getStyle(k)

These references are for convenience only. MapScript doesn’t have any reference counting, and you are certain to run
into trouble if you try to use these references after the parent mapObj has been deleted and freed from memory.

Cloning a Child

A completely independent Layer, Class, or Style can be created using the clone method of layerObj, classObj, and
styleObj:
clone_layer = layer_i.clone()

This instance has no parent, and is self-owned.

6.7. Mapfile Manipulation 249

MapServer Documentation, Release 6.0

New Children

Uninitialized instances of layerObj, classObj, or styleObj can be created with the new constructors:
new_layer = new mapscript.layerObj
new_class = new mapscript.classObj
new_style = new mapscript.styleObj

and are added to a parent object using “insertChild”-like methods of the parent which returns the index at which the
child was inserted:
li = test_map.insertLayer(new_layer)
ci = test_map.getLayer(li).insertClass(new_class)
si = test_map.getLayer(li).getClass(ci).insertStyle(new_style)

The insert* methods create a completely new copy of the object and store it in the parent with all ownership taken on
by the parent.
see the API reference for more details.

Backwards Compatibility

The old style child object constructors with the parent object as a single argument:
new_layer = new mapscript.layerObj(test_map)
new_class = new mapscript.classObj(new_layer)
new_style = new mapscript.styleObj(new_class)

remain in MapServer 4.4.

Removing Children

Child objects can be removed with “removeChild”-like methods of parents, which return independent copies of the
removed object:
# following from the insertion example ...
# remove the inserted style, returns a copy of the original new_style
removed_style = test_map.getLayer(li).getClass(ci).removeStyle(si)
removed_class = test_map.getLayer(li).removeClass(ci)
removed_layer = test_map.removeLayer(li)

6.7.5 Metadata

Map, Layer, and Class metadata are the other arbitrarily numbered elements (well, up to the built-in limit of 41) of a
mapfile.

New API

In MapServer 4.4, the metadata attributes of mapObj.web, layerObj, and classObj are instances of hashTableObj, a
class which functions like a limited dictionary
layer.metadata.set(’wms_name’, ’foo’)
name = layer.metadata.get(’wms_name’) # returns ’foo’

You can iterate over all keys in a hashTableObj like

250 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

key = NULL
while (1):
key = layer.metadata.nextKey(key)
if key == NULL:
break
value = layer.metadata.get(key)
...

See the API Reference (mapscript.txt) for more details.

Backwards Compatibility for Metadata

The old getMetaData and setMetaData methods of mapObj, layerObj, and classObj remain for use by older programs.

6.8 Querying

Author Sean Gillies

Revision $Revision: 8278 $
Date $Date: 2008-12-23 13:34:31 -0800 (Tue, 23 Dec 2008) $

Contents
• Querying
– Introduction
– Querying Overview
– Attribute Queries
– Spatial Queries

6.8.1 Introduction

All examples in this document refer to the mapfile and testing layers distributed with MapServer 4.2+ and found under
mapserver/tests.

Pseudocode

All examples will use a pseudocode that is consistent with the language independent API reference. Each line is a
statement. For object attributes and methods we use the dot, ‘.’, operator. Creation and deletion of objects will be
indicated by ‘new’ and ‘del’ keywords. Other than that, the pseudocode looks a lot like Python.

6.8.2 Querying Overview

The Query Result Set

Map layers can be queried to select features using spatial query methods or the attribute query method. Ignoring for
the moment whether we are executing a spatial or attribute query, results are obtained like so:

6.8. Querying 251

MapServer Documentation, Release 6.0

layer.query() # not an actual method!

results = layer.getResults()

In the case of a failed query or query with zero results, ‘getResults’ returns NULL.

Result Set Members

Individual members of the query results are obtained like:

... # continued

if results:
for i in range(results.numresults): # iterate over results
result = results.getResult(i)

This result object is a handle, of sorts, for a feature of the layer, having ‘shapeindex’ and ‘tileindex’ attributes that can
be used as arguments to ‘getFeature’.

Resulting Features

The previous example code can now be extended to the case of obtaining all queried features:
layer.query()
results = layer.getResults()
if results:
# open layer in preparation of reading shapes
layer.open()

for i in range(results.numresults):
result = results.getResult(i)

layer.getFeature(result.shapeindex, result.tileindex)

... # do something with this feature

# Close when done

layer.close()

Backwards Compatibility

Scripts using the 4.2 API can continue to access query result members through layer methods:
for i in range(layer.getNumResults()):
result = layer.getResult(0)

but should adopt the new API for use in new work.

6.8.3 Attribute Queries

By Attributes

queryByAttributes()

252 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

6.8.4 Spatial Queries

By Rectangle

queryByRect()

By Point

queryByRect()

By Shape

queryByShape()

By Selection

queryByFeatures()

6.9 MapScript Variables

Author Howard Butler

Contact hobu.inc at gmail.com
Revision $Revision: 11476 $
Date $Date: 2011-04-06 04:25:40 -0700 (Wed, 06 Apr 2011) $

6.9. MapScript Variables 253

MapServer Documentation, Release 6.0

Contents
• MapScript Variables
– Version
– Logical Control - Boolean Values
– Logical Control - Status Values
– Map Units
– Layer Types
– Font Types
– Label Positions
– Label Size (Bitmap only)
– Shape Types
– Measured Shape Types
– Shapefile Types
– Query Types
– File Types
– Querymap Styles
– Connection Types
– DB Connection Types
– Join Types
– Line Join Types (for rendering)
– Image Types
– Image Modes
– Symbol Types
– Return Codes
– Limiters
– Error Return Codes

6.9.1 Version

Name Type Value

MS_VERSION character 5.2

6.9.2 Logical Control - Boolean Values

Name Type Value

MS_TRUE integer 1
MS_ON integer 1
MS_YES integer 1
MS_FALSE integer 0
MS_OFF integer 0
MS_NO integer 0

6.9.3 Logical Control - Status Values

Name Type Value

MS_DEFAULT integer 2
MS_EMBED integer 3
MS_DELETE integer 4

254 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

6.9.4 Map Units

Name Type Value

MS_DD integer
MS_FEET integer
MS_INCHES integer
MS_METERS integer
MS_MILES integer
MS_NAUTICALMILES integer
MS_PIXELS integer

6.9.5 Layer Types

Name Type Value

MS_LAYER_POINT integer
MS_LAYER_LINE integer
MS_LAYER_POLYGON integer
MS_LAYER_RASTER integer
MS_LAYER_ANNOTATION integer
MS_LAYER_QUERY integer
MS_LAYER_CIRCLE integer
MS_LAYER_TILEINDEX integer

6.9.6 Font Types

Name Type Value

MS_TRUETYPE integer
MS_BITMAP integer

6.9.7 Label Positions

Name Type Value

MS_UL integer
MS_LL integer
MS_UR integer
MS_LR integer
MS_CL integer
MS_CR integer
MS_UC integer
MS_LC integer
MS_CC integer
MS_AUTO integer

6.9. MapScript Variables 255

MapServer Documentation, Release 6.0

6.9.8 Label Size (Bitmap only)

Name Type Value

MS_TINY integer
MS_SMALL integer
MS_MEDIUM integer
MS_LARGE integer
MS_GIANT integer

6.9.9 Shape Types

Name Type Value

MS_SHAPE_POINT integer
MS_SHAPE_LINE integer
MS_SHAPE_POLYGON integer
MS_SHAPE_NULL integer

6.9.10 Measured Shape Types

Name Type Value

MS_SHP_POINTM integer 21
MS_SHP_ARCM integer 23
MS_SHP_POLYGONM integer 25
MS_SHP_MULTIPOINTM integer 28

6.9.11 Shapefile Types

Name Type Value

MS_SHAPEFILE_POINT integer 1
MS_SHAPEFILE_ARC integer 3
MS_SHAPEFILE_POLYGON integer 5
MS_SHAPEFILE_MULTIPOINT integer 8

6.9.12 Query Types

Name Type Value

MS_SINGLE integer 0
MS_MULTIPLE integer 1

6.9.13 File Types

Name Type Value

MS_FILE_MAP integer
MS_FILE_SYMBOL integer

256 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

6.9.14 Querymap Styles

Name Type Value

MS_NORMAL integer
MS_HILITE integer
MS_SELECTED integer

6.9.15 Connection Types

Name Type Value

MS_INLINE integer
MS_SHAPEFILE integer
MS_TILED_SHAPEFILE integer
MS_SDE integer
MS_OGR integer
MS_POSTGIS integer
MS_WMS integer
MS_ORACLESPATIAL integer
MS_WFS integer
MS_GRATICULE integer
MS_MYGIS integer
MS_RASTER integer

6.9.16 DB Connection Types

Name Type Value

MS_DB_XBASE integer
MS_DB_CSV integer
MS_DB_MYSQL integer
MS_DB_ORACLE integer
MS_DB_POSTGRES integer

6.9.17 Join Types

Name Type Value

MS_JOIN_ONE_TO_ONE integer
MS_JOIN_ONE_TO_MANY integer

6.9.18 Line Join Types (for rendering)

Name Type Value

MS_CJC_NONE integer
MS_CJC_BEVEL integer
MS_CJC_BUTT integer
MS_CJC_MITER integer
MS_CJC_ROUND integer
MS_CJC_SQUARE integer
MS_CJC_TRIANGLE integer

6.9. MapScript Variables 257

MapServer Documentation, Release 6.0

6.9.19 Image Types

Name Type Value

GD/GIF integer
GD/PNG integer
GD/PNG24 integer
GD/JPEG integer
GD/WBMP integer
swf integer
imagemap integer
pdf integer
GDAL/GTiff integer

6.9.20 Image Modes

Name Type Value

MS_IMAGEMODE_PC256 integer
MS_IMAGEMODE_RGB integer
MS_IMAGEMODE_RGBA integer
MS_IMAGEMODE_INT16 integer
MS_IMAGEMODE_FLOAT32 integer
MS_IMAGEMODE_BYTE integer
MS_IMAGEMODE_NULL integer
MS_NOOVERRIDE integer
MS_GD_ALPHA integer 1000

6.9.21 Symbol Types

Name Type Value

MS_SYMBOL_SIMPLE integer
MS_SYMBOL_VECTOR integer
MS_SYMBOL_ELLIPSE integer
MS_SYMBOL_PIXMAP integer
MS_SYMBOL_TRUETYPE integer

6.9.22 Return Codes

Name Type Value

MS_SUCCESS integer
MS_FAILURE integer
MS_DONE integer

6.9.23 Limiters

Name Type Value

MS_MAXSYMBOLS long
MS_MAXVECTORPOINTS long
MS_MAXSTYLELENGTH long
MS_IMAGECACHESIZE long

258 Chapter 6. MapScript

 MapServer Documentation, Release 6.0

6.9.24 Error Return Codes

Name Type Value

MS_NOERR long 0
MS_IOERR long 1
MS_MEMERR long 2
MS_TYPEERR long 3
MS_SYMERR long 4
MS_REGEXERR long 5
MS_TTFERR long 6
MS_DBFERR long 7
MS_GDERR long 8
MS_IDENTERR long 9
MS_EOFERR long 10
MS_PROJERR long 11
MS_MISCERR long 12
MS_CGIERR long 13
MS_WEBERR long 14
MS_IMGERR long 15
MS_HASHERR long 16
MS_JOINERR long 17
MS_NOTFOUND long 18
MS_SHPERR long 19
MS_PARSEERR long 20
MS_SDEERR long 21
MS_OGRERR long 22
MS_QUERYERR long 23
MS_WMSERR long 24
MS_WMSCONNERR long 25
MS_ORACLESPATIALERR long 26
MS_WFSERR long 27
MS_WFSCONNERR long 28
MS_MAPCONTEXTERR long 29
MS_HTTPERR long 30
MS_CHILDERR long 31
MS_WCSERR long 32
MS_NUMERRORCODES long 33
MESSAGELENGTH long 33
ROUTINELENGTH long 33

6.9. MapScript Variables 259

MapServer Documentation, Release 6.0

260 Chapter 6. MapScript

 CHAPTER 7

Data Input

7.1 Vector Data

Author Tyler Mitchell

Contact tmitchell at osgeo.org
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Last Updated 2011-04-11
This work is licensed under the Creative Commons Attribution-ShareAlike License. To view a copy of this license,
visit: http://creativecommons.org/licenses/by-sa/2.0/ca/ or send a letter to Creative Commons, 559 Nathan Abbott
Way, Stanford, California 94305, USA.
What is vector data? This quote from is a good description of what vector data is:
Vector: “An abstraction of the real world where positional data is represented in the form of coordinates.
In vector data, the basic units of spatial information are points, lines and polygons. Each of these units
is composed simply as a series of one or more coordinate points. For example, a line is a collection of
related points, and a polygon is a collection of related lines. Vector images are defined mathematically
as a series of points joined by lines. Vector-based drawings are resolution independent. This means that
they appear at the maximum resolution of the output device, such as a printer or monitor. Each object is
self-contained, with properties such as color, shape, outline, size, and position on the screen.”
From: http://coris.noaa.gov/glossary/glossary_l_z.html#v
The rest of this document is the data format guide. This guide is structured to show the fundamentals of each
MapServer supported data format. Each section discusses one format, ranging from one to several pages in length.
The sections typically start with a summary of the most important information about the format, followed by examples
of file listings, connection methods, ogrinfo usage and MapServer map file syntax examples.
Each section has been designed to stand alone, so you may notice that certain warnings and comments are repeated
or redundant. This is intentional. Each format is presented in rough order of popular use, based on a survey of the
MapServer community.
The following formats are included:

261
MapServer Documentation, Release 6.0

7.1.1 Data Format Types

Each type of data is made up of a data source and (one or more) layers. These two definitions apply to MapServer and
OGR.
Data Source - a group of layers stored in a common repository. This may be a file that handles several layers within
it, or a folder that has several files.
Layer - a sub-set of a data source often containing information in one type of vector format (point, line, polygon).
There are three types of data mapping and GIS data formats. Each type is handled differently. Below are the types and
some example formats:
• File-based- Shapefiles, Microstation Design Files (DGN), GeoTIFF images
• Directory-based - ESRI ArcInfo Coverages, US Census TIGER
• Database connections - PostGIS, ESRI ArcSDE, MySQL

File-based Data

File-based data consists of one or more files stored in any arbitrary folder. In many cases a single file is used (e.g.
DGN) but ESRI Shapefiles, for example, consist of at least 3 files each with a different filename extension: SHP, DBF,
SHX. In this case all 3 files are required because they each perform a different task internally.
Filenames usually serve as the data source name and contain layers that may or may not be obvious from the filename.
In Shapefiles, for example, there is one data source per shapefile and one layer which has the same name as that of the
file.

Directory-based Data

Directory-based data consists of one or more files stored in a particular way within a parent folder. In some cases
(e.g. Coverages) they may also require additional folders in other locations in the file tree in order to be accessed. The
directory itself may be the data source. Different files within the directory often represent the layers of data available.
For example, ESRI ArcInfo Coverages consist of more than one file with an ADF file extension, within a folder. The
PAL.ADF file represents the Polygon data. ARC.ADF holds the arc or line string data. The folder holds the data
source and each ADF file is a layer.

Database Connections

Database Connections are very similar to file and directory-based structures in one respect: they provide geographic
coordinate data for MapServer to interpret. That may be oversimplifying what is happening inside MapServer, but in
essence all you need is access to the coordinates making up the vector datasets.
Database connections provide a stream of coordinate data that is temporarily stored (e.g. in memory) and read by
MapServer to create the map. Other attribute or tabular data may also be required, but the focus of this guide is
coordinate data.
One important distinction between databases must be made. The databases discuss here are spatial databases, those
which can hold geographic data in its own data type. This is opposed to strictly tabular databases which cannot hold
geographic coordinates in the same way. It is possible to store some very simple coordinate data in regular tables, but
for anything but the most simple use a spatial database is required. There are spatial extensions to many databases
(open source and commercial). One of the most robust is the PostGIS extension to the PostgreSQL database. This
database not only allows the storage of geographic data, but also allows the manipulation of that data using SQL
commands. The other open source database with spatial capabilities is MySQL.

262 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Connections to databases usually consist of the following pieces of connection information:

Host - Directions to the server or computer hosting the database.
Database name - The name of the database you wish to access that is running on the host.
User name / passwords - Access privileges are usually restricted by user.
Note: Some databases (e.g. Oracle) use a name service identifier that includes both the host and database names.
Access to specific pieces of coordinate data usually require:
Table/View name - The name of the table or view holding the coordinate data.
Geographic column name - Where the geometry or coordinates are stored.

7.1.2 ArcInfo

ESRI ArcInfo Coverage Files are also known as simply as Coverages and less commonly as ADF files.

File listing

Coverages are made up of a set of files within a folder. The folder itself is the coverage name. The files roughly
represent different layers, usually representing different types of topology or feature types.
> ls /data/coverage/brazil
aat.adf arc.adf arx.adf bnd.adf lab.adf prj.adf tic.adf tol.adf

A folder with the name INFO is also part of the coverage. It sits at the same hierarchical level as the coverage folder
itself. Therefore, to copy a coverage (using regular file system tools) the coverage folder and the INFO folder must
both be copied. The INFO folder holds some catalogue information about the coverage.
> ls /data/coverage/info
arc0000.dat arc0001.dat arc0002.dat arc.dir
arc0000.nit arc0001.nit arc0002.nit

Data Access / Connection Method

• CONNECTIONTYPE OGR must be used. The ability to use coverages is not built into MapServer.
• The path to the coverage folder name is required.
• The layer name (feature type) is specified in the DATA parameter

OGRINFO Examples

The directory is the data source. Layers are found within the directory. Using ogrinfo on a coverage directory:
> ogrinfo /data/coverage/brazil -summary
INFO: Open of ‘brazil’
using driver ‘AVCBin’ successful.
1: ARC (Line String)
2: CNT (Point)
3: LAB (Point)
4: PAL (Polygon)

Using ogrinfo to examine the structure of a layer:

7.1. Vector Data 263

MapServer Documentation, Release 6.0

> ogrinfo /data/coverage/brazil PAL -summary

Had to open data source read-only.
INFO: Open of ‘brazil’
using driver ‘AVCBin’ successful.

Layer name: PAL

Geometry: Polygon
Feature Count: 1
Extent: (1272793.274958, 795381.617050) - (1287078.382785, 807302.747284)
Layer SRS WKT:
(unknown)
ArcIds: IntegerList (0.0)
AREA: Real (18.5)
PERIMETER: Real (18.5)
F_OPER#: Integer (5.0)
F_OPER-ID: Integer (5.0)
OPER: String (2.0)
FCODE: String (10.0)

Map File Example:

LAYER
NAME Brazil_bounds
TYPE POLYGON
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "/data/coverage/brazil"
DATA "PAL"
CLASS
NAME "Brazil Admin Areas"
STYLE
OUTLINECOLOR 153 102 0
SIZE 2
END
END
END

7.1.3 ArcSDE

Spatial Database Engine (SDE) is one of ESRI‘s products which enables spatial data to be stored, managed, and
quickly retrieved from leading commercial database management systems like Oracle, Microsoft SQL Server, Sybase,
IBM DB2, and Informix.

Supported ArcSDE Operations

• Versioned queries (query geometry and attributes from a specified version)

• queryByAttributes (select geometry and attributes based on the values of an attribute)
• Limited join support for within-database tables</li>
• queryByRect (select geometry based on an extent)
• Projection on the fly
• SDE for Coverages (a read-only type of SDE for coverage, shapefile, and ArcStorm/ArcLibrarian repositories)
• SDE 8.1, 8.2, 8.3, 9.0, 9.1, and 9.2

264 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

• Linux, Windows, and Solaris (platforms that have SDE C API support)

Unsupported ArcSDE Operations

• queryByShape (pass in a shape with MapScript and use it for queries)

• Direct Connect (bypass SDE to connect directly to the database with the SDE C API)

How to make a connection to SDE:

• Install the SDE C API client libraries for your platform (preferably matched to the server version you are using,
ie 8.2 client -> 8.2 server, 8.3 client -> 8.3 server)
• Compile MapServer with SDE support MapServer Unix Compilation Howto for specific details)
• Define a LAYER block in a MapFile that uses SDE as the CONNECTIONTYPE
LAYER
NAME states
TYPE POLYGON
CONNECTION "sdemachine.iastate.edu,port:5151,sde,username,password"
CONNECTIONTYPE SDE
DATA "HOBU.STATES_LAYER,SHAPE,SDE.DEFAULT"
FILTER "where MYCOLUMN is not NULL"
PROCESSING "QUERYORDER=ATTRIBUTE" # <-- MapServer 4.10 and above

# Within database one-to-one join support

# MapServer 5.0 and above

PROCESSING "JOINTABLE=SDE_MASTER.GEOSERVWRITE.JOINTABLE"

# MapServer 5.0 and above

CLASSITEM "SDE_MASTER.GEOSERVWRITE.JOINTABLE.VAL"

# MapServer 5.0 and above

FILTER "SDE_MASTER.GEOSERVWRITE.JOINTABLE.AQ_TAG=SDE_MASTER.GEOSERVWRITE.JOINTESTLAYER.AQ_TAG"

# ObjectID column manipulation

# MapServer 5.0 and above
PROCESSING "OBJECTID=OBJECTID"

TEMPLATE ’/where/the/template/file/is/located’
CLASS
STYLE
SYMBOL ’circle’
SIZE 3
COLOR -1 -1 -1
OUTLINECOLOR 0 0 0
END
END
END

CONNECTION - Order is important!

• sdemachine.iastate.edu - The name of the machine you are connecting to. In some instances, this may need
to be the IP address of the machine rather than the name if the server running MapServer is not configured to

7.1. Vector Data 265

MapServer Documentation, Release 6.0

cascade DNS lookups

• port:5151 - The port number of SDE. The port: is important as SDE expects you to define the service in this
slot, and it can be other names like sde:oracle (for direct connect) or esri_sde (for systems with port 5151
defined as esri_sde in /etc/services)
• sde - The database username that the SDE server is using to connect to your database. It is often only important
for SDE setups that are connecting to Oracle (and even then, not so important). Just leave it as sde if you don’t
know what it should be.
• username - The username that will be connecting to SDE. This user must have been granted rights to select
the layer that you will be specifying in the DATA directive. You can use ArcCatalog or the SDE command-line
utilities to grant the appropriate rights to layers.
• password - Password of the user connecting to SDE. Case Sensitive.

DATA - Order is important!

• HOBU.STATES_LAYER - The layer name you are querying. This the full name of the table in which the layer
resides. If you are using Oracle or Microsoft SQL Server as the DB for SDE, the schema name must also be
supplied.
• SHAPE - The column that contains the geometry. SDE technically allows for storage of multiple geometry
types in the same layer, but in practice this isn’t desirable. Also, expect to have problems if there are invalid or
null geometries in the layer (or versions of the layer).
• SDE.DEFAULT - As of MapServer 4.2, you can query against a specific version of the layer. SDE sup-
ports multi-user editing with versions. If a layer has been Registered with the GeoDatabase and Registered
as Versioned (ArcGIS terms), MapServer can query against specified versions of those edits. If not specified,
SDE.DEFAULT will be used for all queries. Case Sensitive.
Note: The version parameter is located in a different spot than MapServer 4.2, which had it on the CONNECTION
string.

TEMPLATE

• /where/the/template/file/is/located - A template directive must be specified (can point to a dummy file) in order
for MapServer to be able to query attributes from SDE. If you are only going to be drawing layers, this directive
is unnecessary and will slow down the query operations of SDE (especially for layers with lots of attribute
columns).

PROCESSING

• PROCESSING “QUERYORDER=ATTRIBUTE” - Allows you to force SDE to use the WHERE clause that
was defined in your FILTER statement first, without attempting to hit the spatial index. Only in very special
cases will you want to do this.
• PROCESSING “OBJECTID=OBJECTID” - If you are having trouble with the SDE driver detecting your
unique ID column, you can override it with this processing parameter. Doing so will also have a slight perfor-
mance benefit because it will save a couple of extra queries to the database.
• PROCESSING “ATTRIBUTE_QUALIFIED=TRUE” - User can set this option to always use fully qualified
attribute names.

266 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Within-database Join Support

MapServer’s SDE driver, as of MapServer 5.0, allows you to join a single attribute table that has no geometries to
the layer that you are rendering. This feature allows you to use the data in the joined table much as you would in a
composite query that was made with something like PostGIS or Oracle Spatial. That is, the columns in the right table
of the join are available for CLASSITEM, LABELITEM and so on. The biggest constraint, however, is that fully
qualified names must be used or it most likely will not work. The join support is activated through PROCESSING
options.
• PROCESSING “JOINTABLE=SDE_MASTER.GEOSERVWRITE.JOINTABLE” - The JOINTABLE
processing option tells the driver which table you are joining the current layer to.
• CLASSITEM “SDE_MASTER.GEOSERVWRITE.JOINTABLE.VAL” - A CLASSITEM or LABELITEM
for a joined table using this mechanism must be fully qualified.
• FILTER “SDE_MASTER.GEOSERVWRITE.JOINTABLE.AQ_TAG=SDE_MASTER.GEOSERVWRITE.JOINTESTLA
- An important part of the join is defining how the join is to be made. Use a FILTER to do so.

7.1.4 DGN

File listing

Data are encapsulated in a single file, usually with the suffix .dgn.
0824t.dgn

Data Access / Connection Method

• Access is available in MapServer through OGR.

• The CONNECTIONTYPE OGR parameter must be used.
• The path to the dgn file is required, file extension is needed.
• All types of features in a DGN file are held in one “layer” of data. The layer is called elements and is the first
and only layer.
• The type of feature to be read from the DGN depends on the TYPE parameter in the map file.
• DGN files typically contain POINT, LINE, POLYGON and ANNOTATION feature types.
• DGN files contain “styling” information - how to color and present the data. This is used, optionally, by speci-
fying the STYLEITEM “AUTO” parameter.
Note: DGN files typically use white as a color for their features and therefore are not visible on maps with white
backgrounds.

OGRINFO Examples

Using ogrinfo on a single DGN file:

> ogrinfo /data/dgn/0824t.dgn
Had to open data source read-only.
INFO: Open of ‘0842t.dgn’
using driver ‘DGN’ successful.
1: elements

7.1. Vector Data 267

MapServer Documentation, Release 6.0

Note: No geometry/feature type for the layer is identified because it can be multiple types.
DGN files are not really GIS data files. They evolved from drafting formats used by computer aided drafting/design
(CADD) programs.
They carry a few key attributes which are usually consistent across all DGN files. Most of the attributes relate to
graphical styling of features for map presentation, such as ColorIndex, Style, etc.
Spatial reference system information is not always encoded into DGN files. This can be a major problem when trying
to adequately reference the DGN data in another mapping program.
Measurement units can be a problem. In some cases the features could be located in kilometres or feet even though
it is not obvious from the output of ogrinfo. Sometimes the only way to identify or correct a problem with units is to
open the file in Microstation software.
Using ogrinfo to examine the structure of the file/layer:
> ogrinfo -summary /data/dgn/0824t.dgn elements
INFO: Open of ’0824t.dgn’
using driver ’DGN’ successful.

Layer name: elements

Geometry: Unknown (any)
Feature Count: 22685
Extent: (-513183.050000, 150292.930000) - (-224583.220000, 407463.360000)
Layer SRS WKT:
(unknown)
Type: Integer (2.0)
Level: Integer (2.0)
GraphicGroup: Integer (4.0)
ColorIndex: Integer (3.0)
Weight: Integer (2.0)
Style: Integer (1.0)
EntityNum: Integer (8.0)
MSLink: Integer (10.0)
Text: String (0.0)

Map File Example:

LAYER
NAME dgn
TYPE LINE
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "dgn/0824t.dgn"
STYLEITEM "AUTO"
CLASS
END
END # Layer

7.1.5 ESRI Personal Geodatabase (MDB)

ESRI Personal Geodatabases are basically Microsoft Access files that contain spatial information. For more informa-
tion see the ESRI description page.

268 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

File listing

Similar to other database formats, the mdb file consists of several tables. The geometry is held in a BLOB table
column.

Data Access / Connection Method

Personal geodatabase access is available through OGR. See the OGR driver page for specific driver information. The
driver is standard in any win32 build of GDAL/OGR version 1.3.2 or later. For Linux/Unix, MDBTools ODBC drivers
can be used for this (with some difficulty).
OGR uses the names of spatial tables within the personal geodatabase (tables with a Shape column) as layers.
The CONNECTION parameter must include the mdb extension, and the DATA parameter should be the name of the
spatial table (or OGR layer).
CONNECTIONTYPE ogr
CONNECTION "pgeodatabase.mdb"
DATA "layername"

OGRINFO Examples

First you should make sure that your GDAL/OGR build contains the personal geodatabase “PGeo” driver, by using
the ‘–formats’ command:
>ogrinfo --formats
Loaded OGR Format Drivers:
...
-> "ODBC" (read/write)
-> "PGeo" (readonly)
-> "PostgreSQL" (read/write)
...

If you don’t have the driver, you might want to try the FWTools or MS4W packages, which include the driver.
Once you have the PGeo driver you are ready to try an ogrinfo command on your database to get a list of spatial tables:
>ogrinfo test.mdb
INFO: Open of ‘test.mdb’
using driver ‘PGeo’ successful.
1: counties

Now use ogrinfo to get information on the structure of the spatial table:
>ogrinfo test.mdb counties -summary
INFO: Open of ‘test.mdb’
using driver ‘PGeo’ successful.

Layer name: counties

Geometry: Unknown (any)
Feature Count: 67
Extent: (-87.634943, 24.543945) - (-80.031369, 31.000975)
Layer SRS WKT:
GEOGCS["GCS_WGS_1984",
DATUM["WGS_1984",
SPHEROID["WGS_1984",6378137.0,298.257223563]],
PRIMEM["Greenwich",0.0],
UNIT["Degree",0.0174532925199433]]

7.1. Vector Data 269

MapServer Documentation, Release 6.0

OBJECTID_1: Integer (10.0)

OBJECTID: Integer (10.0)
NAME: String (32.0)
STATE_NAME: String (25.0)
STATE_FIPS: String (2.0)
CNTY_FIPS: String (3.0)
FIPS: String (5.0)
...

Note that you can also use an ODBC connection to access all of the tables in your geodatabase:
>ogrinfo PGeo:testDSN counties -summary
INFO: Open of ‘testDSN’
using driver ‘PGeo’ successful.

1: counties
2: counties_Shape_Index
...

(where “testDSN” is the name of your System DSN)

Mapfile Example

Direct Access to MDB

LAYER
NAME my_geodatabase
TYPE POLYGON
CONNECTIONTYPE ogr
CONNECTION "test.mdb"
DATA "counties"
STATUS ON
CLASS
NAME "counties"
STYLE
COLOR 255 255 120
END
END
END

Through an ODBC Connection

LAYER
NAME my_geodatabase
TYPE POLYGON
CONNECTIONTYPE ogr
CONNECTION "PGeo:testDSN"
DATA "counties"
STATUS ON
CLASS
NAME "counties"
STYLE
COLOR 255 255 120
END

270 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

END
END

7.1.6 ESRI Shapefiles (SHP)

Also known as ESRI ArcView Shapefiles or ESRI Shapefiles. ESRI is the company that introduced this format.
ArcView was the first product to use shapefiles.

File listing

Shapefiles are made up of a minimum of three similarly named files, with different suffixes:
Countries_area.dbf
Countries_area.shp
Countries_area.shx

Data Access / Connection Method

Shapefile access is built directly into MapServer. It is also available through OGR, but direct access without OGR
is recommended and discussed here. The path to the shapefile is required. No file extension should be specified.
Shapefiles only hold one layer of data, therefore no distinction needs to be made.

OGRINFO Examples

• The directory can serve as a data source.

• Each shapefile in a directory serves as a layer.
• A shapefile can also be a data source. In this case the layer has the same prefix as the shapefile.
Using ogrinfo on a directory with multiple shapefiles:
> ogrinfo /data/shapefiles/
INFO: Open of ‘/data/shapefiles/’
using driver ‘ESRI Shapefile’ successful.
1: wpg_h2o (Line String)
2: wpg_roads (Line String)
3: wpg_roads_dis (Line String)
4: wpgrestaurants (Point)

Using ogrinfo on a single shapefile:

> ogrinfo /data/shapefiles/Countries_area.shp
Had to open data source read-only.
INFO: Open of ‘Countries_area.shp’
using driver ‘ESRI Shapefile’ successful.
1: Countries_area (Polygon)

Using ogrinfo to examine the structure of the file/layer:

> ogrinfo -summary /data/shapefiles/Countries_area.shp Countries_area
Had to open data source read-only.
INFO: Open of ‘Countries_area.shp’
using driver ‘ESRI Shapefile’ successful.

7.1. Vector Data 271

MapServer Documentation, Release 6.0

Layer name: Countries_area

Geometry: Polygon
Feature Count: 27458
Extent: (-180.000000, -90.000000) - (180.000000, 83.627419)
Layer SRS WKT:
(unknown)
FAC_ID: Integer (5.0)
TILE: Integer (3.0)
ARCLIST: String (254.0)
NAM: String (77.0)
PERIMETER: Real (22.17)
POLYGONCOU: Integer (6.0)
NA2DESC: String (45.0)

Map File Example:

LAYER
NAME my_shapefile
TYPE POLYGON
DATA countries_area
STATUS OFF
CLASS
NAME "Countries"
OUTLINECOLOR 0 0 0
END
END

7.1.7 GML

Also known as Geographic Markup Language and GML/XML. GML is a text-based, XML format that can represent
vector and attribute data. This is an Open Geospatial Consortium specification for data interchange. More information
is available at http://www.opengeospatial.org/standards/gml

File listing

GML files are usually a single text file with a GML filename extension. Some may use XML as the filename extension:
coal_dep.gml

XML schema documents often accompany GML files that have been translated from some other format (e.g. using
the ogr2ogr utility).
GML uses sets of nested tags to define attributes and geometry coordinates. Example of text in a GML file:
<gml:featureMember>
<Coal_Deposits fid="1">
<UNKNOWN>0.000</UNKNOWN>
<NA>0.000</NA>
<ID>2</ID>
<ID2>2</ID2>
<MARK>7</MARK>
<COALKEY>110</COALKEY>
<COALKEY2>110</COALKEY2>
<ogr:geometryProperty>
<gml:Point>
<gml:coordinates>78.531,50.694</gml:coordinates>
</gml:Point>

272 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

</ogr:geometryProperty>
</Coal_Deposits>
</gml:featureMember>

Data Access / Connection Method

• GML access is available in MapServer through OGR. More information on OGR GML support is available at
http://www.gdal.org/ogr/drv_gml.html
• The CONNECTIONTYPE OGR parameter must be used.
• The path to the GML file is required, including file extension. There can be multiple layers in a GML file,
including multiple feature types.

OGRINFO Examples

Using ogrinfo on a single GML file:

> ogrinfo /data/gml/coal_dep.gml
Had to open data source read-only.
INFO: Open of ‘coal_dep.gml’
using driver ‘GML’ successful.
1: Coal_Deposits

Using ogrinfo to examine the structure of one layer:

> ogrinfo -summary /data/gml/coal_dep.gml Coal_Deposits
Had to open data source read-only.
INFO: Open of ‘coal_dep.gml’
using driver ‘GML’ successful.

Layer name: Coal_Deposits

Geometry: Unknown (any)
Feature Count: 266
Extent: (23.293650, 37.986340) - (179.272550, 80.969670)
Layer SRS WKT:
(unknown)
UNKNOWN: Real (0.0)
NA: Real (0.0)
ID: Integer (0.0)
ID2: Integer (0.0)
MARK: Integer (0.0)
COALKEY: Integer (0.0)
COALKEY2: Integer (0.0)
LONG: Real (0.0)
LAT: Real (0.0)

Map File Example:

LAYER
NAME coal_deposits
TYPE POINT
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "gml/coal_dep.gml"
CLASS
STYLE

7.1. Vector Data 273

MapServer Documentation, Release 6.0

COLOR 0 0 0
SYMBOL ’circle’
SIZE 6
END
END
END

7.1.8 GPS Exchange Format (GPX)

GPX (the GPS Exchange Format) is a light-weight XML data format containing GPS data (waypoints, routes, and
tracks). For more information see the official GPX site.

File listing

All waypoints, routes, and tracks are stored in a single .gpx file.

Data Access / Connection Method

• GPX access is available through OGR. See the OGR driver page for specific driver information.
• A relative path to the .gpx file can be used in the mapfile LAYER’s CONNECTION string.
• The feature type is specified in the DATA parameter
– the “tracks” feature type will usually be the track line
– the “track_points” feature type will usually be the points that make up the track line

OGRINFO Examples

First you should make sure that your GDAL/OGR build contains the “GPX” driver, by using the ‘–formats’ command:
>ogrinfo --formats
Loaded OGR Format Drivers:
...
-> "CSV" (read/write)
-> "GML" (read/write)
-> "GPX" (read/write)
-> "KML" (read/write)
...

If you don’t have the driver, you might want to try the FWTools or MS4W packages, which include the driver.
Once you have the GPX driver you are ready to try an ogrinfo command on your file to get a list of feature types:
>ogrinfo test.gpx
INFO: Open of ‘test.gpx’
using driver ‘GPX’ successful.
1: waypoints (Point)
2: routes (Line String)
3: tracks (Multi Line String)
4: route_points (Point)
5: track_points (Point)

Now use ogrinfo to get information on one of the feature types:

274 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

>ogrinfo test.gpx track_points -summary

INFO: Open of ‘test.gpx’
using driver ‘GPX’ successful.

Layer name: track_points

Geometry: Point
Feature Count: 661
Extent: (-66.694270, 47.985570) - (-66.675222, 47.990791)
Layer SRS WKT:
GEOGCS["WGS 84",
DATUM["WGS_1984",
SPHEROID["WGS 84",6378137,298.257223563,
AUTHORITY["EPSG","7030"]],
AUTHORITY["EPSG","6326"]],
PRIMEM["Greenwich",0,
AUTHORITY["EPSG","8901"]],
UNIT["degree",0.01745329251994328,
AUTHORITY["EPSG","9122"]],
AUTHORITY["EPSG","4326"]]
track_fid: Integer (0.0)
track_seg_id: Integer (0.0)
track_seg_point_id: Integer (0.0)
ele: Real (0.0)
time: DateTime (0.0)
magvar: Real (0.0)
geoidheight: Real (0.0)
name: String (0.0)
cmt: String (0.0)
desc: String (0.0)
src: String (0.0)
...

Mapfile Example

Since you have confirmed that OGR can read your GPX file, now you can create a MapServer layer:
LAYER
NAME gpx
TYPE POINT
STATUS ON
CONNECTIONTYPE OGR
CONNECTION test.gpx
DATA "track_points"
CLASS
NAME "gpx"
STYLE
SYMBOL ’circle’
COLOR 0 119 255
SIZE 2
END
END
END # layer

7.1. Vector Data 275

MapServer Documentation, Release 6.0

7.1.9 Inline

Inline features refer to coordinates entered directly into the map file. They are not a file or database format and do not
require any DATA or CONNECTION parameters. Instead they use a FEATURE section to define the coordinates.
Inline features can be used to define points, lines and polygons as if taken from an external file. This requires direct
entry of coordinate pairs in the map file using a particular syntax.

Data Access / Connection Method

This is a native MapServer option that doesn’t use any external libraries to support it.

Map File Example

Points

• Each FEATURE..END section defines a feature.

• Multiple points can be defined in a FEATURE section. If multiple points are defined in the same layer, they will
have the same CLASS settings, e.g. for colours and styles.
• Coordinates are entered in the units set in the layer’s projection. In this case it is assuming the map file projection
is using decimal degrees.
LAYER
NAME inline_stops
TYPE POINT
STATUS DEFAULT
FEATURE
POINTS
72.36 33.82
END
TEXT "My House"
END
FEATURE
POINTS
69.43 35.15
71.21 37.95
72.02 38.60
END
TEXT "My Stores"
END
CLASS
STYLE
COLOR 0 0 250
SYMBOL ’circle’
SIZE 6
END
END
END

Lines

Lines are simply a list of points strung together, but the layer must be TYPE LINE instead of TYPE POINT.

276 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

LAYER
NAME inline_track
TYPE LINE
STATUS DEFAULT
MAXSCALE 10000000
FEATURE
POINTS
72.36 33.82
70.85 34.32
69.43 35.15
70.82 36.08
70.90 37.05
71.21 37.95
END
END
CLASS
STYLE
COLOR 255 10 0
SYMBOL ’circle’
SIZE 2
END
END
END

Polygons

Polygons are the same as the line example, just a list of points. They require the TYPE POLYGON parameter.
Polygons also require the final coordinate pair to be the same as the first, making it a closed polygon.

7.1.10 KML - Keyhole Markup Language

Table of Contents
• KML - Keyhole Markup Language
– Links to KML-Related Information
– Data Access / Connection Method
– Example 1: Displaying a .KML file
– Example 2: Displaying a .KMZ file

Keyhole Markup Language (KML) is an XML-based language for managing the display of 3D geospatial data. KML
is a standard maintained by the Open Geospatial Consoritum (OGC).

Links to KML-Related Information

• Google’s KML Reference

• OGC’s KML Specification
• KML Validator
• KML Validator (against OGC KML 2.2)

7.1. Vector Data 277

MapServer Documentation, Release 6.0

Data Access / Connection Method

KML access in MapServer is available through OGR. See the OGR driver page for specific driver information. Read
support was initially added to GDAL/OGR version 1.5.0. A more complete KML reader was added to GDAL/OGR in
version 1.8.0, through the libKML driver (including the ability to read multigeometry, and KMZ files).
The CONNECTION parameter must include the kml or kmz extension, and the DATA parameter should be the name
of the layer.
CONNECTIONTYPE OGR
CONNECTION "filename.kml"
DATA "layername"

Example 1: Displaying a .KML file

OGRINFO

First you should make sure that your GDAL/OGR build contains the “KML” driver, by using the ‘–formats’ command:
>ogrinfo --formats
Loaded OGR Format Drivers:
...
-> "GML" (read/write)
-> "GPX" (read/write)
-> "KML" (read/write)
...

If you don’t have the driver, you might want to try the FWTools or MS4W packages, which include the driver.
Once you have the KML driver you are ready to try an ogrinfo command on your file to get a list of available layers:
>ogrinfo myplaces.kml
INFO: Open of ‘myplaces.kml’
using driver ‘KML’ successful.
1: Layer #0 (Point)

Now use ogrinfo to get information on the structure of the layer:

>ogrinfo fountains-hotel.kml "Layer #0" -summary
Had to open data source read-only.
INFO: Open of ‘fountains-hotel.kml’
using driver ‘KML’ successful.

Layer name: Layer #0

Geometry: Point
Feature Count: 1
Extent: (18.424930, -33.919627) - (18.424930, -33.919627)
Layer SRS WKT:
GEOGCS["WGS 84",
DATUM["WGS_1984",
SPHEROID["WGS 84",6378137,298.257223563,
AUTHORITY["EPSG","7030"]],
AUTHORITY["EPSG","6326"]],
PRIMEM["Greenwich",0,
AUTHORITY["EPSG","8901"]],
UNIT["degree",0.01745329251994328,
AUTHORITY["EPSG","9122"]],
AUTHORITY["EPSG","4326"]]

278 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Name: String (0.0)

Description: String (0.0)

Mapfile Example

LAYER
NAME "kml_example"
TYPE POINT
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "kml/fountains-hotel.kml"
DATA "Layer #0"
LABELITEM "NAME"
CLASS
NAME "My Places"
STYLE
COLOR 250 0 0
OUTLINECOLOR 255 255 255
SYMBOL ’circle’
SIZE 6
END
LABEL
SIZE TINY
COLOR 0 0 0
OUTLINECOLOR 255 255 255
POSITION AUTO
END
END
END

Example 2: Displaying a .KMZ file

OGRINFO

First you should make sure that your GDAL/OGR build contains the “LIBKML” driver, by using the ‘–formats’
command:
>ogrinfo --formats
Loaded OGR Format Drivers:
...
-> "GML" (read/write)
-> "GPX" (read/write)
-> "LIBKML" (read/write)
-> "KML" (read/write)
...

If you don’t have the driver, you might want to try the FWTools or MS4W packages, which include the driver. Or you
can follow the compiling notes for libKML and GDAL/OGR.
Once you have the LIBKML driver you are ready to try an ogrinfo command on your file to get a list of available
layers:
>ogrinfo Lunenburg_Municipality.kmz
INFO: Open of ‘Lunenburg_Municipality.kmz’

7.1. Vector Data 279

MapServer Documentation, Release 6.0

using driver ‘LIBKML’ successful.

1: Lunenburg_Municipality

Now use ogrinfo to get information on the structure of the layer:

>ogrinfo Lunenburg_Municipality.kmz Lunenburg_Municipality -summary
INFO: Open of ‘Lunenburg_Municipality.kmz’
using driver ‘LIBKML’ successful.

Layer name: Lunenburg_Municipality

Geometry: Unknown (any)
Feature Count: 1
Extent: (-64.946433, 44.133207) - (-64.230281, 44.735125)
Layer SRS WKT:
GEOGCS["WGS 84",
DATUM["WGS_1984",
SPHEROID["WGS 84",6378137,298.257223563,
AUTHORITY["EPSG","7030"]],
TOWGS84[0,0,0,0,0,0,0],
AUTHORITY["EPSG","6326"]],
PRIMEM["Greenwich",0,
AUTHORITY["EPSG","8901"]],
UNIT["degree",0.0174532925199433,
AUTHORITY["EPSG","9108"]],
AUTHORITY["EPSG","4326"]]
Name: String (0.0)
description: String (0.0)
timestamp: DateTime (0.0)
begin: DateTime (0.0)
end: DateTime (0.0)
altitudeMode: String (0.0)
tessellate: Integer (0.0)
extrude: Integer (0.0)
visibility: Integer (0.0)

Mapfile Example

LAYER
NAME "lunenburg"
TYPE POLYGON
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "Lunenburg_Municipality.kmz"
DATA "Lunenburg_Municipality"
CLASS
NAME "Lunenburg"
STYLE
COLOR 244 244 16
OUTLINECOLOR 199 199 199
END
END
END # layer

280 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

7.1.11 MapInfo

File listing

The following files are also associated with .TAB files: .DAT, .ID, .MAP. An example is:
border.DAT
border.ID
border.MAP
border.TAB

The term MID/MIF refers to files with .MID and .MIF extension.

Data Access / Connection Method

TAB and MID/MIF access is available in MapServer through OGR.

• The CONNECTIONTYPE OGR parameter must be used.
• The path to the (*.tab or *.mif) file is required, and the file extension is needed.
• The path may be relative to the SHAPEPATH
• MapInfo files already contain styling information. This styling information can be used optionally by specifying
the STYLEITEM “AUTO” parameter in the LAYER object of the map file.
Note: If you use STYLEITEM “AUTO” you must have an empty class in the layer.

OGRINFO Examples

Using ogrinfo on a single TAB file

> ogrinfo elev5_poly.TAB
Had to open data source read-only.
INFO: Open of ‘elev5_poly.TAB’
using driver ‘MapInfo File’ successful.
1: elev5_poly (Polygon)

Using ogrinfo to examine the structure of the file/layer

> ogrinfo elev5_poly.TAB elev5_poly
Had to open data source read-only.
INFO: Open of ‘elev5_poly.TAB’
using driver ‘MapInfo File’ successful.

Layer name: elev5_poly

Geometry: Polygon
Feature Count: 2236
Extent: (-141.000000, 60.000000) - (-124.403310, 69.300251)
Layer SRS WKT:
GEOGCS["unnamed",
DATUM["MIF 0",
SPHEROID["WGS 84 (MAPINFO Datum 0)",6378137.01,298.257223563],
TOWGS84[0,0,0,0,0,0,0]],
PRIMEM["Greenwich",0],
UNIT["degree",0.0174532925199433]]
AREA: Real (0.0)
PERIMETER: Real (0.0)

7.1. Vector Data 281

MapServer Documentation, Release 6.0

ELEV5_: Integer (0.0)

ELEV5_ID: Integer (0.0)
TYPE: Real (4.0)
ELEV5: Real (4.0)
...

Map File Example

LAYER
NAME Elevation_Poly_5
TYPE POLYGON
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "./hypso/elev5_poly.TAB"
STYLEITEM "AUTO"
CLASS
NAME "Elevation Poly 5"
END
END # Layer

7.1.12 MSSQL

Author Tamas Szekeres

Contact szekerest at gmail.com
Last Updated 2010-10-01

Contents
• MSSQL
– Introduction
– Creating spatial data tables in MSSQL 2008
– Connecting to Spatial Data in MSSQL 2008
* Create MapServer Layer
* Selecting the type of the geometry column
* The expected location on the plugin dll
* Using spatial indexes
* Layer Processing Options
– Obtaining binaries containing the MSSQL plugin dll
– More Information

Introduction

Microsoft SQL Server 2008+ supports storing spatial data by using the built in geometry/geography data types.
MapServer supports a driver to access these tables containing spatial columns, which is compiled as a plugin (msplu-
gin_mssql2008.dll).

Creating spatial data tables in MSSQL 2008

There are several ways to create spatial data tables in MSSQL 2008. Having an existing data you can easily upload
that to an MSSQL table by using the ogr2ogr commandline tool and the OGR’s MSSQL Spatial driver

282 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Connecting to Spatial Data in MSSQL 2008

In order to connect to the MSSQL 2008 spatial database you should set up a valid connection string to the database
like the following examples:
Server=.\MSSQLSERVER2008;Database=Maps;Integrated Security=true

Server=55.55.55.55,1433;uid=a_user;pwd=a_password;database=a_database;Integrated Security=True

Server=55.55.55.55\SQLEXPRESS,1433;uid=a_user;pwd=a_password;database=a_database;Integrated Security=

Create MapServer Layer

Once the connection can be established to the server the layer can be configured to access MSSQL2008 as follows:
LAYER
NAME "rivers_mssql_spatial"
TYPE POLYGON
STATUS DEFAULT
CONNECTIONTYPE PLUGIN
PLUGIN "msplugin_mssql2008.dll"
CONNECTION "Server=.\MSSQLSERVER2008;Database=Maps;Integrated Security=true"
DATA "ogr_geometry from rivers USING UNIQUE ogr_fid USING SRID=4326"
...
END

The DATA parameter is used to perform the SQL select statement to access your table in MSSQL. The geometry
column is required in the select statement; in the above example the ogr_geometry column is the geometry column in
the rivers table. The table should also have an unique column (ogr_fid) which is provided for random access to the
features in the feature query operations.
The DATA section should also contain the spatial reference id (SRID) of the features in the data table The SRID is used
when specifying the search shapes during the intersect operations which should match with the SRID of the features
otherwise no features are returned in a particular query. if you omit specifying the SRID value in the DATA section
the diver will use SRID=0 when defining the search shapes.

Selecting the type of the geometry column

For the geometry columns MSSQL supports 2 data types: “geometry” and “geography”. By default the driver con-
siders the type of the geometry column is “geometry”. In case if the type of the geometry column is “geography” we
must specify the data type in the DATA section explicitly, like:
DATA "ogr_geometry(geography) from rivers USING UNIQUE ogr_fid USING SRID=4326"

The expected location on the plugin dll

On Windows platforms the DLLs needed by the program are searched for in the following order:
1. The directory from which the application loaded.
2. The current directory.
3. The system directory. Use the GetSystemDirectory function to get the path of this directory.
4. The 16-bit system directory.

7.1. Vector Data 283

MapServer Documentation, Release 6.0

5. The Windows directory. Use the GetWindowsDirectory function to get the path of this directory.
6. The directories that are listed in the PATH environment variable.

Using spatial indexes

In order to speed up the access to the features a spatial index should be created to the geometry column which could
easily be done with the OGR MSSQL Spatial driver like:
ogrinfo -sql "create spatial index on rivers" "MSSQL:server=.\MSSQLSERVER2008;database=Maps;Integrate

In general we can safely rely on the query optimizer to select the most appropriate index in the sql query operations.
In some cases - however - we should force the optimizer to use the spatial index by specifying the index hint in the
DATA section like:
DATA "ogr_geometry from rivers using index ogr_geometry_sidx USING UNIQUE ogr_fid USING SRID=4326"

Layer Processing Options

We can control the behaviour of the MSSQL driver by using the following PROCESSING options:
• CLOSE_CONNECTION=DEFER - This is where you can enable connection pooling for certain layer types.
Connection pooling will allow MapServer to share the handle to an open database or layer connection throughout
a single map draw process.

Obtaining binaries containing the MSSQL plugin dll

Currently the following binary distributions contain msplugin_mssql2008.dll

• MapServer and GDAL binary and SDK packages
• MS4W distributions

More Information

• OGR MSSQL Spatial driver page (describes the OGR MSSQL support)
• ogr2ogr application (describes the ogr2ogr commandline application)
• Vector Data (MapServer Vector Data Access Guide)

7.1.13 MySQL

Author David Fawcett

Contact david.fawcett at moea.state.mn.us
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-07-29

284 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Contents
• MySQL
– Introduction
– Connecting to Spatial Data in MySQL
* Requirements
* Verify MySQL Support in OGR Build
* Test Connection with ogrinfo
* Create MapServer Layer
– Connecting to non-Spatial Data in MySQL
* Requirements
* Create .ovf file
* Test Connection with ogrinfo
* Create MapServer Layer
– More Information

Introduction

The following methods connect to MySQL through OGR’s MySQL driver, thus avoiding the need to set up an ODBC
connection.

Connecting to Spatial Data in MySQL

This section describes how to display a spatial MySQL table (meaning that the table has a column of type geometry)
in MapServer. OGR’s MySQL driver was expanded in OGR version 1.3.2 to support access to MySQL spatial tables.

Requirements

• MapServer compiled with OGR support

• OGR/GDAL version 1.3.2 or more recent compiled with MySQL support

Verify MySQL Support in OGR Build

You can verify that your local build of OGR contains MySQL support by using the ogrinfo commandline utility, and
making sure that “MySQL” is returned:
ogrinfo --formats

Supported Formats:
-> "ESRI Shapefile" (read/write)
-> "MapInfo File" (read/write)
...
-> "PostgreSQL" (read/write)
-> "MySQL" (read/write)
...

7.1. Vector Data 285

MapServer Documentation, Release 6.0

Test Connection with ogrinfo

MySQL connection strings in OGR are in the following format:

MYSQL:database,host=yourhost,user=youruser,password=yourpass,tables=yourtable

Therefore an example ogrinfo command would be:

> ogrinfo MYSQL:test,user=root,password=mysql,port=3306

which should return a list of all of your tables in the ‘test’ database:
INFO: Open of ‘MYSQL:test,user=root,password=mysql,port=3306’
using driver ‘MySQL’ successful.
1: province (Polygon)

and you can return a summary of the MySQL spatial layer:

> ogrinfo MYSQL:test,user=root,password=mysql,port=3306 province -summary

INFO: Open of ‘MYSQL:test,user=root,password=mysql,port=3306’

using driver ‘MySQL’ successful.

Layer name: province

Geometry: Polygon
Feature Count: 48
Extent: (-13702.315770, 3973784.599548) - (1127752.921471, 4859616.714055)
Layer SRS WKT:
PROJCS["ED50_UTM_zone_30N",
...
FID Column = OGR_FID
Geometry Column = SHAPE
id: Real (2.0)
...

Create MapServer Layer

LAYER
NAME "spain_provinces_mysql_spatial"
TYPE POLYGON
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "MySQL:test,user=root,password=mysql,port=3306"
DATA "SELECT SHAPE,admin_name from province"
LABELITEM "admin_name"
CLASS
NAME "Spain Provinces"
STYLE
COLOR 240 240 240
OUTLINECOLOR 199 199 199
END
LABEL
COLOR 0 0 0
FONT sans
TYPE truetype
SIZE 8

286 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

POSITION AUTO
PARTIALS FALSE
OUTLINECOLOR 255 255 255
END
END
END # layer

The DATA parameter is used to perform the SQL select statement to access your table in MySQL. The geometry
column is required in the select statement; in the above example the SHAPE column is the geometry column in the
province table.

Connecting to non-Spatial Data in MySQL

This section describes how to display a non-spatial MySQL table (meaning the table does not have a column of type
geometry) in MapServer.
Support for this functionality is found in GDAL/OGR 1.2.6 and older on Windows and GDAL/OGR 1.3.2 on Linux.

Requirements

• MySQL database containing a table with fields containing x and y coordinates

• .ovf file, a small xml file you will create
• MapServer compiled with OGR version supporting this functinality

Create .ovf file

Here is the .ovf file named aqidata.ovf

<OGRVRTDataSource>
<OGRVRTLayer name="aqidata">
<SrcDataSource>MYSQL:aqiTest,user=uuuuu,password=ppppp,host=192.170.1.100,port=3306,tables=te
<SrcSQL>SELECT areaID, x, y, sampleValue FROM testdata</SrcSQL>
<GeometryType>wkbPoint</GeometryType>
<GeometryField encoding="PointFromColumns" x="x" y="y"/>
</OGRVRTLayer>
</OGRVRTDataSource>

If you look at the connection string in <SrcDataSource>

• The MySQL database name is ‘aqiTest’
• ‘testdata’ is the table containing the coordinate data
• host and port are for MySQL server
Use the GeometryField element to tell OGR which fields store the x and y coordinate data. Mine are simply named x
and y.

Test Connection with ogrinfo

# usr/local/bin/ogrinfo /maps/aqidata.ovf

ogrinfo returns

7.1. Vector Data 287

MapServer Documentation, Release 6.0

ERROR 4: Update access not supported for VRT datasources.

Had to open data source read-only.
INFO: Open of ‘/maps/aqidata.ovf’
using driver ‘VRT’ successful.
1: aqidata (Point)

Don’t worry about the error, this is just telling you that it is a read-only driver. If it really bugs you, call ogrinfo with
the -ro (read only) flag.
To see the actual data
# usr/local/bin/ogrinfo /maps/aqidata.ovf -al

Create MapServer Layer

LAYER
NAME "MyAqi"
STATUS DEFAULT
TYPE POINT
CONNECTIONTYPE OGR
CONNECTION "aqidata.ovf"
DATA "aqidata"
CLASS
NAME "MyClass"
STYLE
SYMBOL ’circle’
SIZE 15
COLOR 0 255 0
END
END
END

DATA in the LAYER definition should be the same as the name attribute of the OGRVRTLayer element in the ovf file.
For this to draw, you need to have a SYMBOLSET defined in your mapfile and have a symbol called ‘circle’ in your
symbols.sym file.

More Information

• OGR (MapServer OGR document)

• Vector Data (MapServer Vector Data Access Guide)
• MySQL wiki page (describes the deprecated mygis support)

7.1.14 NTF

NTF files are mostly used by the United Kingdom Ordnance Survey (OS). For more on the Ordnance Survey, see their
website at: http://www.ordnancesurvey.co.uk/oswebsite/

File listing

NTF files have an NTF extension.

288 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Data Access / Connection Method

• NTF access requires OGR.

• The path to the NTF file is required in the CONNECTION string. It may be relative to the SHAPEPATH setting
in the map file or the full path.
• The DATA parameter is used to specify the layer to use

OGRINFO Examples

Using ogrinfo on an NTF file to retrieve layer names:

> ogrinfo llcontours.ntf
ERROR 4: NTF Driver doesn’t support update.
Had to open data source read-only.
INFO: Open of ‘llcontours.ntf’
using driver ‘UK .NTF’ successful.
1: LANDLINE_POINT (Point)
2: LANDLINE_LINE (Line String)
3: LANDLINE_NAME (Point)
4: FEATURE_CLASSES (None)

Using ogrinfo to examine the structure of an NTF layer:

> ogrinfo llcontours.ntf LANDLINE_LINE -summary
ERROR 4: NTF Driver doesn’t support update.
Had to open data source read-only.
INFO: Open of ‘llcontours.ntf’
using driver ‘UK .NTF’ successful.

Layer name: LANDLINE_LINE

Geometry: Line String
Feature Count: 491
Extent: (279000.000000, 187000.000000) - (280000.000000, 188000.000000)
Layer SRS WKT:
PROJCS["OSGB 1936 / British National Grid",
GEOGCS["OSGB 1936",
DATUM["OSGB_1936",
SPHEROID["Airy 1830",6377563.396,299.3249646,
AUTHORITY["EPSG","7001"]],
AUTHORITY["EPSG","6277"]],
PRIMEM["Greenwich",0,
AUTHORITY["EPSG","8901"]],
UNIT["degree",0.0174532925199433],
AUTHORITY["EPSG","4277"]],
PROJECTION["Transverse_Mercator"],
PARAMETER["latitude_of_origin",49],
PARAMETER["central_meridian",-2],
PARAMETER["scale_factor",0.999601272],
PARAMETER["false_easting",400000],
PARAMETER["false_northing",-100000],
UNIT["metre",1,
AUTHORITY["EPSG","9001"]],
AUTHORITY["EPSG","27700"]]
LINE_ID: Integer (6.0)
FEAT_CODE: String (4.0)
...

7.1. Vector Data 289

MapServer Documentation, Release 6.0

Map File Example:

LAYER
NAME ntf_uk
TYPE LINE
CONNECTIONTYPE OGR
CONNECTION "./ntf/llcontours.ntf"
DATA "LANDLINE_LINE"
STATUS DEFAULT
CLASS
NAME "Contours"
STYLE
COLOR 0 150 200
END
END
END

7.1.15 OGR

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-10-16

Table of Contents
• OGR
– Introduction
– What is OGR?
– Obtaining and Compiling MapServer with OGR Support
– Integrating OGR Support with MapServer Applications
– STYLEITEM “AUTO” - Rendering Layers Using Style Information from the OGR File
– Sample Sites Using OGR/MapServer
– FAQ / Common Problems

Introduction

Starting with version 3.5, MapServer included the ability to access vector data sets in formats other than Shapefile
in their native format using the OGR library. The following document describes the process for implementing OGR
support within MapServer applications.
Note: Experimental OGR support was included in MapServer version 3.4 but this initial implementation had some
limitations and is not covered in this document.
This document assumes that you are already familiar with certain aspects of MapServer:
• MapServer application development and especially setting up .map files.
• Some compilation skills if you don’t have ready access to a pre-compiled installation and need to compile your
own copy of MapServer with OGR support.
• access to OGR utilities, such as ogrinfo, which are available in the FWTools and MS4W packages.
Readers should also check out the Vector Data Access Guide, which has lots of examples of how to access specific
vector formats.

290 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

What is OGR?

The OGR Simple Features Library is a C++ open source library (and command-line tools) providing read (and some-
times write) access to a variety of vector file formats including ESRI Shapefiles, and MapInfo mid/mif and TAB
formats.
OGR is actually part of the GDAL library, so you will notice that some references point to GDAL (such as the mailing
list).

What Does OGR Add to MapServer?

The OGR Simple Features Library allows MapServer users to display several types of vector data files in their native
formats. For example, MapInfo Mid/Mif and TAB data do not need to be converted to ESRI shapefiles when using
OGR support with MapServer.

What Data Formats are Supported?

See http://www.gdal.org/ogr/ogr_formats.html for the latest list of supported formats. At the date this document was
written, the following formats were supported:
• ArcInfo Binary Coverages
• ArcInfo E00 Coverages
• Atlas BNA
• Comma Separated Value (.csv)
• DODS/OPeNDAP
• ESRI ArcSDE
• ESRI Personal GeoDatabase
• ESRI Shapefiles
• FMEObjects Gateway
• Géoconcept Export
• GeoJSON
• GeoRSS
• GML
• GMT
• GRASS
• GPX
• Informix DataBlade
• INGRES
• INTERLIS
• KML
• MapInfo files
• Memory

7.1. Vector Data 291

MapServer Documentation, Release 6.0

• Microstation DGN files

• MySQL
• ODBC
• OGDI Vectors
• Oracle Spatial
• PostgreSQL
• SDTS
• SQLite
• UK.NTF (National Transfer Format)
• US Census TIGER/Line
• VRT - Virtual Datasource
• X-Plane/Flighgear aeronautical data
Note: Some of the above formats (e.g. OGDI) have external dependencies and are not always included in the
pre-compiled binary distributions of MapServer with OGR support.*
Note: Some of the above formats are not well suited for random access by nature, that’s the case of MapInfo MIF/MID
files which is a TEXT format and will give very poor performance for a web application. On the other hand, some
binary formats such as MapInfo TAB are better suited for random access and will give performance comparable to
native shapefile access in MapServer.*

How to Get More Information on the OGR Project

• More information on the OGR Simple Features Project can be found at http://www.gdal.org/ogr/.
• The GDAL mailing list can be used for OGR related questions. Always search the list archives before sending
new questions.
• The GDAL Wiki has lots of good information for users and developers.
• The #gdal IRC channel on irc.freenode.net might also be of help. For info on IRC see the MapServer IRC page.
The main developer of the OGR library is Frank Warmerdam and the integration of OGR within MapServer was done
by Daniel Morissette.

Obtaining and Compiling MapServer with OGR Support

• Follow the instructions on the OGR page to compile/install OGR/GDAL.

• Obtain the MapServer source.
For UNIX users, see the README.CONFIGURE file in the MapServer source, or see the UNIX Compilation and
Installation. If GDAL/OGR is normally installed it should be sufficient to add –with-ogr to the configure line before
(re)building MapServer. Linux users might want to try FGS, a Linux installer for MapServer.
For Windows users, it is recommended to look for a pre-compiled binary on the MapServer site (MS4W is recom-
mended). If you want to compile your own then see the README.WIN32 file in the MapServer source.

292 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Integrating OGR Support with MapServer Applications

The only change that is needed to integrate OGR support with a MapServer application is with the .map file.
The LAYER’s DATA parameter is expanded to three parameters (CONNECTIONTYPE OGR, CONNECTION and
DATA).
The syntax for this differs depending on the type of data being used (the Vector Data Access Guide is an excellent
resource for this). In OGR, a data source can be either a set of files that share a common basename (e.g. .shp/.shx/.dbf
for ArcView Shapefiles, or .tab/.map/.dat/.ind/.id for MapInfo TAB files) or a whole directory of files (e.g. TIGER).
Let’s call the former “File-based data sources” and the later “Directory-based data sources”. When accessing a file-
based data source you specify the filename of one of the files in the set (e.g. roads.shp or roads.tab) and when
accessing a directory-based data source you specify the directory name and OGR reads all the files in the directory
as a single data source with potentially several layers (e.g. TIGER files).
Some OGR drivers (e.g. SHP, TAB) can have dual behaviors, that is if they’re pointed to a single file then they behave
as a file-based data source and if they’re pointed to a directory then they will behave as a directory-based data source
and then every file in the directory becomes a new layer in the data source.
See the OGR formats page for more info on the specific file format you’re using. (Click on the format name for more
specific driver info on that format)

Using OGR Data Sources in the Map File

The .map file LAYER definition for file-based sources is as follows:

LAYER
...
CONNECTIONTYPE OGR
CONNECTION "<datasource_name>"
DATA "<layer_definition>"
...
END

<datasource_name> is the name of the datasource to read from and is prefixed by the CONNECTION keyword. The
exact organization depends on the format driver in use. The format driver to use is automatically selected by OGR
based on the nature of the string passed as the datasource, and/or the format of the file referenced by it.
• For file based datasources this is the name of the file, including the extension, using an absolute path, or a relative
path. Relative paths are interpreted relative to the SHAPEPATH first, if not found then we try again relative to
the .map file location.
Note: Before version 4.1 the SHAPEPATH was ignored for OGR datasources.
• For directory based datasources, such as TIGER/Line, or Arc/Info Binary Coverages this is the name of the
directory containing the files. If the path is relative it is interpreted relative to the .map file.
• For virtual datasources such as database systems, and OGDI this is the service connection string
and is generally not related to the filesystem. For instance, for Oracle Spatial this might be
“OCI:warmerda/[email protected]”.
<layer_definition> is the name, number or SQL definition of the layer to use from the datasource. It is indicated via
the DATA keyword in the map file.
• Layer Name: The (case insenstive) layer name may be used to select a layer.
• Layer Number: The layer number (starting from 0 for the first layer) may be used to select a layer. Generally
the layer name is preferred to this since it is more self describing.
• Omitted: If no DATA keyword is provided, this is equivalent to selecting layer 0.

7.1. Vector Data 293

MapServer Documentation, Release 6.0

• SQL SELECT: If an SQL SELECT statement is used, it is interpreted in a driver specific manner to try and
generate a temporary pseudo-layer. For some formats this a restricted subset of SQL is interpreted within OGR.
For RDBMS based drivers (such as PostGIS and Oracle) this is passed through to the underlying database.
The OGRINFO utility can be used to find out the list of layers and their names in a data source.

Examples of Layer Definitions Using OGR

Please see the Vector Data Access Guide for details and examples of each data format supported.
Example 1. MapInfo TAB file
LAYER
NAME "Builtup_Areas_tab"
TYPE POLYGON
CONNECTIONTYPE OGR
CONNECTION "data/tab/092b06_builtup_a.tab"
STATUS ON
CLASS
...
END
...
END

Example 2. Microstation DGN file using <layer_index>

The entire DGN file is represented in OGR as one layer (see the DGN driver page for more details):
LAYER
NAME "dgn"
TYPE LINE
CONNECTIONTYPE OGR
CONNECTION "dgn/santabarbara02.dgn"
DATA "0"
STATUS ON
STYLEITEM "AUTO"
CLASS
END
END # Layer

Example 3. TIGER/Line file using <layer_name>

LAYER
NAME "Roads_tig"
TYPE line
CONNECTIONTYPE OGR
CONNECTION "full/path/to/tiger/TGR25001"
DATA "CompleteChain"
STATUS ON
CLASS
...
END
END

Example 4. Directory of Shapefiles using SQL JOIN

LAYER
NAME "Parks_cov"
TYPE POLYGON

294 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

CONNECTIONTYPE OGR
CONNECTION "data/shppoly"
DATA "SELECT eas_id, idlink.Name FROM poly LEFT JOIN idlink ON poly.eas_id = idlink.eas_id"
STATUS ON
CLASSITEM "idlink.Name"
CLASS
...
END
END

How to Use “OGRINFO”

OGRINFO is part of the GDAL/OGR distribution (it is also included in FWTools and MS4W). It is an executable that
can be used to obtain layer information about OGR supported files. The parameters are:
ogrinfo [-ro] [-q] datasource_name [layer [layer...]]
• -ro opens the file as read only (optional)
• -q executes in quiet mode, only the layer idex line will be returned (optional)
• datasource_name is the filename including extension (eg. roads.tab); for TIGER/Line files, data-
source_name is the directory containing the TIGER files (eg. ogrinfo TGR25001)
Example 5. To get the list of layers in a file:
$ ogrinfo popplace.tab

Had to open data source read-only.

INFO: Open of ‘popplace.tab’
using driver ‘MapInfo File’ successful.
1: popplace (Point)

which shows that there is one point layer in the popplace.tab file.
Example 6. To get a dump of a specific layer, including field names, projection, etc:
$ ogrinfo popplace.tab popplace

Had to open data source read-only.

INFO: Open of ‘popplace.tab’
using driver ‘MapInfo File’ successful.

Layer name: popplace

Geometry: Point
Feature Count: 497
Layer SRS WKT: PROJCS["unnamed",GEOGCS["unnamed",DATUM["North ...snipped...
AREA: Real (15.3)
PERIMETER: Real (15.3)
POPPLACE_: Real (11.0)
POPPLACE_I: Real (15.0)
NAME: String (50.0)
OGRFeature(popplace):1
AREA (Real) = 0.000
PERIMETER (Real) = 0.000
POPPLACE_ (Real) = 1
POPPLACE_I (Real) = 1
NAME (String) = Port Hope Simpson
POINT (2437287.249 1153656.751)

7.1. Vector Data 295

MapServer Documentation, Release 6.0

OGRFeature(popplace):2
AREA (Real) = 0.000
PERIMETER (Real) = 0.000
POPPLACE_ (Real) = 2
POPPLACE_I (Real) = 1
NAME (String) = Hopedale

...
...

Example 7. To get a list of layers in a TIGER/Line Directory:

$ ogrinfo TGR25001

Had to open data source read-only.

INFO: Open of ‘TGR25001’
using driver ‘TIGER’ successful.
1: CompleteChain (Line String)
2: AltName (None)
3: FeatureIds (None)
4: ZipCodes (None)
5: Landmarks (Point)
6: AreaLandmarks (None)
7: KeyFeatures (None)
8: Polygon (None)
9: EntityNames (Point)
10: IDHistory (None)
11: PolyChainLink (None)
12: PIP (Point)
13: TLIDRange (None)
14: ZipPlus4 (None)

The above example shows that there are 14 layers in the TGR25001 directory.
Example 8. To get a summary of a specific TIGER layer, including only field names, projection, and extent
$ ogrinfo TGR25001 Landmarks -summary

Had to open data source read-only.

INFO: Open of ‘TGR25001’
using driver ‘TIGER’ successful.

Layer name: Landmarks

Geometry: Point
Feature Count: 777
Extent: (-70.674324, 41.519817) - (-69.969211, 42.046868)
Layer SRS WKT: GEOGCS["NAD83",DATUM["North_American_Datum_1983",
SPHEROID["GRS 1980",6378137,298.257222101]],PRIMEM["Greenwich",0],
UNIT["degree",0.0174532925199433]]
MODULE: String (8.0)
FILE: String (5.0)
STATE: Integer (2.0)
COUNTY: Integer (3.0)
LAND: Integer (10.0)
SOURCE: String (1.0)
CFCC: String (3.0)
LANAME: String (30.0)

296 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Queries Through OGR Format

OGR layers can be queried the same way as regular shapefiles in MapServer.

TILEINDEX with OGR

OGR layers can utilize tile indexes in a similar fashion to Shapefile based layers. The TILEINDEX keyword should
contain the connection string for the tile index file. The tile index file may be any supported OGR format, including
shapefiles.
The TILEITEM keyword in the LAYER definition indicates what attribute from the tile index file should be used as the
datasource location. If omitted, the default TILEITEM value is “location”. The value in the location field should be a
connection string the same as would have been used in the CONNECTION field for OGR layers. The CONNECTION
keyword is not needed (and will be ignored) for layers using the OGR connection type and having the TILEINDEX
keyword.
Tileindex files can be prepared in an external GIS, or using the OGR utility ogrtindex. Details can be found on the
OGR Utilities Page.
The following is a simple example of a point layer using a tile index.
LAYER
NAME "ogr_points"
TYPE POINT
CONNECTIONTYPE OGR
TILEINDEX "PIP_ogr_tiles.shp,0"
STATUS ON
CLASS
NAME "points"
STYLE
SYMBOL "default-circle"
COLOR 255 0 0
SIZE 6
END
END
END

OGR tileindex layers should support all normal query and attribute fetching mechanisms, including from MapScript;
however, this has not been heavily tested as of April/2002. Please report problems via the MapServer Trac. If auto
projection support is used for tileindexed OGR layers, the tileindex is read for the projection (not the component tiles).
Problems may (or may not) be encountered if the component tiles have differing schemas (different sets of attributes).

Connection Pooling

For some OGR supported formats, connecting to the dataset is quite expensive in terms of CPU use and amount of
disk IO. For instance, establishing access to an S-57 dataset results in a complete read into memory of the data files.
Connection pooling control aims at reducing this overhead in situations where the same file is used for several different
map layers.
To ensure that an OGR supported dataset is only opened once per map render (instead of separately for each map
LAYER referencing the dataset, use the CLOSE_CONNECTION PROCESSING option. The default value is for
CLOSE_CONNECTION is NORMAL, but if set to DEFER the dataset will be kept open till the map render is
complete. It will be reused by any other layers with using the same datasource.
Example 9. Preserve S-57 connection for two layers

7.1. Vector Data 297

MapServer Documentation, Release 6.0

In this example, we are using the same dataset (NO410810.000) for two layers. To avoid re-reading the dataset, we
mark the first layer to defer closing the connection till layer. In the second (or last) layer we request NORMAL
connection handling (though this could have been left out as normal handling is the default).
LAYER
NAME "AdminAreas"
TYPE POLYGON
CONNECTIONTYPE OGR
CONNECTION "NO410810.000"
DATA "ADMARE"
PROCESSING "CLOSE_CONNECTION=DEFER"
STATUS ON
...
END
LAYER
NAME "Land Areas"
TYPE POLYGON
CONNECTIONTYPE OGR
CONNECTION "NO410810.000"
DATA "LNDARE"
PROCESSING "CLOSE_CONNECTION=NORMAL"
STATUS ON
...
END

1. The text of the CONNECTION keyword must match exactly between layers for the connection to be reused.
2. Some dataset connections are quite memory expensive, and keeping them open may result in increased memory
use.
3. If all layers rendered for a particular connection defer closing the connection, it will remain open till MapServer
terminates. For normal cgi or MapScript use this is likely OK.
4. This use of CLOSE_CONNECTION handling is unique to OGR layers, and may be changed at some point in
the future as part of a broader implementation of connection pooling in MapServer.

STYLEITEM “AUTO” - Rendering Layers Using Style Information from the OGR File

Note: This feature is only supported with MapInfo TAB and Microstation DGN files at the moment, but eventually
other formats that carry colors and styles at the shape-level may also be supported through OGR.*
In MapServer, ArcView, and other shapefile-based applications, colors and styles are usually defined at the layer level.
This means that all the shapes in a given layer are usually rendered using the same color and styles.
On the other hand, some formats supported by OGR such as MapInfo TAB do have color and style information attached
to each shape. OGR adds support for the ‘STYLEITEM “AUTO”’ layer parameter which allows you to request that the
shapes in a layer be rendered using colors and styles coming from the data source instead of being driven by CLASSes
as was traditionally done with MapServer.

How to Implement

In order to have a layer rendered using colours and styles coming from the OGR data source, your must do the
following:
• Your layer definition must contain the STYLEITEM “AUTO” parameter.

298 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

• Your layer definition needs to contain at least one CLASS (which may be empty) and optionally a CLASSITEM
to match the expressions if your CLASS contains an expression. The empty CLASS in the layer will be updated
dynamically at runtime to contain colours and styles coming from the data source for each shape.

Examples

Example 10. Layer Definition Using STYLEITEM “AUTO” without a CLASSITEM

LAYER
NAME "test_dgn"
STATUS ON
TYPE POLYGON
CONNECTIONTYPE OGR
CONNECTION "../data/dgn/test.dgn"

# This enables use of colors and styles from the source file.
STYLEITEM "AUTO"

# Define an empty class that will be filled at runtime from the color and
# styles read on each shape in the source file.
CLASS
END
END # layer

Example 11. Layer Definition Using STYLEITEM “AUTO” with a CLASSITEM

LAYER
NAME "Builtup_Areas_tab"
TYPE POLYGON
CONNECTIONTYPE OGR
CONNECTION "data/tab/092b06_builtup_a.tab"
STATUS ON

# This enables use of colors and styles from the source file.
STYLEITEM "AUTO"

# Define an empty class that will be filled at runtime from the color and
# styles read on each shape in the source file.
CLASSITEM "CATEGORY"
CLASS
EXPRESSION "1"
END
END

Please Note:
CLASS EXPRESSIONs are still working, so it is still possible to query and classify layers that are using STYLEITEM
“AUTO”. The only difference is that instead of using static class definitions, the colors and style will be read from the
data file.

Important Notes

NOTE 1 Even though MapInfo and other OGR data sources may support layers with mixed geometry
types (e.g. points, lines and polygons in the same file) this is not yet supported in MapServer. So
you still have to define a layer ‘TYPE’ and make sure that all the shapes in the OGR data source

7.1. Vector Data 299

MapServer Documentation, Release 6.0

are compatible with that layer type, otherwise MapServer may produce an error about incompatible
geometry types at runtime.
NOTE 2 Due to the dynamic nature of this feature, it is not compatible with the labelcache, so the label-
cache is automatically disabled for layers that make use of ‘STYLEITEM “AUTO”’.
NOTE 3 When you use STYLEITEM AUTO, MapServer tries to match symbol names returned by OGR
to names in your symbol file. For a quick solution, try using the following symbol file:
http://demo.mapserver.org/ogr-demos/yk_demo/etc/symbols_mapinfo.txt
The name of the symbols returned by OGR to MapServer depends on the file format. In the case of MapInfo files, it
will be:
• For “old-style” symbols (default MapInfo 3.0 symbols numbered 32 to 67) the symbol name will be ‘mapinfo-
sym-##’ where ‘##’ is the symbol number, e.g. ‘mapinfo-sym-32’.
• For “Font Symbols”, the symbol name is also ‘mapinfo-sym-##’ where ‘##’ is the symbol number in the font.
In this case, the name of the font itself is ignored by MapServer.
• MapInfo also supports “custom symbols” (bitmap symbols)... I’m not sure what you would get from OGR for
this, but I’m pretty sure that MapServer doesn’t do anything useful with them.
The OGRINFO utility can be used to find out exactly which symbol names OGR will return to MapServer. Look at
the “Style” string in the ogrinfo output for each shape that is read.

Mapping of OGR Style Info to the MapServer CLASS Members

Here is the list of style parameters that are currently supported from OGR data sources and how they’re mapped in
MapServer:
Line color The line colour is mapped to CLASS.COLOR
Line thickness The line thickness is mapped to CLASS.STYLE.WIDTH. The default will be 1 pixel line (as it always
is with MapServer).
Polygon fill color Polygon fill color is mapped directly to CLASS.COLOR
Note that at this time, transparent polygons are not supported (they’re always opaque).
Polygon outline If a polygon has an outline color and thickness defined in the data source then the same rule as for
line color and thickness above will apply, except that the outline color is mapped to CLASS.OUTLINECOLOR
Point symbols Point symbol color is directly mapped to CLASS.COLOR Point symbol size is directly mapped to
CLASS.SIZE
If your symbolset contains a symbol called “default-marker” then this symbol will be used, otherwise the default
will be CLASS.SYMBOL=0 (i.e. a 1 pixel dot)
It is also possible (with a bit of work) to control which symbol gets used in rendering point symbols. OGR
provides MapServer with symbol names, and if the symbol name returned by OGR to MapServer matches the
name of one of the symbols in your symbolset then this symbol will be used.
For MapInfo point symbols (numbered 32 to 67 in the MapInfo MIF spec), the name returned by OGR is
“mapinfo-sym-X” where X should be replaced with the MapInfo symbol number (e.g. “mapinfo-sym-35” is the
star symbol).
Text labels The text string is mapped to CLASS.TEXT
Text color is mapped to CLASS.LABEL.COLOR
Text background color is mapped to CLASS.LABEL.BACKGROUNDCOLOR

300 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Text height is mapped to CLASS.LABEL.SIZE

Text angle is mapped to CLASS.LABEL.ANGLE
Text font mapping follows the following rules:
1. If TTF fonts are supported:
(a) If the native font name (e.g. “Arial”) is found in your fontset then this font will be used.
(b) If 1a. failed and a font called “default” is present in your fontset then this “default” font will be used.
2. If TTF fonts are not supported or if all above cases failed, then BITMAP MEDIUM font will be used.

Accessing OGR STYLEITEMAUTO Label Styles Through MapScript

OGR STYLEITEMAUTO label styles can be accessed through MapScript, such as PHP/MapScript’s getshape() or
getvalue() methods, by setting the LAYER’s PROCESSING parameter to “GETSHAPE_STYLE_ITEMS=all”. There-
fore, the LAYER may contain:
LAYER
...
PROCESSING "GETSHAPE_STYLE_ITEMS=all"
...
END

The following label styles are supported:

Label Style Description MapServer Version
Implemented
OGR:LabelFont Comma-delimited list of fonts names 5.4
OGR:LabelSize Numeric value with units 5.2.0
OGR:LabelText Label text string 5.2.0
OGR:LabelAngle Rotation angle (in degrees) 5.2.0
OGR:LabelFColorForeground color 5.4
OGR:LabelBColorBackground color 5.4
OGR:LabelPlacement
How is the text drawn relative to the feature’s geometry 5.4
OGR:LabelAnchorA value from 1 to 12 defining the label’s position relative to the 5.4
point to which it is attached.
OGR:LabelDx X offset 5.4
OGR:LabelDy Y offset 5.4
OGR:LabelPerp Perpendicular offset 5.4
OGR:LabelBold Bold text 5.4
OGR:LabelItalic Italic text 5.4
OGR:LabelUnderline
Underlined text 5.4
OGR:LabelPriorityNumeric value defining the order in which style parts should be 5.4
drawn.
OGR:LabelStrikeout
Strike out text (gdal >= 1.4.0) 5.4
OGR:LabelStretchStretch factor changes the width of all characters in the font by 5.4
factor percent. (gdal >= 1.4.0)
OGR:LabelAdjHorHorizontally adjacent text (gdal >= 1.4.0) 5.4
OGR:LabelAdjVertVertically adjacent text (gdal >= 1.4.0) 5.4
OGR:LabelHColorShadow color (gdal >= 1.4.0) 5.4
OGR:LabelOColorOutline color (gdal > 1.6.0) 5.4
Please see the OGR Feature Style Specification document for more details on those specific styles.

7.1. Vector Data 301

MapServer Documentation, Release 6.0

Sample Sites Using OGR/MapServer

The following sites use OGR’s STYLEITEM “AUTO” feature:

• http://demo.mapserver.org/ogr-demos/yk_demo/demo_init.html
• http://demo.mapserver.org/ogr-demos/nfld_demo/demo_init.html
The following site uses OGR, as well as MapInfo’s ‘Seamless Map Layers’ feature:
• http://demo.mapserver.org/ogr-demos/ro_demo/demo_init.html
The following site uses OGR to display TIGER 2000 files:
• http://demo.mapserver.org/ogr-demos/tig_demo/demo_init.html

FAQ / Common Problems

Q What Does “OGR” Stand For?

A Basically, OGR does not stand for anything. For a detailed explanation of how OGR was named, see
GDAL’s FAQ at http://trac.osgeo.org/gdal/wiki/FAQ.

Q When using STYLEITEM AUTO, what should I have in my .sym symbols file?
A When you use STYLEITEM AUTO, MapServer tries to match symbol names returned by OGR to
names in your symbol file. For a quick solution, try using the following symbol file:
http://demo.mapserver.org/ogr-demos/yk_demo/etc/symbols_mapinfo.txt
The name of the symbols returned by OGR to MapServer depends on the file format. In the case of
MapInfo files, it will be:
• For “old-style” symbols (default MapInfo 3.0 symbols numbered 32 to 67) the symbol name
will be ‘mapinfo-sym-##’ where ‘##’ is the symbol number, e.g. ‘mapinfo-sym-32’.
• For “Font Symbols”, the symbol name is also ‘mapinfo-sym-##’ where ‘##’ is the symbol
number in the font. In this case, the name of the font itself is ignored by MapServer.
• MapInfo also supports “custom symbols” (bitmap symbols)... I’m not sure what you would get
from OGR for this, but I’m pretty sure that MapServer doesn’t do anything useful with them.
The OGRINFO utility can be used to find out exactly which symbol names OGR will return to
MapServer. Look at the “Style” string in the ogrinfo output for each shape that is read.

7.1.16 Oracle Spatial

Author Bart van den Eijnden

Last Updated 2005/12/12

302 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Table of Contents
• Oracle Spatial
– What MapServer 5.2 with Oracle Spatial
– Binaries
– Installation
– Two options for using Oracle Spatial with MapServer
– Mapfile syntax for native Oracle Spatial support
– Using subselects in the DATA statement
– Additional keywords - [FUNCTION]
– Additional keywords - [VERSION]
– More information
– Example of a LAYER
– Mapfile syntax for OGR Oracle Spatial support

Oracle Spatial is a spatial cartridge for the Oracle database. Remember that all Oracle databases come with Locator,
which has less features than Oracle Spatial. The differences between Locator and Spatial can be found in the Oracle
Spatial FAQ.
You can also see the original OracleSpatial wiki page that this document was based on.

What MapServer 5.2 with Oracle Spatial

• mode=map
• query modes: query, nquery, itemnquery
• MapScript query functions such as querybyattributes
• OGC:WMS: GetCapabilities, GetMap, GetFeatureInfo, DescribeLayer
• OGC:WFS, GetCapabilities, DescribeFeatureType, GetFeature

Binaries

MapServer Windows plugins with Oracle spatial support can be downloaded from MS4W. But you need Oracle client
software in the server on which you are running MapServer. Oracle client software can be obtained for development
purposes from the Oracle website, but you need to register, which by the way is free. The most recent version is
Oracle Database 10g Release 1 Client. The ORACLE TECHNOLOGY NETWORK DEVELOPMENT LICENSE
AGREEMENT applies to this software. The instant client will be satisfactory, and you can download the instant
client. Make sure though your MapServer is compiled against the same version as your Oracle client, for compiling
you need a full client install, not just the instant client.

Installation

See Oracle Installation for more configuration and installation information for MapServer’s native Oracle support
Note: If you receive error messages like “Error: .”. It’s likely related to MapServer being unable access or locate the
ORACLE_HOME.

Two options for using Oracle Spatial with MapServer

Oracle Spatial layers in MapServer can be used through 2 interfaces:

7.1. Vector Data 303

MapServer Documentation, Release 6.0

• The native built-in support through maporaclespatial.c

• OGR, but watch out: OGR is not compiled with Oracle Spatial support so it won’t work without compiling in
OCI (Oracle client) yourself. This requires both recompiling GDAL/OGR as well as recompiling MapServer
itself against the new GDAL/OGR !!!!

Mapfile syntax for native Oracle Spatial support

The DATA statement for a LAYER of CONNECTIONTYPE oraclespatial can now have 4 options. This change is
backwards compatible, i.e. the old ways of specifying DATA still work. The new options are an extension to the
old DATA statements, as they needed to include identification of the primary key to be used for the query modes
(UNIQUE).
The following options are valid DATA statements:
"[geom_column]
FROM
[table]| [(
SELECT [...]
FROM [table]|[Spatial Operator]
[WHERE condition] )]
[USING [UNIQUE column]
| [SRID #srid]
| [FUNCTION]
| [VERSION #version]
]"

Example 1

The most simple DATA statement, in this case you only need to define one geometry column and one table. This
option assumes you do not have an SRID defined.
LAYER
...
CONNECTIONTYPE oraclespatial
DATA "MYGEOMETRY FROM MYTABLE"
...
END

Example 2

It’s composed of the first option plus the USING UNIQUE parameter. These new features are necessary when you
want to use any query function. When it is used you must pass a numeric column type. This option assumes you do
not have an SRID defined.
LAYER
...
CONNECTIONTYPE oraclespatial
DATA "MYGEOMETRY FROM MYTABLE USING UNIQUE MYTABLE_ID"
...
END

304 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Example 3

This option is an extension to the first option. In this mode you must define the USING SRID parameter when the
SRID value in your data is different from NULL.
LAYER
...
CONNECTIONTYPE oraclespatial
DATA "MYGEOMETRY FROM MYTABLE USING SRID 90112"
...
END

Example 4

This option is a combination of examples 2 and 3.

LAYER
...
CONNECTIONTYPE oraclespatial
DATA "MYGEOMETRY FROM MYTABLE USING UNIQUE MYTABLE_ID SRID 90112"
...
END

Using subselects in the DATA statement

It is possible to define the source of the date as a subselect and not only as a table. As source of data, used in FROM
token, you can define any SQL, table, function, or operator that returns a SDO_GEOMETRY. For example:
DATA "[geom_column] FROM (SELECT [columns] FROM [table]|[Spatial function])"

If the LAYER definition contains a CLASSITEM, LABELITEM or FILTER, it is necessary that the fields used are
returned by the query. When you define CLASSITEM you can use an expression without any problems.

Additional keywords - [FUNCTION]

You can add three keywords to the DATA statement for [FUNCTION] option that influence the query which will be
executed in Oracle:

USING FILTER

"[geom_column] FROM [table]|([Subselect]) USING FILTER"

Using this keyword triggers MapServer to use the Oracle Spatial SDO_FILTER operator. This operator executes only
the Oracle Spatial primary filter over your query data. In the Oracle User guide they explain: The primary filter
compares geometric approximations, it returns a superset of exact result. The primary filter therefore should be as
efficient (that is, selective yet fast) as possible. This operator uses the spatial index, so you need to define your spatial
index correctly to retrieve an exact result. If the result of the query is not exact you can try the next option.

7.1. Vector Data 305

MapServer Documentation, Release 6.0

USING RELATE

"[geom_column] FROM [table]|([Subselect]) USING RELATE"

Using this keyword triggers MapServer to use the Oracle Spatial SDO_RELATE operator. This operator applies the
primary and secondary Oracle Spatial filters. It’s performance can be slightly slow but the result is extremely correct.
You can use this mode when you want a perfect result or when you can’t readjust the spatial index.

USING GEOMRELATE

"[geom_column] FROM [table]|([Subselect]) USING GEOMRELATE"

Using this keyword triggers MapServer to use the geometry function SDO_GEOM.RELATE, a function that searches
the relations between geometries. SDO_GEOM.RELATE does not use the spatial index and your performance is more
slow than operators but it’s very accurate. You can use this mode when you can’t use the spatial index or when it
doesn’t exist.

USING NONE

"[geom_column] FROM [table]|([Subselect]) USING NONE"

Using this keyword triggers MapServer to don’t use any geometry function or spatial operator. So, the internal SQL
don’t retrict, bases in the extent, the data from source. All the data from source will be returned for MapServer. The
NONE token is very useful when the source of the data don’t contains any spatial index. It’s usually occur when the
source is a function like SDO_BUFFER, SDO_XOR, SDO_INTERSECTION...... So this mode is recomended when
you can’t use the spatial index or when it doesn’t exist.

Additional keywords - [VERSION]

You can define what version of database you are using to improve the internal sql. This is very useful when using
geodetic SRIDs and MapServer functions that retrieve the extent from data.

USING VERSION 8i

"[geom_column] FROM [table]|([Subselect]) USING VERSION 8i"

This indicates MapServer to use a internal SQL that it’s compatible with Oracle 8i version.

USING VERSION 9i

"[geom_column] FROM [table]|([Subselect]) USING VERSION 9i"

The second indicates MapServer to use 9i version, is recommended to use this parameter if you are using 9i version
because the internal SQL will use specific spatial functions that is need to retrieve data correctly from 9i Oracle Spatial
versions.

306 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

USING VERSION 10g

"[geom_column] FROM [table]|([Subselect]) USING VERSION 10g"

This indicates MapServer to use a internal SQL that it’s compatible with Oracle 10g version.

More information

• You can define any PROJECTION to your LAYER without problem, can be used for data with or without an
SRID in Oracle.
• The native support for Oracle Spatial supports the defaults definition for SDO_GEOMETRY in database, the
Oracle Spatial SDO package.
• Information about the primary and secondary Oracle Spatial filters can be found in the Oracle Spatial User Guide
(the “Query Model” section). Information about the SDO_FILTER and SDO_RELATE operators can be found
in the “Spatial Operators” section, and information about the SDO_GEOM.RELATE function can be found in
the “Geometry Function” section of the Oracle Spatial User Guide.

Example of a LAYER

LAYER
NAME kwadranten
TYPE POLYGON
CONNECTIONTYPE oraclespatial
CONNECTION "user/pwd"
DATA "GEOMETRIE FROM KWADRANTEN USING SRID 90112"
DUMP TRUE
CLASS
STYLE
OUTLINECOLOR 0 0 0
COLOR 0 128 128
END
END
END

You can specify the SID for your database, the SID alias needs to be supplied in the tnsnames.ora file of the Oracle
client, e.g.
Example for tnsnames.ora:
MYDB =
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = TCP)(HOST = server_ip)(PORT = 1521))
)
(CONNECT_DATA =
(SERVICE_NAME = DB1)
)
)

So after this you can define you layer connection as:

CONNECTION "user/pwd@MYDB"

7.1. Vector Data 307

MapServer Documentation, Release 6.0

Mapfile syntax for OGR Oracle Spatial support

Syntax for your MAP file:

CONNECTION "OCI:user/pwd@service"
CONNECTIONTYPE OGR
DATA "Tablename"

Note: Make sure you set the wms_extent METADATA for the LAYER, as otherwise the “Getcapabilities” request
takes a lot of time.

7.1.17 PostGIS/PostgreSQL

Table of Contents
• PostGIS/PostgreSQL
– PostGIS/PostgreSQL
– Data Access /Connection Method
– OGRINFO Examples
– Mapfile Example
– Support for SQL/MM Curves
* Example#1: CircularString in MapServer
* Example#2: CompoundCurve in MapServer
* Example#3: CurvePolygon in MapServer
* Example#4: MultiCurve in MapServer
* Example#5: MultiSurface in MapServer
* Using MapServer < 6.0

PostGIS/PostgreSQL

PostGIS spatially enables the Open Source PostgreSQL database.

The PostGIS wiki page may include additional information.

Data Access /Connection Method

PostGIS is supported directly by MapServer and must be compiled into MapServer to work.
The PostgreSQL client libraries (libpq.so or libpq.dll) must be present in the system’s path environment for function-
ality to be present.
The CONNECTIONTYPE parameter must be set to POSTGIS.
The CONNECTION parameter is used to specify the parameters to connect to the database. CONNECTION param-
eters can be in any order. Most are optional. dbname is required. user is required. host defaults to localhost, port
defaults to 5432 (the standard port for PostgreSQL).
The DATA parameter is used to specify the data used to draw the map. The form of DATA is “[geometry_column]
from [table_name|sql_subquery] using unique [unique_key] using srid=[spatial_reference_id]”. The “using unique”
and “using srid=” clauses are optional, but using them improves performance.
Here is a simple generic example:

308 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

CONNECTIONTYPE POSTGIS
CONNECTION "host=yourhostname dbname=yourdatabasename user=yourdbusername password=yourdbpassword por
DATA "geometrycolumn from yourtablename"

This example shows specifying the unique key and srid in the DATA line:
CONNECTIONTYPE POSTGIS
CONNECTION "dbname=yourdatabasename user=yourdbusername"
DATA "the_geom from the_database using unique gid using srid=4326"

This example shows using a SQL subquery to perform a join inside the database and map the result in MapServer.
Note the “as subquery” string in the statement – everything between “from” and “using” is sent to the database for
evaluation:
CONNECTIONTYPE POSTGIS
CONNECTION "dbname=yourdatabasename user=yourdbusername"
DATA "the_geom from (select g.gid, g.the_geom, a.attr1, a.attr2 from geotable g join attrtable a on g

This example shows using a geometry function and database sort to limit the number of features and vertices returned
to MapServer:
CONNECTIONTYPE POSTGIS
CONNECTION "dbname=yourdatabasename user=yourdbusername"
DATA "the_geom from (select g.gid, ST_Simplify(g.the_geom, 10.0) as the_geom from geotable g order by

This example shows the use of the !BOX! substitution string to over-ride the default inclusion of the map bounding
box in the SQL. By default the spatial box clause is appended to the SQL in the DATA clause, but you can use !BOX!
to insert it anywhere you like in the statement. In general, you won’t need to use !BOX!, because the PostgreSQL
planner will generate the optimal plan from the generated SQL, but in some cases (complex sub-queries) a better plan
can be generated by placing the !BOX! closer to the middle of the query:
CONNECTIONTYPE POSTGIS
CONNECTION "dbname=yourdatabasename user=yourdbusername"
DATA "the_geom from (select g.gid, ST_Union(g.the_geom, 10.0) as the_geom from geotable g where ST_In

OGRINFO Examples

OGRINFO can be used to read out metadata about PostGIS tables directly from the database.
First you should make sure that your GDAL/OGR build contains the PostgreSQL driver, by using the ‘–formats’
command:
>ogrinfo --formats
Loaded OGR Format Drivers:
...
-> "PGeo" (readonly)
-> "PostgreSQL" (read/write)
-> "MySQL" (read/write)
...

If you don’t have the driver, you might want to try the FWTools or MS4W packages, which include the driver.
Once you have the driver you are ready to try an ogrinfo command on your database to get a list of spatial tables:
>ogrinfo PG:"host=127.0.0.1 user=postgres password=postgres dbname=canada port=5432"
using driver ‘PostgreSQL’ successful.
1: province (Multi Polygon)

Now use ogrinfo to get information on the structure of the spatial table:

7.1. Vector Data 309

MapServer Documentation, Release 6.0

>ogrinfo PG:"host=127.0.0.1 user=postgres password=postgres dbname=canada port=5432" province -summar

INFO: Open of ‘PG:host=127.0.0.1 user=postgres password=postgres dbname=canada’
using driver ‘PostgreSQL’ successful.

Layer name: province

Geometry: Multi Polygon
Feature Count: 1068
Extent: (-2340603.750000, -719746.062500) - (3009430.500000, 3836605.250000)
Layer SRS WKT:
(unknown)
FID Column = gid
Geometry Column = the_geom
area: Real (0.0)
island: String (30.0)
island_e: String (30.0)
island_f: String (30.0)
name: String (30.0)
...

Mapfile Example

LAYER
NAME "province"
STATUS ON
TYPE POLYGON
CONNECTIONTYPE POSTGIS
CONNECTION "host=127.0.0.1 port=5432 dbname=canada user=postgres password=postgres"
DATA "the_geom from province"
CLASS
...
END
END

For more info about PostGIS and MapServer see the PostGIS docs: http://postgis.org/documentation/

Support for SQL/MM Curves

PostGIS is able to store circular interpolated curves, as part of the SQL Multimedia Applications Spatial specification
(read about the SQL/MM specification). For more information about PostGIS’ support, see the SQL-MM Part 3 section
in the PostGIS documentation, such as here.
As of MapServer 6.0, the PostGIS features CircularString, CompoundCurve, CurvePolygon, MultiCurve, and Multi-
Surface can be drawn through MapServer directly.

Example#1: CircularString in MapServer

The following is the Well Known Text of the feature loading into PostGIS:
INSERT INTO test ( g, id ) VALUES ( ST_GeomFromText(’CIRCULARSTRING(0 0, 4 0, 4 4, 0 4, 0 0)’, -1), 2

An example MapServer layer might look like:

LAYER
NAME "curves_poly"
STATUS DEFAULT

310 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

TYPE POLYGON
CONNECTIONTYPE postgis
CONNECTION "user=postgres password=postgres dbname=curves host=localhost port=5432"
DATA "g from test using SRID=-1 using unique id"
CLASS
STYLE
COLOR 128 128 128
ANTIALIAS true
END
END
END

And testing with shp2img should produce a map image of:

Example#2: CompoundCurve in MapServer

The following is the Well Known Text of the feature loading into PostGIS:
INSERT INTO test ( g, id ) VALUES ( ST_GeomFromText(’COMPOUNDCURVE(CIRCULARSTRING(0 0, 1 1, 1 0),(1 0

An example MapServer layer might look like:

LAYER
NAME "curves_poly"
STATUS DEFAULT
TYPE POLYGON
CONNECTIONTYPE postgis
CONNECTION "user=postgres password=postgres dbname=curves host=localhost port=5432"
DATA "g from test using SRID=-1 using unique id"
CLASS
STYLE
COLOR 128 128 128
ANTIALIAS true
END
END
END

And testing with shp2img should produce a map image of:

7.1. Vector Data 311

MapServer Documentation, Release 6.0

Example#3: CurvePolygon in MapServer

The following is the Well Known Text of the feature loading into PostGIS:
INSERT INTO test ( g, id ) VALUES ( ST_GeomFromText(’CURVEPOLYGON(CIRCULARSTRING(0 0, 4 0, 4 4, 0 4,

An example MapServer layer might look like:

LAYER
NAME "curves_poly"
STATUS DEFAULT
TYPE POLYGON
CONNECTIONTYPE postgis
CONNECTION "user=postgres password=postgres dbname=curves host=localhost port=5432"
DATA "g from test using SRID=-1 using unique id"
CLASS
STYLE
COLOR 128 128 128
ANTIALIAS true
END
END
END

And testing with shp2img should produce a map image of:

Example#4: MultiCurve in MapServer

The following is the Well Known Text of the feature loading into PostGIS:
INSERT INTO test ( g, id ) VALUES ( ST_GeomFromText(’MULTICURVE((0 0, 5 5),CIRCULARSTRING(4 0, 4 4, 8

An example MapServer layer might look like:

LAYER
NAME "curves_poly"
STATUS DEFAULT

312 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

TYPE POLYGON
CONNECTIONTYPE postgis
CONNECTION "user=postgres password=postgres dbname=curves host=localhost port=5432"
DATA "g from test using SRID=-1 using unique id"
CLASS
STYLE
COLOR 128 128 128
ANTIALIAS true
END
END
END

And testing with shp2img should produce a map image of:

Example#5: MultiSurface in MapServer

The following is the Well Known Text of the feature loading into PostGIS:
INSERT INTO test ( g, id ) VALUES ( ST_GeomFromText(’MULTISURFACE(CURVEPOLYGON(CIRCULARSTRING(0 0, 4

An example MapServer layer might look like:

LAYER
NAME "curves_poly"
STATUS DEFAULT
TYPE POLYGON
CONNECTIONTYPE postgis
CONNECTION "user=postgres password=postgres dbname=curves host=localhost port=5432"
DATA "g from test using SRID=-1 using unique id"
CLASS
STYLE
COLOR 128 128 128
ANTIALIAS true
END

7.1. Vector Data 313

MapServer Documentation, Release 6.0

END
END

And testing with shp2img should produce a map image of:

Using MapServer < 6.0

If you cannot upgrade to MapServer 6.0, then you can use the PostGIS function ST_CurveToLine() in your MapServer
LAYER to draw the curves (note that this is much slower however):
LAYER
NAME "curves_poly"
STATUS DEFAULT
TYPE POLYGON
CONNECTIONTYPE postgis
CONNECTION "user=postgres password=postgres dbname=curves host=localhost port=5432"
DATA "wkb_geometry from (select c.id, ST_CurveToLine(c.g) as wkb_geometry from c) as subquery using
CLASS
STYLE
COLOR 128 128 128
ANTIALIAS true

314 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

END
END
END

7.1.18 SDTS

This is a United States Geological Survey (USGS) format. SDTS has a raster and a vector format. The raster format
is not supported in MapServer. Only the vector formats are supported, including VTP and DLG files.

File listing

• SDTS files are often organized into state-sized pieces. For example, all of the state of Maryland (MD), U.S.A.
• Files are also available for multiple types of features including hydrography, transportation and administrative
boundaries.
This example uses transportation data, which consists of 35 separate files, each with the suffix DDF:
MDTRAHDR.DDF MDTRARRF.DDF MDTRCATS.DDF
MDTRDQCG.DDF MDTRFF01.DDF MDTRLE02.DDF
MDTRNA03.DDF MDTRNO03.DDF MDTRSPDM.DDF
MDTRAMTF.DDF MDTRBFPS.DDF MDTRCATX.DDF
MDTRDQHL.DDF MDTRIDEN.DDF MDTRLE03.DDF
MDTRNE03.DDF MDTRPC01.DDF MDTRSTAT.DDF
MDTRARDF.DDF MDTRBMTA.DDF MDTRDDSH.DDF
MDTRDQLC.DDF MDTRIREF.DDF MDTRNA01.DDF
MDTRNO01.DDF MDTRPC02.DDF MDTRXREF.DDF
MDTRARDM.DDF MDTRCATD.DDF MDTRDQAA.DDF
MDTRDQPA.DDF MDTRLE01.DDF MDTRNA02.DDF
MDTRNO02.DDF MDTRPC03.DDF

Data Access / Connection Method

• SDTS access is available in MapServer through OGR.

• The CONNECTIONTYPE OGR parameter must be used.
• The path (which can be relative) to the catalog file (????CATD.DDF) is required, including file extension.
• There are multiple layers in the SDTS catalog, some of which are only attributes and have no geometries.
• The layer name is specified with the DATA parameter

OGRINFO Examples

Using ogrinfo on a catalog file (note that the first 7 layers do not have geometries):
> ogrinfo /data/sdts/MD/MDTRCATD.DDF
Had to open data source read-only.
INFO: Open of ‘MDTRCATD.DDF’
using driver ‘SDTS’ successful.
1: ARDF (None)
2: ARRF (None)
3: AMTF (None)
4: ARDM (None)
5: BFPS (None)

7.1. Vector Data 315

MapServer Documentation, Release 6.0

6: BMTA (None)
7: AHDR (None)
8: NE03 (Point)
9: NA01 (Point)
10: NA02 (Point)
11: NA03 (Point)
12: NO01 (Point)
13: NO02 (Point)
14: NO03 (Point)
15: LE01 (Line String)
16: LE02 (Line String)
17: LE03 (Line String)
18: PC01 (Polygon)
19: PC02 (Polygon)
20: PC03 (Polygon)

Using ogrinfo to examine the structure of the file/layer:

> ogrinfo /data/sdts/MD/MDTRCATD.DDF LE01 -summary
Had to open data source read-only.
INFO: Open of ‘MDTRCATD.DDF’
using driver ‘SDTS’ successful.

Layer name: LE01

Geometry: Line String
Feature Count: 780
Extent: (-80.000289, 36.999774) - (-74.999711, 40.000225)
Layer SRS WKT:
GEOGCS["NAD27",
DATUM["North_American_Datum_1927",
SPHEROID["Clarke 1866",6378206.4,294.978698213901]],
PRIMEM["Greenwich",0],
UNIT["degree",0.0174532925199433]]
RCID: Integer (0.0)
SNID: Integer (0.0)
ENID: Integer (0.0)
ENTITY_LABEL: String (7.0)
ARBITRARY_EXT: String (1.0)
RELATION_TO_GROUND: String (1.0)
VERTICAL_RELATION: String (1.0)
OPERATIONAL_STATUS: String (1.0)
ACCESS_RESTRICTION: String (1.0)
OLD_RAILROAD_GRADE: String (1.0)
WITH_RAILROAD: String (1.0)
COVERED: String (1.0)
HISTORICAL: String (1.0)
LIMITED_ACCESS: String (1.0)
PHOTOREVISED: String (1.0)
LANES: Integer (2.0)
ROAD_WIDTH: Integer (3.0)
BEST_ESTIMATE: String (1.0)
ROUTE_NUMBER: String (7.0)
ROUTE_TYPE: String (9.0)

Map File Example:

LAYER
NAME sdts_maryland
TYPE LINE

316 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

CONNECTIONTYPE OGR
CONNECTION "data/sdts/MD/MDTRCATD.DDF"
DATA "LE01"
STATUS DEFAULT
CLASS
STYLE
COLOR 0 0 0
END
END
END

7.1.19 S57

Also known as S57. The IHO S-57 format is a vector interchange format used for maritime charts. It was developed by
the International Hydrographic Organisation (IHO). For more information about the IHO see: http://www.iho.shom.fr/

File listing

Individual S57 data files have an extension of *.000. For example:

US1BS02M.000

Data Access / Connection Method

• S57 access in MapServer occurs through OGR, CONNECTIONTYPE OGR must be used.
• Specify a full path or a relative path from the SHAPEPATH to the .000 file for the CONNECTION
• Use the DATA parameter to specify the s57 layer name

Special Notes

The underlying OGR code requires two files from your GDAL/OGR installation when reading S57 data in
MapServer : s57objectclasses.csv and s57attributes.csv. These files can be found in the /GDAL/data/ folder (unix:
/usr/local/share/gdal windows: /ms4w/gdaldata). If you receive an error in MapServer such as:
msDrawMap(): Image handling error. Failed to draw layer named ’s57’.
msOGRFileOpen(): OGR error. xxx failed for OGR connection

you may have to point MapServer to these files using the CONFIG parameter in the main section of your map file:
CONFIG GDAL_DATA "C:\ms4w\gdaldata"

OGRINFO Examples

Using ogrinfo on an S57 file to get the layer name:

> ogrinfo us1bs02m.000
ERROR 4: S57 Driver doesn’t support update.
Had to open data source read-only.
INFO: Open of ‘us1bs02m.000’
using driver ‘IHO S-57 (ENC)’ successful.
1: ADMARE (Polygon)

7.1. Vector Data 317

MapServer Documentation, Release 6.0

2: CBLSUB (Line String)

3: CTNARE
4: COALNE (Line String)
5: DEPARE
6: DEPCNT (Line String)
7: LNDARE
8: LNDELV
9: LNDRGN
10: LNDMRK
11: LIGHTS (Point)
12: OBSTRN
13: RDOSTA (Point)
14: SEAARE
15: SBDARE
16: SLCONS
17: SOUNDG (Multi Point)
18: UWTROC (Point)
19: WATTUR
20: WRECKS
21: M_COVR (Polygon)
22: M_NPUB (Polygon)
23: M_NSYS (Polygon)
24: M_QUAL (Polygon)
25: C_ASSO (None)

Using ogrinfo to examine the structure of an S57 layer:

> ogrinfo us1bs02m.000 DEPARE -summary
ERROR 4: S57 Driver doesn’t support update.
Had to open data source read-only.
INFO: Open of ‘us1bs02m.000’
using driver ‘IHO S-57 (ENC)’ successful.

Layer name: DEPARE

Geometry: Unknown (any)
Feature Count: 297
Extent: (165.666667, 48.500000) - (180.000000, 60.750000)
Layer SRS WKT:
GEOGCS["WGS 84",
DATUM["WGS_1984",
SPHEROID["WGS 84",6378137,298.257223563]],
PRIMEM["Greenwich",0],
UNIT["degree",0.0174532925199433]]
GRUP: Integer (3.0)
OBJL: Integer (5.0)
RVER: Integer (3.0)
AGEN: Integer (2.0)
FIDN: Integer (10.0)
FIDS: Integer (5.0)
LNAM: String (16.0)
LNAM_REFS: StringList (16.0)
DRVAL1: Real (0.0)
DRVAL2: Real (0.0)
QUASOU: String (0.0)
SOUACC: Real (0.0)
VERDAT: Integer (0.0)
INFORM: String (0.0)
NINFOM: String (0.0)

318 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

NTXTDS: String (0.0)

SCAMAX: Integer (0.0)
SCAMIN: Integer (0.0)
TXTDSC: String (0.0)
RECDAT: String (0.0)
RECIND: String (0.0)
...

Map File Example:

LAYER
NAME s57
TYPE POLYGON
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "./s57/us1bs02m.000"
DATA "DEPARE"
CLASS
STYLE
COLOR 247 237 219
OUTLINECOLOR 120 120 120
END
END
END # Layer

7.1.20 USGS TIGER

TIGER/Line files are created by the US Census Bureau and cover the entire US. They are often referred simply as
TIGER files. For more information see: http://www.census.gov/geo/www/tiger/.

File listing

TIGER/Line files are text files and directory-based data sources. For example, one county folder TGR06059 contains
several associated files:
TGR06059.RT1 TGR06059.RT2 TGR06059.RT4 TGR06059.RT5
TGR06059.RT6 TGR06059.RT7 TGR06059.RT8 TGR06059.RTA
TGR06059.RTC TGR06059.RTH TGR06059.RTI TGR06059.RTP
TGR06059.RTR TGR06059.RTS TGR06059.RTT TGR06059.RTZ

Data Access / Connection Method

• TIGER/Line access occurs through an OGR CONNECTION

• The full path to the directory containing the associated files is required in the CONNECTION string.
• The feature type is specified in the DATA parameter

OGRINFO Examples

Using ogrinfo on a TIGER directory to retrieve feature types:

7.1. Vector Data 319

MapServer Documentation, Release 6.0

> ogrinfo TGR06059 (NOTE that this is a directory)

ERROR 4: Tiger Driver doesn’t support update.
Had to open data source read-only.
INFO: Open of ‘TGR06059’
using driver ‘TIGER’ successful.
1: CompleteChain (Line String)
2: AltName (None)
3: FeatureIds (None)
4: ZipCodes (None)
5: Landmarks (Point)
6: AreaLandmarks (None)
7: Polygon (None)
8: PolygonCorrections (None)
9: EntityNames (Point)
10: PolygonEconomic (None)
11: IDHistory (None)
12: PolyChainLink (None)
13: PIP (Point)
14: TLIDRange (None)
15: ZeroCellID (None)
16: OverUnder (None)
17: ZipPlus4 (None)

Using ogrinfo to examine the structure of the TIGER feature type CompleteChain:
> ogrinfo TGR06059 CompleteChain -summary
ERROR 4: Tiger Driver doesn’t support update.
Had to open data source read-only.
INFO: Open of ‘TGR06059’
using driver ‘TIGER’ successful.

Layer name: CompleteChain

Geometry: Line String
Feature Count: 123700
Extent: (-118.125898, 33.333992) - (-117.412987, 33.947512)
Layer SRS WKT:
GEOGCS["NAD83",
DATUM["North_American_Datum_1983",
SPHEROID["GRS 1980",6378137,298.257222101]],
PRIMEM["Greenwich",0],
UNIT["degree",0.0174532925199433]]
MODULE: String (8.0)
TLID: Integer (10.0)
SIDE1: Integer (1.0)
SOURCE: String (1.0)
FEDIRP: String (2.0)
FENAME: String (30.0)
FETYPE: String (4.0)
FEDIRS: String (2.0)
CFCC: String (3.0)
FRADDL: String (11.0)
TOADDL: String (11.0)
FRADDR: String (11.0)
TOADDR: String (11.0)
FRIADDL: String (1.0)
TOIADDL: String (1.0)
FRIADDR: String (1.0)
TOIADDR: String (1.0)

320 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

ZIPL: Integer (5.0)

Map File Example:

LAYER
NAME Complete_Chain
TYPE LINE
STATUS DEFAULT
CONNECTIONTYPE OGR
CONNECTION "/path/to/data/tiger/TGR06059"
DATA "CompleteChain"
CLASS
STYLE
COLOR 153 102 0
END
END
END # Layer

7.1.21 Virtual Spatial Data

Table of Contents
• Virtual Spatial Data
– Types of Databases
– Types of Flat Files
– Steps for Display

This is an OGR extension to MapServer. It allows you to connect to databases that do not explicitly hold spatial data,
as well as flat text files. Your data must have an X and a Y column, and the data may be accessed through an ODBC
connection or a direct pointer to a text file.
The original VirtualSpatialData wiki page may contain additional information.

Types of Databases

The VirtualSpatialData OGR extension has been tested with the following databases and should, in theory, support all
ODBC data sources.
• Oracle
• MySQL
• SQL Server
• Access
• PostgreSQL

Types of Flat Files

Comma, tab or custom delimited text/flat files work with VirtualSpatialData.

7.1. Vector Data 321

MapServer Documentation, Release 6.0

Steps for Display

1. Create the Datasource Name (DSN)

• Specific notes about creating a DSN on Windows and Linux can be found by searching the MapServer reference
documents site
• On some Windows systems you must create a SYSTEM DSN.

2. Test your Connection

Test your connection with ogrinfo. The syntax for this command is:
> ogrinfo ODBC:user/pass@DSN table

Windows users may not be required to specify a user/password, so the syntax would be:
> ogrinfo ODBC:@DSN table

Example: Accessing a comma separated text file through ODBC using ogrinfo
The following is a snippet of the flat text file coal_dep.txt containing lat/long points:
unknown,na,id,id2,mark,coalkey,coalkey2,long,lat
0.000,0.000,1,1,7,87,87,76.90238,51.07161
0.000,0.000,2,2,7,110,110,78.53851,50.69403
0.000,0.000,3,3,3,112,112,83.22586,71.24420
0.000,0.000,4,4,6,114,114,80.79896,73.41175

If the DSN name is Data_txt, the ogrinfo command to see a list of applicable files in the directory is:
> ogrinfo ODBC:jeff/test@Data_txt
INFO: Open of ‘ODBC:jeff/test@Data_txt’
using driver ‘ODBC’ successful.
1: coal_dep.csv
2: coal_dep.txt
3: coal_dep_nf.txt
4: coal_dep_trim.txt
5: Copy of coal_dep.txt
6: deposit.csv
7: maruia.asc
8: oahuGISbathy.csv
9: oahuGISbathy.txt
10: on_pts.txt
11: on_pts_utm.txt
12: test.txt
13: utm_test.txt

Username and password may be optional, so the following may also be valid:
> ogrinfo ODBC:@Data_txt

Therefore, the command to see more information about one of the specific layers is:
> ogrinfo ODBC:@Data_txt coal_dep.txt
INFO: Open of ‘ODBC:@Data_txt’
using driver ‘ODBC’ successful.

322 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Layer name: coal_dep.txt

Geometry: Unknown (any)
Feature Count: 266
Layer SRS WKT:
(unknown)
UNKNOWN: String (255.0)
NA: String (255.0)
ID: String (255.0)
ID2: String (255.0)
MARK: String (255.0)
COALKEY: String (255.0)
COALKEY2: String (255.0)
LONG: String (255.0)
LAT: String (255.0)
OGRFeature(coal_dep.txt):0
UNKNOWN (String) = 0.000
....

3. Create a Virtual Data File

This is a file with an ovf extension and looks like the following:
<OGRVRTDataSource>
<OGRVRTLayer name="mylayer">
<SrcDataSource>ODBC:user/pass@DSN</SrcDataSource>
<SrcLayer>tablename</SrcLayer>
<GeometryType>wkbPoint</GeometryType>
<LayerSRS>WGS84</LayerSRS>
<GeometryField encoding="PointFromColumns" x="x" y="y"/>
</OGRVRTLayer>
</OGRVRTDataSource>

More information on ovf files can be found at: http://www.gdal.org/ogr/drv_vrt.html

Example ovf file for coal_dep.txt:
<OGRVRTDataSource>
<OGRVRTLayer name="coal-test">
<SrcDataSource>ODBC:Data_txt</SrcDataSource>
<SrcLayer>coal_dep.txt</SrcLayer>
<GeometryField encoding="PointFromColumns" x="Long" y="Lat"/>
<GeometryType>wkbPoint</GeometryType>
</OGRVRTLayer>
</OGRVRTDataSource>

4. Test Virtual Data File with ogrinfo

Use ogrinfo to test your new ovf file, such as:

> ogrinfo coal.ovf coal-test
ERROR 4: Update access not supported for VRT datasources.
Had to open data source read-only.
INFO: Open of ‘myfile.ovf’
using driver ‘VRT’ successful.

Layer name: coal_dep.txt

7.1. Vector Data 323

MapServer Documentation, Release 6.0

Geometry: Unknown (any)

Feature Count: 266
Layer SRS WKT:
(unknown)
UNKNOWN: String (255.0)
NA: String (255.0)
ID: String (255.0)
ID2: String (255.0)
MARK: String (255.0)
...

5. Mapfile Layer

Using an ovf file your layer may look like:

LAYER
CONNECTION "coal.ovf"
CONNECTIONTYPE OGR
DATA "coal-test"
METADATA
"wms_srs" "4326"
"wms_title" "coal-test"
END
NAME "coal-test"
SIZEUNITS PIXELS
STATUS ON
TOLERANCE 0
TOLERANCEUNITS PIXELS
TYPE POINT
UNITS METERS
CLASS
STYLE
COLOR 255 0 0
MAXSIZE 100
MINSIZE 1
SIZE 6
SYMBOL "star"
END
END
END

Or you may specify the ovf contents inline such as:

LAYER
CONNECTION "<OGRVRTDataSource>
<OGRVRTLayer name=’coal-test’>
<SrcDataSource>ODBC:@Data_txt</SrcDataSource>
<SrcLayer>coal_dep.txt</SrcLayer>
<GeometryField encoding=’PointFromColumns’ x=’Long’ y=’Lat’/>
<GeometryType>wkbPoint</GeometryType>
</OGRVRTLayer>
</OGRVRTDataSource>"
CONNECTIONTYPE OGR
DATA "coal-test"
METADATA
"wms_srs" "4326"
"wms_title" "coal-test"

324 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

END
NAME "coal-test"
SIZEUNITS PIXELS
STATUS ON
TOLERANCE 0
TOLERANCEUNITS PIXELS
TYPE POINT
UNITS METERS
CLASS
STYLE
COLOR 255 0 0
MAXSIZE 100
MINSIZE 1
SIZE 6
SYMBOL "star"
END
END
END

6. Test your Mapfile

The first thing you should try is to use the shp2img utility:
shp2img -m mymapfile.map -o test.png

Once you successfully created a map image, then try your application. Note Windows users may come across a
problem where shp2img works but their application throws an error similar to this:
Warning: [MapServer Error]: msOGRFileOpen(): Open failed for OGR connection ‘coal.ovf’.
Unable to initialize ODBC connection to DSN for jeff/test@Data_txt,
[Microsoft][ODBC Driver Manager] Data source name not found
and no default driver specified in D:\ms4w\Apache\htdocs\quickmap.php on line 40

If that happens you should make sure you have created a System DSN.

7.1.22 WFS

WFS is an Open Geospatial Consortium (OGC) specification. For more information about the format itself, see:
http://www.opengeospatial.org/standards/wfs
WFS allows a client to retrieve geospatial data encoded in Geography Markup Language (GML) from multiple Web
Feature Services. GML is built on the standard web language XML.
WFS differs from the popular Web Map Service (WMS) specification in that WFS returns a subset of the data in valid
GML format, not just a graphic image of data.

Capabilities

Requesting the capabilities using the GetCapabilities request to a WFS server returns an XML document showing
what layers and projections are available, etc. Example of a WFS GetCapabilities URL:
http://demo.mapserver.org/cgi-bin/wfs?SERVICE=WFS&VERSION=1.0.0&REQUEST=GetCapabilities
Example of the Resulting XML from GetCapabilties:

7.1. Vector Data 325

MapServer Documentation, Release 6.0

...
<FeatureTypeList>
<Operations>
<Query/>
</Operations>
<FeatureType>
<Name>continents</Name>
<Title>World continents</Title>
<SRS>EPSG:4326</SRS>
<LatLongBoundingBox minx="-180" miny="-90" maxx="180" maxy="83.6274"/>
</FeatureType>
<FeatureType>
<Name>cities</Name>
<Title>World cities</Title>
<SRS>EPSG:4326</SRS>
<LatLongBoundingBox minx="-178.167" miny="-54.8" maxx="179.383" maxy="78.9333"/>
</FeatureType>
</FeatureTypeList>
...

Data Access / Connection Method

• WFS access is a core MapServer feature. MapServer currently supports WFS version 1.0.0
• The CONNECTIONTYPE WFS parameter must be used.
• WFS layers can be requested through a layer in a map file, or you can request the GML directly through the
browser with a GetFeature request. You can specify a specific layer with the TypeName request. In a map file
the name/value pairs should be put into a METADATA object.
• You can limit the number of features returned in the GML by using the MaxFeatures option (e.g. &MAXFEA-
TURES=100).
Example of a WFS Request Directly Through the Browser:
The following URL requests the GML for the layer continents. (see the GetCapabilities above for the possible layers
available on this test server) . The URL is all one line, broken up here for readability (click here for a working link).
http://demo.mapserver.org/cgi-bin/wfs?
SERVICE=WFS&
VERSION=1.0.0&
REQUEST=getfeature&
TYPENAME=continents&
MAXFEATURES=100

Map File Example

LAYER
NAME "continents"
TYPE POLYGON
STATUS ON
CONNECTION "http://demo.mapserver.org/cgi-bin/wfs?"
CONNECTIONTYPE WFS
METADATA
"wfs_typename" "continents"
"wfs_version" "1.0.0"
"wfs_connectiontimeout" "60"
"wfs_maxfeatures" "10"

326 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

END
PROJECTION
"init=epsg:4326"
END
CLASS
NAME "Continents"
STYLE
COLOR 255 128 128
OUTLINECOLOR 96 96 96
END
END
END # Layer

7.2 Raster Data

Author Frank Warmerdam

Contact warmerdam at pobox.com
Revision $Revision: 9436 $
Date $Date: 2009-10-09 14:49:13 -0700 (Fri, 09 Oct 2009) $
Last Updated 2007/12/09

Table of Contents
• Raster Data
– Introduction
– How are rasters added to a Map file?
– Supported Formats
– Rasters and Tile Indexing
– Raster Warping
– 24bit RGB Rendering
– Special Processing Directives
– Raster Query
– Raster Display Performance Tips
– Preprocessing Rasters
– Georeference with World Files

7.2.1 Introduction

MapServer supports rendering a variety of raster file formats in maps. The following describes some of the supported
formats, and what capabilities are supported with what formats.
This document assumes that you are already familiar with setting up MapServer Mapfile, but does explain the raster
specific aspects of mapfiles.

7.2.2 How are rasters added to a Map file?

A simple raster layer declaration looks like this. The DATA file is interpreted relative to the SHAPEPATH, much like
shapefiles.

7.2. Raster Data 327

MapServer Documentation, Release 6.0

LAYER
NAME "JacksonvilleNC_CIB"
DATA "Jacksonville.tif"
TYPE RASTER
STATUS ON
END

Though not shown rasters can have PROJECTION, METADATA, PROCESSING, MINSCALE, and MAXSCALE
information. It cannot have labels, CONNECTION, CONNECTIONTYPE, or FEATURE information.

Classifying Rasters

Rasters can be classified in a manner similar to vectors, with a few exceptions.

There is no need to specify a CLASSITEM. The raw pixel value itself (“[pixel]”) and, for paletted images, the red,
green and blue color associated with that pixel value (“[red]”, “[green]” and “[blue]”) are available for use in classifi-
cations. When used in an evaluated expression the pixel, red, green and blue keywords must be in lower case.
LAYER
NAME "JacksonvilleNC_CIB"
DATA "Jacksonville.tif"
TYPE RASTER
STATUS ON
CLASSITEM "[pixel]"
# class using simple string comparison, equivelent to ([pixel] = 0)
CLASS
EXPRESSION "0"
STYLE
COLOR 0 0 0
END
END
# class using an EXPRESSION using only [pixel].
CLASS
EXPRESSION ([pixel] >= 64 AND [pixel] < 128)
STYLE
COLOR 255 0 0
END
END
# class using the red/green/blue values from the palette
CLASS
NAME "near white"
EXPRESSION ([red] > 200 AND [green] > 200 AND [blue] > 200)
STYLE
COLOR 0 255 0
END
END
# Class using a regular expression to capture only pixel values ending in 1
CLASS
EXPRESSION /*1/
STYLE
COLOR 0 0 255
END
END
END

As usual, CLASS definitions are evaluated in order from first to last, and the first to match is used. If a CLASS has
a NAME attribute it may appear in a LEGEND. Only the COLOR, EXPRESSION and NAME parameters within a
CLASS definition are utilized for raster classifications. The other styling or control information is ignored.

328 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Raster classifications always take place on only one raster band. It defaults to the first band in the referenced file, but
this can be altered with the BANDS PROCESSING directive. In particular this means that including even a single
CLASS declaration in a raster layer will result in the raster layer being rendered using the one band classification rules
instead of other rules that might have applied (such as 3 band RGB rendering).

Classifying Non-8bit Rasters

As of MapServer 4.4 support has been added for classifying non-8bit raster inputs. That is input rasters with values
outside the range 0-255. Mostly this works transparently but there are a few caveats and options to provide explicit
control.
Classifying raster data in MapServer is accomplished by pre-classifying all expected input values and using that table
of classification results to lookup each pixel as it is rendered. This is done because evaluating a pixel value against a
series of CLASS definitions is relatively expensive to do for the hundreds of thousands of pixels in a typical rendered
image.
For simple 8bit inputs, only 256 input values need to be pre-classified. But for non-8bit inputs more values need to be
classified. For 16bit integer inputs all 65536 possible input values are pre-classified. For floating point and other input
data types, up to 65536 values are pre-classified based on the maximum expected range of input values.
The PROCESSING directive can be used to override the range of values to be pre-classified, or the number of val-
ues (aka Buckets) in that range to classify. The SCALE=min,max PROCESSING directive controls the range. The
SCALE_BUCKETS PROCESSING directive controls the number of buckets. In some cases rendering can be accel-
erated considerable by selecting a restricted range of input values and a reduced number of scaling values (buckets).
The following example classifies a floating raster, but only 4 values over the range -10 to 10 are classified. In particular,
the values classified would be -7.5, -2.5, 2.5, and 7.5 (the middle of each “quarter” of the range). So those four value
are classified, and one of the classification results is selected based on which value is closest to the pixel value being
classified.
LAYER
NAME grid1
TYPE raster
STATUS default
DATA data/float.tif
PROCESSING "SCALE=-10,10"
PROCESSING "SCALE_BUCKETS=4"
CLASS
NAME "red"
EXPRESSION ([pixel] < -3)
STYLE
COLOR 255 0 0
END
END
CLASS
NAME "green"
EXPRESSION ([pixel] >= -3 and [pixel] < 3)
STYLE
COLOR 0 255 0
END
END
CLASS
NAME "blue"
EXPRESSION ([pixel] >= 3)
STYLE
COLOR 0 0 255
END

7.2. Raster Data 329

MapServer Documentation, Release 6.0

END
END

7.2.3 Supported Formats

What raster formats are supported by MapServer is largely controlled by configuration time options. Some formats
are considered to be built-in while the remainder are handled by the optional GDAL raster library.
More information on GDAL can be found at http://www.gdal.org, including the supported formats list. Some of the
advanced MapServer raster features, such as resampling, RGB color cube generation and automatic projection capture
only work with raster formats used through GDAL. GDAL is normally built and installed separately from MapServer,
and then enabled during the build of MapServer using the –with-gdal configuration switch.
To find out what is built into a particular MapServer executable, use the -v flags to discover what build options are
enabled. To find out what GDAL formats are available, the “gdalinfo –formats” command may be used. For example:
warmerda@gdal2200[124]% mapserv -v
MapServer version 4.4.0-beta2 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT
SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER SUPPORTS=FASTCGI
INPUT=EPPL7 INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE DEBUG=MSDEBUG
warmerda@gdal2200[18]% gdalinfo --formats
Supported Formats:
GRASS (ro): GRASS Database Rasters (5.7+)
GTiff (rw+): GeoTIFF
NITF (rw+): National Imagery Transmission Format
HFA (rw+): Erdas Imagine Images (.img)
SAR_CEOS (ro): CEOS SAR Image
...

The following formats are potential builtins:

TIFF/GeoTIFF If built with INPUT=TIFF MapServer will have builtin support for reading TIFF or GeoTIFF files.
The builtin TIFF support has some limitations with regard to the organization of files that can be read (no tiled,
16bit, RGB, or odd color models). This driver supports world files, or simple builtin GeoTIFF coordinates for
georeferencing.
Full featured TIFF/GeoTIFF support is available through GDAL. Note that only GDAL supports tiled TIFF files
and TIFF files with overviews. Tiled TIFF files with overviews pre-built are one of the highest performance
ways of serving large raster images.
GIF If GD is configured with GIF (OUTPUT=GIF) support, then MapServer will also be able to read GIF files for
raster layers. The only way to georeference GIF files is with a world file.
If GD is not configured with GIF support, it may still be available in GDAL.
PNG If GD is configured with PNG (OUTPUT=PNG) support, then MapServer will also be able to read PNG files
for raster layers. The only way to georeference PNG files is with a world file.
If GD is not configured with PNG support, it may still be available in GDAL.
JPEG If MapServer is built with JPEG (INPUT=JPEG) support then greyscale JPEG files may be rendered in raster
layers. RGB files (the more common kind) will not be able to be displayed. Georeferencing is via world files.
If MapServer is not built with native JPEG support, GDAL may still support the format. In this case RGB files
are also supported (via the RGB color cube mechanism). Georeferencing is still via world file.
Erdas .LAN/.GIS If configured with INPUT=EPPL7 (the default) MapServer will support one band eight bit Erdas
LAN/GIS files. The .trl file is read for a colormap, and if not found the layer is treated as greyscale. Georefer-
encing is read from the file.

330 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

If MapServer is built with GDAL it is generally better to access all possible formats through GDAL rather than via the
built-in drivers. The built-in drivers are less featureful, and not as well maintained.

7.2.4 Rasters and Tile Indexing

When handling very large raster layers it is often convenient, and higher performance to split the raster image into a
number of smaller images. Each file is a tile of the larger raster mosaic available for display. The list of files forming a
layer can be stored in a shapefile with polygons representing the footprint of each file, and the name of the files. This
is called a TILEINDEX and works similarly to the same feature in vector layers. The result can be represented in the
MAP file as one layer, but MapServer will first scan the tile index, and ensure that only raster files overlapping the
current display request will be opened.
The following example shows a simple example. No DATA statement is required because MapServer will fetch the
filename of the raster files from the Location attribute column in the hp2.dbf file for records associated with polygons
in hp2.shp that intersect the current display region. The polygons in hp2.shp should be rectangles representing the
footprint of the corresponding file. Note that the files do not have to be all the same size, the formats can vary and they
can even overlap (later files will be drawn over earlier ones); however, they must all be in the same coordinate system
(projection) as the layer.
LAYER
NAME "hpool"
STATUS ON
TILEINDEX "hp2.shp"
TILEITEM "Location"
TYPE RASTER
END

The filenames in the tileindex are searched for relative to the SHAPEPATH or map file, not relative to the tileindex.
Great care should be taken when establishing the paths put into the tileindex to ensure they will evaluate properly
in use. Often it is easiest to place the tileindex in the SHAPEPATH directory, and to create the tileindex with a path
relative to the SHAPEPATH directory. When all else fails, absolute paths can be used in tileindex, but then they cannot
be so easily moved from system to system.
While there are many ways to produce TILEINDEX shapefiles for use with this command, one option is the gdaltindex
program, part of the GDAL utility suite. The gdaltindex program will automatically generate a tile index shapefile from
a list of GDAL supported raster files passed on the command line.
Usage: gdaltindex [-tileindex field_name] index_file [gdal_file]*

% gdaltindex doq_index.shp doq/*.tif

Tile Index Notes

• The shapefile (index_file) will be created if it doesn’t already exist.

• The default tile index field is ‘location’.
• Simple rectangular polygons are generated in the same coordinate system as the rasters.
• Raster filenames will be put in the file exactly as they are specified on the commandline.
• Many problems with tile indexes relate to how relative paths in the tile index are evaluated. They should be
evaluated relative to the SHAPEPATH if one is set, otherwise relative to the tileindex file. When in doubt
absolute paths may avoid path construction problems.
The gdaltindex program is built as part of GDAL. Prebuilt binaries for GDAL including the gdaltindex program can
be downloaded as part of the OSGeo4W, FWTools and MS4W distributions.

7.2. Raster Data 331

MapServer Documentation, Release 6.0

See Also:
Tile Indexes

7.2.5 Raster Warping

MapServer is able to resample GDAL rasters on the fly into new projections. Non-GDAL rasters may only be up or
down sampled without any rotation or warping.
Raster warping kicks in if the projection appears to be different for a raster layer than for the map being generated.
Warped raster layers are significantly more expensive to render than normal raster layers with rendering time being
perhaps 2-4 times long than a normal layer. The projection and datum shifting transformation is computed only at
selected points, and generally linearly interpolated along the scanlines (as long as the error appears to be less than
0.333 pixels.
In addition to reprojecting rasters, the raster warping ability can also apply rotation to GDAL rasters with rotational
coefficients in their georeferencing information. Currently rotational coefficients won’t trigger raster warping unless
the map and layer have valid (though matching is fine) projection definitions.

7.2.6 24bit RGB Rendering

Traditionally MapServer has been used to produce 8 bit pseudo-colored map displays generated from 8bit greyscale
or pseudocolored raster data. However, if the raster file to be rendered is actually 24bit (a red, green and blue band)
then additional considerations come into play. Currently rendering of 24bit imagery is only supported via the GDAL
renderer. The built-in PNG, JPEG and other drivers do not support 24bit input images.
If the output is still 8bit pseudo-colored (the IMAGEMODE is PC256 in the associated OUTPUT format declaration)
then the full 24bit RGB colors for input pixels will be converted to a color in the colormap of the output image. By
default a color cube is used. That is a fixed set of 175 colors providing 5 levels of red, 7 levels of green and 5 levels
of blue is used, plus an additional 32 greyscale color entries. Colors in the input raster are mapped to the closest color
in this color cube on the fly. This substantial degrades color quality, especially for smoothly changing images. It also
fills up the colors table, limited to 256 colors, quite quickly.
A variation on this approach is to dither the image during rendering. Dithering selects a color for a pixel in a manner
that “diffuses error” over pixels. In an area all one color in the source image, a variety of output pixel colors would
be selected such that the average of the pixels would more closely approximate the desired color. Dithering also takes
advantage of all currently allocated colors, not just those in the color cube. Dithering requires GDAL 1.1.9 or later, and
is enabled by providing the PROCESSING “DITHER=YES” option in the mapfile. Dithering is more CPU intensive
than using a simple color cube, and should be avoided if possible in performance sensitive situations.
The other new possibility for handling 24bit input imagery in MapServer 4.0 or later, is to produce 24bit output images.
The default “IMAGETYPE png24” or “IMAGETYPE jpeg” declaration may be used to produce a 24bit PNG output
file, instead of the more common 8bit pseudo-colored PNG file. The OUTPUTFORMAT declaration provides for
detailed control of the output format. The IMAGEMODE RGB and IMAGEMODE RGBA options produce 24bit and
32bit (24bit plus 8bit alpha/transparency) for supported formats.

7.2.7 Special Processing Directives

As of MapServer 4.0, the PROCESSING parameter was added to the LAYER of the Mapfile. It is primarily used to
pass specialized raster processing options to the GDAL based raster renderer. The following processing options are
supported in MapServer 4.0 and newer.
DITHER=YES This turns on error diffusion mode, used to convert 24bit images to 8bit with error diffusion to get
better color results.

332 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Example:
PROCESSING "DITHER=YES"

BANDS=red_or_grey[,green,blue[,alpha]] This directive allows a specific band or bands to be selected from a raster
file. If one band is selected, it is treated as greyscale. If 3 are selected, they are treated as red, green and blue. If
4 are selected they are treated as red, green, blue and alpha (opacity).
Example:
PROCESSING "BANDS=4,2,1"

SCALE[_n]=AUTO or min,max This directive instructs the GDAL reader to pre-scale the incoming raster data. It
is primarily used to scale 16bit or floating point data to the range 0-255, but can also be used to constrast stretch
8bit data. If an explicit min/max are provided then the input data is stretch (or squished) such that the minimum
value maps to zero, and the maximum to 255. If AUTO is used instead, a min/max is automatically computed.
To control the scaling of individual input bands, use the SCALE_1, SCALE_2 and SCALE_3 keywords (for
red, green and blue) instead of SCALE which applies to all bands.
Example:
PROCESSING "SCALE=AUTO"
or
PROCESSING "SCALE_1=409,1203"
PROCESSING "SCALE_2=203,296"
PROCESSING "SCALE_3=339,1004"

LUT[_n]=<lut_spec> This directive (MapServer 4.9+) instructs the GDAL reader to apply a custom LUT (lookup
table) to one or all color bands as a form of on the fly color correction. If LUT is used, the LUT is applied to all
color bands. If LUT_n is used it is applied to one color band (n is 1 for red, 2 for green, 3 for blue, 4 for alpha).
The LUT can be specified inline in the form:
<lut_spec> = <in_value>:<out_value>[,<in_value>:<out_value>]*

This essentially establish particular input values which are mapped to particular output values. The list implicitly
begins with 0:0, and 255:255. An actual 256 entry lookup table is created from this specification, linearly
interpolating between the values. The in values must be in increasing order. The LUT specification may also be
in a text file with the <lut_spec> being the filename. The file contents should be in the same syntax, and the file
is searched relative to the mapfile.
Example:
PROCESSING "LUT_1=red.lut"
PROCESSING "LUT_2=green.lut"
PROCESSING "LUT_3=blue.lut"
or
PROCESSING "LUT=100:30,160:128,210:200"

As a special case there is also support for GIMP format curve files. That is the text files written out by the
Tools->Color->Curves tool. If this is specified as the filename then it will be internally converted into linear
segments based on the curve control points. Note that this will not produce exactly the same results as the
GIMP because linear interpolation is used between control points instead of curves as used in the GIMP. For a
reasonable number of control points the results should be similar. Also note that GIMP color curve files include
an overall “value” curve, and curves for red, green, blue and alpha. The value curve and the appropriate color
curve will be composed internally to produce the final LUT.
Example:

7.2. Raster Data 333

MapServer Documentation, Release 6.0

PROCESSING "LUT=munich.crv"

COLOR_MATCH_THRESHOLD=n Alter the precision with which colors need to match an entry in the color
table to use it when producing 8bit colormapped output (IMAGEMODE PC256). Normally colors from a raster
colormap (or greyscale values) need to match exactly. This relaxes the requirement to being within the specified
color distance. So a COLOR_MATCH_THRESHOLD of 3 would mean that an existing color entry within 3
(sum of difference in red, green and blue) would be used instead of creating a new colormap entry. Especially
with greyscale raster layers, which would normally use all 256 color entries if available, this can be a good way
to avoid “stealing” your whole colormap for a raster layer. Normally values in the range 2-6 will give good
results.
Example:
PROCESSING "COLOR_MATCH_THRESHOLD=3"

RESAMPLE=NEAREST/AVERAGE/BILINEAR This option can be used to control the resampling kernel used
sampling raster images. The default (and fastest) is NEAREST. AVERAGE will perform compute the average
pixel value of all pixels in the region of the disk file being mapped to the output pixel (or possibly just a sampling
of them). BILINEAR will compute a linear interpolation of the four pixels around the target location. This topic
is discussed in more detail in MS RFC 4: MapServer Raster Resampling.
Resampling options other than NEAREST result in use of the generalized warper and can dramatically slow
down raster processing. Generally AVERAGE can be desirable for reducing noise in dramatically downsampled
data, and can give something approximating antialiasing for black and white linework. BILINEAR can be
helpful when oversampling data to give a smooth appearance.
Example (chose one):
PROCESSING "RESAMPLE=NEAREST"
PROCESSING "RESAMPLE=AVERAGE"
PROCESSING "RESAMPLE=BILINEAR"

LOAD_FULL_RES_IMAGE=YES/NO This option affects how image data is loaded for the resampler when repro-
jecting or otherwise going through complex resampling (as opposed to the fast default image decimation code
path). This forces the source image to be loaded at full resolution if turned on (default is NO). This helps work
around problems with default image resolution selection in when radical warping is being done. It can result in
very slow processing if the source image is large.
LOAD_WHOLE_IMAGE=YES/NO This option affects how image data is loaded for the resampler (as above). This
option, if turned on, will cause the whole source image to be loaded and helps make up for problem identifying
the area required, usually due to radical image reprojection near a dateline or projection “horizon”. The default
is NO. Turning this on can dramatically affect rendering performance and memory requirements.

7.2.8 Raster Query

A new feature added in MapServer 4.4 is the ability to perform queries on rasters in a manner similar to queries
against vector layers. Raster queries on raster layers return one point feature for each pixel matching the query. The
point features will have attributes indicating the value of different bands at that pixel, the final rendering color and
the class name. The resulting feature can be directly access in MapScript, or processed through templates much like
normal vector query results. Only raster layers with a query TEMPLATE associated can be queried, even for the query
methods that don’t actually use the query template (much like vector data).
Raster query supports QueryByPoint, QueryByRect, and QueryByShape. QueryByPoint supports single and multiple
result queries. Other query operations such as QueryByIndex, QueryByIndexAdd, QueryByAttributes and Query-
ByFeature are not supported for raster layers.
Raster layers do not support saving queries to disk, nor query maps.

334 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Raster queries return point features with some or all of the following attributes:
x georeferenced X location of pixel.
y georeferenced Y location of pixel.
value_list a comma separated list of the values of all selected bands at the target pixel.
value_n the value for the n’th band in the selected list at this pixel (zero based). There is one value_n
entry for each selected band.
class Name of the class this pixel is a member of (classified layers only).
red red component of the display color for this pixel.
green green component of the display color for this pixel.
blue blue component of the display color for this pixel.
The red, green and blue attribute are intended to be the final color the pixel would be rendered with, but in some subtle
cases it can be wrong (ie. classified floating point results). The selected bands are normally the band that would be
used to render the layer. For a pure query-only layer BANDS PROCESSING directive can be used to select more
bands than could normally be used in a render operation. For instance for a 7 band landsat scene a PROCESSING
“BANDS=1,2,3,4,5,6,7” directive could be used to get query results for all seven bands in results to a query operation.
Care should be taken to avoid providing a large query area (selecting alot of pixels) as each selected pixel requires
over 100 bytes of memory for temporary caching. The RASTER_QUERY_MAX_RESULT PROCESSING item can
be used to restrict the maximum number of query results that will be returned. The default is one million which would
take on the order of 100MB of RAM.
Query results can be returned as HTML via the normal substitution into query template HTML. Query results are also
accessible via WMS GetFeatureInfo calls, and from MapScript. The following example shows executing a feature
query from Python MapScript and fetching back the results:
map = mapscript.Map(’rquery.map’)
layer = map.getLayer(0)

pnt = mapscript.Point()
pnt.x = 440780
pnt.y = 3751260

layer.queryByPoint( map, pnt, mapscript.MS_MULTIPLE, 180.0 )

layer.open()
for i in range(1000):
result = layer.getResult( i )
if result is None:
break

print ’(%d,%d)’ % (result.shapeindex, result.tileindex)

s = layer.getShape( result.shapeindex, result.tileindex )

for i in range(layer.numitems):
print ’%s: %s’ % (layer.getItem(i), s.getValue(i))

layer.close()

This following is a simple example query TEMPLATE file. The raster pixel attributes will be substituted in before the
query result is returned to the user as HTML.
Pixel:<br>
values=[value_list]<br>

7.2. Raster Data 335

MapServer Documentation, Release 6.0

value_0=[value_0]<br>
value_1=[value_1]<br>
value_2=[value_2]<br>
RGB = [red],[green],[blue]<p>
Class = [class]<br>

Internally raster query results are essentially treated as a set of temporary features cached in RAM. Issuing a new
query operation clears the existing query cache on the layer. The transitory in-memory representation of raster query
results is also responsible for the inability to save raster query results since saved query results normally only contain
the feature ids, not the entire features. Some addition information is available in the RasterQuery Wiki topic.

7.2.9 Raster Display Performance Tips

• Build overview levels for large rasters to ensure only a reasonable amount of data needs to be touched to display
an overview of a large layer. Overviews can be implemented as a group of raster layers at different resolutions,
using MINSCALE, and MAXSCALE to control which layers are displayed at different resolutions. Another,
perhaps easier way, is to build overviews for GDAL supported formats using the gdaladdo utility.
• When using tileindexes to manage many raster files as a single file, it is especially important to have an overview
layer that kicks in at high scales to avoid having to open a large number of raster files to fulfill the map request.
• Preprocess RGB images to eightbit with a colormap to reduce the amount of data that has to be read, and the
amount of computation to do on the fly.
• For large images use tiling to reduce the overhead for loading a view of a small area. This can be accomplished
using the TILEINDEX mechanism of the mapfile, or by creating a tiled format file (ie. TIFF with GDAL).
• Ensure that the image is kept on disk in the most commonly requested projection to avoid on-the-fly image
warping which is fairly expensive.
• If you are getting debug output from MapServer in your web server log file, check to see if the message msRe-
sampleGDALToMap in effect appears. If so, the raster layer is being resampled. If you don’t think it should be
resampled carefully review your map file to ensure that the layer projection exactly matches the map projection
or that the layer has no projection definition.

7.2.10 Preprocessing Rasters

The following operations use GDAL commandline utilities, some of which are python scripts. They are generally
available on any GDAL installation with python support.

Producing Tiled Datasets

The TIFF and Erdas Imagine formats support internal tiling within files, and will generally give better display speed
for local map requests from large images. To produce a GeoTIFF file in internally tiled format using the TILED=YES
creation option with the gdal_translate utility:
gdal_translate -co TILED=YES original.tif tiled.tif

Erdas Imagine (HFA) files are always tiled, and can be larger than 4GB (the GeoTIFF limit). Use a command like the
following to translate a raster to Imagine format:
gdal_translate -of HFA original.tif tiled.img

336 Chapter 7. Data Input

 MapServer Documentation, Release 6.0

Reducing RGB to 8bit

Rendering and returning 24bit images (especially as PNG) can be quite expensive in render/compress time and band-
width. Pre-reducing raster data to 8bit can save disk space, processing time, and bandwidth. However, such a color
reduction also implicitly reduces the quality of the resulting map. The color reduction can be done on the fly by
MapServer but this requires even more processing. A faster approach is to pre-reduce the colors of 24bit imagery to
8bit. This can be accomplished with the GDAL rgb2pct.py script like this:
rgb2pct.py original.tif 8bit.tif

By default images will be reduced to 256 colors but this can mean there are not enough colors to render other colors
in the map. So it may be desired to reduce to even less colors:
rgb2pct.py -n 200 original.tif 8bit.tif

Downsampling to 8bit should be done before internal tiling and overview building. The rgb2pct.py script tries to
compute an optimal color table for a given image, and then uses error diffusion during the 24bit to 8bit reduction.
Other packages (such as ImageMagick or Photoshop) may have alternative color reduction algorithms that are more
appropriate for some uses.

Building Internal Overviews

Most GDAL supported raster formats can have overviews pre-built using the gdaladdo utility. However, a few formats,
such as JPEG2000, MrSID, and ECW already contain implicit overviews in the format themselves and will not gener-
ally benefit from external overviews. For other formats (such as GeoTIFF, and Erdas Imagine format) use a command
like the following to build overviews:
gdaladdo tile.tif 2 4 8 16 32 64 128

The above would build overviews at x2 through x128 decimation levels. By default it uses “nearest neighbour”
downsampling. That is one of the pixels in the input downsampled area is selected for each output pixel. For some
kinds of data averaging can give much smoother overview results, as might be generated with this command:
gdaladdo -r average tiled.tif 2 4 8 16 32 64 128

Note that overview building should be done after translating to a final format. Overviews are lost in format conversions
using gdal_translate. Also, nothing special needs to be done to make MapServer use GDAL generated overviews. They
are automatically picked up by GDAL when mapserver requests a reduced resolution map.

Building External Overviews

When working with large collections of raster files using a MapServer tileindex, it is desirable to build reduced reso-
lution overview layers that kick in at high scales (using MINSCALE/MAXSCALE to control which layer activates).
Preparing the overviews can be a somewhat complex process. One approach is to use the gdal_merge.py script to
downsample and mosaic all the images. For instance if we want to produce an overview of many 1meter ortho photos
with 250 meter pixels we might do something like:
gdal_merge.py -o overview.tif -ps 250 250 ortho_*.tif

The gdal_merge.py utility suffers from a variety of issues, including no support for different resampling kernels. With
GDAL 1.3.2 or later it should be able to accomplish something similar with the more flexible gdalwarp utility:
gdalwarp -rc -tr 250 250 ortho_*.tif overview.tif

In some cases the easiest way of generating an overview is to let MapServer do it using the shp2img utility. For
instance if the tileindex layer is called ‘’orthos” we could do something like:

7.2. Raster Data 337

MapServer Documentation, Release 6.0

shp2img -m ortho.map -l orthos -o overview.png

Note that the overview will be generated with the extents and size in the .map file, so it may be necessary to temporarily
adjust the map extents and size values to match the raster extents and the desired output size. Also, if using this method,
don’t leave large files in PNG (or GIF or JPEG) format as they are slow formats to extract subareas from.

7.2.11 Georeference with World Files

World files are a simple mechanism for associating georeferencing (world coordinates) information with raster files.
ESRI was the first company to propagate the use of world files, and they often used with TIFF instead of embedding
georeferencing information in the file itself.
The world file contents look like the following. The first coefficient is the X pixel size. The second and third are
rotational/shear coefficients (and should normally be 0.0). The fourth is the Y pixel size, normally negative indicating
that Y decreases as you move down from the top left origin. The final two values are the X and Y location of the
center of the top left pixel. This example is for an image with a 2m x 2m pixel size, and a top left origin at (356800E,
5767999N):
2
0.0000000000
0.0000000000
-2
356800.00
5767999.00

The name of the world file is based on the file it relates to. For instance, the world file for aerial.tif might be aerial.tfw.
Conventions vary for appropriate endings, but with MapServer the extension .wld is always OK for world files.
Since the GDAL/OGR library is used for vector and raster access in MapServer, many more formats are supported, so
please see the OGR (vector) and GDAL (raster) formats pages.

338 Chapter 7. Data Input

 CHAPTER 8

Output Generation

8.1 AGG Rendering Specifics

Author Thomas Bonfort

Contact thomas.bonfort at gmail
Revision $Revision: 8295 $
Date $Date: 2008-12-26 21:08:04 -0800 (Fri, 26 Dec 2008) $
Last Updated 2008/11/24

Table of Contents
• AGG Rendering Specifics
– Introduction
– Setting the OutputFormat
– New Features
– Modified Behavior

8.1.1 Introduction

MapServer 5.0 released with a new rendering backend. This howto details the changes and new functionality that this
adds to map creation. This howto assumes you already now the basics of mapfile syntax. If not, you should probably
be reading the mapfile syntax.

8.1.2 Setting the OutputFormat

24 bit png (high quality, large file size):

OUTPUTFORMAT
NAME ’AGG’
DRIVER AGG/PNG
IMAGEMODE RGB
END

339
MapServer Documentation, Release 6.0

24 bit png, transparent background:

OUTPUTFORMAT
NAME ’AGGA’
DRIVER AGG/PNG
IMAGEMODE RGBA
END

24 bit jpeg (jpeg compression artifacts may appear, but smaller file size):
OUTPUTFORMAT
NAME ’AGG_JPEG’
DRIVER AGG/JPEG
IMAGEMODE RGB
END

png output, with number of colors reduced with quantization.

OUTPUTFORMAT
NAME ’AGG_Q’
DRIVER AGG/PNG
IMAGEMODE RGB
FORMATOPTION "QUANTIZE_FORCE=ON"
FORMATOPTION "QUANTIZE_DITHER=OFF"
FORMATOPTION "QUANTIZE_COLORS=256"
END

8.1.3 New Features

• All rendering is now done antialiased by default. All ANTIALIAS keywords are now ignored, as well as
TRANSPARENCY ALPHA. Pixmaps and fonts are now all drawn respecting the image’s internal alpha channel
(unless a backgroundcolor is specified).
• As with GD in ver. 4.10, using a SYMBOL of type ELLIPSE to draw thick lines isn’t mandatory anymore. To
draw a thick line just use:
STYLE
WIDTH 5
COLOR 0 0 255
END

• A line symbolizer has been added, that works with vector or pixmap symbols, to draw textured lines. This
happens by default if a line’s style is given a symbol of type vector or pixmap. To enable “shield” symbolization,
i.e. a marker placed only on some points of the line, you must add a GAP parameter to your symbol definition.
This GAP value is scaled w.r.t the style’s SIZE parameter. Specify a positive gap value for symbols always
facing north (optionally rotated by the ANGLE of the current style), or a negative value for symbols that should
follow the line orientation

340 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

• This happens by default if a line’s style is given a symbol of type vector or pixmap. To enable “shield” sym-
bolization, i.e. a marker placed only on some points of the line, you must add a GAP parameter to your symbol
definition. This GAP value is scaled w.r.t the style’s SIZE parameter - specify a positive gap value for symbols
always facing north (optionally rotated by the ANGLE of the current style), or a negative value for symbols that
should follow the line orientation
• Pixmap and font symbols can now be rotated without loosing their transparency
• For POLYGON layers with no specific SYMBOL, the WIDTH keyword specifies the width of the outline, if
an OUTLINECOLOR was specified. This is a shorthand that avoids having to create multiple styles for basic
rendering, and will provide a marginal performance gain. Note that in this case, the width of the outline is /not/
scale dependent.

8.1.4 Modified Behavior

• When specifying a SYMBOL for a polygon shape, the GAP parameter of the symbol is used as a separation
between each rendered symbol. This works for symbols of type vector, pixmap and ellipse. For example a
symbol defined by
SYMBOL
NAME ’triangle’
TYPE VECTOR
FILLED TRUE
POINTS
0 1
.5 0
1 1
0 1
END
GAP 1
END

that is rendered in a class where SIZE is 15 will be rendered like

• layers of type CIRCLE support hatch type symbol filling

• the ENCODING keyword for labels is now enforced. If unset, MapServer will treat your label text byte-by-byte
(resulting in corrupt special characters).

8.2 AntiAliasing with MapServer

Author Pericles Nacionales

Contact naci0002 at umn.edu
Revision $Revision: 11477 $
Date $Date: 2011-04-06 04:51:20 -0700 (Wed, 06 Apr 2011) $
Last Updated 2009/01/17

8.2. AntiAliasing with MapServer 341

MapServer Documentation, Release 6.0

Note: For quality antialiased output from mapserver, it is highly recommended to use the AGG rendering. This
document applies only if you whish to stick to the GD rendering, or if you are using a version predating the 5.0 release
of mapserver.

Table of Contents
• AntiAliasing with MapServer
– What needs to be done

8.2.1 What needs to be done

1. Change (or add) IMAGETYPE keyword in MAP object to PNG24 (24-bit PNG output) or JPEG
MAP
...
IMAGETYPE P N G 24
...
END

2. Add TRANSPARENCY to the LAYER object and set value to ALPHA

MAP
...
IMAGETYPE P N G 24
...

LAYER
...
TRANSPARENCY ALPHA
...
END
END

3. Add ANTIALIAS keyword to the STYLE object within the CLASS object within the LAYER and set value to
TRUE
MAP
...
IMAGETYPE P N G 24
...

LAYER
...
TRANSPARENCY ALPHA
...
CLASS
...
STYLE
...
ANTIALIAS TRUE
...
END
\.\.\.
END # end class
END # end layer
END # end map

342 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Note: Don’t use the SYMBOL or the SIZE keywords within the CLASS object, instead use WIDTH to specify width
of line or polygon outline. Don’t use WIDTH unless you have to. If you must define a SYMBOL, use symbol of type
ELLIPSE–it supports antialiasing.
Here’s an example of a real-world mapfile:
Note: From Mapserver 6, symbol type CARTOLINE is no longer supported. You have to use AGG rendering and
STYLE PATTERN to achieve dashed lines. Therefore, the following example does not work anymore.
1 MAP
2 NAME ’ms101’
3 EXTENT -2198022.00 -2444920.25 2707932.00 1234545.25 # CONUS LAEA (US)
4 SIZE 640 480
5 SHAPEPATH ’data’
6 SYMBOLSET ’symbols/symbols.txt’
7

8 IMAGETYPE P N G 24
9

10 PROJECTION
11 "init=epsg:2163"
12 END
13

14 # The layer below will be rendered as 1-pixel wide, antialiased line

15 # If you’d like to change the line thickness add the WIDTH keyword
16 # in the STYLE object with a value of 3 or greater.
17 LAYER # begin antialiased country boundary (line) layer
18 NAME ’country_line’
19 DATA ’shapefile/WorldCountryBorders’
20 TYPE LINE
21 STATUS ON
22 TRANSPARENCY ALPHA
23

24 PROJECTION
25 "init=epsg:4326"
26 END
27

28 CLASS
29 NAME ’Country Boundary’
30 STYLE
31 COLOR 96 96 96
32 ANTIALIAS TRUE
33 END
34 END
35 END # end country boundary layer
36

37 # The layer below shows one way to draw a polygon with antialiased outline
38 LAYER # begin antialiased country boundary (polygon) layer
39 NAME ’country_line’
40 DATA ’shapefile/Countries_area’
41 TYPE POLYGON
42 STATUS ON
43 TRANSPARENCY ALPHA
44

45 PROJECTION
46 "init=epsg:4326"
47 END
48

49 CLASS
50 NAME ’Country Boundary’

8.2. AntiAliasing with MapServer 343

 MapServer Documentation, Release 6.0

51 STYLE
52 COLOR 212 212 212
53 OUTLINECOLOR 96 96 96
54 WIDTH 3
55 ANTIALIAS TRUE
56 END
57 END
58 END # end country boundary polygon layer
59

60 # The layer below shows one way to draw a polygon with antialiased outline
61 LAYER # begin antialiased state boundary (line) layer
62 NAME ’state_line’
63 DATA ’shapefile/us_states’
64 TYPE LINE
65 STATUS ON
66 TRANSPARENCY ALPHA
67

68 PROJECTION
69 "init=epsg:4326"
70 END
71

72 CLASS
73 NAME ’State Boundary’
74 STYLE
75 COLOR 144 144 144
76 SYMBOL ’cartoline’
77 ANTIALIAS TRUE
78 END
79 END
80 END # end state line layer
81 END # end of map file

Here’s how the ‘cartoline’ symbol is defined:

Note: From Mapserver 6, symbol type CARTOLINE is not available. You have to use AGG rendering and STYLE
PATTERN to achieve dashed lines. Therefore, the following symbol can not be used anymore.
SYMBOL
NAME ’cartoline’
TYPE CARTOLINE
LINECAP "round"
LINEJOIN "round"
LINEJOINMAXSIZE 3
END

Note: The examples provided here are for illustrative purposes only–keep your map file definitions simple. Antialias-
ing adds computing overhead on the server and could slow/degrade its performance. Don’t use it unless you must and
certainly don’t use symbols with it unless you really have to.

8.3 Dynamic Charting

Author Thomas Bonfort

Contact thomas.bonfort at gmail.com
Revision $Revision: 8439 $
Date $Date: 2009-01-17 07:27:10 -0800 (Sat, 17 Jan 2009) $

344 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Last Updated 2009/01/17

Table of Contents
• Dynamic Charting
– Setup
– Adding a Chart Layer to a Mapfile
– Pie Charts
– Bar Graphs

Starting with version 5.0, MapServer included the ability to automatically draw pie or bar graphs whose values are
taken and adjusted from attributes of a datasource.
This document assumes that you are already familiar with MapServer application development and especially setting
up Mapfile s. You can also check out the Vector Data Access Guide, which has lots of examples of how to access
specific data sources.

8.3.1 Setup

Supported Renderers

Dynamic charts are supported solely with the GD and AGG renderers.
Attempting to add a chart layer with any other renderer (e.g. PDF or SWF) will result in an error. Rendering quality
with the GD renderer is less than optimal, especially with small graphs, due to the lack of subpixel rendering functions.

Output from AGG and GD Renderers

MapServer AGG Rendering

MapServer GD Rendering

8.3.2 Adding a Chart Layer to a Mapfile

Layer Type

A new type of layer has been added to the mapfile syntax. To specify a chart layer, use

8.3. Dynamic Charting 345

MapServer Documentation, Release 6.0

LAYER
...
TYPE CHART
...
END

No other specific keywords have been added in order to keep the number of different keywords to a minimum in the
mapfile syntax, therefore all the chart specific configuration is determined by PROCESSING directives.

Specifying the Size of each Chart

The size of each chart is specified by the CHART_SIZE directive. If two values are given for this parameter, this will
specify the width and height of each chart (this only applies for bar graphs). By default, the charts are 20x20 pixels.
LAYER
TYPE CHART
PROCESSING "CHART_SIZE=21" # specify size of the chart for pie or bar graphs
#PROCESSING "CHART_SIZE=20 10" # specify width and height for bar graphs
...
END

From version 5.2 and onwards, the diameter of a pie chart can be bound to an attribute,using the
CHART_SIZE_RANGE PROCESSING attribute:
PROCESSING "CHART_SIZE_RANGE = itemname minsize maxsize minval maxval"

where:
• itemname is the name of the attribute that drives the chart size (e.g. total_sales)
• minsize and maxsize are the minimum and maximum chart size values in pixels (e.g. “10 100”)
• minval and maxval are the minimum values of the attribute that correspond to chart sizes of minsize and maxsize
(e.g. 10000 1000000).
If the attribute value is smaller than ‘minval’ then the chart size will be minsize pixels, and if the attribute value is
larger than maxval, the chart size will be maxsize pixels.

Specifying the Values to be Plotted

Each value to be plotted (i.e. a slice in a pie chart, or a bar in a par graph) is specified in a CLASS of the chart layer.
The value to be plotted is taken from the SIZE keyword from the first STYLE block of the class. This is semantically
a bit awkward, but keeps the number of different keywords to a minimum in the mapfile syntax. The value given to
the SIZE keyword could of course be given a static value, but dynamic charting really only makes sense with attribute
binding.
LAYER
...
CLASS
# include a NAME keyword if you want this class to be included
# in the legend
NAME "value 1"
STYLE
# specify which value from the data source will be used as the
# value for the graph
SIZE [ a t t r i b u t e ]
...
END

346 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

END
CLASS
...
END
...
END

At least 2 CLASS blocks must be specified before charting can occur (but you already knew this if you want your
charts to convey at least some information ;) ).

Specifying Style

The styling of each value in the charts is specified by the usual MapServer syntax. Only one style per class is supported,
any other STYLE block will be silently ignored. Only a subset of the styling keywords are supported:
STYLE
SIZE [ a t t r i b u t e ]
# specify the fill color
COLOR r g b

# if present will draw an outline around the corresponding bar or slice

OUTLINECOLOR r g b

#specify the width of the outline if OUTLINECOLOR is present (defaults to 1)

WIDTH w

# only for pie charts. ’a’ is the number of pixels the corresponding
# slice will be offset relative to the center of the pie. This is useful
# for emphasizing a specific value in each chart. ’b’ is required by the
# mapfile parser but is ignored.
OFFSET a b
END

8.3.3 Pie Charts

This is the default type of chart that is rendered. This can also be specifically set with a PROCESSING keyword in the
layer attributes:
PROCESSING "CHART_TYPE=PIE"

For each shape in the layer’s datasource, the STYLE SIZE is used to set the relative size (value) of each pie slice, with
the angles of the slices that are automatically computed so as to form a full pie. For example:
1 LAYER
2 NAME "Ages"
3 TYPE CHART
4 CONNECTIONTYPE postgis
5 CONNECTION "blabla"
6 DATA "the_geom from demo"
7 PROCESSING "CHART_TYPE=pie"
8 PROCESSING "CHART_SIZE=30"
9 STATUS ON
10 CLASS
11 NAME "Population Age 0-19"
12 STYLE
13 SIZE [ v 1006]

8.3. Dynamic Charting 347

 MapServer Documentation, Release 6.0

14 COLOR 255 244 237

15 END
16 END
17 CLASS
18 NAME "Population Age 20-39"
19 STYLE
20 SIZE [ v 1007]
21 COLOR 255 217 191
22 END
23 END
24 CLASS
25 NAME "Population Age 40-59"
26 STYLE
27 SIZE [ v 1008]
28 COLOR 255 186 140
29 END
30 END
31 END

In the example above, if for a given shape we have v1006=1000, v1007=600 and v1008=400 then the actual pie slices
for each class will be respectively 50%, 30% and 20% of the total pie size.

8.3.4 Bar Graphs

Bar graph drawing is set with a PROCESSING keyword in the layer attributes:
PROCESSING "CHART_TYPE=BAR"

For each shape in the layer’s datasource, the STYLE SIZE is used to set the relative size (value) of each bar in the
graph. By default, the vertical axis of each bar graph is scaled for the values of the corresponding shape, and will
always include the origin (=0). For example
• a shape whose STYLE SIZEs contains values {5,8,10,3} will be plotted on a graph whose vertical axis spans 0
to 10.
• a shape whose STYLE SIZEs contains values {-5,-8,-10,-3} will be plotted on a graph whose vertical axis spans
-10 to 0.
• a shape whose STYLE SIZEs contains values {-5,-8,10,3} will be plotted on a graph whose vertical axis spans
-8 to 10.
Additional PROCESSING directives are used to optionally specify the bounds of vertical axes so that the graphs for
all the shapes can be plotted with the same scale:
PROCESSING "CHART_BAR_MINVAL=val"
PROCESSING "CHART_BAR_MAXVAL=val"

Values in the datasource that are above CHART_BAR_MAXVAL or below CHART_BAR_MINVAL will be clipped
respectively to these values. If only one of these directives is included, the other will be automatically adjusted for
each shape to include at least the origin, i.e. the graphs for all the shapes will be in the same scale only if all the values
are of the same sign (positive or negative).

8.4 Flash Output

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com

348 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Author Yewondwossen Assefa

Contact assefa at dmsolutions.ca
Revision $Revision: 8370 $
Date $Date: 2008-12-31 12:32:02 -0800 (Wed, 31 Dec 2008) $
Last Updated 2008/07/15

Table of Contents
• Flash Output
– Introduction
– Installing MapServer with Flash Support
– How to Output SWF Files from MapServer
– What is Currently Supported and Not Supported

8.4.1 Introduction

Since MapServer 4.0, MapServer can output Flash files, in SWF format (or “Shockwave Flash Format”). The following
document outlines how to enable Flash output in MapServer.

Links to Flash-Related Information

• Open Source Flash Viewer

• Flash maps demo

8.4.2 Installing MapServer with Flash Support

To check that your mapserv executable includes Flash support, use the “-v” command-line switch and look for “OUT-
PUT=SWF”.
$ ./mapserv -v
MapServer version 5.2.0-rc1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
OUTPUT=PDF OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=AGG
SUPPORTS=FREETYPE SUPPORTS=ICONV SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT
SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER
SUPPORTS=SOS_SERVER SUPPORTS=FASTCGI SUPPORTS=THREADS
SUPPORTS=GEOS SUPPORTS=RGBA_PNG INPUT=JPEG INPUT=POSTGIS
INPUT=ORACLESPATIAL INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE

Using Pre-compiled Binaries

Windows users can use MS4W, which supports SWF output.

Compiling MapServer with Flash Support

The library chosen to output SWF files is the Ming library. Ming is a C library for generating SWF (“Flash”) format
movies, and it contains a set of wrappers for using the library from C++ and popular scripting languages like PHP,
Python, and Ruby.

8.4. Flash Output 349

MapServer Documentation, Release 6.0

Building on Windows

• download the Ming library (the version currently supported is 0.2a)

• as of Ming 0.3 there was no makefile for Windows available in the distribution yet, but you can download a MS
VC++ makefile (makefile.vc) from here (contains makefile and also libming.lib)
• copy makefile.vc under the src directory (ming-0.2/src)
• execute:
nmake /f makefile.vc

• at this point you should have a libming.lib that will be linked with MapServer
• edit the nmake.opt in your MapServer directory and uncomment the MING=-DUSE_MING_FLASH flag, and
point MING_DIR to your Ming directory.
• build MapServer as usual

Building on Unix

Use the “–with-ming” configure flag to enable MING support on Unix. “–with-ming=dir” will try to find the include
files and library in the indicated directory.
Note: compiling MapServer 4.4.2 with flash support (mingbeta version 0.3) requires
the -DMING_VERSION_03 option otherwise the make fails. This option should be included
in the configure.in after -DUSE_MING_FLASH as below:

MING ENABLED= "-DUSE_MING_FLASH -DMING_VERSION_03"

8.4.3 How to Output SWF Files from MapServer

SWF output is specified by using the OUTPUTFORMAT object. There are 2 possible output types:
1. A single movie containing the raster output for all the layers. To enable this, declare the following in the map
file:
OUTPUTFORMAT
NAME swf
MIMETYPE "application/x-shockwave-flash"
DRIVER swf
IMAGEMODE PC256
FORMATOPTION "OUTPUT_MOVIE=SINGLE"
END

2. A movie for every layer (vector movies for vector layers and raster movies for raster layers). To enable this,
declare the following in the map file:
OUTPUTFORMAT
NAME swf
MIMETYPE "application/x-shockwave-flash"
DRIVER swf
IMAGEMODE PC256
FORMATOPTION "OUTPUT_MOVIE=MULTIPLE"
END

350 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Other OutputFormat Options

• FORMATOPTION “FULL_RESOLUTION=FALSE”
The FULL_RESOLUTION applies only for vector layers. If set to FALSE, filtering will be applied to the vector
elements. It results in a smaller SWF file. The default value is TRUE.
• FORMATOPTION “LOAD_AUTOMATICALLY=OFF”
Setting this option to OFF will not load the SWF files for each layer. The default value is ON.

Composition of the Resulting SWF Files

Several SWF Files will be produced from a single map file: there will be one SWF file for each layer defined in the
map file and one ‘main’ SWF file containing critical information on the map file and the layers produced.
• The ‘main’ SWF File will contain Action Script (AS) code that gives critical information on the map file and
the SWF layers produced. Basically there will be an object called mapObj containing the height, width, extent,
scale, number of layers, etc. Here is an example (in AS) of the contents of this main movie:
mapObj = new Object ();
mapObj.name = "DEMO_SWF";
mapObj.width = 400;
mapObj.height = 300;
mapObj.extent = "-2594561.353333,3467361.353333,3467361.353333,3840000.000000"; ;
mapObj.numlayers = 4;
mapObj.layers = new Array ();
function LayerObj (name, type, fullname, relativename) {
this.name = name;
this.type = type;
this.fullname = fullname;
this.relativename = relativename;
}
mapObj.layers[0] = new LayerObj ("park", "2", "c:/tmp/ms_tmp/102389536132841_layer_0.swf", "10
mapObj.layers[1] = new LayerObj ("popplace", "4", "c:/tmp/ms_tmp/102389536132841_layer_1.swf",
mapObj.layers[2] = new LayerObj ("rail", "1", "c:/tmp/ms_tmp/102389536132841_layer_2.swf", "10
mapObj.layers[3] = new LayerObj ("road", "1", "c:/tmp/ms_tmp/102389536132841_layer_3.swf", "10

This example is produced based on a mapfile with two layers defined in it. We create a layer class object
containing useful information on a layer. The parameters are:
– Name : the name found in the map file
– Type : the type of layer ( 0 = Point Layer; 1=Line; 2=Polygon; 3=Raster; 4=Annotation; 6=Circle)
– Fullname : Full name of the file with path included
– Relative name : Relative Name
For example you can use mapObj.layers[0].name to extract the name of the first layer.
Note: All map parameters from MapServer are not exported at this time. We should come up with a list of
information of what we want to output. Note that this information can be used in a Flash application to load the
SWF file, to build a legend, to build a scale bar, etc.
• SWF Files for each layer
Each layer defined in the mapfile will have an associated SWF file created. The names of these SWF files are
based on the name of the main file with an addition of ‘layer_X’ at the end of the name (where X is the layer
index).

8.4. Flash Output 351

MapServer Documentation, Release 6.0

These SWF files will contain vector and raster data as well as some Action Script depending on the layer and
some configurations in the map file. We will see these configurations in detail in the following section.

Exporting Attributes

Exporting attributes works on a layer basis (it is only available for Vector Layers). To be able to export attributes to
the SWF files, you have to define a metadata item called SWFDUMPATTRIBUTES in the layer section of the mapfile.
Here is an example :
...
LAYER
NAME park
METADATA
"DESCRIPTION""Parks"
"RESULT_FIELDS" "NAME_E YEAR_EST AREA_KMSQ"
"SWFDUMPATTRIBUTES" "NAME_E,AREA_KMSQ "
END
TYPE POLYGON
STATUS ON
DATA park
...

In the above example, the values for the attributes NAME_E and AREA_KMSQ will be exported for each element in
the layer.
The resulting SWF File will have the values of these attributes (written in Action Script). Here is an example related
to the above layer:
nAttributes= 2;
Attributes = new Array();
Attributes[0] = "NAME_E";
Attributes[1] = "AREA_KMSQ";
Element = new Array ();
Element[0] = new Array();
Element[0][0] = "Ellesmere Island National Park Reserve";
Element[0][1] = "1500";
Element[1][0] = " Aulavik National park";
Element[1][1] = "1500";

Events and Highlights

Here is what is currently implemented concerning events (events here refer to mouse events happening on an element.
The available events are MOUSEUP, MOUSEDOWN, MOUSEOVER, MOUSEOUT):
• Events are only available for layers that have defined attributes exported (using SWFDUMPATTRIBUTES).
This is like defining that a certain layer is queryable.
• When a mouse event happens on one of the elements, there is an Action Script call that is made:
_root.ElementSelecetd(LayerId, ShapeId, Event) . The Flash application who wants to receive these events
should define the function ElementSelected and use the information received to do actions like retrieving the
attribute values from the specific SWF for the specified shape and display it.
In order to have highlighting, it has to be defined when the SWF is produced (basically highlighting means that the
shape is redrawn using a different color).
As of MapServer 5.0, highlighting is available on queryable layers by using the QueryMap object in the map file to
extract the color and do a highlight when on MOUSEOVER. The current implementation will highlight all objects that
are in a layer that uses SWFDUMPATTRIBUTES, using the COLOR set in the QueryMap object in the mapfile.

352 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Before MapServer 5.0, all objects that are in a layer that uses SWFDUMPATTRIBUTES are highlighted using a red
color.

Fonts

Ming uses a special type of font called FDB files. It does not yet support Truetype fonts. Please refer to ming
documentation on how to produce FDB files.

Outputting Raster SWF for Vector Layers

One mechanism would be to use the metadata for layer objects to define a raster output for vector layers. We could
use something like “SWFOUTPUT” “RASTER”. If this sounds desirable, please file an enhancement ticket with this
request, specifying the “Output-SWF” component.

8.4.4 What is Currently Supported and Not Supported

1. Vector layers
• Layer Point (case MS_LAYER_POINT) : done
– msDrawMarkerSymbol
– msDrawLabel
• Layer line (case MS_LAYER_LINE) : done
– msDrawLineSymbol
– msDrawLabel
• Layer circle ( case MS_LAYER_CIRCLE) : not done (should be done easily but missing data for testing)
– omsCircleDrawLineSymbol
– omsCircleDrawShadeSymbol
• Layer annotation (case MS_LAYER_ANNOTATION): done
– omsDrawMarkerSymbol
– omsDrawLabel
• Layer Polygon (MS_SHAPE_POLYGON): done
– omsDrawShadeSymbol
– omsDrawLineSymbol
– omsDrawLabel
• Vector Low Level functions
– omsDrawMarkerSymbol

* case(MS_SYMBOL_TRUETYPE) : done
* case(MS_SYMBOL_PIXMAP) : done
* case(MS_SYMBOL_ELLIPSE) : done
* case(MS_SYMBOL_VECTOR) : done
– omsDrawLineSymbol

8.4. Flash Output 353

MapServer Documentation, Release 6.0

* case : simple line : done

• drawing with the symbols : not done
– omsDrawShadeSymbol

* case : solid fill polygon : done

* case : filled with symbols : cannot be implemented for now (tried to create a GD image to fill the
shape but files created were huge)
– omsCircleDrawLineSymbol : not done
– omsCircleDrawShadeSymbol : not done
– omsDrawLabel : done
– omsDrawLabelCache : done
– obillboard (shadow for texts) : not done
2. Raster Layer
• msDrawRasterLayer: done
3. WMS Layer
• msDrawWMSLayer: done
4. Surround components (Legend, scalebar) : not supported

8.5 HTML Legends with MapServer

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Revision $Revision: 8278 $
Date $Date: 2008-12-23 13:34:31 -0800 (Tue, 23 Dec 2008) $
Last Updated 2006/01/09

Contents
• HTML Legends with MapServer
– Introduction
– Sample Site Using the HTML Legend

8.5.1 Introduction

The HTML legend is an alternative to the traditional GIF legend in MapServer. The following document describes
the process for implementing an HTML legend in MapServer CGI applications (NOTE: MapServer version > 3.5 is
required).
This document assumes that you are already familiar with certain aspects of MapServer:
• Setting up MapServer mapfiles and templates.

354 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Implementation

Key components for generating HTML legends are 1) a template parameter in the legend object, 2) a CGI [legend] tag
in the HTML file, and 3) an HTML legend template file. So that means that if the HTML page has the CGI [legend]
parameter set, and the mapfile has a LEGEND object with its TEMPLATE set to a valid HTML legend file then an
HTML legend will be returned. The following sections discuss these components.

Legend Object of Mapfile

The HTML legend is enabled by a new TEMPLATE parameter in the Legend Object of the mapfile. If TEMPLATE
is set in the Legend Object, then the HTML legend template file is read and used to generate an HTML legend which
will be inserted at the location of the [legend] tag in the main HTML template. Similar to other MapServer templates,
the HTML legend template filename MUST end with an “.html” extension.
Example 1. Sample Legend Object with the new TEMPLATE parameter
...
# LEGEND object
LEGEND
STATUS ON
KEYSIZE 18 12
# LABEL object
LABEL
TYPE BITMAP
SIZE MEDIUM
COLOR 0 0 89
END
TEMPLATE "legend.html" ### HTML template file
END
...

If TEMPLATE is not set, then the [legend] tag produces a regular image in a GIF/PNG image (the traditional be-
haviour).

CGI [legend] tag

The traditional CGI [legend] tag returns the URL of an image, so it is usually used inside an <IMG SRC=[legend]> tag
in the HTML file. The new HTML [legend] tag returns a block of HTML, so when converting an existing application
template from using a traditional image legend to the new HTML legend, you have to remove the IMG tag in the
main application template. Also note that if legend mode is specified in the URL, then MapServer will return a gif
containing the whole legend if no template is specified.
See the CGI Reference doc for more information on CGI parameters.
Example 2. [legend] tag in the main HTML template (with TEMPLATE set)
...
<FONT SIZE=+1><B>Legend</B></FONT><BR><HR>[legend]<HR>
...

Example 3. [legend] tag in the main HTML template (with TEMPLATE not set)
...
<FONT SIZE=+1><B>Legend</B></FONT><BR><HR><IMG SRC="[legend]"><HR>
...

8.5. HTML Legends with MapServer 355

MapServer Documentation, Release 6.0

HTML Legend Template File

The HTML legend template file is a separate file that contains 0 or 1 of each of the following tags that define blocks
of HTML to use in building the legend:
[leg_group_html] ... [/leg_group_html]
[leg_layer_html <OPTIONAL PARAMS>] ... [/leg_layer_html]
[leg_class_html <OPTIONAL PARAMS>] ... [/leg_class_html]

Note
Any text or HTML tags outside the [leg_*_html] tag pairs in the legend template file are ignored by the template
parser.
The following example shows what an HTML legend TEMPLATE file could look like:
Example 4. An HTML legend TEMPLATE file
[leg_group_html]
<tr>
<td colspan=3 bgcolor=#cccccc><b>[leg_group_name]</b></td>
</tr>
[/leg_group_html]

[leg_layer_html order_metadata=legend_order opt_flag=5]

<tr>
<td>
<input type=checkbox name="map_[leg_layer_name]_status"
value=1 [if name=layer_status oper=eq value=2]CHECKED[/if]>
</td>
<td colspan=2>
<a href="[metadata name=href]">[metadata name=layer_title]</a>
</td>
</tr >
[/leg_layer_html]

[leg_class_html]
<tr>
<td width=15> </td>
<td>
<img src="[leg_icon width=20 height=10]" width=20 height=10>
</td>
<td>
[leg_class_name]
</td>
</tr>
[/leg_class_html]

Supported Tags for the TEMPLATE file:

HEADER block

Tag [leg_header_html]...[/leg_header_html]
Description HTML block to use as the header of the legend.

356 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

FOOTER block

Tag [leg_footer_html]...[/leg_footer_html]
Description HTML block to use as the footer of the legend.
Example 5. HTML Legend File Using Header/Footer Blocks
[leg_header_html]
<p><b>my header</b></p>
[/leg_header_html]

[leg_layer_html]
...
[/leg_layer_html]

[leg_footer_html]
<p><b>my footer</b></p>
[/leg_footer_html]

GROUP block

Tag [leg_group_html <OPTIONAL PARAMS>]...[/leg_group_html]

Description HTML block to use for layer group headers if layers should be grouped in the legend. If not
set then layers are not grouped in the legend.
When the [leg_group_html] tag is used, then layers that don’t belong to any group (i.e. LAYER GROUP not set in
the mapfile) and their classes will not show up at all in the legend. The group list is decided by the order_metadata
parameter, which is explained later.
SUPPORTED PARAMETERS:
Parameter opt_flag=<bit_mask>
Description Control the group’s display, by adding the following values (default is 15). The opt_flag is
applied on all layers in the group. If at least one layer matches the flag, the group will show up in
the legend.
1 If set, show group even if all layers in group are out of scale (default: hide
groups out of scale).
2 If set, show group even if all layers in group have status OFF (default: hide
groups with STATUS OFF).
4 If set, show group even if all layers in group are of type QUERY (default: hide
group of TYPE QUERY)
8 If set, show group even if all layers in group are of type ANNOTATION (default:
hide groups of TYPE ANNOTATION)
e.g. opt_flag=12 (shown below) means show all layer types, including QUERY and AN-
NOTATION layers (4 + 8)

[leg_group_html opt_flag=12]
...
[/leg_group_html]

SUPPORTED TAGS:

8.5. HTML Legends with MapServer 357

MapServer Documentation, Release 6.0

Tag [leg_group_name]
Description Returns the group’s name.
Tag [layer_status]
Description Returns the status of the first layer in the group.
Tag [leg_icon width=<optional_width> height=<optional_height>]
Description In the group context, the [leg_icon] tag returns the URL of a legend icon for the first class
in the first layer that’s part of this group.
Tag [metadata name=<metadata_field_to_display>]
Description Returns specified metadata value from web’s metadata.
e.g. the group block below simply displays the name of the group in the legend:
[leg_group_html]
<tr><td colspan=2><b>[leg_group_name]</b></td></tr>
[/leg_group_html]

LAYER block

Tag [leg_layer_html <OPTIONAL PARAMS>] ... [/leg_layer_html]

Description HTML block to use for layer header. If not set then no layer headers are displayed (could
allow a legend with only classes in it).
SUPPORTED PARAMETERS:
Parameter order_metadata=<field_to_order_by>
Description Specifies that the value of the layer metadata <field_to_order_by> controls the order and
visibility of the layers in the legend.
• Layers with <field_to_order_by> >= 0 are sorted in order of this value, with multiple layers with
same value being accepted, in which case the map layer orderapplies between those layers.
• Layers with <field_to_order_by> < 0 are always hidden in the legend.
Parameter opt_flag=<bit_mask>
Description Control the layer display process. Add the values below to acquire the desired options
(default is 15):
1 If set, show layer even if out of scale (default: hide layers out of scale).
2 If set, show layer even if status is OFF (default: hide layers with STATUS OFF).
4 If set, show layer even if type is QUERY (default: hide layers of TYPE QUERY)
8 If set, show layer even if type is ANNOTATION (default: hide layers of TYPE
ANNOTATION)
e.g. opt_flag=14 (shown below) means do not show layers in the legend that are out of
scale.

[leg_layer_html opt_flag=14]
...
[/leg_layer_html]

SUPPORTED TAGS:

358 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Tag [leg_layer_group]
Description Returns the group name of the layer. This was added to MapServer v4.8.
Tag [leg_layer_index]
Description Returns the mapfile index value of the layer, which is useful for ordering. This was added to
MapServer v4.8.
Tag [leg_layer_maxscale]
Description Returns the maximum scale set for the layer. This was added to MapServer v4.8.
Tag [leg_layer_minscale]
Description Returns the minimum scale set for the layer. This was added to MapServer v4.8.
Tag [leg_layer_name]
Description Returns the current LAYER NAME value.
Tag [leg_icon width=<optional_width> height=<optional_height>]
Description In the layer context, the [leg_icon] tag returns the URL of a legend icon for the first class in
this layer.
Tag [metadata name=<metadata_field_to_display>]
Description Returns specified metadata value from this layer’s metadata and web’s metadata.
e.g. the layer block below simply displays an icon of the layer’s class and the layer name:
[leg_layer_html]
<tr><td><img src=[leg_icon width=15 height=15]><b>[leg_layer_name]</b></td></tr>
[/leg_layer_html]

CLASS block

Tag [leg_class_html <OPTIONAL PARAMS>] ... [/leg_class_html]

Description HTML block to use for classes. If not set then no classes are displayed (could allow a legend
with only layer headers in it). Note that classes with NULL (i.e. empty) NAMEs are not displayed.
SUPPORTED PARAMETERS:
Parameter opt_flag=<bit_mask>
Description Control the layer (i.e. class) display process. Add the values below to acquire the desired
options (default is 15). Note that using this parameter for the CLASS block has the same effect as
using the opt_flag parameter in the LAYER block.
1 If set, show layer even if out of scale (default: hide layers out of scale).
2 If set, show layer even if status is OFF (default: hide layers with STATUS OFF).
4 If set, show layer even if type is QUERY (default: hide layers of TYPE QUERY)
8 If set, show layer even if type is ANNOTATION (default: hide layers of TYPE
ANNOTATION)
e.g. opt_flag=14 (shown below) means do not show classes in the legend that are out of
scale.

8.5. HTML Legends with MapServer 359

MapServer Documentation, Release 6.0

[leg_class_html opt_flag=14]
...
[/leg_class_html]

SUPPORTED TAGS:
Tag [leg_class_index]
Description Returns the mapfile index value of the class, which is useful for ordering and legend icon
creation. This was added to MapServer v4.8.
Tag [leg_class_maxscale]
Description Returns the maximum scale set for the class. This was added to MapServer v4.8.
Tag [leg_class_minscale]
Description Returns the minimum scale set for the class. This was added to MapServer v4.8.
Tag [leg_class_name]
Description Returns the CLASS NAME value.
Tag [leg_class_title]
Description Returns the CLASS TITLE value.
Tag [leg_layer_name]
Description Returns the parent layer name. This was added to MapServer v4.8.
Tag [leg_icon width=<optional_width> height=<optional_height>]
Description In the layer context, the [leg_icon] tag returns the URL of a legend icon for the first class in
this layer.
Tag [metadata name=<metadata_field_to_display>]
Description Returns specified metadata value from the metadata of the layer to which this class belongs
and web’s metadata.
e.g. the class block below simply displays an icon of the layer’s class and the class name:
[leg_class_html]
<tr><td><img src=[leg_icon width=15 height=15]><b>[leg_class_name]</b></td></tr>
[/leg_class_html]

CONDITIONAL text

[if] tags can be used in any of the [leg_*_html] tags above to place conditional text. The syntax is:
[if name=<field_to_check> oper=<eq|neq|isset|isnull> value=<to_compare_with_field>] ... [/if]

Note:
Nested IF’s are supported. Parameter “oper” can be “eq” for equal, “neq” for not equal, “isset” (self-explanatory), or
“isnull” (self-explanatory). The default value is equal.
Example 6. [if] tag can be used to maintain the state of a layer checkbox
[leg_layer_html order_metadata=legend_order opt_flag=5]
<tr>
<td>

360 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

<input type=checkbox name="map_[leg_layer_name]_status"

value=1 [if name=layer_status oper=eq value=2]CHECKED[/if] >
</td>
<td colspan=2>
<a href="[metadata name=href]">[metadata name=layer_title]</a>
</td>
</tr >
[/leg_layer_html]

The possible values that can be tested in an [if] tag depend on the context in which the [if] tag is used. At the moment,
the number of values that can be tested is limited, but new values may be added as needed.
Note that the order of the items in the following [if] contexts are listed by their order of precedence. The rule is always
that special keywords have top priority (e.g. layer_status, etc.), followed by layer-level metadata, and ending with
map-level metadata. The possible values that can be tested are as follows:
In a [leg_group_html] context:
• [if name=layer_status value=...] ... [/if]
value is the layer status of the first layer that belongs to the group in integer format: 0=OFF, 1=ON,
2=DEFAULT
• [if name=layer_visible value=...] ... [/if]
value is the visibility of the first layer in the group: 0=NOT VISIBLE, 1=VISIBLE
• [if name=group_name value=...] ... [/if]
• [if name=any_layer_metadata value=...] ... [/if]
Uses metadata value from the first layer in the mapfile that belongs to that group
• [if name=any_web_metadata value=...] ... [/if]
In a [leg_layer_html] context:
• [if name=layer_status value=...] ... [/if]
value is the layer’s status in integer format: 0=OFF, 1=ON, 2=DEFAULT
• [if name=layer_type value=...] ... [/if]
value is the layer’s type in integer format: 0=POINT, 1=LINE, 2=POLYGON, 3=RASTER, 4=AN-
NOTATION, 5=QUERY, 6=CIRCLE
• [if name=layer_name value=...] ... [/if]
value is the layer’s name in string format
• [if name=layer_group value=...] ... [/if]
value is the layer’s group name in string format
• [if name=layer_visible value=...] ... [/if]
value is the visibility of the layer: 0=NOT VISIBLE, 1=VISIBLE
• [if name=any_layer_metadata value=...] ... [/if]
• [if name=any_web_metadata value=...] ... [/if]
In a [leg_class_html] context:
• [if name=layer_status value=...] ... [/if]
value is the status of the layer in which the class is located

8.5. HTML Legends with MapServer 361

MapServer Documentation, Release 6.0

• [if name=layer_type value=...] ... [/if]

value is the layer’s type in integer format: 0=POINT, 1=LINE, 2=POLYGON, 3=RASTER, 4=AN-
NOTATION, 5=QUERY, 6=CIRCLE
• [if name=layer_name value=...] ... [/if]
value is the layer’s name in string format
• [if name=layer_group value=...] ... [/if]
value is the layer’s group name in string format
• [if name=layer_visible value=...] ... [/if]
value is the visibility of the layer: 0=NOT VISIBLE, 1=VISIBLE
• [if name=class_name value=...] ... [/if]
• [if name=any_layer_metadata value=...] ... [/if]
• [if name=any_web_metadata value=...] ... [/if]

8.5.2 Sample Site Using the HTML Legend

http://www2.dmsolutions.ca/msapps/itasca_legend/demo_init.html
This demo is based on the MapServer Itasca demo and contains several variations of HTML Legends, some of which
are listed below:
• “HTML Legend 1” - displays classes only, similar to the traditional legends:
[leg_class_html opt_flag=15]
<img src=[leg_icon]> [leg_class_name]<br>
[/leg_class_html]

• “HTML Legend 2” - displays layer titles with HREF links and classes:
[leg_layer_html order_metadata=WMS_ORDER visibility_flag=15]
<a href="[leg_layer_name]">[metadata name=WMS_TITLE]</a><BR>
[/leg_layer_html]

[leg_class_html visibility_flag=15]
<img src=[leg_icon]> [leg_class_name]<br>
[/leg_class_html]

• “HTML Legend 3” - displays layers by group, with checkboxes to turn layers on/off:
[leg_group_html]
<tr><td colspan=2><b>[leg_group_name]</b></td></tr>
[/leg_group_html]

[leg_layer_html order_metadata=WMS_ORDER opt_flag=15]

<tr>
<td><input type=checkbox name=layer value=[leg_layer_name]
[if name=layer_status value=1]CHECKED[/if]>
[if name=layer_type value=4]
<img src=[leg_icon width=22 height=18]>
[/if]
[if name=layer_type oper=neq value=4]<img src=[leg_icon]>[/if]
</td>
<td>

362 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

<a href="[leg_layer_name]">[metadata name=WMS_TITLE]</a>

</td>
</tr>
[/leg_layer_html]

8.6 HTML Imagemaps

Author David Fawcett

Contact david.fawcett at gmail.com
Last Updated 2008-10-08

Contents
• HTML Imagemaps
– Introduction
– Mapfile Layer Definition
– Templates
– Request URL
– Additional Notes
– More Information

8.6.1 Introduction

The shpxy method of creating imagemaps uses MapServer query functionality to build a html imagemap. Just like a
regular MapServer query, you send a query request and MapServer uses the templates to build a block of html that it
sends back to the browser. The first example shows you how to build an imagemap based on a point layer. An example
template for a polygon layer is also included.
Components
• MapServer mapfile
• query template file
• query header template
• query footer template

8.6.2 Mapfile Layer Definition

Here is a simple mapfile for our example

MAP
NAME "myMapFile
STATUS ON
SIZE 200 200
EXTENT 178784 4804000 772653 5483346

UNITS METERS
STATUS ON
SHAPEPATH "/web/maps/data"
IMAGECOLOR 255 255 255

8.6. HTML Imagemaps 363

MapServer Documentation, Release 6.0

WEB
IMAGEPATH "/web/maps/tmp/"
IMAGEURL "/maps/tmp/"
END

QUERYMAP
STATUS ON
STYLE NORMAL
END

LAYER
NAME "sites"
STATUS DEFAULT
TYPE point
DATA ’aqiAreas’
TEMPLATE "bodytemplate.html"
HEADER "imapheader.html"
FOOTER "imapfooter.html"
END
END

You can see that we have a mapfile with one point layer, and that it contains references to three query templates.

8.6.3 Templates

In MapServer, the query header and footers get processed only once. The main query template, ‘bodytemplate.html’
in this example, gets processed once for each record in the record set returned by the query.

Point Layers

Here is the query header, ‘imapheader.html’. It creates the opening tag for your html imagemap.
<map id="mymap" name="mymap">

Here is the query template, ‘bodytemplate.html’. It creates the body of the html imagemap.
<area shape="circle" coords="[shpxy precision=0 proj=image yf=",7" xf=","]" href="http://my.url/mypag

This template is used to create circular imagemap elements for a point layer. NAME is a fieldname in the data source,
the value for NAME for each individual record gets substituted as the template is processed. The href specifies the
URL link if the element is clicked. Title and alt will display the value when an element is moused over.
The resulting html element looks like
<area shape="circle" coords="80,103,7" href="http://my.url/mypage.cfm?region=Northern" >

The key part here is

coords="[shpxy precision=0 proj=image xf="," yf=",7"]"

This is where MapServer will substitute the image coordinates for that query record. With Precision=0, the coordinates
will be integers.
You also see shpxy template formatting options ‘xf’ and ‘yf’. ‘xf=”,” tells MapServer to place a comma after the x
coordinate. ‘yf=”,7” after the y coordinate. This is done to specify a radius of 7 pixels for the circle. More options can
be found in the Template Reference.

364 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

The query footer template simply adds the closing tag for the html imagemap
</map>

Polygon Layers

Here is a query template for a polygon layer

<area shape="poly" coords="[shpxy precision=0 proj=image]" href="http://my.url/mypage.cfm?ID=[SITE_ID

8.6.4 Request URL

To get the imagemap, you need to send a GET or POST request to MapServer with several URL variables defined. The
below URL tells MapServer where the mapfile is located, what layer we are querying, and that we are using nquery
mode to return multiple results.
http://myurl/cgi-bin/mapserv?map=/web/maps/demoimap.map&qlayer=sites&mode=nquery&searchmap=true

8.6.5 Additional Notes

If you use separate map files to generate your imagemap and your map image, make sure that the EXTENT and SIZE
specified in both mapfiles are identical. If they are not, your features will not align properly.

8.6.6 More Information

Steve Lime’s SHPXY Example

8.7 OGR Output

Author Frank Warmerdam

Contact warmerdam at pobox.com
Last Updated 2010-10-25

Table of Contents
• OGR Output
– Introduction
– OUTPUTFORMAT Declarations
– LAYER Metadata
– MAP / WEB Metadata
– Geometry Types Supported
– Attribute Field Definitions
– Return Packaging
– Test Suite Example

8.7. OGR Output 365

MapServer Documentation, Release 6.0

8.7.1 Introduction

OGR output support was added to MapServer 6.0. It provides an output driver to produce feature style output suitable
as a return result from WMS GetFeatureInfo or WFS GetFeature requests. OGR feature output depends on MapServer
being built against the GDAL/OGR library. The OGR output driver should be enabled in MapServer 6.0 or newer
when INPUT=OGR appears in the version string.

8.7.2 OUTPUTFORMAT Declarations

Details of OGR output formats allowed are controlled by an OUTPUTFORMAT declaration. The declarations define
the OGR format driver to be used, creation options specific to that driver, and more general instructions to MapServer
on how to package multi-file results and whether to try and build the result on disk or in memory.
Examples:
OUTPUTFORMAT
NAME "CSV"
DRIVER "OGR/CSV"
MIMETYPE "text/csv"
FORMATOPTION "LCO:GEOMETRY=AS_WKT"
FORMATOPTION "STORAGE=memory"
FORMATOPTION "FORM=simple"
FORMATOPTION "FILENAME=result.csv"
END

OUTPUTFORMAT
NAME "OGRGML"
DRIVER "OGR/GML"
FORMATOPTION "STORAGE=filesystem"
FORMATOPTION "FORM=multipart"
FORMATOPTION "FILENAME=result.gml"
END

OUTPUTFORMAT
NAME "SHAPEZIP"
DRIVER "OGR/ESRI Shapefile"
FORMATOPTION "STORAGE=memory"
FORMATOPTION "FORM=zip"
FORMATOPTION "FILENAME=result.zip"
END

The OGR format driver to be used is determined by the name appearing after “OGR/” in the DRIVER argument.
This name should match one of the formats listed as supported for the “-f” argument to ogr2ogr in the ogr2ogr usage
message.
The IMAGEMODE for OGR output is FEATURE, but this is implicit and does not need to be explicitly stated for
OGR output driver declarations.
The OGR renderer will support the following FORMATOPTION declarations:
DSCO:* Anything prefixed by DSCO: is used as a dataset creation option with the OGR driver. See the OGR web
page for the particular format driver to see layer creation options available.
LCO:* Anything prefixed by LCO: is used as a layer creation option. See the OGR web page for the particular format
driver to see layer creation options available.w
FORM=simple/zip/multipart Indicates whether the result should be a simple single file (single), a mime multipart
attachment (multipart) or a zip file (zip). “zip” is the default.

366 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

STORAGE=memory/filesystem/stream Indicates where the datasource should be stored while being written. “file”
is the default.
If “memory” then it will be created in /vsimem/ - but this is only suitable for drivers supporting VSI*L which
we can’t easily determine automatically.
If “file” then a temporary directory will be created under the IMAGEPATH were the file(s) will be written and
then read back to stream to the client.
If “stream” then the datasource will be created with a name “/vsistdout” as an attempt to write directly to stdout.
Only a few OGR drivers will work properly in this mode (ie. CSV, perhaps kml, gml).
FILENAME=name Provides a name for the datasource created, default is “result.dat”.

8.7.3 LAYER Metadata

The OGR output driver utilizes several items from the LAYER level METADATA object. Some of these were origi-
nally intended for GML output or are primarily intended to support WFS.
wfs_getfeature_formatlist (Optional) A comma delimited list of formats supported for WFS GetFeature responses.
The OUTPUTFORMAT NAME values should be listed.
"wfs_getfeature_formatlist" "OGRGML,SHAPEZIP,CSV"

gml_include_items (Optional) A comma delimited list of items to include, or keyword “all”. You can enable full
exposure by using the keyword “all”.
"gml_include_items" "all"

You can specify a list of attributes (fields) for partial exposure, such as:
"gml_include_items" "Name,ID"

The new default behaviour is to expose no attributes at all.

gml_include_items (Optional) A comma delimited list of items to include, or keyword “all”. As of MapServer 4.6,
you can control how many attributes (fields) you expose for your data layer with this metadata. The previous
behaviour was simply to expose all attributes all of the time. You can enable full exposure by using the keyword
“all”, such as:
"gml_include_items" "all"

You can specify a list of attributes (fields) for partial exposure, such as:
"gml_include_items" "Name,ID"

The new default behaviour is to expose no attributes at all.

gml_[item name]_alias (Optional) An alias for an attribute’s name. The resulting file will refer to this attribute by
the alias. Here is an example:
"gml_province_alias" "prov"

gml_[item name]_type (Optional) Specifies the type of the attribute. Valid values are Inte-
ger|Real|Character|Date|Boolean.
gml_[item name]_width (Optional) Specifies the width of the indicated field for formats where this is significant,
such as Shapefiles.
gml_[item name]_precision (Optional) Specifies the precision of the indicated field for formats where this is signif-
icant, such as Shapefiles. Precision is the number of decimal places, and is only needed for “Real” fields.

8.7. OGR Output 367

MapServer Documentation, Release 6.0

gml_types (Optional) If this field is “auto” then some input feature drivers (ie. OGR, and native shapefiles) will
automatically populate the type, width and precision metadata for the layer based on the source file.
"gml_types" "auto"

ows/wfs_geomtype (Optional) Set the geometry type of OGR layers created from this MapServer LAYER. One of
“Point”, “LineString”, “Polygon”, “MultiPoint”, “MultiLineString”, “MultiPolygon”, “GeometryCollection”,
“Geometry”, or “None”. Most are fairly obvious, but “Geometry” can be used to represent a mix of geometry
types, and “None” is sometimes suitable for layers without geometry. Note that layers which are a mix of
polygon and multipolygon would normally have to be described as “Geometry”.
"ows_geomtype" "Polygon"

8.7.4 MAP / WEB Metadata

wms_feature_info_mime_type In order for WMS GetFeatureInfo to allow selection of OGR output formats, the
mime type associated with the OUTPUTFORMAT must be listed in this metadata item.
"wms_feature_info_mime_type" "text/csv"

8.7.5 Geometry Types Supported

In MapServer we have POINT, LINE and POLYGON layers which also allow for features with multiple points, lines
or polygons. However, in the OGC Simple Feature geometry model used by OGR a point and multipoint layer are
quite distinct. Likewise for a LineString and MultiLineString and Polygon an MultiPolygon layer type.
To work around the mismatches between the MapServer and OGR geometry models, there is a mechanism to specify
the geometry type to be used when exporting through OGR. This is the “wfs/ows_geomtype” metadata item on the
layer. It may be one of one of “Point”, “LineString”, “Polygon”, “MultiPoint”, “MultiLineString”, “MultiPolygon”,
“GeometryCollection”, “Geometry”, or “None”.
If this item is not specified, then “Point”, “LineString” or “Polygon” will be used depending on the TYPE of the
LAYER. In cases of mixed geometry types (ie. polygons and multipolygons) the geometry type should be set to
“Geometry” which means any geometry type.
"ows_geomtype" "Geometry"

8.7.6 Attribute Field Definitions

For OGR output it is highly desirable to be able to create the output fields with the appropriate datatype, width and
precision to reflect the source feature definition.
It is possible to set the gml_[item]_type, gml_[item]_width and gml_[item]_precision metadata on the layer to provide
detailed field definitions:
METADATA
"gml_ID_type" "Integer"
"gml_ID_width" "8"
"gml_AREA_type" "Real"
"gml_AREA_width" "15"
"gml_AREA_precision" "6"
"gml_NAME_type" "Character"
"gml_NAME_width" "64"
...

368 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

However, doing this manually is tedious and error prone. For that reason some feature sources (at least OGR, Shape-
files, POSTGIS and ORACLESPATIAL) support a mechanism to automatically populate this information from the
source datastore. To accomplish this specify:
"gml_types" "auto"

If no effort is made to set type, width and precision information for attribute fields, they will all be treated as variable
length character fields when writing through OGR.

8.7.7 Return Packaging

One of the challenges returning generalized feature formats is that many such formats consists of multiple files which
must be returned in the result. There are three approaches taken to this based on the FORM FORMATOPTION in the
OUTPUTFORMAT declaration.
simple In this case a single result is returned. This is suitable for format drivers that produce a single file. The
return result will have the mimetype listed in the OUTPUTFORMAT declaration. Note that if the OGR driver
actually returns multiple files, only the primary one (the one with a name matching the filename passed into the
OGR CreateDataSource call) will be returned. The return result will have a suggested filename based on the
FILENAME FORMATOPTION.
multipart In this case all the files produced are returned as a multipart mime result. In this case the MIMETYPE of
the OUTPUTFORMAT is ignored. All component files are returned with a mime type of “application/binary”
and the whole package is “multipart/mixed”.
zip In this case all the files produced are bundled into one .zip file and this zip file is returned with a mimetype of
“application/zip”. The OUTPUTFORMAT MIMETYPE is ignored.
One caveat with “zip” results is that this option is only available if the GDAL/OGR version is 1.8 or newer (or a 1.8
development later than approximately Oct 15, 2010). Earlier versions of GDAL/OGR lacked the zipping capability
needed.

8.7.8 Test Suite Example

The MSAutoTest test suite contains a test case for use of OGR Output from WFS. The mapfile is at:
http://svn.osgeo.org/mapserver/trunk/msautotest/wxs/wfs_ogr.map
The comments at the start of the file have a variety of sample requests that can be run against the map, as long as [MAP-
FILE] is replaced with the mapfile name. They requests should be run against mapserv sitting in the msautotest/wxs
directory.

8.8 PDF Output

Author Yewondwossen Assefa

Contact yassefa at dmsolutions.ca
Revision $Revision: 8278 $
Date $Date: 2008-12-23 13:34:31 -0800 (Tue, 23 Dec 2008) $
Last Updated 2006/01/12

8.8. PDF Output 369

MapServer Documentation, Release 6.0

Table of Contents
• PDF Output
– Introduction
– What is currently supported and not supported
– Implementing PDF Output
– PHP/MapScript and PDF Output

8.8.1 Introduction

PDF output support was added to MapServer 3.7. Previous versions of MapServer had support for pdf output using a
utility program (shp2pdf) to output a pdf file given a MapServer mapfile.
The difference in this new version is that the output to PDF can now be directly specified in the mapfile using the
IMAGETYPE or the OUTPUTFORMAT parameters in the mapfile. Additionally, raster layers are now supported for
pdf output.

8.8.2 What is currently supported and not supported

1. Vector Layers
• Layer Point: supported
• Layer Line: supported
• Layer Polygon: supported
• Layer Circle : not supported
• Layer Annotation: supported
Note: Note: Dashed lines are supported with PDFlib version 6 or greater.
Note: Polygons filled with symbols are not supported.
2. Raster Layers
Raster layers are supported. Note that at this point all raster layers are transformed to jpeg format
before being written to the PDF file.
3. WMS Layers
Not yet supported
4. Surround components
Legend, scalebar are not supported.
5. Fonts
Standard PostScript fonts are supported. For use of other fonts (such as truetype), see the pdflib
documentation for use of UPR description files (some notes on it are here).

8.8.3 Implementing PDF Output

Note that the following instructions were developed for MapServer 3.7 and pdflib 4.0.3, but the general steps should
be similar for recent versions of both.

370 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Build the PDF Library

In order to have access to the PDF support in MapServer, you should download and build the PDF library from
http://www.pdflib.com/products/pdflib/. Please follow the instructions on the PDFLib site to build on your specific
platforms.
Here are some quick notes on how to build on windows:
• download and extract the source code from http://www.pdflib.com/products/pdflib/
• open the project PDFlib.dsw in MS Visual C++
• build the project pdflib_dll
• after a sucessful build, you should have a pdflib.lib and pdblib.dll under the pdflib directory
• copy the pdflib.dll under your system directory (ex : c:/winnt/system32)
• the pdflib.lib will be used while building mapserver with the PDF support

Build MapServer with PDF support

Windows platform
Edit the makefile.vc and uncomment the following lines (make sure that the paths are adapted to your installation):
PDF_LIB=../pdflib-4.0.3/pdflib/pdflib.lib

PDF_INC=-I../pdflib-4.0.3/pdflib

PDF=-DUSE_PDF

See the Windows Compilation document for general MapServer compile instructions.
Unix platforms
Add with-pdf to your configure command line before compiling.
See the Unix Compilation document for general MapServer compile instructions.

Mapfile definition

The IMAGETYPE parameter in the Mapfile should be set to pdf in order to output to PDF:
NAME pdf-test
STATUS ON
...
IMAGETYPE pdf
..

WEB
...
END

LAYER
...
END

END

8.8. PDF Output 371

MapServer Documentation, Release 6.0

You can also specify the output using the OUTPUTFORMAT tag (this tag was introduced in mapserver 3.7) :
OUTPUTFORMAT
NAME pdf
MIMETYPE "application/x-pdf"
DRIVER pdf
FORMATOPTION "OUTPUT_TYPE=RASTER" ##not mandatory
END

If the OUTPUT_TYPE=RASTER all the layers will be rendered as rasters. Note that when WMS layers are in-
cluded in the mapfile, this option should be set since there is a problem with transparency and wms layers. See the
OUTPUTFORMAT object in the Mapfile reference for parameter explanations.

Testing

The easiest way to test your pdf output mapfile is with the MapServer shp2img utility. Windows users can find this
utility in MS4W, as well as FWTools.
You simply pass a mapfile to the executable and a name for the output pdf, and a pdf file is generated:
shp2img -m gmap_pdf.map -o test.pdf

Possible Errors

PDFlib I/O error: Resource configuration file ’pdflib.upr’ not found

This is related to fonts. If you remove the LABEL object from your mapfile you will see this error go away. The pdf
error is described here. Basically, until this issue is ‘fixed’, if you want to use a font other than the included standard
PostScript fonts in pdf output (such as truetype fonts), consult the PDFlib documentation.

8.8.4 PHP/MapScript and PDF Output

MapServer can render to PDF directly, another option is to render to a PNG and insert that into a PDF document. This
is not the only way to create a PDF document of course. You will need to have support for PDFLib compiled into your
PHP install.
This example shows the key parts of the process, you will need to furnish parts of the script yourself (depending on
your app) and repeat the process for each map element that you want to include.
Refer to the PHP/MapScript Reference wherever necessary.

How does it work?

In brief, we will pass parameters required to render a map to a PHP script that will:
• create a PDF document
• render a PNG view at a suitably higher resolution
• insert the PNG
• buffer it and send it to the user

372 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Create the PDF document

Here is an example similiar to the one given on the PHP website to create a new PDF document:
$my_pdf = pdf_new();
...

Get this stage and section 4.5 working before you try inserting MapServer elements.

Render PNG views at a suitable resolution

Work back from the assumption that you will need no more than 300 dpi on your page for your map to look presentable.
For an A4 map, I am using 150 dpi for an 8’ x 8’ main map, which is 1200 x 1200 pixels.
$map->set(width,1200);
$map->set(height,1200);

Of course, our map will not be very useful unless it is zoomed in to the extent our user requested, and the layers they
selected are switched on. Maintain arrays in your application that record:
- The current extent (say $ext[])
- Layer status (say $layer[])

Open your map file and pass these back through to set the map file into the state the user is expecting, something like:
$map->setextent($ext[0], $ext[1], $ext[2], $ext[3]);

while($layer[]) {
$layer=$map->getLayer($n);
if($layer[$n]==1) {
$layer->set(status,1);
} else {
$layer->set(status,0);
}
}

Now you will need to save a rendered view to a PNG file.

$img = $map->draw();
$url = $img->saveWebImage(MS_PNG, 0, 0, 0);

Use the same method for all your map elements, such as drawReferenceMap?(), drawScaleBar?() and drawLegend().

Insert the PNG elements into your PDF document

This is really easy, use the pdf_open_image_file() function to import the map elements into your PDF document:
$element = pdf_open_image_file($my_pdf, "png", "$webroot/$url");
pdf_place_image($my_pdf, $element, $xpos, $ypos);
pdf_close_image($my_pdf, $element);

Repeat as needed for any map elements you created.

Buffer the PDF and send it to the user

Assuming we have been creating the document $my_pdf, when we are done, we merely buffer it and send it to the
user using echo():

8.8. PDF Output 373

MapServer Documentation, Release 6.0

<?php

....
pdf_close($my_pdf);

$data = pdf_get_buffer($my_pdf);

header(’Content-type: application/pdf’);
header(’Content-disposition: inline; filename=my_pdf.pdf’);
header(’Content-length: ’ . strlen($data) );

echo $data;

?>

Gotcha: remember that you cannot send headers if you have at any stage outputed text to the browser.

Additional stuff to try

Rendering everything as PNG can look ugly, so I step through the key and extract labels so I can render them using
PDF’s text functions.
This can be done for other map element, such as map titles, layer descriptions, or anything else that can be read from
the mapfile.

8.9 SVG

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Revision $Revision: 8365 $
Date $Date: 2008-12-31 07:49:02 -0800 (Wed, 31 Dec 2008) $
Last Updated 2005/12/13

Table of Contents
• SVG
– Introduction
– Feature Types and SVG Support Status
– Testing your SVG Output
– goSVG

8.9.1 Introduction

SVG (or Scalable Vector Graphics) is a standardized XML language for describing 2D graphics via vector graphics,
text and raster graphics. As of version 4.5, MapServer can output SVG v1.1 maps. The following documentation is
based on the World Wide Web Consortium’s (W3C) Scalable Vector Graphics (SVG) 1.1 Specification.
This document assumes that you are already familiar with certain aspects of MapServer:
• MapServer application development and setting up map files.

374 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Links to SVG-Related Information

• SVG 1.1 specification

• SVG Discussion Paper
• G-XML Project Page
• SVG Tiny Profile
• MapFile Reference Doc

8.9.2 Feature Types and SVG Support Status

Annotation Layers

Annotation layers are supported (see the Text Features section below for details).

Circle Layers

Circle layers are not yet supported.

Line Layers

The following items describe how line layers are handled by MapServer for SVG output:
• Lines are converted to SVG polyline elements.
• The STYLE object’s WIDTH parameter is used for SYMBOL 0 for line thickness.
• The STYLE object’s SIZE parameter is used for other symbols for line thickness.
• All lines are drawn without symbols - only line thickness changes.
• If a style uses a symbol and this symbol has a dashed style, it will be transformed into an SVG stroke-dasharray
element.

Point Layers

The following items describe how point layers are handled by MapServer for SVG output:
• VECTOR, ELLIPSE, and TRUETYPE symbols are supported.
• PIXMAP symbols are not currently supported.
• Labels attached with the symbols are supported (see the Text Features section below for details).

Polygon Layers

The following items describe how polygon layers are handled by MapServer for SVG output:
• Polygons are converted to SVG polygon elements.
• The STYLE’s COLOR is used for the fill.
• The STYLE’s OUTLINECOLOR is used for the stroke.
• SVG patterns are not currently supported.

8.9. SVG 375

MapServer Documentation, Release 6.0

Raster Layers

The following items describe how raster layers are handled by MapServer for SVG output:
• Temporary image is created through the GD library, and GD functions are used to draw the layer.
• You must have at least PNG or JPEG support compiled in MapServer.
• You must have the WEB object’s IMAGEPATH and IMAGEURL set properly in your mapfile.

Text Features

The following items describe how text features are handled by MapServer for SVG output:
• Text is converted to SVG text element.
• Only TRUETYPE fonts are supported.
• Supports labels with ENCODING (output as UTF-8 hexadecimal values).
• The FONT name used in MapServer is parsed to form the SVG font-family, font-style, and font-weight.

WMS Layers

WMS layers are not yet supported.

Setting up a Mapfile for SVG Output
• You must have valid IMAGEPATH and IMAGEURL parameters set in the WEB object of the mapfile.
• To be able to output a valid SVG file, the user needs to define an OUTPUTFORMAT object in the map file and
set the IMAGETYPE parameter to svg. Here is an example:
MAP
...
IMAGETYPE svg
...
OUTPUTFORMAT
NAME svg
MIMETYPE "image/svg+xml"
DRIVER svg
FORMATOPTION "COMPRESSED_OUTPUT=TRUE"
FORMATOPTION "FULL_RESOLUTION=TRUE"
END
...
WEB
IMAGEPATH "/tmp/ms_tmp/"
IMAGEURL "/ms_tmp/"
END
...
LAYER
...
END
END

Note:
If FORMATOPTION “COMPRESSED_OUTPUT=TRUE” is set MapServer will produce a compressed SVG
file (svgz). By default this option is FALSE. Note that to be able to create compressed output, MapServer must
be built with the compile flag USE_ZLIB.

376 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

If FORMATOPTION “FULL_RESOLUTION=TRUE” is set MapServer will not eliminate duplicate points and
collinear lines when outputting SVG. By default this option is set to FALSE.

8.9.3 Testing your SVG Output

• The easiest way to test your SVG mapfile is to use MapServer CGI. For example, you might enter the following
URL in a browser:
http://127.0.0.1/cgi-bin/mapserv.exe?map=my/path/to/my-svg.map&mode=map&layers=layer1 layer2

• You can also use PHP/MapScript to test your SVG mapfile. Your php file might look like the following:
<?php

dl("php_mapscript_45.dll");

$oMap = ms_newmapObj("my/path/to/my-svg.map");

$img = $oMap->draw();

header("Content-type: image/svg+xml");

$url = $img->saveImage("");

?>

An SVG file should be created in your IMAGEPATH directory. If you open the SVG file in a text editor you can see
that it is an XML file. Below is a sample SVG file of a point layer with labels:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11-flat.dtd
<svg version="1.1" width="400" height="300" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://ww

<!-- START LAYER popplace -->

<ellipse cx="252" cy="130" rx="3" ry="3" fill="#000000" />
<ellipse cx="37" cy="227" rx="3" ry="3" fill="#000000" />
<ellipse cx="127" cy="239" rx="3" ry="3" fill="#000000" />
<ellipse cx="255" cy="282" rx="3" ry="3" fill="#000000" />
<polygon fill="#000000" stroke-width="1" points=" 267,263 270,263 271,260 272,263 275,263 273,265
<ellipse cx="288" cy="247" rx="3" ry="3" fill="#000000" />
<ellipse cx="313" cy="243" rx="3" ry="3" fill="#000000" />
<ellipse cx="328" cy="233" rx="3" ry="3" fill="#000000" />
<ellipse cx="331" cy="245" rx="3" ry="3" fill="#000000" />
<ellipse cx="366" cy="196" rx="3" ry="3" fill="#000000" />
<ellipse cx="161" cy="246" rx="3" ry="3" fill="#000000" />
<ellipse cx="92" cy="208" rx="3" ry="3" fill="#000000" />
<ellipse cx="40" cy="125" rx="3" ry="3" fill="#000000" />
<ellipse cx="108" cy="146" rx="3" ry="3" fill="#000000" />
<text x="40" y="143" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="43" y="121" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="34" y="205" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="164" y="258" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="316" y="190" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="334" y="258" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="249" y="230" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="241" y="242" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="223" y="260" font-family="fritqat-italic" font-size="8pt" fill="#ff0000" stroke="#ffffff"
<text x="210" y="279" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke

8.9. SVG 377

MapServer Documentation, Release 6.0

<text x="82" y="234" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-

<text x="40" y="223" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="214" y="125" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
</svg>

You can now view the SVG file in a supported browser (see the official list of SVG implementations for possible SVG
viewers). The Adobe Viewer plugin is very popular.

8.9.4 goSVG

goSVG is now supported as a vector output format in MapServer 4.5 (and later).

Definition

This definition of goSVG was obtained from here.

goSVG is short for “G-XML over SVG” and “g-contents over SVG”. This is a subset for mobiles specified within
the G-XML (a Japanese Spatial Information Format which is an XML based protocol with the ability to describe,
communicate and exchange Spatial Information and Electric Maps), and is a Spatial Information Exchanging format
that determines the method to expand spatial information and connect to the backend system(G-XML standard mark
format). goSVG is an expanded SVG Tiny profile (a Mobile profile of SVG 1.1. suited for cellular phones) that adds
functions that are useful for Spatial Information Services (SVG Map Service).

Support for Specific goSVG Elements

• Name space extension: supported

• Content Area Definition (bounding box): supported
• Geographic Coordinate System: supported
• Map Request Protocol: supported

Setting up a Mapfile for goSVG Output

Requirements

• A valid MapServer Mapfile.

• Valid IMAGEPATH and IMAGEURL parameters set in the WEB object of the mapfile.
• A PROJECTION object defined beneath the MAP object, using an EPSG code. For example:
MAP
...
WEB
IMAGEPATH "/tmp/ms_tmp/"
IMAGEURL "/ms_tmp/"
END
...
PROJECTION
"init=epsg:42304"
END
...
LAYER

378 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

...
END
END

Setting the OUTPUTFORMAT

To be able to output a valid goSVG file, you must define an OUTPUTFORMAT object in the mapfile and set the
IMAGETYPE to svg. Here is an example:
MAP
...
IMAGETYPE svg
...
OUTPUTFORMAT
NAME svg
MIMETYPE "image/svg+xml"
DRIVER svg
FORMATOPTION "GOSVG=TRUE"
FORMATOPTION "GOSVG_ZoomInTH=20"
FORMATOPTION "GOSVG_ZoomOutTH=40"
FORMATOPTION "GOSVG_ScrollTH=60"
END
...
WEB
IMAGEPATH "/tmp/ms_tmp/"
IMAGEURL "/ms_tmp/"
END
...
PROJECTION
"init=epsg:42304"
END
...
LAYER
...
END
END

Specific FORMATOPTIONs Related to goSVG

GOSVG should be set to TRUE. The default is false.

GOSVG_ZoomInTH controls the zoomin threshold when outputting the Map Request Protocol. If it is not defined
the default value is set to 70.
GOSVG_ZoomOutTH controls the zoomout threshold when outputting the Map R equest Protocol. If it is not
defined the default value is set to 100.
GOSVG_ScrollTH controls the scrolling threshold when outputting the Map Request Protocol. If it is not defined
the default value is set to 10.

Testing your goSVG Output

Refer to the section Testing your SVG Output to generate and test your goSVG output. goSVG can be read by regular
SVG viewers (they will just ignore the goSVG headers).

8.9. SVG 379

MapServer Documentation, Release 6.0

Sample goSVG File Produced by MapServer

Below is a sample goSVG file of a point layer with labels:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11-flat.dtd
<svg version="1.1" width="400" height="300" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://ww
<title>DEMO</title>
<metadata>
<rdf:RDF xmlns:rdf = "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:crs = "http://www.ogc.org/crs" xmlns:svg="http://wwww.w3.org/2000/svg">
<rdf:Description>
<crs:CoordinateReferenceSystem svg:transform="matrix(0.000066,0.000000,0.000000,-0.000066,171.243002,
rdf:resource="http://www.opengis.net/gml/srs/epsg.xml#42304"/>
</rdf:Description>
</rdf:RDF>
<au:lbs protocol="maprequest">
<au:zoomin th="20" xlink:href="."/>
<au:zoomout th="40" xlink:href="."/>
<au:scroll th="60" xlink:href="."/>
</au:lbs>
</metadata>

<!-- START LAYER popplace -->

<ellipse cx="252" cy="130" rx="3" ry="3" fill="#000000" />
<ellipse cx="37" cy="227" rx="3" ry="3" fill="#000000" />
<ellipse cx="127" cy="239" rx="3" ry="3" fill="#000000" />
<ellipse cx="255" cy="282" rx="3" ry="3" fill="#000000" />
<polygon fill="#000000" stroke-width="1" points=" 267,263 270,263 271,260 272,263 275,263 273,265
<ellipse cx="288" cy="247" rx="3" ry="3" fill="#000000" />
<ellipse cx="313" cy="243" rx="3" ry="3" fill="#000000" />
<ellipse cx="328" cy="233" rx="3" ry="3" fill="#000000" />
<ellipse cx="331" cy="245" rx="3" ry="3" fill="#000000" />
<ellipse cx="366" cy="196" rx="3" ry="3" fill="#000000" />
<ellipse cx="161" cy="246" rx="3" ry="3" fill="#000000" />
<ellipse cx="92" cy="208" rx="3" ry="3" fill="#000000" />
<ellipse cx="40" cy="125" rx="3" ry="3" fill="#000000" />
<ellipse cx="108" cy="146" rx="3" ry="3" fill="#000000" />
<text x="40" y="143" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="43" y="121" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="34" y="205" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="164" y="258" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="316" y="190" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="334" y="258" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="249" y="230" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="241" y="242" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="223" y="260" font-family="fritqat-italic" font-size="8pt" fill="#ff0000" stroke="#ffffff"
<text x="210" y="279" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
<text x="82" y="234" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="40" y="223" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke-
<text x="214" y="125" font-family="fritqat" font-size="8pt" fill="#000000" stroke="#ffffff" stroke
</svg>

8.10 Tile Mode

Author Paul Ramsey

380 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Contact pramsey at cleverelephant.ca

Revision $Revision: 9817 $
Date $Date: 2010-02-20 11:35:53 -0800 (Sat, 20 Feb 2010) $
Last Updated 2008/04/30

Table of Contents
• Tile Mode
– Introduction
– Configuration
– Utilization

8.10.1 Introduction

MapServer can feed tile-based map clients directly using the CGI “tile mode”. Tile-based map clients work by dividing
the map of the world up into a discrete number of zoom levels, each partitioned into a number of identically sized
“tiles”. Instead of accessing a map by requesting a bounding box, a tile client builds a map by accessing individual
tiles.

8.10.2 Configuration

Tile requests are handled by the ‘mapserv’ CGI program. In order to return tiles in the correct projection, MapServer
must be built with the –use-proj option turned on. You can check if your version of ‘mapserv’ has projection support
by running it with the ‘-v’ option and looking for ‘SUPPORTS=PROJ’.
Example 1. On Unix:
$ ./mapserv -v
MapServer version 4.6.1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=PDF
OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER
SUPPORTS=WMS_CLIENT SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER
INPUT=JPEG INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE DEBUG=MSDEBUG

Example 2. On Windows:
C:\apache\cgi-bin> mapserv -v
MapServer version 4.6.1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=PDF
OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER
SUPPORTS=WMS_CLIENT SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER
INPUT=JPEG INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE DEBUG=MSDEBUG

MapServer requires that each LAYER in your map file have a valid PROJECTION block to support reprojection.
Because the tile mode uses reprojection, you will have to ensure each LAYER has a valid PROJECTION block.
Configuration checklist:
• MapServer compiled with PROJ support
• Map file with a PROJECTION defined for every LAYER
As of MapServer 6.0, there are two extra parameters available for configuring tile mode.
• tile_map_edge_buffer renders the tile into a buffered rendering frame, then clips out the final tile. This will
reduce edge effects when large symbols or wide lines are drawn. Recommended value: the size of the largest
symbol or line width in your map file.

8.10. Tile Mode 381

 MapServer Documentation, Release 6.0

• tile_metatile_level renders the into into a fixed metatile, then clips out the final tile. This will reduce label
repetition, at the expense of much higher rendering cost. Recommended value: 1 if you are doing labelling of
large features in your layer. 0 otherwise.
If you use both tile_map_edge_buffer and tile_metatile_level at the same time, the buffer will be applied at the meta-tile
level.

8.10.3 Utilization

The MapServer tile support adds three new directives to the CGI interface:
• mode=tile tells the server to generate tiles based on the other tile mode parameters
• tilemode=gmap tells the server use the Google Maps tile scheme for the tiles
• tile=x+y+z tells the server what tile you want to retrieve, using the Google Maps tile addressing system
• tilemode=ve tells the server use the Virtual Earth tile naming scheme for the tiles
• tile=10231 tells the server what tile you want to retrieve, using the Virtual Earth tile addressing system

About Spherical Mercator

Spherical Mercator (also called “web mercator” by some) is a world projection. All the major tile-based map interfaces
(Google Maps, Microsoft Virtual Earth, Yahoo Maps, OpenLayers) use the spherical mercator system to address tiles.
A spherical mercator set of tiles has the following properties:
• The map has been reprojected to mercator using a spherical mercator algorithm
• There is one tile in the top zoom level, zoom level zero
• Each successive zoom level (z) has 2^z tiles along each axis
• Tiles are 256x256 in size
Google Maps and Virtual Earth both use spherical mercator as their underlying tile projection, but use different formats
to address the individual tiles.
Google Maps uses an “x”, “y”, “zoom” format. The zoom indicates which level to pull tiles from, and the “x” and “y”
indicate while tile in that zoom level to pull.
Virtual Earth uses a single string to address each tile. The top zoom level in Virtual Earth has four tiles (equivalent
to Google’s zoom level 1). The top left tile in the Virtual Earth top zoom level is addessed as “0”, top right as “1”,
bottom left as “2” and bottom right as “3”. Each tile the next level is addressed by first referencing the top level tile
that contains it, then its address relative to that tile. So the top left tile in the second zoom level is “00” and the bottom
right one is “33”. See the Virtual Earth site for more details: http://msdn.microsoft.com/en-us/library/bb545006.aspx

Using Google Maps

The Google Maps API includes support for using alternative tile sets as overlays, or as alternate base maps. Here is an
example of an GTileLayerOverlay
1 <!DOCTYPE html
2 PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
3 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
4 <html xmlns="http://www.w3.org/1999/xhtml">
5 <head>
6 <meta http-equiv="content-type" content="text/html; charset=utf-8"/>

382 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

7 <title>Google/MapServer Tile Example</title>

8 <script src="http://maps.google.com/maps?file=api&v=2&key=[YOUR KEY HERE]"
9 type="text/javascript"></script>
10 <script type="text/javascript">
11

12 function load() {
13 if (GBrowserIsCompatible()) {
14 var urlTemplate = ’http://localhost/cgi-bin/mapserv?’;
15 urlTemplate += ’map=/var/map.map&’;
16 urlTemplate += ’layers=layer1 layer2&’;
17 urlTemplate += ’mode=tile&’;
18 urlTemplate += ’tilemode=gmap&’;
19 urlTemplate += ’tile={X}+{Y}+{Z}’;
20 var myLayer = new GTileLayer(null,0,18,{
21 tileUrlTemplate:urlTemplate,
22 isPng:true,
23 opacity:1.0 });
24 var map = new GMap2(document.getElementById("map"));
25 map.addControl(new GLargeMapControl());
26 map.addControl(new GMapTypeControl());
27 map.setCenter(new GLatLng(35.35, -80.55), 15);
28 map.addOverlay(new GTileLayerOverlay(myLayer));
29 }
30 }
31

32 </script>
33 </head>
34 <body onload="load()" onunload="GUnload()">
35 <div id="map" style="width: 500px; height: 500px"></div>
36 </body>
37 </html>

Note the format of the tileUrlTemplate: a valid URL, with {X}, {Y} and {Z} substitution tokens that Google Maps
will replace with the tile coordinates and zoom level on the fly to retrieve tiles from your server.
You can also use a MapServer tile layer as an alternate base map:
1 <!DOCTYPE html
2 PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
3 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
4 <html xmlns="http://www.w3.org/1999/xhtml">
5 <head>
6 <meta http-equiv="content-type" content="text/html; charset=utf-8"/>
7 <title>Google/MapServer Tile Example</title>
8 <script src="http://maps.google.com/maps?file=api&v=2&key=[YOUR KEY HERE]"
9 type="text/javascript"></script>
10 <script type="text/javascript">
11

12 function load() {
13 if (GBrowserIsCompatible()) {
14 var urlTemplate = ’http://localhost/cgi-bin/mapserv?’;
15 urlTemplate += ’map=/var/map.map&’;
16 urlTemplate += ’layers=layer1 layer2&’;
17 urlTemplate += ’mode=tile&’;
18 urlTemplate += ’tilemode=gmap&’;
19 urlTemplate += ’tile={X}+{Y}+{Z}’;
20 var myLayer = new GTileLayer(null,0,18,{
21 tileUrlTemplate:urlTemplate,
22 isPng:true,

8.10. Tile Mode 383

 MapServer Documentation, Release 6.0

23 opacity:0.3 });
24 var map = new GMap2(document.getElementById("map"));
25 map.addControl(new GLargeMapControl());
26 map.addControl(new GMapTypeControl());
27 map.setCenter(new GLatLng(35.35, -80.55), 15);
28 var myMapType = new GMapType([myLayer], new GMercatorProjection(18), ’MapServer’);
29 map.addMapType(myMapType);
30 }
31 }
32

33 </script>
34 </head>
35 <body onload="load()" onunload="GUnload()">
36 <div id="map" style="width: 500px; height: 500px"></div>
37 </body>
38 </html>

The only change from the previous example is that we don’t create a GTileLayerOverlay, we create a GMapType, and
use addMapType(), instead of addOverlay().

Using Virtual Earth

The Virtual Earth API also includes support for using alternative tile sets as overlays, or as alternate base maps. Here
is an example:
1 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-stri
2 <html xmlns="http://www.w3.org/1999/xhtml">
3 <head>
4 <meta http-equiv="content-type" content="text/html; charset=utf-8"/>
5 <title>Virtual Earth Example</title>
6 <script type="text/javascript" src="http://dev.virtualearth.net/mapcontrol/mapcontrol.ashx?v=6.1"><
7 <script type="text/javascript">
8

9 var map = null;

10

11 function OnLoadMap () {
12 map = new VEMap("myMap");
13 map.LoadMap();
14

15 var url = "http://localhost/cgi-bin/mapserv?";

16 url += "map=/var/map.map&";
17 url += "mode=tile&";
18 url += "layers=layer1 layer2&";
19 url += "tilemode=ve&";
20 url += "tile=%4";
21

22 var tileSourceSpec = new VETileSourceSpecification( "myLayer", url );

23 tileSourceSpec.Opacity = 0.3;
24 map.AddTileLayer(tileSourceSpec, true);
25 }
26

27 </script>
28 </head>
29 <body onload="OnLoadMap();">
30 <div id="myMap" style="position:relative; width:500px; height:500px;"></div>
31 </body>
32 </html>

384 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

8.11 Template-Driven Output

Author Chris Hodgson

Contact chodgson at refractions.net
Last Updated 2011-04-13

Table of Contents
• Template-Driven Output
– Introduction
– OUTPUTFORMAT Declarations
– Template Substitution Tags
– Examples

8.11.1 Introduction

RFC 36 added support for defining template-driven OUTPUTFORMATs for use with feature queries, including WMS
GetFeatureInfo and WFS GetFeature. This allows for custom text-oriented output such as GeoJSON, KML, or XML.
The templates are essentially the same as with the standard MapServer query Templating, however there are some
additional tags to allow for template definition in a single file instead of the standard header/template/footer.
Note: There are other, simpler, ways to output some of these formats using MapServer. However, template-driven
output provides maximal flexibility and customization of the output, at the cost of additional complexity and configu-
ration.

8.11.2 OUTPUTFORMAT Declarations

Details of template-driven output formats are controlled by an OUTPUTFORMAT declaration. The declarations define
the template file to be used, as well as other standard OUTPUTFORMAT options.
Examples:
OUTPUTFORMAT
NAME "kayml"
DRIVER "TEMPLATE"
MIMETYPE "application/vnd.google-earth.kml+xml"
FORMATOPTION "FILE=myTemplate.kml"
FORMATOPTION "ATTACHMENT=queryResults.kml"
END

OUTPUTFORMAT
NAME "geojson"
DRIVER "TEMPLATE"
FORMATOPTION "FILE=myTemplate.json"
END

OUTPUTFORMAT
NAME "customxml"
DRIVER "TEMPLATE"
FORMATOPTION "FILE=myTemplate.xml"
END

8.11. Template-Driven Output 385

MapServer Documentation, Release 6.0

The template file to be used is determined by the “FILE=...” FORMATOPTION. The template filename is relative to
the mapfile’s path. As is standard with MapServer template files, the file must containt the magic string ‘mapserver
template’ in the first line of the file, usually within a comment, but this line is not output to the client.
Note that both the MIMETYPE and FORMATOPTION “ATTACHMENT=...” parameters are very useful for control-
ling how a web browser handles the output file.

8.11.3 Template Substitution Tags

These tags only work in query result templates, and their purpose is primarily to simplify the templating to a single
file for custom ouput formats.
[include src=”otherTemplate.txt”] Includes another template file; the path to the template file is relative to the map-
file path.
[resultset name=layername]...[/resultset] Defines the location of the results for a given layer.
[feature]...[/feature] Defines the loop around the features returned for a given layer.
Optional arguments:
• limit= specifies the maximum number of features to output for this layer.
• trimlast= specifies a string to be trimmed off of the end of the final feature that is output. This is intended
to allow for trailing record delimiters to be removed. See the examples below.
[join name=join1]...[/join] defines the loop around the features join from another layer.

8.11.4 Examples

This example shows how to emulate the old 3-file system using the new system, to compare the usage:
<!-- mapserver template -->
[include src="templates/header.html"]
[resultset name=lakes]
... old layer HEADER stuff goes here, if a layer has no results this block disappears...
[feature]
...repeat this block for each feature in the result set...
[join name=join1]
...repeat this block for each joined row...
[/join]
[/feature]
...old layer FOOTER stuff goes here...
[/resultset]
[resulset name=streams]
... old layer HEADER stuff goes here, if a layer has no results this block disappears...
[feature]
...repeat this block for each feature in the result set...
[/feature]
...old layer FOOTER stuff goes here...
[/resultset]
[include src="templates/footer.html"]

A specific GML3 example:

<!-- mapserver template -->
<?xml version="1.0" encoding="ISO-8859-1"?>
[resultset layer=mums]
<MapServerUserMeetings xmlns="http://localhost/ms_ogc_workshop" xmlns:xlink="http://www.w3.org/1999/x

386 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

<gml:description>This is a GML document which provides locations of all MapServer User Meeting that
<gml:name>MapServer User Meetings</gml:name>
<gml:boundedBy>
<gml:Envelope>
<gml:coordinates>-93.093055556,44.944444444 -75.7,45.4166667</gml:coordinates>
</gml:Envelope>
</gml:boundedBy>
[feature]
<gml:featureMember>
<Meeting>
<gml:description>[desc]</gml:description>
<gml:name>[name]</gml:name>
<gml:location>
<gml:Point srsName="http://www.opengis.net/gml/srs/epsg.xml#4326">
<gml:pos>[x] [y]</gml:pos>
</gml:Point>
</gml:location>
<year>[year]</year>
<venue>[venue]</venue>
<website>[url]</website>
</Meeting>
</gml:featureMember>
[/feature]
<dateCreated>2007-08-13T17:17:32Z</dateCreated>
</MapServerUserMeetings>
[resultset]

A GeoJSON example:
// mapserver template
[resultset layer=mums]
{
"type": "FeatureCollection",
"features": [
[feature trimlast=’,’]
{
"type": "Feature",
"id": "[id]",
"geometry": {
"type": "PointLineString",
"coordinates": [
{
"type": "Point",
"coordinates": [[x], [y]]
}
]
},
"properties": {
"description": "[description]",
"venue": "[venue]",
"year": "[year]"
}
},
[/feature]
]
}
[/resultset]

A more complicated KML example. Note the use of [shpxy] to support multipolygons with holes, and also that a point

8.11. Template-Driven Output 387

MapServer Documentation, Release 6.0

placemark is included with each feature using [shplabel]:

<!--MapServer Template-->
<?xml version="1.0" encoding="UTF-8"?>
<kml xmlns="http://www.opengis.net/kml/2.2" xmlns:gx="http://www.google.com/kml/ext/2.2" xmlns:kml="h
<Document>
<Style id="parks_highlight">
<IconStyle>
<scale>1.4</scale>
<Icon>
<href>http://maps.google.com/mapfiles/kml/shapes/parks.png</href>
</Icon>
<hotSpot x="0.5" y="0" xunits="fraction" yunits="fraction"/>
</IconStyle>
<LineStyle>
<color>ffff5500</color>
<width>4.2</width>
</LineStyle>
<PolyStyle>
<color>aaaaaaaa</color>
</PolyStyle>
<BalloonStyle>
<text>
<![CDATA[
<p ALIGN="center"><b>$[name]</b></p>
$[description]
]]>
</text>
</BalloonStyle>
</Style>
<Style id="parks_normal">
<IconStyle>
<scale>1.2</scale>
<Icon>
<href>http://maps.google.com/mapfiles/kml/shapes/parks.png</href>
</Icon>
<hotSpot x="0.5" y="0" xunits="fraction" yunits="fraction"/>
</IconStyle>
<LineStyle>
<color>ffff5500</color>
<width>4.2</width>
</LineStyle>
<PolyStyle>
<color>ff7fff55</color>
</PolyStyle>
<BalloonStyle>
<text>
<![CDATA[
<p ALIGN="center"><b>$[name]</b></p>
$[description]
]]>
</text>
</BalloonStyle>
</Style>
<StyleMap id="parks_map">
<Pair>
<key>normal</key>
<styleUrl>#parks_normal</styleUrl>
</Pair>

388 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

<Pair>
<key>highlight</key>
<styleUrl>#parks_highlight</styleUrl>
</Pair>
</StyleMap>
[resultset layer=parks]
<Folder>
<name>Parks</name>
[feature trimlast=’,’ limit=1]
<Placemark>
<name>[NAME]</name>
<Snippet/>
<description>
<![CDATA[
<p>Year Established: [YEAR_ESTABLISHED]</p>
<p>Area: [AREA_KILOMETERS_SQUARED] sq km</p>
]]>
</description>
<styleUrl>#parks_map</styleUrl>
<ExtendedData>
<Data name="Year Established">[YEAR_ESTABLISHED]</Data>
<Data name="Area">[AREA_KILOMETERS_SQUARED]</Data>
</ExtendedData>
<MultiGeometry>
<Point>
<coordinates>[shplabel proj=epsg:4326 precision=10],0</coordinates>
</Point>
[shpxy ph="<Polygon><tessellate>1</tessellate>" pf="</Polygon>" xf="," xh=" " yh=" " yf=",0 " orh="<o
</MultiGeometry>
</Placemark>
[/feature]
</Folder>
[/resultset]
</Document>
</kml>

8.12 Kml Output

Last Updated 2010/11/26

Authors Dvaid Kana (david.kana at gmail.com)
Authors Thomas.Bonfort (thomas.bonfort at gmail.com)
Authors Yewondwossen Assefa (yassefa at dmsolutions.ca)
Authors Michael Smith (michael.smith at usace.army.mil)
Version MapServer 6.0
Id $

8.12.1 Introduction

This purpose of this docuement is to describethe KML/KMZ output support in MapServer 6.0.
The main goal of the KML driver is to generate KML output used mainly by Google Earth application.

8.12. Kml Output 389

MapServer Documentation, Release 6.0

8.12.2 General Functionnality

Kml support is provided by using a kml or kmz image type in the map file. Output can then be generated using
MapServer cgi (example mode=map) or through a WMS request.

8.12.3 Output format

The default name of the output format is kml or kmz, and this name can be used to set the imagetype parameter in the
map file.
The format can also be defined in the map file:
OUTPUTFORMAT
NAME kml
DRIVER "KML"
MIMETYPE "application/vnd.google-earth.kml+xml"
IMAGEMODE RGB
EXTENSION "kml"
FORMATOPTION ’ATTACHMENT=gmap75.kml’ #name of kml file returned
FORMATOPTION "maxfeaturestodraw=100"
END

OUTPUTFORMAT
NAME kmz
DRIVER "KMZ"
MIMETYPE "application/vnd.google-earth.kmz"
IMAGEMODE RGB
EXTENSION "kmz"
FORMATOPTION ’ATTACHMENT=gmap75.kmz’ #name of kmz file returned
END

8.12.4 Build

• On windows: there is a flag KML in nmake.opt

• On Linux: –with-kml
• AGG driver is necessary for the kml driver
• To be able to get kmz support, MapServer needs to be build aginst GDAL 1.8

8.12.5 Limiting the number of features

The number of vector features drawn by default is set to 1000 per layer. To control the number of features, users can
set:
• layer level metadata that only applies to the layer: “maxfeaturestodraw” “100”
• map level metada that applies to all layers: “maxfeaturestodraw” “100”
• output format option that applies to all layers: FORMATOPTION “maxfeaturestodraw=100”

8.12.6 Map

In terms for Kml object, the MapServer KML output will produce a <Document> element to include all the layers that
are part of the MapServer map file. Features supported for the Document are:

390 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Docu- Sup- MapServer equivalence/Notes

ment ported
element
name Yes Name in the map file
visibility No Can be supported if needed. Default is 1
open No Can be supported if needed. Default is 0
address No Could be supported for example using ows_address if available
AdressDe- No
tails
pho- No Could be supported using ows_contactvoicetelephone is available
neNumber
Snippet No
description No
Ab- No
stractView
TimePrim- No
itive
styleURL No
StyleSe- Yes Style element will be supported. All different styles from the layers will be stored here
lector and referenced from the folders using a styleUrl.In addition to the Styles related to
features, there is a ListStyle element added at the document level. This allows to control
the way folders are presented.See Layers section (styleUrl) setting for more details.
Region No
Metadata No
Extended- No
Data

8.12.7 Layers

Each layer of the MapServer map file will be inside a Kml <Folder> element. Supported Folder elements are:
Folder element Sup- MapServer equivalence/Notes
ported
name Yes Name of the layer. If not available the name will be Layer concatenated
with the layer’s index (Layer1)
visibility Yes Always set to 1
open No Default is 0
atom:authoratom:linkaddressAddressDetailsphoneNumberSnippet
No
description No Could be supported using ows_description
AbstarctView No
TimePrimitive No
styleUrl Yes The user can use the kml_folder_display layer or map level metedata to
choose a setting. Possible values are ‘check’ (default), ‘radioFolder’,
‘checkOffOnly’, ‘checkHideChildren’.
RegionMetadataExtended- No
Data
Each element in the Layer will be inside a Kml <Placemark> element. As described in the Kml reference : « A
Placemark is a Feature with associated Geometry. In Google Earth, a Placemark appears as a list item in the Places
panel. A Placemark with a Point has an icon associated with it that marks a point on the Earth in the 3D viewer. (In
the Google Earth 3D viewer, a Point Placemark is the only object you can click or roll over. Other Geometry objects
do not have an icon in the 3D viewer. To give the user something to click in the 3D viewer, you would need to create
a MultiGeometry object that contains both a Point and the other Geometry object.) »
For Polygon and Line layers, when a feature is associated with a label, a MultiGeometry element containing a point

8.12. Kml Output 391

MapServer Documentation, Release 6.0

geometry and the geometry of the feature is created. The point feature will be located in the middle of the polygon or
line
<Folder>
<name>park</name>
<visibility>1</visibility>
<styleUrl>#LayerFolder_check</styleUrl>
<Placemark>
<name>Ellesmere Island National Park Reserve</name>
<styleUrl>#style_line_ff787878_w4.0_polygon_ff00ffc8_label_ff0000ff</styleUrl>
<MultiGeometry>
<Polygon>
<outerBoundaryIs>
<LinearRing>
<coordinates>
...
<Point>
<coordinates>
-70.86810858,82.12291871
</coordinates>
</Point>
</MultiGeometry>
</Placemark>
</Folder>

Supported Features in the Placemark element are:

General notes on layers

• Labelcache is turned off on each layer

• Projection block should be set. If not set It will be assumed that the data is in lat/lon projection (a debug message
will be sent to the user: Debug Message 1)
• Possible to output vector layers as raster using metadata: “KML_OUTPUTASRASTER” “true”
• The user can use the KML_FOLDER_DSIPLAY layer or map level metedata to choose a setting. Possible
values are ‘check’ (default), ‘radioFolder’, ‘checkOffOnly’, ‘checkHideChildren’.
• The user can use metadata KML/OWS_DESCRIPTION or KML/OWS_INCLUDE_ITEMS to define the de-
scription attached to each feature. If KML/OWS_DESCRIPTION are defined, the <description> tag of the
Placemark will be used. If KML/OWS_INCLUDE_ITEMS, the <ExtendedData> tag will be used.
• The user can use the metadata KML_NAME_ITEM to indicate the field name to be used a a name tag for each
feature.
• The user can use metadata KML_ALTITUDEMODE to specify how altitude components in the <co-
ordinates> element are interpreted. Possible values are: absolute, relativeToGround, clampToGround.
http://code.google.com/apis/kml/documentation/kmlreference.html#altitudemode
• The user can use metedata KML_EXTRUDE to specify whether to connect the LinearRing to the ground.
http://code.google.com/apis/kml/documentation/kmlreference.html#tessellate
• The user can use metedata KML_TESSELLATE to specify whether to allow the LinearRing to follow the terrain.
http://code.google.com/apis/kml/documentation/kmlreference.html#extrude
• The user can specify an attribute to be used as the elevation value using a layer level metedata:
‘kml_elevation_attribute’ ‘<Name of attribute>’. The value will be used as the z value when the coordinates are
written.

392 Chapter 8. Output Generation

 MapServer Documentation, Release 6.0

Point Layers

• Each layer will be inside a Folder element.

• Each feature will be represented by a Placemark.
• The Geometry element for a Point layer would be represented as a Point geometry element in Kml. Supported
elements are:

Line Layers

• Each layer will be inside a Folder element.

• Each feature in the layer would be represented by a Placemark.
• If a label is attached to the line, the Geometry element would be represented as a MultiGeometry that includes a
LineString element and a Point element representing the position of the label. If no label is attached to a feature,
the Geometry element will be a LineString.

Polygon Layers

• Each layer will be inside a Folder element.

• Each feature will be represented by a Placemark.
• If a label is attached to the polygon, the Geometry element would be represented as a MultiGeometry that
includes a Polygon element and a Point element representing the position of the label.

Annotation Layers

Not supported.

Raster Layers

• Each layer will be inside a Folder element.

• A GroundOverlay feature is created for the layer, that includes an href link to the raster image generated and
LatLongBox for the extents (map extents).
• The href is generated using the imagepath and imageurl settings in the map file.

8.12.8 Styling

As described in Section 4, all different styles from the layers will be stored at the Document level and referenced from
the folders using a styleUrl.

Point Layers

Point layers will be styled using the IconStyle styling element of kml. An image representing the symbol will be
created and referenced from the IconStyle object. If a label is attached to the point, a LabelStyle element will also be
used. The LabelStyle will have the color parameter set.

8.12. Kml Output 393

MapServer Documentation, Release 6.0

<Style id="style_label_ff0000ff_symbol_star_13.0_ff000000">
<IconStyle>
<Icon>
<href>>http://localhost/ms_tmp/4beab862_19bc_0.png</href>
</Icon>
</IconStyle>
<LabelStyle>
<color>ff0000ff</color>
</LabelStyle>
</Style>

Line Layers

Line layers will be styled using the LineStyle styling element of kml. Color and width parameters of the LineStyle
will be used. If a label is attached to the layer, a LabelStyle element will also be used.

Polygon Layers

Polygon layers will be styled using the PolyStyle styling element of kml. Color parameter of the PolyStyle will be
used. If an outline is defined in the map file, an addition LineStyle will be used. If a label is attached to the layer, a
LabelStyle element will also be used.

8.12.9 Attributes

As described in section on Layers, two ways of defining the description:

• kml/ows_description
• kml/ows_include_items

8.12.10 Coordinate system

The map level projection should be set to epsg:4326. If not set, the driver will automatically set it. Layers are expected
to have projection block if their projection is different from epsg:4326.

8.12.11 Warning and Error Messages

• When the projection of the map file is not set or is different from a a lat/lon projection, the driver automatically
sets the projection to espg:4326. If the map is is debug mode, the following message is sent: « ”KmlRen-
derer::checkProjection: Mapfile projection set to epsg:4326 »
• If imagepath and imageurl are not set in the web object, the following message is sent in debug mode: « Kml-
Renderer::startNewLayer: imagepath and imageurl should be set in the web object »

394 Chapter 8. Output Generation

 CHAPTER 9

OGC Support and Configuration

Interoperability is increasingly becoming a focus point for organizations that distribute and share data over the In-
ternet. The Open Geospatial Consortium (OGC) focuses on the development of publicly available geospatial web
standards. MapServer supports numerous OGC standards, allowing users to publish and consume data and services in
an application neutral implementation manner.

9.1 WMS Server

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-10-01

Table of Contents
• WMS Server
– Introduction
– Setting Up a WMS Server Using MapServer
– WMS 1.3.0 Support
– Reference Section
– FAQ / Common Problems

9.1.1 Introduction

A WMS (or Web Map Server) allows for use of data from several different servers, and enables for the creation of a
network of Map Servers from which clients can build customized maps. The following documentation is based on the
Open Geospatial Consortium’s (OGC) Web Map Server Interfaces Implementation Specification v1.1.1.
MapServer v3.5 or more recent is required to implement WMS features. At the time this document was written,
MapServer supports the following WMS versions: 1.0.0, 1.0.7, 1.1.0 (a.k.a. 1.0.8), 1.1.1 and 1.3.0
This document assumes that you are already familiar with certain aspects of MapServer:
• MapServer application development and setting up .map files.

395
MapServer Documentation, Release 6.0

• Familiarity with the WMS spec would be an asset. A link to the WMS specification document is included in the
“WMS-Related Information” section below.

Links to WMS-Related Information

• MapServer WMS Client Howto

• WMS 1.1.1 specification
• WMS 1.3.0 specification
• Open Geospatial Consortium (OGC) home page
• WMS-Dev mailing list and archive
• WMS Cookbook
• MapServer OGC Web Services Workshop package
• MapServer Styled Layer Descriptor (SLD) Howto
• MapServer WMS Time Support Howto

How does a WMS Work

WMS servers interact with their clients via the HTTP protocol. In most cases, a WMS server is a CGI program. This
is also the case with MapServer.
The WMS specification defines a number of request types, and for each of them a set of query parameters and associ-
ated behaviors. A WMS-compliant server MUST be able to handle at least the following 2 types of WMS requests:
1. GetCapabilities: return an XML document with metadata of the Web Map Server’s information
2. GetMap: return an image of a map according to the user’s needs.
And support for the following types is optional:
1. GetFeatureInfo: return info about feature(s) at a query (mouse click) location. MapServer supports 3 types of
responses to this request:
• text/plain output with attribute info.
• text/html output using MapServer query templates specified in the CLASS template parameter. The
MIME type returned by the Class templates defaults to text/html and can be controlled using the meta-
data “wms_feature_info_mime_type”.
• application/vnd.ogc.gml, GML.1 or GML for GML features.
2. DescribeLayer: return an XML description of one or more map layers. To execute this:
• for vector layers: to have a valid return the user needs to setup wfs_onlineresource (or ows_onlineresource)
metadata either at the map level or at the layer level (the layer level metadata is the one which is used if
both are defined)
• for raster layers: the metadata is wcs_onlineresource with the same logic as above.
3. GetLegendGraphic: returns a legend image (icon) for the requested layer, with label(s). More information on
this request can be found in the GetLegendGraphic section later in this doc.
With respect to MapServer specifically, it is the “mapserv” CGI program that knows how to handle WMS requests.
So setting up a WMS server with MapServer involves installing the mapserv CGI program and a setting up a mapfile
with appropriate metadata in it. This is covered in the rest of this document.

396 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

9.1.2 Setting Up a WMS Server Using MapServer

Install the Required Software

WMS requests are handled by the mapserv CGI program. Not all versions of the mapserv program do include WMS
support (it is included by default when you compile together with the PROJ library), so the first step is to check that
your mapserv executable includes WMS support. One way to verify this is to use the “-v” command-line switch and
look for “SUPPORTS=WMS_SERVER”.
(Unix users should refer to the Compiling on Unix document for any compiling instructions, and Windows users might
want to use MS4W, which comes ready with WMS/WFS support)
Example 1. On Unix:
$ ./mapserv -v
MapServer version 4.6.1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=PDF
OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER
SUPPORTS=WMS_CLIENT SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER
INPUT=JPEG INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE DEBUG=MSDEBUG

Example 2. On Windows:
C:\apache\cgi-bin> mapserv -v
MapServer version 4.6.1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=PDF
OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER
SUPPORTS=WMS_CLIENT SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER
INPUT=JPEG INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE DEBUG=MSDEBUG

Setup a Mapfile For Your WMS

Each instance of WMS server that you setup needs to have its own mapfile. It is just a regular MapServer mapfile in
which some parameters and some metadata entries are mandatory. Most of the metadata is required in order to produce
a valid GetCapabilites output.
Here is the list of parameters and metadata items that usually optional with MapServer, but are required (or strongly
recommended) for a WMS configuration:
At the MAP level:
• Map NAME
• Map PROJECTION
• Map Metadata (in the WEB Object):
– wms_title
– wms_onlineresource
– wms_srs (unless PROJECTION object is defined using “init=epsg:...”)
And for each LAYER:
• Layer NAME
• Layer PROJECTION
• Layer METADATA
– wms_title
– wms_srs (optional since the layers inherit the map’s SRS value)

9.1. WMS Server 397

MapServer Documentation, Release 6.0

• Layer STATUS
– Layers set to STATUS DEFAULT will always be sent to the client.
– Layers set to STATUS ON or STATUS OFF can be requested by the client.
• Layer TEMPLATE (required for GetFeatureInfo requests)
• Layer DUMP TRUE (only required for GetFeatureInfo GML requests)
Let’s go through each of these paramters in more detail:
• Map Name and wms_title:
WMS Capabilities requires a Name and a Title tag for every layer. The Map’s NAME and wms_title metadata
will be used to set the root layer’s name and title in the GetCapabilities XML output. The root layer in the WMS
context corresponds to the whole mapfile.
• Layer Name and wms_title metadata:
Every individual layer needs its own unique name and title. Layer names are also used in GetMap and GetFea-
tureInfo requests to refer to layers that should be included in the map output and in the query. Layer names must
start with a letter when setting up a WMS server (layer names should not start with a digit or have spaces in
them).
• Map PROJECTION and wms_srs metadata:
WMS servers have to advertise the projection in which they are able to serve data using EPSG projection codes
(see http://www.epsg.org/ for more background on EPSG codes). Recent versions of the PROJ4 library come
with a table of EPSG initialization codes and allow users to define a projection like this:
PROJECTION
"init=epsg:4269"
END

(Note that “epsg” has to be in lowercase when used in the PROJ4 ‘init’ directive.)
If the MAP PROJECTION block is provided in the format “init=epsg:xxxx” then MapServer will also use this
information to generate a <BoundingBox> tag for the top-level layer in the WMS capabilities document. Note
that the BoundingBox is an optional element of WMS capabilities, but it is good practice to allow MapServer to
include it when possible.
The above is sufficient for MapServer to recognize the EPSG code and include it in SRS tags in the capabilities
output (wms_srs metadata is not required in this case). However, it is often impossible to find an EPSG code
to match the projection of your data. In those cases, the “wms_srs” metadata is used to list one or more EPSG
codes that the data can be served in, and the PROJECTION object contains the real PROJ4 definition of the
data’s projection.
Here is an example of a server whose data is in an Lambert Conformal Conic projection (42304). It’s capabilities
output will advertize EPSG:4269 and EPSG:4326 projections (lat/lon), but the PROJECTION object is set to
the real projection that the data is in:
NAME "DEMO"
...

WEB
...
METADATA
"wms_title" "WMS Demo Server"
"wms_onlineresource" "http://my.host.com/cgi-bin/mapserv?map=wms.map&"
"wms_srs" "EPSG:4269 EPSG:4326"
END
END

398 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

PROJECTION
"init=epsg:42304"
END
...
END

In addition to EPSG:xxxx projections, a WMS server can advertize projections in the AUTO:xxxx namespace.
AUTO projections 42001 to 42005 are internally supported by MapServer. However, AUTO projections are
useful only with smart WMS clients, since the client needs to define the projection parameters in the WMS
requests to the server. For more information see Annex E of the WMS 1.1.1 specification and section 6.5.5.2 of
the same document. See also the FAQ on AUTO projections at the end of this document.
• Layer PROJECTION and wms_srs metadata:
By default layers inherit the SRS of their parent layer (the map’s PROJECTION in the MapServer case). For
this reason it is not necessary (but still strongly recommended) to provide PROJECTION and wms_srs for every
layer. If a layer PROJECTION is not provided then the top-level map projecion will be assumed.
Layer PROJECTION and wms_srs metadata are defined exactly the same way as the map’s PROJECTION and
wms_srs metadata.
For vector layers, if a PROJECTION block is provided in the format “init=epsg:xxxx” then MapServer will also
use this information to generate a <BoundingBox> tag for this layer in the WMS capabilities document. Note
that the BoundingBox is an optional element of WMS capabilities, but it is good practice to allow MapServer to
include it when possible.
• “wms_onlineresource” metadata:
The wms_onlineresource metadata is set in the map’s web object metadata and specifies the URL that should
be used to access your server. This is required for the GetCapabilities output. If wms_onlineresource is not
provided then MapServer will try to provide a default one using the script name and hostname, but you shouldn’t
count on that too much. It is strongly recommended that you provide the wms_onlineresource metadata.
See section 6.2.2 of the WMS 1.1.1 specification for the whole story about the online resource URL. Basically,
what you need is a complete HTTP URL including the http:// prefix, hostname, script name, potentially a
“map=” parameter, and and terminated by “?” or “&”.
Here is a valid online resource URL:
http://my.host.com/cgi-bin/mapserv?map=mywms.map&

By creating a wrapper script on the server it is possible to hide the “map=” parameter from the URL and then
your server’s online resource URL could be something like:
http://my.host.com/cgi-bin/mywms?

This is covered in more detail in the section “More About the Online Resource URL” below.
• Configuring for GetFeatureInfo Requests:
You must set the layer TEMPLATE parameter for the layer to be queryable by GetFeatureInfo requests. For
requests of type “text/html” you should also set the layer HEADER and FOOTER parameters.
For GetFeatureInfo requests of GML you must set the layer to DUMP TRUE in the mapfile. As of MapServer
4.6 you must also set the gml_* metadata for the layer attributes to be served (see the Layer Object metadata in
the Reference Section later in this document).
Here are working examples of GetFeatureInfo requests: text/plain / text/html / gml (for gml, your browser might
ask you to save the file, if so save it locally as a .gml file and view it in a text editor)

9.1. WMS Server 399

MapServer Documentation, Release 6.0

Test Your WMS Server

Validate the Capabilities Metadata

OK, now that we’ve got a mapfile, we have to check the XML capabilities returned by our server to make sure nothing
is missing.
Using a web browser, access your server’s online resource URL to which you add the parameters “SER-
VICE=WMS&VERSION=1.1.1&REQUEST=GetCapabilities” to the end, e.g.
http://my.host.com/cgi-bin/mapserv?map=mywms.map&SERVICE=WMS&VERSION=1.1.1&REQUEST=GetCapabilities

Here is a working GetCapabilities request (note that the SERVICE parameter is required for all GetCapabilities re-
quests):
http://demo.mapserver.org/cgi-bin/wms?SERVICE=WMS&VERSION=1.1.1&REQUEST=GetCapabilities
This should return a document of MIME type application/vnd.ogc.wms_xml, so your browser is likely going to prompt
you to save the file. Save it and open it in a text editor (Emacs, Notepad, etc.), and you will see the returned XML
from the WMS server.
If you get an error message in the XML output then take necessary actions. Common problems and solutions are listed
in the FAQ at the end of this document.
If everything went well, you should have a complete XML capabilities document. Search it for the word “WARN-
ING”... MapServer inserts XML comments starting with “<!–WARNING: ” in the XML output if it detects missing
mapfile parameters or metadata items. If you notice any warning in your XML output then you have to fix all of them
before you can register your server with a WMS client, otherwise things are likely not going to work.
Note that when a request happens, it is passed through WMS, WFS, and WCS in MapServer (in that order) until one
of the services respond to it.

Test With a GetMap Request

OK, now that we know that our server can produce a valid XML GetCapabilities response we should test the GetMap
request. MapServer only checks for a few of the required GetMap parameters, so both of the minimum MapServer
parameters and a valid GetMap request will be explained below.
The following is a list of the required GetMap parameters according to the WMS spec:
VERSION=version: Request version
REQUEST=GetMap: Request name
LAYERS=layer_list: Comma-separated list of one or more map layers. Optional if SLD parameter is
present.
STYLES=style_list: Comma-separated list of one rendering style per requested layer. Optional if SLD
parameter is present. NOTE that MapServer does not support named styles, so most times you would
specify “STYLES=” with an empty value. MapServer does support STYLES when used with an SLD.
SRS=namespace:identifier: Spatial Reference System.
BBOX=minx,miny,maxx,maxy: Bounding box corners (lower left, upper right) in SRS units.
WIDTH=output_width: Width in pixels of map picture.
HEIGHT=output_height: Height in pixels of map picture.
FORMAT=output_format: Output format of map.

400 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Note: WMS Servers only advertise supported formats that are part of the gd / gdal libraries.
A valid example would therefore be:
http://my.host.com/cgi-bin/mapserv?map=mywms.map&SERVICE=WMS&VERSION=1.1.1&REQUEST=GetMap&LAYERS=prov

Here is a working valid request.

Test with a Real Client

If you have access to a WMS client, then register your new server’s online resource with it and you should be off and
running.
If you don’t have your own WMS client installed already, here are a few pointers:
• MapServer itself can be used as a WMS client, see the MapServer WMS Client Howto.
• Quantum GIS is a full GIS package which includes WMS client support. (recommended)
• OpenJUMP is a desktop GIS package which includes WMS client support.
• uDig is a desktop package that allows users to add WMS layers.
• Deegree provides a WMS client.
• owsview Viewer Client Generator is an online application that allows users to add WMS layers.
This list is not exhaustive, there are several Open Source or proprietary packages that offer WMS support and could
be used to interact with your new MapServer WMS server instance.

More About the Online Resource URL

As mentioned in the section “Setup a Mapfile / wms_onlineresource metadata” above, the following Online Resource
URL is perfectly valid for a MapServer WMS according to section 6.2.2 or the WMS 1.1.1 specification:
http://my.host.com/cgi-bin/mapserv?map=mywms.map&

However, some people will argue that the above URL contains mandatory vendor-specific parameters and that this is
illegal. First we would like to point that “map=...” is not considered a vendor-specific parameter in this case since it
is part of the Online Resource URL which is defined as an opaque string terminated by “?” or “&” (See WMS 1.1.1
section 6.2.2).
But anyway, even if it’s valid, the above URL is still ugly. And you might want to use a nicer URL for your WMS
Online Resource URL. Here are some suggestions:
1. On Unix servers, you can setup a wrapper shell script that sets the MS_MAPFILE environment variable and
then passes control to the mapserv executable... that results on a cleaner OnlineResource URL:
#! /bin/sh
MS_MAPFILE=/path/to/demo.map
export MS_MAPFILE
/path/to/mapserv

2. Another option is to use the “setenvif” feature of Apache: use symbolic links that all point to a same mapserv
binary, and then for each symbolic link test the url, and set the MAP environment accordingly.
For Windows and Apache users the steps are as follows (this requires Apache 1.3 or newer):
• Copy mapserv.exe to a new name for your WMS, such as “mywms.exe”.
• In httpd.conf, add:

9.1. WMS Server 401

MapServer Documentation, Release 6.0

SetEnvIf Request_URI "/cgi-bin/mywms" MS_MAPFILE=/path/to/mymap.map

3. On IIS servers (Windows), you can use the following ASP script:
Note: The script below, while functional, is intended only as an example of using ASP to filter
MapServer requests. Using ASP in a production WMS server will likely require additional ASP
especially in the area of error handling and setting timeouts.*
<%
Server.ScriptTimeout = 360

Select Case Request.ServerVariables("REQUEST_METHOD")

Case "GET" strRequest = Request.QueryString
Case "POST" strRequest = Request.Form
End Select

strURL = "http://myserver/cgi-bin/mapserv.exe?map=C:\Inetpub\wwwroot\workshop\itasca.map&" &

Dim objHTTP
Set objHTTP = Server.CreateObject("MSXML2.ServerXMLHTTP")
objHTTP.open "GET", strURL, false
objHTTP.send ""

Response.ContentType = objHTTP.getResponseHeader("content-type")
Response.BinaryWrite objHTTP.responseBody

Set objHTTP = Nothing

%>

4. Some OGC services (WFS, SOS) support both GET and POST requests. Here, you can use a minimal MapScript
WxS wrapper. Here’s a Python example:
#!/usr/bin/python

import mapscript

req = mapscript.OWSRequest()
req.loadParams()
map = mapscript.mapObj(’/path/to/config.map’)
map.OWSDispatch(req)

GetLegendGraphic Request

This request returns a legend image (icon) for the specified layer. The request will draw an icon and a label for all
classes defined on the layer. If the requested layername is a GROUP-name, all included layers will be returned in the
legend-icon.

Requirements

The following are required in the WMS server mapfile to enable this request:
• a LEGEND object.
• a CLASS object for each layer.
• a NAME in the CLASS object.
• the STATUS of each LAYER must be set to ON.

402 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Parameters

The following are valid parameters for this request:

• LAYER - (Required) Name of the WMS layer to return the legend image of. Note that this is the <Name>
parameter of the Layer in the GetCapabilities.
• FORMAT - (Required) Format of the legend image (e.g. “image/png”).
• WIDTH - (Optional) Width of the legend image. Note that the Width parameter is only used when the Rule
parameter is also used in the request.
• HEIGHT - (Optional) Height of the legend image. Note that the Height parameter is only used when the Rule
parameter is also used in the request.
• SLD - (Optional) The URL to the SLD. Applies the SLD on the layer and the legend is drawn after the SLD is
applied (using the classes specfied by the SLD). Note here that you need to put a <Name>class1</Name> inside
the Rule element so that a class name is created from the SLD and therefore a correct legend image.
• SLD_BODY - (Optional) The body (code) of the SLD, instead of specifying a URL (as in the ‘SLD’ parameter).
• SCALE - (Optional) Specify a scale so that only layers that fall into that scale will have a legend.
• RULE - (Optional) Specify the name of the CLASS to generate the legend image for (as opposed to generating
an icon and label for ALL classes for the layer).
Note: All rules that are used to draw the legend in normal CGI mode apply here. See the CGI Reference doc if
necessary.
The CLASS object’s KEYIMAGE parameter can also be used to specify a legend image for a CLASS. See the MapFile
Reference doc if necessary. Example Request
An example request might look like:
http://127.0.0.1/cgi-bin/mapserv.exe?SERVICE=WMS&VERSION=1.1.1&layer=park&
REQUEST=getlegendgraphic&FORMAT=image/png

9.1.3 WMS 1.3.0 Support

MapServer 5.4 adds support for WMS 1.3.0. Although the general mechanism in Mapserver to support this new
specification are the same, there are some notable upgrades.

Major features related to the WMS 1.3.0 support

• Support WMS 1.3.0 basic operations: GetCapabilities, GetMap and GetFeatureInfo.

• Implement the Styled Layer Descriptor profile of the Web Map Service Implementation Specification. This
specification extends the WMS 1.3.0 and allows to advertise styling capabilities (Styled Layer Descriptor (SLD)
support). It also defines two addition operations GetLegendGraphic and DescribeLayer
• Implement the Symbology Encoding Implementation Specification, which is the new version of the SLD. Read
support was added for Point, Line, Polygon, Raster symbolizers
• Upgrade the generation of SLD to version 1.1.0 (SLD generated through through the GetStyles operation or
through MapScript)

9.1. WMS Server 403

MapServer Documentation, Release 6.0

Coordinate Systems and Axis Orientation

The most notable changes introduced in WMS 1.3.0 are the:

• the axis changes
• the introduction of new coordinate reference systems
• the use of CRS parameter (instead of SRS)
The axis order in previous versions of the WMS specifications was to always use easting (x or lon ) and northing (y or
lat). WMS 1.3.0 specifies that, depending on the particular CRS, the x axis may or may not be oriented West-to-East,
and the y axis may or may not be oriented South-to-North. The WMS portrayal operation shall account for axis order.
This affects some of the EPSG codes that were commonly used such as ESPG:4326. MapServer 5.x makes sure that
coordinates passed to the server (as part of the GetMap BBOX parameter) as well as those advertised in the capabilities
document reflect the inverse axe orders for EPSG codes between 4000 and 5000.
MapServer 6.0 and up holds a list of epsg codes with inverted axis order. It is currently based on EPSG database
version 7.6. It is also possible to define the axis order at build time for a specific ESPG code(see #3582). This allows
for example to use the “normal” axis order for some of EPSG codes between 4000 and 5000.
In addition, the WMS 1.3.0 defines a series of new coordinate system. These are the once that are currently supported
in MapServer:
• CRS:84 (WGS 84 longitude-latitude)
• CRS:83 (NAD83 longitude-latitude)
• CRS:27 (NAD27 longitude-latitude)
• AUTO2:420001 (WGS 84 / Auto UTM )
• AUTO2:420002 (WGS 84 / Auto Tr. Mercator)
• AUTO2:420003 (WGS 84 / Auto Orthographic)
• AUTO2:420004 (WGS 84 / Auto Equirectangular)
• AUTO2:420005 (WGS 84 / Auto Mollweide)

Example of requests

Users can use the CRS:84 coordinate system and order the BBOX coordinates as long/lat: -
...&CRS=CRS:84&BBOX=-180.0,-90.0,180.0,90.0&... (example request)
Users can also use the ESPG:4326 coordinates and use the axis odering of lat/long: - ...&EPSG:4326&BBOX=-90.0,-
180.0,90,180.0&... (example request)

Other notable changes

• valid values for the EXCEPTIONS parameter in a GetMap request are XML, INIMAGE, BLANK
• valid value for the EXCEPTIONS parameter in a GetFeatureInfo request is XML
• LayerLimit is introduced, allowing a server to advertise and limit the number of layers a client is allowed to
include in a GetMap request

404 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Some Missing features

• WMS 1.3.0 Post request should be an XML document containing the different operations and parameters.
• SLD documents containing elements form the Feature Encoding 1.1 specification could potentially use ESPG
projections with some filters. It is not yet clear nor implemented if the axis ordering should be taken into account
in these specific cases.

OCG compliance tests

As of version 5.4, MapServer passes all the basic and query tests of the OGC CITE test suite for WMS 1.3.0.

9.1.4 Reference Section

The following metadata are available in the setup of the mapfile:

(Note that each of the metadata below can also be referred to as ‘ows_*’ instead of ‘wms_*’. MapServer tries the
‘wms_*’ metadata first, and if not found it tries the corresponding ‘ows_*’ name. Using this reduces the amount of
duplication in mapfiles that support multiple OGC interfaces since “ows_*” metadata can be used almost everywhere
for common metadata items shared by multiple OGC interfaces.)

Web Object Metadata

ows_http_max_age
• Description: (Optional) an integer (in seconds) to specify how long a given map response should be con-
sidered new. Setting this directive allows for aware WMS clients to use this resulting HTTP header
value as a means to optimize (and minimize) requests to a WMS Server. More info is available at
http://www.mnot.net/cache_docs/#CACHE-CONTROL
ows_updatesequence
• Description: (Optional) The updateSequence parameter can be used for maintaining the consistency of a client
cache of the contents of a service metadata document. The parameter value can be an integer, a timestamp in
[ISO 8601:2000] format, or any other number or string.
ows_sld_enabled
• Description: (Optional) A value (true or false) which, when set to “false”, will ignore SLD and SLD_BODY
parameters in order to disable remote styling of WMS layers. Also, SLD is not advertised in WMS Capabilities
as a result
ows_schemas_location
• Description: (Optional) (Note the name ows_schemas_location and not wms_... this is because all OGC Web
Services (OWS) use the same metadata) Root of the web tree where the family of OGC WMS XMLSchema
files are located. This must be a valid URL where the actual .xsd files are located if you want your WMS output
to validate in a validating XML parser. Default is http://schemas.opengis.net. See http://ogc.dmsolutions.ca for
an example of a valid schema tree.
wms_abstract
• WMS TAG Name: Abstract (WMS1.1.1, sect. 7.1.4.2)
• Description: (Optional) A blurb of text providing more information about the WMS server.
wms_accessconstraints
• WMS TAG Name: AccessConstraints (WMS1.1.1, sect. 7.1.4.2)

9.1. WMS Server 405

MapServer Documentation, Release 6.0

• Description: (Optional) Access constraints information. Use the reserved word “none” if there are no access
constraints.
wms_addresstype, wms_address, wms_city, wms_stateorprovince, wms_postcode, wms_country
• WMS TAG Name: ContactAddress and family (WMS1.1.1, sect. 7.1.4.2)
• Description: Optional contact address information. If provided then all six metadata items are required.
wms_attribution_logourl_format
• Description: (Optional) The MIME type of the logo image. (e.g. “image/png”). Note that the other
wms_attribution_logourl_* metadata must also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_logourl_height
• Description: (Optional) Height of the logo image in pixels. Note that the other wms_attribution_logourl_*
metadata must also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_logourl_href
• Description: (Optional) URL of the logo image. Note that the other wms_attribution_logourl_* metadata must
also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_logourl_width
• Description: (Optional) Width of the logo image in pixels. Note that the other wms_attribution_logourl_*
metadata must also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_onlineresource
• Description: (Optional) The data provider’s URL.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_title
• Description: (Optional) Human-readable string naming the data provider.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_contactelectronicmailaddress
• WMS TAG Name: ContactElectronicMailAddress (WMS1.1.1, sect. 7.1.4.2)
• Description: Optional contact Email address.
wms_contactfacsimiletelephone
• WMS TAG Name: ContactFacsimileTelephone (WMS1.1.1, sect. 7.1.4.2)
• Description: Optional contact facsimile telephone number.
wms_contactperson, wms_contactorganization, wms_contactposition
• WMS TAG Name: ContactInformation, ContactPerson, ContactOrganization, ContactPosition (WMS1.1.1, sect.
7.1.4.2)
• Description: Optional contact information. If provided then all three metadata items are required.
wms_contactvoicetelephone

406 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• WMS TAG Name: ContactVoiceTelephone (WMS1.1.1, sect. 7.1.4.2)

• Description: Optional contact voice telephone number.
wms_encoding
• WMS TAG Name: Encoding
• Description: Optional XML capabilities encoding type. The default is ISO-8859-1.
wms_feature_info_mime_type
• WMS TAG Name: Feature_info_mime_type
• Description:
– Used to specify an additional MIME type that can be used when responding to the GetFeature request.
For example if you want to use the layer’s HTML template as a base for its response, you need to add
“WMS_FEATURE_INFO_MIME_TYPE” “text/html”. Setting this will have the effect of advertizing
text/html as one of the MIME types supported for a GetFeature request. You also need to make sure that the
layer points to a valid html template. The client can then call the server with INFO_FORMAT=text/html.
– If not specified, MapServer by default has text/plain and GML implemented.
– Note that for GML to be returned the layer in the wms-server mapfile must be set to DUMP TRUE.
wms_fees
• WMS TAG Name: Fees (WMS1.1.1, sect. 7.1.4.2)
• Description: (Optional) Fees information. Use the reserved word “none” if there are no fees.
wms_getcapabilities_version
• Description: (Optional) Default version to use for GetCapabilities requests that do not have a version parameter.
If not set, the latest supported version will be returned.
wms_getlegendgraphic_formatlist
• Description: (Optional) A comma-separated list of valid formats for a WMS GetLegendGraphic request.
wms_getmap_formatlist
• Description: (Optional) A comma-separated list of valid formats for a WMS GetMap request.
wms_keywordlist
• WMS TAG Name: KeywordList (WMS1.1.1, sect. 7.1.4.2)
• Description: (Optional) A comma-separated list of keywords or keyword phrases to help catalog searching. As
of WMS 1.1.0 no controlled vocabulary has been defined.
wms_onlineresource
• WMS TAG Name: OnlineResource (WMS1.1.1, sect. 6.2.2)
• Description: (Recommended) The URL that will be used to access this WMS server. This value is used in the
GetCapabilities response.
• See Also: sections “Setup a Mapfile / wms_onlineresource metadata” and “More About the Online Resource
URL” above.
wms_resx, wms_resy
• WMS TAG Name: BoundingBox (WMS1.1.1, sect. 6.5.6)
• Description: (Optional) Used in the BoundingBox tag to provide info about spatial resolution of the data, values
are in map projection units.

9.1. WMS Server 407

MapServer Documentation, Release 6.0

wms_service_onlineresource
• Description: (Optional) Top-level onlineresource URL. MapServer uses the onlineresource metadata (if pro-
vided) in the following order:
1. wms_service_onlineresource
2. ows_service_onlineresource
3. wms_onlineresource (or automatically generated URL, see the onlineresource section of this document)
wms_srs
• WMS TAG Name: SRS (WMS1.1.1, sect. 6.5.5)
• Description: (Recommended) Contains a list of EPSG projection codes that should be advertized as being
available for all layers in this server. The value can contain one or more EPSG:<code> pairs separated by spaces
(e.g. “EPSG:4269 EPSG:4326”) This value should be upper case (EPSG:42304.....not epsg:42304) to avoid
problems with case sensitive platforms.
• See Also: section “Setup a Mapfile / Map PROJECTION and wms_srs metadata” above.
wms_timeformat
• Description: The time format to be used when a request is sent. (e.g. “wms_timeformat” “%Y-%m-%d %H,
%Y-%m-%d %H:%M”). Please see the WMS Time Support Howto for more information.
wms_title
• WMS TAG Name: Title (WMS1.1.1, sect. 7.1.4.1)
• Description: (Required) A human-readable name for this Layer.
wms_rootlayer_title
• WMS TAG Name: Title (WMS1.1.1, sect. 7.1.4.1)
• Description: (Optional) Same as wms_title, applied to the root Layer element. If not set, then wms_title will be
used.
wms_rootlayer_abstract
• WMS TAG Name: Abstract (WMS1.1.1, sect. 7.1.4.2)
• Description: (Optional) Same as wms_abstract, applied to the root Layer element. If not set, then wms_abstract
will be used.
wms_rootlayer_keywordlist
• WMS TAG Name: KeywordList (WMS1.1.1, sect. 7.1.4.2)
• Description: (Optional) Same as wms_keywordlist, applied to the root Layer element. If not set, then
wms_keywordlist will be used.
wms_layerlimit
• WMS TAG Name: LayerLimit (WMS1.3.0, sect. 7.2.4.3)
• Description: (Optional) The maximum number of layers a WMS client can specify in a GetMap request. If not
set, then no limit is imposed.

Layer Object Metadata

gml_exclude_items

408 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• Description: (Optional, applies only to GetFeatureInfo GML requests) A comma delimited list of items to
exclude. As of MapServer 4.6, you can control how many attributes (fields) you expose for your data layer with
metadata. The previous behaviour was simply to expose all attributes all of the time. The default is to expose
no attributes at all. An example excluding a specific field would be:
"gml_include_items" "all"
"gml_exclude_items" "Phonenum"

gml_groups
• Description: (Optional, applies only to GetFeatureInfo GML requests) A comma delimited list of group names
for the layer.
gml_[group name]_group
• Description: (Optional, applies only to GetFeatureInfo GML requests) A comma delimited list of attributes in
the group. Here is an example:
"gml_include_items" "all"
"gml_groups" "display"
"gml_display_group" "Name_e,Name_f"

gml_include_items
• Description: (Optional, applies only to GetFeatureInfo GML requests) A comma delimited list of items to
include, or keyword “all”. As of MapServer 4.6, you can control how many attributes (fields) you expose for
your data layer with this metadata. The previous behaviour was simply to expose all attributes all of the time.
You can enable full exposure by using the keyword “all”, such as:
"gml_include_items" "all"

You can specify a list of attributes (fields) for partial exposure, such as:
"gml_include_items" "Name,ID"

The new default behaviour is to expose no attributes at all.

gml_[item name]_alias
• Description: (Optional, applies only to GetFeatureInfo GML requests) An alias for an attribute’s name. The
served GML will refer to this attribute by the alias. Here is an example:
"gml_province_alias" "prov"

gml_[item name]_type
• Description: (Optional, applies only to GetFeatureInfo GML requests) Specifies the type of the attribute. Valid
values are Integer|Real|Character|Date|Boolean.
gml_xml_items
• Description: (Optional, applies only to GetFeatureInfo GML requests) A comma delimited list of items that
should not be XML-encoded.
gml_geometries
• Description: (Optional, applies only to GetFeatureInfo GML requests) Provides a name for geometry elements.
The value is specified as a string to be used for geometry element names. By default, GML geometries are
not written in GML GetFeatureInfo output, unless gml_geometries and gml_[name]_type are both set. By
default, only the bounding box is written. If gml_geometries is set to “none”, neither the bounding box
nor the geometry are written.
gml_[name]_type

9.1. WMS Server 409

MapServer Documentation, Release 6.0

• Description: (Optional, applies only to GetFeatureInfo GML requests) When employing gml_geometries, it
is also necessary to specify the geometry type of the layer. This is accomplished by providing a value for
gml_[name]_type, where [name] is the string value specified for gml_geometries, and a value which is
one of:
– point
– multipoint
– line
– multiline
– polygon
– multipolygon
wms_abstract
• Same as wms_abstract in the Web Object.
wms_attribution_logourl_format
• Description: (Optional) The MIME type of the logo image. (e.g. “image/png”). Note that the other
wms_attribution_logourl_* metadata must also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_logourl_height
• Description: (Optional) Height of the logo image in pixels. Note that the other wms_attribution_logourl_*
metadata must also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_logourl_href
• Description: (Optional) URL of the logo image. Note that the other wms_attribution_logourl_* metadata must
also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_logourl_width
• Description: (Optional) Width of the logo image in pixels. Note that the other wms_attribution_logourl_*
metadata must also be specified.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_onlineresource
• Description: (Optional) The data provider’s URL.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_attribution_title
• Description: (Optional) Human-readable string naming the data provider.
• refer to section 7.1.4.5.11 of the WMS 1.1.1 spec.
wms_authorityurl_name, wms_authorityurl_href - Description: (Optional) AuthorityURL is used in tan-
dem with Identifier values to provide a means of linking identifier information back to a web service.
The wms_identifier_authority should provide a string that matches a declared wms_authorityurl_name. Both
wms_authorityurl_name and wms_authorityurl_href must be present for an AuthorityURL tag to be written to the
capabilities. - refer to section 7.1.4.5.12 of the WMS 1.1.1 spec.

410 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

wms_identifier_authority, wms_identifier_value - Description: (Optional) Identifier is used in tandem with

AuthorityURL end points to provide a means of linking identifier information back to a web service. The
wms_identifier_authority should provide a string that matches a declared wms_authorityurl_name. Both
wms_identifier_authority and wms_identifier_value must be present for an Identifier tag to be written to the capa-
bilities. - refer to section 7.1.4.5.12 of the WMS 1.1.1 spec.
wms_dataurl_format
• Description: (Optional) Non-standardized file format of the metadata. The layer metadata wms_dataurl_href
must also be specified.
• refer to section 7.1.4.5.14 of the WMS 1.1.1 spec.
wms_dataurl_href
• Description: (Optional) The URL to the layer’s metadata. The layer metadata wms_dataurl_format must also
be specified.
• refer to section 7.1.4.5.14 of the WMS 1.1.1 spec.
wms_extent
• WMS TAG Name: BoundingBox (WMS1.1.1, sect. 6.5.6)
• Description: (Optional) Used for the layer’s BoundingBox tag for cases where it is impossible (or very ineffi-
cient) for MapServer to probe the data source to figure its extents. The value for this metadata is “minx miny
maxx maxy” separated by spaces, with the values in the layer’s projection units. If wms_extent is provided then
it has priority and MapServer will NOT try to read the source file’s extents.
For Rasters served through WMS, MapServer can now use the wms_extent metadata parameter to register the
image. If a .wld file cannot be found, MapServer will then look for the wms_extent metadata parameter and use
the extents of the image and the size of the image for georegistration.
wms_getfeatureinfo_formatlist
• Description: (Optional) Comma-separted list of formats that sould be valid for a GetFeatureInfo request. If
defined, only these formats are advertised through in the Capabilities document.
wms_getlegendgraphic_formatlist
• Description: (Optional) Comma-separted list of image formats that sould be valid for a GetLegendGraphic
request. If defined, only these formats are advertised through in the Capabilities document.
wms_getmap_formatlist
• Description: (Optional) Comma-separted list of image formats that sould be valid for a GetMap request. If
defined, only these formats are advertised through in the Capabilities document.
wms_group_abstract
• Description: (Optional) A blurb of text providing more information about the group. Only one layer for the
group needs to contain wms_group_abstract, MapServer will find and use the value. The value found for the
first layer in the group is used. So if multiple layers have wms_group_abstract set then only the first value is
used.
wms_group_title
• WMS TAG Name: Group_title (WMS1.1.1, sect. 7.1.4.1)
• Description: (Optional) A human-readable name for the group that this layer belongs to. Only one layer for the
group needs to contain wms_group_title, MapServer will find and use the value. The value found for the first
layer in the group is used. So if multiple layers have wms_group_title set then only the first value is used.
wms_keywordlist

9.1. WMS Server 411

MapServer Documentation, Release 6.0

• Same as wms_keywordlist in the Web Object .

wms_layer_group
• Description: (Optional) Can be used to assign a layer to a number of hierarchically nested groups. This grouped
hierarchy will be expressed in the capabilities.
WMS_LAYER_GROUP is different from the GROUP keyword in that it does not publish the name of the
group in the capabilities, only the title. As a consequence the groups set with WMS_LAYER_GROUP can
not be requested with a GetMap or GetFeatureInfo request (see section 7.1.4.5.2 of the WMS implementation
specification version 1.1.1. (OGC 01-068r2)). Another difference is that GROUP does not support nested
groups. The purpose of this metadata setting is to enable making a WMS client aware of layer grouping.
All group names should be preceded by a forward slash (/). It is not allowed to use both the
WMS_LAYER_GROUP setting and the GROUP keyword for a single layer.
LAYER
NAME "mylayer"
DATA "mylayer"
TYPE LINE
CLASS
STYLE
COLOR 100 100 255
END
END
METADATA
"WMS_LAYER_GROUP" "/rootgroup/subgroup"
END
END

wms_metadataurl_format
• Description: (Optional) The file format MIME type of the metadata record (e.g. “text/plain”). The layer
metadata wms_metadataurl_type and wms_metadataurl_href must also be specified.
• refer to section 7.1.4.5.10 of the WMS 1.1.1 spec.
wms_metadataurl_href
• Description: (Optional) The URL to the layer’s metadata. The layer metadata wms_metadataurl_format and
wms_metadataurl_type must also be specified.
• refer to section 7.1.4.5.10 of the WMS 1.1.1 spec.
wms_metadataurl_type
• Description: (Optional) The standard to which the metadata complies. Currently only two types are valid:
“TC211” which refers to [ISO 19115], and “FGDC” which refers to [FGDC-STD-001-1988]. The layer meta-
data wms_metadataurl_format and wms_metadataurl_href must also be specified.
• refer to section 7.1.4.5.10 of the WMS 1.1.1 spec.
wms_opaque
• WMS TAG Name: Opaque (WMS1.1.1, sect. 7.1.4.6.3)
• Description: (Optional) Set this metadata to “1” to indicate that the layer represents an area-filling coverage of
space (e.g. a bathymetry and elevation layer). This should be taken by the client as a hint that this layer should
be placed at the bottom of the stack of layers.
wms_srs
• Same as wms_srs in the Web Object .
wms_style

412 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• Description: (Optional) The LegendURL style name. Requires the following meta-
data: wms_style_<style’s_name>_width, wms_style_<style’s_name>_legendurl_height,
wms_style_<style’s_name>_legendurl_format, wms_style_<style’s_name>_legendurl_href
• refer to section 7.1.4.5.4 of the WMS 1.1.1 spec.
wms_style_<style’s_name>_legendurl_height
• Description: (Optional) The height of the legend image in pixels. Requires the following
metadata: wms_style_<style’s_name>_width, wms_style, wms_style_<style’s_name>_legendurl_format,
wms_style_<style’s_name>_legendurl_href.
• refer to section 7.1.4.5.4 of the WMS 1.1.1 spec.
wms_style_<style’s_name>_legendurl_href
• Description: (Optional) The URL to the layer’s legend. Requires the following
metadata: wms_style_<style’s_name>_width, wms_style_<style’s_name>_legendurl_height,
wms_style_<style’s_name>_legendurl_format, wms_style.
• refer to section 7.1.4.5.4 of the WMS 1.1.1 spec.
wms_style_<style’s_name>_legendurl_format
• Description: (Optional) The file format MIME type of the legend image. Requires the follow-
ing metadata: wms_style_<style’s_name>_width, wms_style_<style’s_name>_legendurl_height, wms_style,
wms_style_<style’s_name>_legendurl_href.
• refer to section 7.1.4.5.4 of the WMS 1.1.1 spec.
wms_style_<style’s_name>_legendurl_width
• Description: (Optional) The width of the legend image in pixels. Requires the following
metadata: wms_style_<style’s_name>_format, wms_style_<style’s_name>_legendurl_height, wms_style,
wms_style_<style’s_name>_legendurl_href.
• refer to section 7.1.4.5.4 of the WMS 1.1.1 spec.
wms_timedefault
• Description: (Optional for Time Support) This value is used if it is defined and the Time value is missing in the
request. Please see the WMS Time Support Howto for more information.
wms_timeextent
• Description: (Mandatory for Time Support) This is used in the capabilities to return the valid time values for
the layer. The value defined here should be a valid time range. Please see the WMS Time Support Howto for
more information.
wms_timeitem
• Description: (Mandatory for Time Support) This is the name of the field in the DB that contains the time values.
Please see the WMS Time Support Howto for more information.
wms_title
• Same as wms_title in the Web Object.

Sample WMS Server Mapfile

The following is a very basic WMS Server mapfile:

9.1. WMS Server 413

 MapServer Documentation, Release 6.0

1 NAME "WMS-test"
2 STATUS ON
3 SIZE 400 300
4 SYMBOLSET ../ e t c / s y m b o l s . s y m
5 EXTENT -2200000 -712631 3072800 3840000
6 UNITS METERS
7 SHAPEPATH "../data"
8 IMAGECOLOR 255 255 255
9 FONTSET ../ e t c / f o n t s . t x t
10

11 WEB
12 IMAGEPATH "/ms4w/tmp/ms_tmp/"
13 IMAGEURL "/ms_tmp/"
14 METADATA
15 "wms_title" "WMS Demo Server" ##required
16 "wms_onlineresource" "http://yourpath/cgi-bin/mapserv.exe?" ##required
17 "wms_srs" "EPSG:42304 EPSG:42101 EPSG:4269 EPSG:4326" ##recommended
18 END
19 END
20

21 PROJECTION
22 "init=epsg:42304" ##required
23 END
24

25 #
26 # Start of layer definitions
27 #
28

29 LAYER
30 NAME "park"
31 METADATA
32 "wms_title" "Parks" ##required
33 END
34 TYPE POLYGON
35 STATUS OFF
36 DATA p a r k
37 PROJECTION
38 "init=epsg:42304" ##recommended
39 END
40 CLASS
41 NAME "Parks"
42 STYLE
43 COLOR 200 255 0
44 OUTLINECOLOR 120 120 120
45 END
46 END
47 END # Layer
48

49 LAYER
50 NAME p o p p l a c e
51 METADATA
52 "wms_title" "Cities" ##required
53 END
54 TYPE POINT
55 STATUS ON
56 DATA p o p p l a c e
57 PROJECTION
58 "init=epsg:42304" ##recommended
59 END
60 CLASS
61 NAME "Cities"
62
414 STYLE Chapter 9. OGC Support and Configuration
63 SYMBOL 2
64 SIZE 8
65 COLOR 0 0 0
 MapServer Documentation, Release 6.0

9.1.5 FAQ / Common Problems

Q How can I find the EPSG code for my data’s projection?

A If you know the parameters of your data’s projection, then you can browse the “epsg” file that comes
with PROJ4 and look for a projection definition that matches your data’s projection. It’s a simple
text file and the EPSG code is inside brackets (<...>) at the beginning of every line.
The “epsg” file is usually located in /usr/local/share/proj/ on Unix systems and in C:/PROJ/ or
C:/PROJ/NAD in Windows systems (depending on the installation). MS4W users will find the epsg
file in /MS4W/proj/nad/.

Q My WMS server produces the error “msProcessProjection(): no system list, errno: ..”
A That’s likely PROJ4 complaining that it cannot find the “epsg” projection definition file. Make sure
you have installed PROJ 4.4.3 or more recent and that the “epsg” file is installed at the right location.
On Unix it should be under /usr/local/share/proj/, and on Windows PROJ looks for it under C:/PROJ/
or C:/PROJ/NAD (depending on the installation). You should also check the error documentation to
see if your exact error is discussed.
If you don’t have the “epsg” file then you can get it as part of the PROJ4 distribution at
http://trac.osgeo.org/proj/ or you can download it at http://www.maptools.org/dl/proj4-epsg.zip.

Q How do AUTO projections work?

A When a WMS client calls a WMS server with an auto projection, it has to specify the SRS in the form:
AUTO: proj_id,unit_id,lon0,lat0 where:
• proj_id is one of 42001, 42002, 42003, 42004, or 42005 (only five auto projections are currently
defined).
• unit_id is always 9001 for meters. (It is uncertain whether anyone supports any other units.)
• lon0 and lat0 are the coordinates to use as the origin for the projection.
When using an AUTO projection in WMS GetCapabilities, you include only the “AUTO:42003”
string in your wms_srs metadata, you do not include the projection parameters. Those are added by
the application (client) at runtime depending on the map view. For example:
NAME "DEMO"
...

WEB
...
METADATA
"wms_title" "WMS Demo Server"
"wms_onlineresource" "http://my.host.com/cgi-bin/mapserv?map=wms.map&"
"wms_srs" "AUTO:42001 AUTO:42002"
END
END

The above server advertises the first two auto projections.

9.2 WMS Client

Author Jeff McKenna

9.2. WMS Client 415

MapServer Documentation, Release 6.0

Contact jmckenna at gatewaygeomatics.com

Last Updated 2010-10-01

Table of Contents
• WMS Client
– Introduction
– Compilation / Installation
– MapFile Configuration
– Limitations/TODO

9.2.1 Introduction

A WMS (or Web Map Server) allows for use of data from several different servers, and enables for the creation
of a network of Map Servers from which clients can build customized maps. The following document contains
information about using MapServer’s WMS connection type to include layers from remote WMS servers in MapServer
applications.
This document assumes that you are already familiar with certain aspects of MapServer:
• MapServer application development and setting up .map files.
• Familiarity with the WMS spec would be an asset. A link to the WMS specification document is included below.

WMS-Related Information

• MapServer WMS Server HowTo

• WMS 1.1.1 specification
• MapServer OGC Web Services Workshop package

9.2.2 Compilation / Installation

The WMS connection type is enabled by the –with-wmsclient configure switch. It requires PROJ4, GDAL and libcurl
version 7.10.1 or more recent. Windows users who do not want to compile MapServer should use MS4W (which
comes ready for WMS/WFS client and server use), or check for the availability of other Windows binaries with WMS
support.
• For PROJ4 and GDAL installation, see the MapServer Compilation HowTo (Compiling on Unix / Compiling on
Win32)
• For libcurl, make sure you have version 7.10.1 or more recent installed on your system. You can find out your
libcurl version using curl-config –version. (if your system came with an older version of libcurl preinstalled
then you MUST uninstall it prior to installing the new version)
Once the required libraries are installed, then configure MapServer using the –with-wmsclient switch (plus all the other
switches you used to use) and recompile. This will give you a new set of executables (and possibly php_mapscript.so
if you requested it). See the MapServer Compilation HowTo (links above) for installation details.

416 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Check your MapServer executable

To check that your mapserv executable includes WMS support, use the “-v” command-line switch and look for “SUP-
PORTS=WMS_CLIENT”.
Example 1. Mapserv Version Info on Unix:
$ ./mapserv -v
MapServer version 5.2.0-rc1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
OUTPUT=PDF OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=AGG
SUPPORTS=FREETYPE SUPPORTS=ICONV SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT
SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER
SUPPORTS=SOS_SERVER SUPPORTS=FASTCGI SUPPORTS=THREADS
SUPPORTS=GEOS SUPPORTS=RGBA_PNG INPUT=JPEG INPUT=POSTGIS
INPUT=ORACLESPATIAL INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE

Example 2. Mapserv Version Info on Windows:

C:\ms4w\apache\cgi-bin> mapserv -v
MapServer version 5.2.0-rc1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
OUTPUT=PDF OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=AGG
SUPPORTS=FREETYPE SUPPORTS=ICONV SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT
SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER
SUPPORTS=SOS_SERVER SUPPORTS=FASTCGI SUPPORTS=THREADS
SUPPORTS=GEOS SUPPORTS=RGBA_PNG INPUT=JPEG INPUT=POSTGIS
INPUT=ORACLESPATIAL INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE

Install Optional PROJ4 EPSG Codes

(Note: installing these PROJ4 codes is optional, install only if you need them)
Some Canadian WMS servers will use some non-standard projection codes not included in the default distribution
(e.g. EPSG:42304, etc.). If you are planning to use MapServer to connect to Canadian WMS servers then you might
want to download a custom Canadian epsg file with those codes, and unzip it in the /usr/local/share/proj directory (or
/ms4w/proj/nad/ for MS4W users).
Finally, ESRI WMS servers also come with their own series of non-standard codes. If you are planning to connect
to ESRI WMS servers then you might want to get a custom epsg file that contains the canadian codes and the ESRI
codes, allowing you to connect to any server. Download the custom ESRI epsg file and unzip it in /usr/local/share/proj
(or /ms4w/proj/nad/ for MS4W users).
Q But why not always install and distribute the proj4-epsg-with-42xxx-and-esri.zip file then since it’s
more complete?
A You should install only the epsg projection codes that you need, the epsg file with all ESRI codes in it
is 20% larger than the default one, so it comes with extra overhead that you may not need. Also note
that when creating WMS servers, in order to be really interoperable, only EPSG codes that are part
of the standard EPSG list should be used. i.e. it is a bad idea for interoperability to use the custom
canadian codes or the custom ESRI codes and we do not want to promote their use too much.

9.2.3 MapFile Configuration

Note: A PROJECTION must be set in the mapfile for the MAP unless you are sure that all your WMS layers support
only a single projection which is the same as the PROJECTION of the map. The MAP PROJECTION can be set using
“init=epsg:xxxx” codes or using regular PROJ4 parameters. Failure to set a MAP PROJECTION may result in blank
maps coming from remote WMS servers (because of inconsistent BBOX+SRS combination being used in the WMS
connection URL).

9.2. WMS Client 417

MapServer Documentation, Release 6.0

Storing Temporary Files

Before version 6.0, and in version 6.0 when wms_cache_to_disk metadata is turned on, you have to set the IM-
AGEPATH value in the WEB object of your mapfile to point to a valid and writable directory. MapServer will use this
directory to store temporary files downloaded from the remote servers. The temporary files are automatically deleted
by MapServer so you won’t notice them.
Example 3. Setting IMAGEPATH Parameter in Mapfile
MAP
...
WEB
IMAGEPATH "/tmp/ms_tmp/"
IMAGEURL ...
END
...
END

If you want to keep this temporary file for debugging purposes, you should add the following statement to the LAYER
object of your mapfile:
LAYER
....
DEBUG ON
...
END

Adding a WMS Layer

WMS layers are accessed via the WMS connection type in the Mapfile. Here is an example of a layer using this
connection type:
LAYER
NAME "country_bounds"
TYPE RASTER
STATUS ON
CONNECTION "http://demo.mapserver.org/cgi-bin/wms?"
CONNECTIONTYPE WMS
METADATA
"wms_srs" "EPSG:4326"
"wms_name" "country_bounds"
"wms_server_version" "1.1.1"
"wms_format" "image/gif"
END
END

Required Layer Parameters and Metadata

• CONNECTIONTYPE WMS
• CONNECTION - this is the remote server’s online resource URL, just the base URL without any of the WMS
parameters. The server version, image format, layer name, etc. will be provided via metadata, see below.
Note: Note that if the CONNECTION parameter value is not set the the value of the “wms_onlineresource” metadata
will be used. If both CONNECTION and “wms_onlineresource” are set then the “wms_onlineresource” metadata
takes precedence.

418 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• “wms_format” metadata - the image format to use in GetMap requests.

Note: If wms_formatlist is provided then wms_format is optional and MapServer will pick the first supported format
in wms_formatlist for use in GetMap requests. If both wms_format and wms_formatlist are provided then wms_format
takes precedence. Also note that WMS Servers only advertize supported formats that are part of the GD/GDAL
libraries.
• “wms_name” metadata - comma-separated list of layers to be fetched from the remote WMS server. This
value is used to set the LAYERS and QUERY_LAYERS WMS URL parameters.
• “wms_server_version” metadata - the version of the WMS protocol supported by the remote WMS server and
that will be used for issuing GetMap requests.
• “wms_srs” metadata - space-delimited list of EPSG projection codes supported by the remote server. You
normally get this from the server’s capabilities output. This value should be upper case (EPSG:4236.....not
epsg:4236) to avoid problems with case sensitive platforms. The value is used to set the SRS WMS URL
parameter.

Optional Layer Parameters and Metadata

• MINSCALE, MAXSCALE - if the remote server’s capabilities contains a ScaleHint value for this layer then
you might want to set the MINSCALE and MAXSCALE in the LAYER object in the mapfile. This will allow
MapServer to request the layer only at scales where it makes sense
• PROJECTION object - it is optional at this point. MapServer will create one internally if needed. Including
one may
allow MapServer to avoid looking up a definition in the PROJ.4 init files.
• “wms_auth_username” metadata - msEncrypt-style authorization string. Empty strings are also accepted.
METADATA
"wms_auth_username" "foo"
"wms_auth_password" "{FF88CFDAAE1A5E33}"
END

• “wms_auth_type” metadata - Authorization type. Supported types include:

– basic
– digest
– ntlm
– any (the underlying http library picks the best among the opotions supported by the remote server)
– anysafe (the underlying http library picks only safe methods among the options supported by the remote
server)
METADATA
"wms_auth_type" "ntlm"
END

• “wms_connectiontimeout” metadata - the maximum time to wait for a remote WMS layer to load, set in
seconds (default is 30 seconds). This metadata can be added at the layer level so that it affects only that
layer, or it can be added at the map level (in the web object) so that it affects all of the layers. Note that
wms_connectiontimeout at the layer level has priority over the map level.
METADATA
...
"wms_connectiontimeout" "60"

9.2. WMS Client 419

MapServer Documentation, Release 6.0

...
END

• “wms_exceptions_format” metadata - set the format for exceptions (as of MapServer 4.6). MapServer de-
faults to application/vnd.ogc.se_inimage (the exception will be in a picture format). You can check the GetCa-
pabilities of the server to see what formats are available for exceptions. The application/vnd.ogc.se_inimage
exception format is actually a non-required exception format in the WMS 1.1.1 spec, so there are servers out
there which don’t support this format. In that case you would use:
LAYER
...
METADATA
"wms_exceptions_format " a p p l i c a t i on/ v n d . o g c . s e _ x m l "
END
...
END

Which would return this xml exception in the MS_ERRORFILE:

Tue Jan 17 18:05:13 2006 - msDrawWMSLayerLow(): WMS server error. WMS GetMap
request got XML exception for layer ’prov_bound’: <?xml version=’1.0’
encoding="ISO-8859-1" standalone="no" ?><!DOCTYPE ServiceExceptionReport
SYSTEM "http://schemas.opengis.net/wms/1.1.1/exception_1_1_1.dtd">
<ServiceExceptionReport version="1.1.1"><ServiceException code="LayerNotDefined">
msWMSLoadGetMapParams(): WMS server error. Invalid layer(s) given in the LAYERS parameter.
</ServiceException>
</ServiceExceptionReport>

• “wms_force_separate_request” metadata - set this to “1” to force this WMS layer to be requested using its
own separate GetMap request. By default MapServer will try to merge multiple adjacent WMS layers from the
same server into a single multi-layer GetMap request to reduce the load on remote servers and improve response
time. This metadata is used to bypass that behavior.
• “wms_formatlist” metadata - comma-separated list of image formats supported by the remote WMS server.
Note that wms_formatlist is used only if wms_format is not set. If both wms_format and wms_formatlist are
provided then wms_format takes precedence.
• “wms_latlonboundingbox” metadata - the bounding box of this layer in geographic coordinates in the format
“lon_min lat_min lon_max lat_max”. If it is set then MapServer will request the layer only when the map view
overlaps that bounding box. You normally get this from the server’s capabilities output.
METADATA
"wms_latlonboundingbox" "-124 48 -123 49"
END

• “wms_proxy_auth_type” metadata - the authorization type to use for a proxy connection. Supported types
include:
– basic
– digest
– ntlm
– any (the underlying http library picks the best among the opotions supported by the remote server)
– anysafe (the underlying http library picks only safe methods among the options supported by the remote
server)
METADATA
"wms_proxy_auth_type" "ntlm"
END

420 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• “wms_proxy_host” metadata - the hostname of the proxy to use, in “dot-quad” format, with an optional port
component (e.g. ‘192.168.2.10:8080’).
METADATA
"wms_proxy_host" "192.168.2.10"
END

• “wms_proxy_port” metadata - the port to use for a proxy connection.

METADATA
"wms_proxy_port" "8080"
END

• “wms_proxy_type” metadata - the type of the proxy connection. Valid values are ‘http’ and ‘socks5’, which
are case sensitive.
METADATA
"wms_proxy_type" "http"
END

• “wms_proxy_username” metadata - msEncrypt-style string for a proxy connection. Empty strings are also
accepted.
METADATA
"wms_proxy_username" "foo"
"wms_proxy_password" "{FF88CFDAAE1A5E33}"
END

• “wms_sld_body” metadata - can be used to specify an inline SLD document.

• “wms_sld_url” metadata - can be used to specify a link to an SLD document.
• “wms_style” metadata - name of style to use for the STYLES parameter in GetMap requests for this layer.
• “wms_style_<stylename>_sld” metadata URL of a SLD to use in GetMap requests. Replace <stylename> in
the metadta name with the name of the style to which the SLD applies.
METADATA
...
"wms_style" "mystyle"
"wms_style_mystyle_sld" "http://my.host.com/mysld.xml"
...
END

For more information on SLDs in MapServer see the SLD HowTo document.
• “wms_time” metadata - value to use for the TIME parameter in GetMap requests for this layer. Please see the
WMS Time HowTo for more information.
• “wms_bgcolor” metadata - specifies the color to be used as the background of the map. The general format of
BGCOLOR is a hexadecimal encoding of an RGB value where two hexadecimal characters are used for each of
Red, Green, and Blue color values. The values can range between 00 and FF for each (0 and 255, base 10). The
format is 0xRRGGBB; either upper or lower case characters are allowed for RR, GG, and BB values. The “0x”
prefix shall have a lower case “x”.
• “wms_transparent” metadata - specifies whether the map background is to be made transparent or not.
TRANSPARENT can take on two values, “TRUE” or “FALSE”. If not specified, MapServer sets default to
“TRUE”
• “wms_cache_to_disk” metadata - set this to “1” to force MapServer to write fetched images to disk. Writing
to disk is necessary to take advantage of MapServer’s caching logic to avoid refetching WMS requests made
previously. This feature is new to MapServer 6.0 - previously results were always written to disk.

9.2. WMS Client 421

MapServer Documentation, Release 6.0

• “wms_nonsquare_ok” metadata - set this to “0” to indicate that the remote WMS only supports requests
for square pixels. In this case MapServer will be careful to only make square pixel requests even if it means
oversampling in one dimension compared to the resolution of image data required. This feature is new to
MapServer 6.0.
• “wms_extent” metadata - If there is exactly one SRS supported by this layer (as listed in the wms_srs meta-
data), and if the wms_extent metadata item (or an extent specified via the EXTENT keyword) is set then
MapServer will take care to only making requests within this area. This can short circuit requests completely
outside the layer, reduce processing for layers that only partially overlap the target map area and avoid poor
behaviors with reprojection in some areas. The contents of this metadata item should be of the form “minx miny
maxx maxy”. This feature is new to MapServer 6.0.
Note: Note that each of the above metadata can also be referred to as ‘ows_*’ instead of ‘wms_*’. MapServer tries
the ‘wms_*’ metadata first, and if not found it tries the corresponding ‘ows_*’ name. Using this reduces the amount of
duplication in mapfiles that support multiple OGC interfaces since “ows_*” metadata can be used almost everywhere
for common metadata items shared by multiple OGC interfaces.

Old CONNECTION parameter format from version 3.5 and 3.6 (deprecated)

In MapServer version 3.5 and 3.6, the CONNECTION parameter had to include at a minimum the VERSION, LAY-
ERS, FORMAT and TRANSPARENT WMS parameters. This mode of operation is still supported but is deprecated
and you are encouraged to use metadata items for those parameters as documented in the previous section above.
Here is an example of a layer definition using this deprecated CONNECTION parameter format:
LAYER
NAME "bathymetry_elevation"
TYPE RASTER
STATUS ON
CONNECTIONTYPE WMS
CONNECTION "http://demo.mapserver.org/cgi-bin/wms?VERSION=1.1.0&LAYERS=bluemarble&FORMAT=image/png"
PROJECTION
"init=epsg:4326"
END
END

9.2.4 Limitations/TODO

1. GetFeatureInfo is not fully supported since the output of getFeatureInfo is left to the discretion of the remote
server. A method layer.getWMSFeatureInfoURL() has been added to MapScript for applications that want to
access featureInfo results and handle them directly.
2. MapServer does not attempt to fetch the layer’s capabilities. Doing so at every map draw would be extremely
inefficient. And caching that information does not belong in the core of MapServer. This is better done at
the application level, in a script, and only the necessary information is passed to the MapServer core via the
CONNECTION string and metadata.

9.3 WMS Time

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Revision $Revision: 9071 $

422 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Date $Date: 2009-06-02 16:17:26 -0700 (Tue, 02 Jun 2009) $

Last Updated 2006/06/26

Table of Contents
• WMS Time
– Introduction
– Enabling Time Support in MapServer
– Future Additions
– Limitations and Known Bugs

9.3.1 Introduction

A WMS server can provide support to temporal requests. This is done by providing a TIME parameter with a time
value in the request. MapServer 4.4 and above provides support to interpret the TIME parameter and transform the
resulting values into appropriate requests.

Links to WMS-Related Information

• MapServer WMS Server HowTo

• MapServer WMS Client HowTo
• WMS 1.1.1 specification
• MapServer OGC Web Services Workshop

9.3.2 Enabling Time Support in MapServer

Time Patterns

WMS specifies that the basic format used for TIME requests is based on the ISO 8601:1988(E) “extended” format.
MapServer supports a limited set of patterns that are defined in the ISO 8601 specifications, as well as few other
patterns that are useful but not compliant to ISO 8601. Here is a list of patterns currently supported:
Table 1. Supported Time Patterns
Time Patterns Examples
YYYYMMDD 20041012
YYYY-MM-DDTHH:MM:SSZ 2004-10-12T13:55:20Z
YYYY-MM-DDTHH:MM:SS 2004-10-12T13:55:20
YYYY-MM-DD HH:MM:SS 2004-10-12 13:55:20
YYYY-MM-DDTHH:MM 2004-10-12T13:55
YYYY-MM-DD HH:MM 2004-10-12 13:55
YYYY-MM-DDTHH 2004-10-12T13
YYYY-MM-DD HH 2004-10-12 13
YYYY-MM-DD 2004-10-12
YYYY-MM 2004-10
YYYY 2004
THH:MM:SSZ T13:55:20Z
THH:MM:SS T13:55:20

9.3. WMS Time 423

MapServer Documentation, Release 6.0

Setting Up a WMS Layer with Time Support

To have a valid WMS layer with time support, the user has to define the following metadata at the layer level:
• wms_timeextent: (Mandatory) this is used in the capabilities document to return the valid time values for the
layer. The value defined here should be a valid time range. (more on this in ‘Specifying Time Extents’ below)
• wms_ timeitem: (Mandatory) this is the name of the field in the DB that contains the time values.
• wms_timedefault: (Optional) this value is used if it is defined and the TIME value is missing in the request.

Specifying Time Extents

Time Extents can be declared with the following syntax for the wms_timeextent metadata (see Annex C.3 in the WMS
1.1.1 specification document for a full description):
1. value - a single value. This is not directly supported in MapServer but there is an easy workwound by specifying
the same value as min and max.
2. value1,value2,value3,... - a list of multiple values.
3. min/max/resolution - an interval defined by its lower and upper bounds and its resolution. This is supported in
MapServer (note that the resolution is not supported however).
4. min1/max1/res1,min2/max2/res2,... - a list of multiple intervals.

Example WMS-Server Layer

LAYER
NAME "earthquakes"
METADATA
"wms_title" "Earthquakes"
"wms_timeextent" "2004-01-01/2004-02-01"
"wms_timeitem" "TIME"
"wms_timedefault" "2004-01-01 14:10:00"
END
TYPE POINT
STATUS ON
DATA "quakes"
CLASS
..
END
END

GetCapabilities Output

If your layer is set up properly, requesting the capabilities on the server outputs a Dimension element. Here is an
example of a GetCapabilities result for a layer configured for time support:
<Layer queryable="0" opaque="0" cascaded="0">
<Name>earthquakes</Name>
<Title>Earthquakes</Title>
<SRS>EPSG:4326</SRS>
<LatLonBoundingBox minx="-131.02" miny="24.84" maxx="-66.59" maxy="48.39" />
<BoundingBox SRS="EPSG:4326"
minx="-131.02" miny="24.84" maxx="-66.59" maxy="48.39" />

424 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

<Dimension name="time" units="ISO8601"/>

<Extent name="time" default="2004-01-01 14:10:00" nearestValue="0">2004-01-01/2004-02-01</Extent>
</Layer>

Supported Time Requests

When sending a request with the TIME parameter, different types of time values can be specified. The following are
supported by MapServer:
• single value: for example: ...&TIME=2004-10-12&...
• multiple values: for example: ...&TIME=2004-10-12, 2004-10-13, 2004-10-14&...
• single range value: for example: ...&TIME=2004-10-12/2004-10-13&...
• multiple range values: for example: ...&TIME=2004-10-12/2004-10-13, 2004-10-15/2004-10-16&...

Interpreting Time Values

When MapServer receives a request with a TIME parameter, it transforms the time requests into valid expressions that
are assigned to the filter parameter on layers that are time-aware. Here are some examples of how different types of
requests are treated (wms_timeitem is defined here as being “time_field”):
• single value (2004-10-12) transforms to (‘[time_field]‘ eq ‘2004-10-12‘)
• multiple values (2004-10-12, 2004-10-13) transform to (‘[time_field]‘ eq ‘2004-10-12‘ OR ‘[time_field]‘ eq
‘2004-10-13‘)
• single range : 2004-10-12/2004-10-13 transforms to ((‘[time_field]‘ ge ‘2004-10-12‘) AND (‘[time_field]‘ le
‘2004-10-13‘))
• multiple ranges (2004-10-12/2004-10-13, 2004-10-15/2004-10-16) transform to ((‘[time_field]‘ ge ‘2004-10-
12‘ AND ‘[time_field]‘ le ‘2004-10-13‘) OR (‘[time_field]‘ ge ‘2004-10-15‘ AND ‘[time_field]‘ le ‘2004-10-
16‘))
As shown in the above examples, all fields and values are written inside back tics (‘) - this is the general way of
specifying time expressions inside MapServer.
Exceptions to this rule:
1. When dealing with layers that are not Shapefiles nor through OGR, the expression built has slightly different
syntax. For example, the expression set in the filter for the first example above would be ([time_field] = ‘2004-
10-12’).
2. For PostGIS/PostgreSQL layers, the time expression built uses the date_trunc function available in PostgreSQL.
For example, if the user passes a time value of ‘2004-10-12’, the expression set in the filter is date_trunc(‘day’,
time_field) = ‘2004-10-12’. The use of the date_trunc function allows requests to use the concept of time
resolution. In the example above, for a request of ‘2004-10-12’, MapServer determines that the resolution is
“day” by parsing the time string and the result gives all records matching the date 2004-10-12 regardless of the
values set for Hours/Minutes/Seconds in the database. For more information on the date_trunc function, please
refer to the PostgreSQL documentation.

Limiting the Time Formats to Use

The user has the ability to define the time format(s) to be used when a request is sent, in metadata at the WEB level.
For example, the user can define the following two formats:

9.3. WMS Time 425

 MapServer Documentation, Release 6.0

"wms_timeformat" "YYYY-MM-DDTHH, YYYY-MM-DDTHH:MM"

Another example is for a WMS layer that is based on time data that contains precise time values taken every minute
(e.g., 2004-10-12T13:55, 2004-10-12T13:56, 2004-10-12 T13:57, ...). Normally, a valid request on such a layer
would require the time value to be as complete as the data underneath. By defining a set of patterns to use, MapServer
introduces the notion of resolution to be used when doing a query. Using the example above, a request TIME= 2004-
10-12T13:55 would be valid and a request TIME= 2004-10-12T13 would also be valid and would return all elements
taken for that hour.
Note that this functionality is only available on layers based on Shapefiles and OGR.

Example of WMS-T with PostGIS Tile Index for Raster Imagery

This example currently requires latest 4.9 CVS build!

Here is an example mapfile snippet for a raster WMS-T instance using a PostGIS tileindex. This example shows
US Nexrad Base Reflectivity running at Iowa State U at http://mesonet.agron.iastate.edu/cgi-bin/wms/nexrad/n0r-
t.cgi?SERVICE=WMS&request=GetCapabilities
1 # Tile Index
2 LAYER
3 STATUS ON
4 NAME "time_idx"
5 TYPE POLYGON
6 DATA "the_geom from nexrad_n0r_tindex"
7 METADATA
8 "wms_title" "TIME INDEX"
9 "wms_srs" "EPSG:4326"
10 "wms_extent" "-126 24 -66 50"
11 "wms_timeextent" "2003-08-01/2006-12-31/PT5M"
12 "wms_timeitem" "datetime" #column in postgis table of type timestamp
13 "wms_timedefault" "2006-06-23T03:10:00Z"
14 END
15 CONNECTION "dbname=postgis host=10.10.10.20"
16 CONNECTIONTYPE postgis
17 END
18

19 # raster layer
20 LAYER
21 NAME "nexrad-n0r-wmst"
22 TYPE RASTER
23 STATUS ON
24 DEBUG ON
25 DUMP TRUE
26 PROJECTION
27 "init=epsg:4326"
28 END
29 METADATA
30 "wms_title" "NEXRAD BASE REF WMS-T"
31 "wms_srs" "EPSG:4326"
32 "wms_extent" "-126 24 -66 50"
33 "wms_timeextent" "2003-08-01/2006-12-31/PT5M"
34 "wms_timeitem" "datetime" #datetime is a column in postgis table of type timestamp
35 "wms_timedefault" "2006-06-23T03:10:00Z"
36 END
37 OFFSITE 0 0 0
38 TILEITEM "filepath" #filepath is a column in postgis table with varchar of the filepath to each imag

426 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

39 TILEINDEX "time_idx"
40 END

You can find more information on Time and tileindexes in the WCS documentation.

9.3.3 Future Additions

• Support for a special time value: “current”.

9.3.4 Limitations and Known Bugs

• Pattern “YYYYMMDD” does not work on Windows. (Bug#970)

9.4 Map Context

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-10-07

Contents
• Map Context
– Introduction
– Implementing a Web Map Context

9.4.1 Introduction

The term ‘map context’ comes from the Open Geospatial Constortium’s (OGC) Web Map Context Specification v1.0.0,
which coincides with the OGC Web Map Server Specification (WMS) v1.1.1. A map context is a XML document
that describes the appearance of layers from one or more WMS servers, and can be transferred between clients while
maintaining startup views, the state of the view (and its layers), and storing additional layer information.
Support for OGC Web Map Context was added to MapServer in version 3.7/4.0. This allows client applications to load
and save a map configuration in a standard XML format. MapServer can read context documents of versions 0.1.2,
0.1.4, 0.1.7, 1.0.0, 1.1.0 and can export contents in versions 0.1.4, 0.1.7, 1.0.0, 1.1.0. Web Map Context 1.1.0 support
was added to MapServer 4.10
This document assumes that you are already familiar with certain aspects of MapServer:
• MapServer application development and setting up mapfiles.
• Familiarity with the WMS spec would be an asset. Please see the following section for links to associated
sources.

Links to WMS / Map Context Related Information

• MapServer WMS Client HowTo

• Open Geospatial Consortium (OGC) home page

9.4. Map Context 427

MapServer Documentation, Release 6.0

• WMS 1.1.1 specification

• Map Context 1.0.0 specification
• MapServer OGC Web Services Workshop

9.4.2 Implementing a Web Map Context

Special Build Considerations

Map Context support requires PROJ4, GDAL/OGR and PHP support libraries.
Build/install the above libraries on your system and then build MapServer with the ‘–with-wmsclient –with-proj –with-
ogr –with-gdal –with-php’ configure options. Also make sure that your build uses the USE_WMS_LYR and USE_OGR
flags. For more details on MapServer compilation see the appropriate HowTo: Unix / Windows
Windows users can use MS4W, which is ready for Map Context use.

Map Context Mapfile

A map context document can ONLY contain WMS layers (e.g. CONNECTIONTYPE WMS). Please refer to the
MapServer WMS Client HowTo for more information on declaring WMS layers.

MapFile Metadata

The following mapfile metadata are used by MapServer to handle map context information:
(Note that some parameters have width, height, format, and href, and some only have format and href. This is because
width and height are only used for images and parameters that do not have them are text or html. For consistency with
the spec MapServer supports height and width for all parameters, but they should only be used for images)

Web Object Metadata

• ows_schemas_location : Location of XML schema document. Default is http://schemas.opengis.net. See

http://ogc.dmsolutions.ca for an example of a valid schema tree.
• wms_abstract : A blurb of text providing more information about the WMS server.
• wms_address : If provided must also then provide wms_addresstype, wms_city, wms_stateorprovince,
wms_postcode, and wms_country)
• wms_addresstype : If provided must also then provide wms_address, wms_city, wms_stateorprovince,
wms_postcode, and wms_country)
• wms_city : If provided must also then provide wms_address, wms_addresstype, wms_stateorprovince,
wms_postcode, and wms_country)
• wms_contactelectronicmailaddress : contact Email address.
• wms_contactfacsimiletelephone : contact facsimile telephone number.
• wms_contactorganization :
• wms_contactperson :
• wms_contactposition :

428 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• wms_contactvoicetelephone : contact voice telephone number.

• wms_context_fid : the feature id of the context. Set to 0 when saving if not specified.
• wms_context_version : the version of the map context specification.
• wms_country : If provided must also then provide wms_address, wms_city, wms_stateorprovince,
wms_postcode, and wms_addresstype.
• wms_descriptionurl_format : Format of the webpage which contains relevant information to the view.
• wms_descriptionurl_href : Reference to a webpage which contains relevant information to the view.
• wms_keywordlist : A comma-separated list of keywords or keyword phrases to help catalog searching.
• wms_logourl_width : Width of the context logo.
• wms_logourl_height : Height of the context logo.
• wms_logourl_format : Format of the context logo.
• wms_logourl_href : Location of the context logo.
• wms_postcode : If provided must also then provide wms_address, wms_city, wms_stateorprovince,
wms_addresstype, and wms_country.
• wms_stateorprovince : If provided must also then provide wms_address, wms_city, wms_addresstype,
wms_postcode, and wms_country.
• wms_title : (Required) A human-readable name for this Layer (this metadata does not exist beyond version
0.1.4)

Layer Object Metadata

• wms_abstract : A blurb of text providing more information about the WMS server.
• wms_dataurl_href : Link to an online resource where data corresponding to the layer can be found.
• wms_dataurl_format : Format of the online resource where data corresponding to the layer can be found.
• wms_dimension : Current dimension used. New in version 4.10.
• wms_dimensionlist : List of available dimensions. New in version 4.10.
• wms_dimension_%s_default : Default dimension value. MapServer will check for wms_time and
wms_timedefault metadata when this is not specified. %s = the name of the dimension. New in version 4.10.
• wms_dimension_%s_multiplevalues : Multiple dimension values. %s = the name of the dimension. New in
version 4.10.
• wms_dimension_%s_nearestvalue : Nearest dimension value. The default value is 0. %s = the name of the
dimension. New in version 4.10.
• wms_dimension_%s_units : Units for the dimension values. The default value is ISO8601. %s = the name of
the dimension. New in version 4.10.
• wms_dimension_%s_unitsymbol : Symbol for dimension units. The default value is t. %s = the name of the
dimension. New in version 4.10.
• wms_dimension_%s_uservalue : User dimension value. MapServer will check for wms_time and
wms_timedefault metadata when this is not specified. %s = the name of the dimension. New in version 4.10.
• wms_format : Current format used.
• wms_formatlist : List of available formats for this layer.

9.4. Map Context 429

 MapServer Documentation, Release 6.0

• wms_metadataurl_href : Link to an online resource where descriptive metadata of the corresponding layer can
be found.
• wms_metadataurl_format : Format of the online resource where descriptive metadata of the corresponding layer
can be found.
• wms_name : Name of the WMS layer on the server.
• wms_onlineresource : Required URL to access the server.
• wms_server_version : The version of the web map server specification.
• wms_server_title : The title of the web map server.
• wms_stylelist : Current style used.
• wms_style_%s_legendurl_width : Width of an image describing the style. %s = the name of the style.
• wms_style_%s_legendurl_height : Height of an image describing the style. %s = the name of the style.
• wms_style_%s_legendurl_format : Format of an image describing the style. %s = the name of the style.
• wms_style_%s_legendurl_href : Location of an image describing the style. %s = the name of the style.
• wms_style_%s_sld : URL to the SLD document of this style. %s = the name of the style.
• wms_style_%s_sld_body : SLD_BODY document of this style. %s = the name of the style.
• wms_style_%s_title : Title of the layer. %s = the name of the style.
• wms_title : (Required) A human-readable name for this Layer.

Sample Map Context Mapfile

1 MAP
2

3 NAME "mapcontext"
4 STATUS ON
5 SIZE 400 300
6 SYMBOLSET "../etc/symbols.txt"
7 EXTENT -180 -90 180 90
8 UNITS DD
9 SHAPEPATH "../data"
10 IMAGECOLOR 255 255 255
11 FONTSET "../etc/fonts.txt"
12

13

14 #
15 # Start of web interface definition
16 #
17 WEB
18 IMAGEPATH "/ms4w/tmp/ms_tmp/"
19 IMAGEURL "/ms_tmp/"
20 METADATA
21 "wms_abstract" "Demo for map context document. Blah blah..."
22 "wms_title" "Map Context demo" #### REQUIRED
23 END
24 END
25

26 PROJECTION
27 "init=epsg:4326"
28 END

430 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

29

30 #
31 # Start of layer definitions
32 #
33

34 LAYER
35 NAME "country_bounds"
36 TYPE RASTER
37 STATUS ON
38 CONNECTION "http://demo.mapserver.org/cgi-bin/wms?"
39 CONNECTIONTYPE WMS
40 METADATA
41 "wms_title" "World Country Boundaries" #### REQUIRED
42 "wms_onlineresource" "http://demo.mapserver.org/cgi-bin/wms?" #### REQUIRED
43 "wms_srs" "EPSG:4326"
44 "wms_name" "country_bounds"
45 "wms_server_version" "1.1.1"
46 "wms_format" "image/gif"
47 "wms_dimensionlist" "time,width"
48 "wms_dimension" "time"
49 "wms_dimension_time_unitsymbol" "t"
50 "wms_dimension_time_units" "ISO8601"
51 "wms_dimension_time_uservalue" "1310"
52 "wms_dimension_time_default" "1310"
53 "wms_dimension_time_multiplevalues" "1310,1410"
54 "wms_dimension_time_nearestvalue" "0"
55 END
56 END
57

58 END # Map File

Testing Map Context Support

1. The first thing to do is to save your mapfile using the saveMapContext function available from the
PHP/MapScript library. An example script is shown below:
<?php
if (!extension_loaded("MapScript")) dl(MODULE);
$oMap = ms_newMapObj("mapcontext.map");
$oMap->saveMapContext("mapcontext_output.xml");
?>

2. Scan the XML output to look for <!– WARNING: ... –> comments. Then make the necessary changes to fix
every warning that you encounter. At the end of this you should have a mapfile compatible with the Map Context
specification.
3. Now you can load your new Map Context document into an application using the loadMapContext function
from the PHP/MapScript library.

Sample Map Context Document

The following is a sample Map Context document:

1 <?xml version=’1.0’ encoding="ISO-8859-1" standalone="no" ?>
2 <ViewContext version="1.1.0" id="mapcontext" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance
3 <General>

9.4. Map Context 431

 MapServer Documentation, Release 6.0

4 <Window width="400" height="300"/>

5 <!-- Bounding box corners and spatial reference system -->
6 <BoundingBox SRS="EPSG:4326" minx="-180.000000" miny="-90.000000" maxx="180.000000" maxy="90.
7 <!-- Title of Context -->
8 <Title>Map Context demo</Title>
9 <Abstract>Demo for map context document. Blah blah...</Abstract>
10 <ContactInformation>
11 </ContactInformation>
12 </General>
13 <LayerList>
14 <Layer queryable="0" hidden="0">
15 <Server service="OGC:WMS" version="1.1.1" title="World Country Boundaries">
16 <OnlineResource xlink:type="simple" xlink:href="http://demo.mapserver.org/cgi-bin/wms?"/>
17 </Server>
18 <Name>country_bounds</Name>
19 <Title>World Country Boundaries</Title>
20 <SRS>EPSG:4326</SRS>
21 <FormatList>
22 <Format current="1">image/gif</Format>
23 </FormatList>
24 <DimensionList>
25 <Dimension name="time" units="ISO8601" unitSymbol="t" userValue="1310" default="1310" mul
26 </DimensionList>
27 </Layer>
28 </LayerList>
29 </ViewContext>

Map Context Support Through CGI

MapServer CGI allows you to load a map context through the use of a CONTEXT parameter, and you can point
this parameter to a locally stored context file or a context file accessible through a URL. For more information on
MapServer CGI see the CGI Reference.

Support for Local Map Context Files

There is a new cgi parameter called CONTEXT that is used to specify a local context file. The user can then use
MapServer to request a map using the following syntax:
http://localhost/mapserver.cgi?MODE=map&MAP=/path/to/mapfile.map&CONTEXT=
/path/to/contextfile.xml&LAYERS=layer_name1 layers_name2

Note: All layers created from a context file have their status set to ON. To be able to display layers, the user needs to
add the LAYERS argument in the URL.

Support for Context Files Accessed Through a URL

The syntax of using a web accessible context file would be similar to accessing a local context file:
http://localhost/mapserver.cgi?MODE=map&MAP=/path/to/mapfile.map&CONTEXT=
http://URL/path/to/contextfile.xml&LAYERS=layers_name1 layer_name2

Due to security concerns loading a file from a URL is disabled by default. To enable this functionality, the user needs
to set a CONFIG parameter called CGI_CONTEXT_URL in the default mapfile that will allow this functionality. Here
is an example of a map file with the CONFIG parameter:

432 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

# Start of map file

NAME "map-context"
STATUS ON
SIZE 400 300
EXTENT -2200000 -712631 3072800 3840000
UNITS METERS
IMAGECOLOR 255 255 255
IMAGETYPE png
CONFIG "CGI_CONTEXT_URL" "1"
...
WEB
...
END
LAYER
...
END

END

Default Mapfile

To smoothly run a MapServer CGI application with a Map Context, the application administrator needs to provide a
default mapfile with at least the basic required parameters that will be used with the Context file. This default mapfile
can contain as little information as the imagepath and imageurl or contain a list of layers. Information coming from
the context (e.g.: layers, width, height, ...) would either be appended or will replace values found in the mapfile.
Here is an example of a default map file containing the minimum required parameters:
1 NAME "CGI-CONTEXT-DEMO"
2 STATUS ON
3 SIZE 400 300
4 EXTENT -2200000 -712631 3072800 3840000
5 UNITS METERS
6 IMAGECOLOR 255 255 255
7 IMAGETYPE png
8 #
9 # Start of web interface definition
10 #
11 WEB
12 MINSCALE 2000000
13 MAXSCALE 50000000
14 #
15 # On Windows systems, /tmp and /tmp/ms_tmp/ should be created at the root
16 # of the drive where the .MAP file resides.
17 #
18 IMAGEPATH "/ms4w/tmp/ms_tmp/"
19 IMAGEURL "/ms_tmp/"
20 END
21 END # Map File

Map Context Support Through WMS

MapServer can also output your WMS layers as a Context document. MapServer extends the WMS standard by adding
a request=GetContext operation that allows you to retrieve a context for a WMS-based mapfile with a call like:

9.4. Map Context 433

MapServer Documentation, Release 6.0

http://localhost/mapserver.cgi?map=/path/to/mapfile.map&service=WMS&
request=GetContext&version=1.1.0

The VERSION parameter controls the version of context document to return.

GetContext is disabled by default because it could be considered a security issue: it could publicly expose the URLs
of WMS layers used (cascaded) by a mapfile.
To enable it, set the “wms_getcontext_enabled” web metadata to “1” in your WMS server’s mapfile.

9.5 WFS Server

Author Jean-François Doyon

Contact jdoyon at nrcan.gc.ca
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-10-07

Contents
• WFS Server
– Introduction
– Configuring your MapFile to Serve WFS layers
– Reference Section
– To-do Items and Known Limitations

9.5.1 Introduction

A WFS ( Web Feature Service ) publishes feature-level geospatial data to the web. This means that instead of re-
turning an image, as MapServer has traditionally done, the client now obtains fine-grained information about specific
geospatial features of the underlying data, at both the geometry AND attribute levels. As with other OGC specifica-
tions, this interface uses XML over HTTP as it’s delivery mechanism, and, more precisely, GML (Geography Markup
Language), which is a subset of XML.

WFS-Related Information

Here are some WFS related links (including a newly added OGC services workshop with MapServer). Since these
are highly detailed technical specifications, there is no need to read through them in their entirety to get a MapServer
WFS up and running. It is still recommended however to read them over and get familiar with the basics of each of
them, in order to understand how it all works:
• The OGC Web Feature Service Implementation Specification.
• The Geography Markup Language Implementation Specification.
• MapServer OGC Web Services Workshop package.
Working knowledge of MapServer is of course also required.

434 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Software Requirements

In order to enable MapServer to serve WFS, it MUST be compiled against certain librairies:
• PROJ.4: The reprojection library. Version 4.4.3 or greater is required.
• GDAL/OGR: I/O support libraries. Version 1.1.8 or greater is required.
Please see the MapServer UNIX Compilation and Installation HowTo for detailed instructions on compiling mapserver
with support for these libraries and features. For Windows users, the MS4W installer comes ready to serve both WFS
and WMS.

Version of GML Supported

MapServer can output both GML2 and GML3. By default MapServer serves GML2. You can test this by adding an
‘OUTPUTFORMAT’ parameter to a GetFeature request, such as:
• GML2 request output
• GML3 request output
For a detailed discussion on the versions supported, see bug#884.

9.5.2 Configuring your MapFile to Serve WFS layers

Much as in the WMS support, WFS publishing is enabled by adding certain magic METADATA keyword/value pairs
to a MapFile.
MapServer will serve and include in its WFS capabilities only the layers that meet the following conditions:
• Data source is of vector type (Shapefile, OGR, PostGIS, SDE, SDO, ...)
• LAYER NAME must be set. Layer names must start with a letter when setting up a WFS server (layer names
should not start with a digit or have spaces in them).
• LAYER TYPE is one of: LINE, POINT, POLYGON
• LAYER DUMP parameter set to TRUE
• The “wfs_onlineresource” metadata:
The wfs_onlineresource metadata is set in the map’s web object metadata and specifies the URL that should be
used to access your server. This is required for the GetCapabilities output. If wfs_onlineresource is not provided
then MapServer will try to provide a default one using the script name and hostname, but you shouldn’t count
on that too much. It is strongly recommended that you provide the wfs_onlineresource metadata.
See section 12.3.3 of the WFS 1.0.0 specification for the whole story about the online resource URL. Basically,
what you need is a complete HTTP URL including the http:// prefix, hostname, script name, potentially a
“map=” parameter, and and terminated by “?” or “&”.
Here is a valid online resource URL:
http://my.host.com/cgi-bin/mapserv?map=mywfs.map&

By creating a wrapper script on the server it is possible to hide the “map=” parameter from the URL and then
your server’s online resource URL could be something like:
http://my.host.com/cgi-bin/mywfs?

This is covered in more detail in the “More About the Online Resource URL” section of the WMS Server
document.

9.5. WFS Server 435

MapServer Documentation, Release 6.0

Example WFS Server Mapfile

The following is an example of a bare minimum WFS Server mapfile. Note the comments for the required parameters.
MAP

NAME "WFS_server"
STATUS ON
SIZE 400 300
SYMBOLSET "../etc/symbols.txt"
EXTENT -180 -90 180 90
UNITS DD
SHAPEPATH "../data"
IMAGECOLOR 255 255 255
FONTSET "../etc/fonts.txt"

#
# Start of web interface definition
#
WEB
IMAGEPATH "/ms4w/tmp/ms_tmp/"
IMAGEURL "/ms_tmp/"
METADATA
"wfs_title" "WFS Demo Server for MapServer" ## REQUIRED
"wfs_onlineresource" "http://demo.mapserver.org/cgi-bin/wfs?" ## Recommended
"wfs_srs" "EPSG:4326 EPSG:4269 EPSG:3978 EPSG:3857" ## Recommended
"wfs_abstract" "This text describes my WFS service." ## Recommended
END
END

PROJECTION
"init=epsg:4326"
END

#
# Start of layer definitions
#

##################
# World Continents
##################
LAYER
NAME "continents"
METADATA
"wfs_title" "World continents" ##REQUIRED
"wfs_srs" "EPSG:4326" ## REQUIRED
"gml_include_items" "all" ## Optional (serves all attributes for layer)
"gml_featureid" "ID" ## REQUIRED
END
TYPE POLYGON
STATUS ON
DATA ’shapefile/countries_area’
PROJECTION
"init=epsg:4326"
END
DUMP TRUE ## REQUIRED
CLASS
NAME ’World Continents’

436 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

STYLE
COLOR 255 128 128
OUTLINECOLOR 96 96 96
END
END
END #layer

END #mapfile

Rules for Handling SRS in MapServer WFS

Contrary to WMS, the OGC WFS specification doesn’t allow a layer (feature type) to be advertised in more than one
SRS. Also, there is no default SRS that applies to all layers by default in the OGC WFS spec. However, it is possible
to have every layer in a WFS server advertised in a different SRS.
Here is how MapServer decides the SRS to advertise and use for each layer in your WFS:
• If a top-level map SRS is defined* then this SRS is used and applies to all layers (feature types) in this WFS. In
this case the SRS of individual layers is simply ignored even if it’s set.
• If there is no top-level map SRS defined* then each layer is advertised in its own SRS in the capabilities.
• By “SRS is defined”, we mean either the presence of a PROJECTION object defined using an EPSG code, or of
a “wfs_srs” metadata at this level.
Note that at the map top-level the “wfs_srs” metadata value takes precedence over the contents of the PROJECTION
block.
At the layer level, if both the wfs_srs metadata and the PROJECTION object are set to different values, then the
wfs_srs metadata defines the projection to use in advertising this layer (assuming there is no top-level map SRS), and
the PROJECTION value is assumed to be the projection of the data. So this means that the data would be reprojected
from the PROJECTION SRS to the one defined in the wfs_srs metadata before being served to WFS clients.
Confusing? As a rule of thumb, simply set the wfs_srs at the map level (in web metadata) and never set the wfs_srs
metadata at the layer level and things will work fine for most cases.

Axis Orientation in WFS 1.1

The axis order in previous versions of the WFS specifications was to always use easting (x or lon ) and northing (y or
lat). WMS 1.1 specifies that, depending on the particular SRS, the x axis may or may not be oriented West-to-East, and
the y axis may or may not be oriented South-to-North. The WFS portrayal operation shall account for axis order. This
affects some of the EPSG codes that were commonly used such as ESPG:4326. The current implementation makes
sure that coordinates returned to the server for the GetFeature request reflect the inverse axis orders for EPSG codes
between 4000 and 5000.

Test Your WFS Server

Validate the Capabilities Metadata

OK, now that we’ve got a mapfile, we have to check the XML capabilities returned by our server to make sure nothing
is missing.
Using a web browser, access your server’s online resource URL to which you add the parameter “RE-
QUEST=GetCapabilities” to the end, e.g.
http://demo.mapserver.org/cgi-bin/wfs?SERVICE=WFS&VERSION=1.0.0&REQUEST=GetCapabilities

9.5. WFS Server 437

MapServer Documentation, Release 6.0

If everything went well, you should have a complete XML capabilities document. Search it for the word “WARN-
ING”... MapServer inserts XML comments starting with “<!–WARNING: ” in the XML output if it detects missing
mapfile parameters or metadata items. If you notice any warning in your XML output then you have to fix all of them
before you can register your server with a WFS client, otherwise things are likely not going to work.
Note that the SERVICE parameter is required for all WFS requests. When a request happens, it is passed through
WMS, WFS, and WCS in MapServer (in that order) until one of the services respond to it.

Test With a GetFeature Request

OK, now that we know that our server can produce a valid XML GetCapa-
bilities response we should test the GetFeature request. Simply adding “SER-
VICE=WFS&VERSION=1.0.0&REQUEST=GetFeature&TYPENAME=yourlayername1,yourlayername2” to
your server’s URL should return the GML associated with those layers.
http://demo.mapserver.org/cgi-bin/wfs?SERVICE=WFS&VERSION=1.0.0&REQUEST=getfeature&TYPENAME=continents&MAXF

Test with a Real Client

If you have access to a WFS client, then register your new server’s online resource with it and you should be off and
running.
If you don’t have your own WFS client installed already, here are a few pointers:
• MapServer itself can be used as a WFS client, see the WFS Client HowTo.
• Quantum GIS is a full GIS package which includes WFS client support. (recommended)
• Deegree provides a WFS client.
• uDig can add layers from WMS/WFS servers.
• The owsview Viewer Client Generator is an online application that allows users to validate WFS Capabilities
XML (it does not allow you to view WFS data).

Support for GET and POST Requests

Starting from version 4.2 MapServer supports XML-encoded POST requests and GET requests. The default in
MapServer is POST.

Support for Filter Encoding

Starting from version 4.2 MapServer supports Filter Encoding (FE) in WFS GetFeature requests. For more information
on the server side of Filter Encoding see the Filter Encoding HowTo.

MapServer WFS Extensions

STARTINDEX In addition to the MAXFEATURES=n keyword, MapServer also supports a STARTINDEX=n key-
word in WFS GetFeature requests. This can be used to skip some features in the result set and in combination
with MAXFEATURES provides for the ability to use WFS GetFeature to page through results. Note that
STARTINDEX=1 means start with the first feature, skipping none.

438 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

OUTPUTFORMAT Normally OUTPUTFORMAT should be GML2 for WFS 1.0 and either “text/xml; sub-
type=gml/2.1.2” or “text/xml; subtype=gml/3.1.1” for WFS 1.1. However as an extension to the specification, it
is also possible to configure MapServer for a variety of other feature output formats. This is discussed in some
detail in the OGR Output document.

9.5.3 Reference Section

The following metadata are available in the setup of the WFS Server mapfile:
Note
Each of the metadata below can also be referred to as ‘ows_*’ instead of ‘wfs_*’. MapServer tries the ‘wfs_*’
metadata first, and if not found it tries the corresponding ‘ows_*’ name. Using this reduces the amount of duplication
in mapfiles that support multiple OGC interfaces since “ows_*” metadata can be used almost everywhere for common
metadata items shared by multiple OGC interfaces.

Web Object Metadata

ows_updatesequence
• Description: (Optional) The updateSequence parameter can be used for maintaining the consistency of a client
cache of the contents of a service metadata document. The parameter value can be an integer, a timestamp in
[ISO 8601:2000] format, or any other number or string.
ows_schemas_location
• Description: (Optional) (Note the name ows_schemas_location and not wfs/_... this is because all OGC Web
Services (OWS) use the same metadata) Root of the web tree where the family of OGC WFS XMLSchema files
are located. This must be a valid URL where the actual .xsd files are located if you want your WFS output to
validate in a validating XML parser. Default is http://schemas.opengis.net. See http://ogc.dmsolutions.ca for an
example of a valid schema tree.
wfs_abstract
• WFS TAG Name: Abstract (WFS 1.0.0, sect. 12.3.3)
• Description: (Optional) Descriptive narrative for more information about the server.
wfs_accessconstraints
• WFS TAG Name: Accessconstraints (WFS 1.0.0, sect. 12.3.3)
• Description: (Optional) Text describing any access constraints imposed by the service provider on the WFS or
data retrieved from this service.
wfs_encoding
• Description: (Optional) XML encoding for all XML documents returned by the server. The default is ISO-
8859-1.
wfs_fees
• WFS TAG Name: Fees (WFS 1.0.0, sect. 12.3.3)
• Description: (Optional) Any fees imposed by the service provider for usage of this service or for data retrieved
from the WFS.
wfs_getcapabilities_version
• Description: (Optional) Default version to use for GetCapabilities requests that do not have a version parameter.
If not set, the latest supported version will be returned.

9.5. WFS Server 439

MapServer Documentation, Release 6.0

wfs_keywordlist
• WFS TAG Name: Keyword (WFS 1.0.0, sect. 12.3.3)
• Description: (Optional) List of words to aid catalog searching.
wfs_maxfeatures
• Description: (Optional) The number of elements to be returned by the WFS server. This has priority over the
‘maxfeatures’ parameter passed by the user. If the not set the current behaviour is not changed.
wfs_namespace_prefix
• Description: (Optional) User defined namespace prefix to be used in the response of a WFS GetFeature request.
e.g. “wfs_namespace_prefix” “someprefix”.
wfs_namespace_uri
• Description: (Optional) User defined namespace URI to be used in the response of a WFS GetFeature request.
e.g. “wfs_namespace_uri” “http://somehost/someurl“.
wfs_onlineresource
• WFS TAG Name: Onlineresource (WFS 1.0.0, sect. 12.3.3)
• Description: (Recommended) The URL prefix for HTTP GET requests.
wfs_service_onlineresource
• Description: (Optional) Top-level onlineresource URL. MapServer uses the onlineresource metadata (if pro-
vided) in the following order:
1. wfs_service_onlineresource
2. ows_service_onlineresource
3. wfs_onlineresource (or automatically generated URL, see the onlineresource section of this document)
wfs_title
• WFS TAG Name: Title (WFS 1.0.0, sect. 12.3.3)
• Description: (Required) Human readable title to identify server.
wfs_srs
• Description: (Recommended) The SRS to use for all layers in this server. (e.g. EPSG:4326) See the notes below
about the SRS rules in the WFS.
wfs_feature_collection
• Description: Replaces the default name of the feature-containing element (<msFeatureCollection>) with a user-
defined value.

Layer Object

gml_exclude_items
• Description: (Optional) A comma delimited list of items to exclude. As of MapServer 4.6, you can control how
many attributes (fields) you expose for your data layer with metadata. The previous behaviour was simply to
expose all attributes all of the time. The default is to expose no attributes at all. An example excluding a specific
field would be:
"gml_include_items" "all"
"gml_exclude_items" "Phonenum"

440 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

gml_featureid
• Description: (Required for MapServer 4.10) Field to be used for the ID of the feature in the output GML.
wfs_featureid or ows_feature_id can be specified instead.
gml_groups
• Description: (Optional) A comma delimited list of group names for the layer.
gml_[group name]_group
• Description: (Optional) A comma delimited list of attributes in the group. Here is an example:
"gml_include_items" "all"
"gml_groups" "display"
"gml_display_group" "Name_e,Name_f"

gml_include_items
• Description: (Optional) A comma delimited list of items to include, or keyword “all”. As of MapServer 4.6,
you can control how many attributes (fields) you expose for your data layer with this metadata. The previous
behaviour was simply to expose all attributes all of the time. You can enable full exposure by using the keyword
“all”, such as:
"gml_include_items" "all"

You can specify a list of attributes (fields) for partial exposure, such as:
"gml_include_items" "Name,ID"

The new default behaviour is to expose no attributes at all.

gml_[item name]_alias
• Description: (Optional) An alias for an attribute’s name. The served GML will refer to this attribute by the alias.
Here is an example:
"gml_province_alias" "prov"

gml_[item name]_type
• Description: (Optional) Specifies the type of the attribute. Valid values are Integer|Real|Character|Date|Boolean.
gml_[item name]_width (Optional) Specifies the width of the indicated field for formats where this is significant,
such as Shapefiles.
gml_[item name]_precision (Optional) Specifies the precision of the indicated field for formats where this is sig-
nificant, such as Shapefiles. Precision is the number of decimal places, and is only needed for “Real” fields.
Currently this is only used for OGR based output formats, not the WFS GML2/GML3 output.
gml_types (Optional) If this field is “auto” then some input feature drivers (ie. OGR, POSTGIS, ORACLESPATIAL
and native shapefiles) will automatically populate the type, width and precision metadata for the layer based on
the source file. Currently this is only used for OGR based output formats, not the WFS GML2/GML3 output.
"gml_types" "auto"

gml_xml_items
• Description: (Optional) A comma delimited list of items that should not be XML-encoded.
gml_geometries
• Description: provides a name other than the default “msGeometry” for geometry elements. The value is speci-
fied as a string to be used for geometry element names.

9.5. WFS Server 441

MapServer Documentation, Release 6.0

gml_[name]_type
• Description: When employing gml_geometries, it is also necessary to specify the geometry type of the layer.
This is accomplished by providing a value for gml_[name]_type, where [name] is the string value specified for
gml_geometries, and a value which is one of:
– point
– multipoint
– line
– multiline
– polygon
– multipolygon
gml_[name]_occurances
• Description: MapServer applies default values of 0 and 1, respectively, to the “minOccurs” and “maxOccurs”
attributes of geometry elements, as can be seen in the preceding examples. To override these defaults, a value is
assigned to a gml_[name]_occurances layer metadata item, where again [name] is the string value specified for
gml_geometries, and the value is a comma-delimited pair containing the respective lower and upper bounds.
wfs_abstract
• Same as wfs_abstract in the Web Object.
wfs_extent
• Description: (Optional) Used for the layer’s BoundingBox tag for cases where it is impossible (or very ineffi-
cient) for MapServer to probe the data source to figure its extents. The value for this metadata is “minx miny
maxx maxy” separated by spaces, with the values in the layer’s projection units. If wfs_extent is provided then
it has priority and MapServer will NOT try to read the source file’s extents.
wfs_featureid
• Description: (Required for MapServer 4.10) Field to be used for the ID of the feature in the output GML.
gml_featureid or ows_feature_id can be specified instead.
wfs_getfeature_formatlist
• Description: (Optional) Comma-separted list of formats that sould be valid for a GetFeature request. If defined,
only these formats are advertised through in the Capabilities document.
wfs_keywordlist
• Same as wfs_keywordlist in the Web Object.
wfs_metadataurl_format
• Description: (Optional) The file format of the metadata record. Valid values are “XML”, “SGML”, or “HTML”.
The layer metadata wfs_metadataurl_type and wfs_metadataurl_href must also be specified.
• refer to section 12.3.5 of the WFS 1.0.0 spec.
wfs_metadataurl_href
• Description: (Optional) The URL to the layer’s metadata. The layer metadata wfs_metadataurl_type and
wfs_metadataurl_format must also be specified.
• refer to section 12.3.5 of the WFS 1.0.0 spec.
wfs_metadataurl_type

442 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• Description: (Optional) The standard to which the metadata complies. Currently only two types are valid:
“TC211” which refers to [ISO 19115], and “FGDC” which refers to [FGDC CSDGM]. The layer metadata
wfs_metadataurl_format and wfs_metadataurl_href must also be specified.
• refer to section 12.3.5 of the WFS 1.0.0 spec.
wfs_srs
• Description: If there is no SRS defined at the top-level in the mapfile then this SRS will be used to advertize
this feature type (layer) in the capabilities. See the note below about the SRS rules in the WFS.
wfs_title
• Same as wfs_title in the Web Object.

9.5.4 To-do Items and Known Limitations

• This is just a basic WFS (read-only): transaction requests are not supported and probably never will given the
nature of MapServer. GeoServer or TinyOWS is recommended for those needing WFS-T support.
• WFS spec. seems to require that features of a given feature type must all be of the same geometry type (point,
line, polygon). This works fine for shapefiles, but some data source formats supported by MapServer allow
mixed geometry types in a single layer and this goes against the WFS spec. Suggestions on how to handle this
are welcome (send suggestions to the mapserver-dev mailing list.

9.6 WFS Client

Author Jean-François Doyon

Contact jdoyon at nrcan.gc.ca
Author Jeff McKenna
Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-11-07

Contents
• WFS Client
– Introduction
– Setting up a WFS-client Mapfile
– TODO / Known Limitations

9.6.1 Introduction

MapServer can retrieve and display data from a WFS server. The following document explains how to display data
from a WFS server using MapServer.
A WFS ( Web Feature Service ) publishes feature-level geospatial data to the web. This means that it is possible to
use this data as a data source to render a map. In effect, this is not unlike having a shapefile accessible over the web,
only it’s not a shapefile, it’s XML-Encoded geospatial data (GML to be exact), including both geometry AND attribute
information.

9.6. WFS Client 443

MapServer Documentation, Release 6.0

WFS-Related Information

Although in-depth understanding of WFS and GML is neither necessary nor required in order to implement a
MapServer application that reads remote WFS data, it is recommended to at least get aquainted with the concepts and
basic functionality of both. Here are the official references (including a newly added OGC workshop with MapServer):
• OGC Web Feature Service Implementation Specification.
• Geography Markup Language Implementation Specification.
• MapServer OGC Web Services Workshop package.

Software Requirements

In order to enable MapServer to serve WFS, it MUST be compiled against certain libraries:
• PROJ.4: The reprojection library. Version 4.4.3 or greater is required.
• GDAL/OGR: I/O support librairies. Version 1.1.8 or greater is required.
• LibCURL: Used to help MapServer act as an HTTP client. Version 7.10 or greater is required.
Please see the MapServer UNIX Compilation and Installation HOWTO for detailed instructions on compiling
mapserver with support for these librairies and features. For Windows users, look on the MapServer website to
see if there are any binaries available that meet these requirements.

9.6.2 Setting up a WFS-client Mapfile

Storing Temporary Files

You must set the IMAGEPATH parameter in your mapfile since MapServer uses this directory to store temporary
files downloaded from the remote WFS server. Windows users must specify a full path for IMAGEPATH, such as:
IMAGEPATH “C:/tmp/ms_tmp/”
MAP
...
WEB
IMAGEPATH "/tmp/ms_tmp/"
IMAGEURL ...
END
...
END

WFS Layer

A WFS layer is a regular mapfile layer, which can use CLASS objects, with expressions, etc.
As of MapServer 4.4, the suggested method to define a WFS Client layer is through the CONNECTION parameter
and the layer’s METADATA. The necessary mapfile parameters are defined below:
• CONNECTIONTYPE: must be “wfs”
• CONNECTION: The URL to the WFS Server. e.g. http://demo.mapserver.org/cgi-bin/wfs? The
path to the mapfile on the WFS server is required if it was required in the GetCapabilities re-
quest e.g. you would have to specify the MAP parameter in the CONNECTION for the following
server: http://map.ns.ec.gc.ca/MapServer/mapserv.exe?MAP=/mapserver/services/envdat/config.map &SER-
VICE=WFS&VERSION=1.0.0&REQUEST=GetCapabilities

444 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• METADATA: The LAYER’s must contain a METADATA object with the following parameters:
– wfs_connectiontimeout (optional): The maximum time to wait for a remote WFS layer to load, set in
seconds (default is 30 seconds). This metadata can be added at the layer level so that it affects only that
layer, or it can be added at the map level (in the web object) so that it affects all of the layers. Note that
wfs_connectiontimeout at the layer level has priority over the map level.
– wfs_filter: This can be included to include a filter encoding parameter in the getFeature request (see the
Filter Encoding Howto for more information on filtering). The content of the wfs_filter is a valid filter
encoding element.
...
METADATA
"wfs_filter" "<PropertyIsGreaterThan><PropertyName>POP_RANGE</PropertyName><Literal>
</ L i t e r a l ></ P r o p e r t y I s G r e a t e r T h a n >"
END
...

– wfs_latlongboundingbox (optional): The bounding box of this layer in geographic coordinates in the for-
mat “lon_min lat_min lon_max lat_max”. If it is set then MapServer will request the layer only when the
map view overlaps that bounding box. You normally get this from the server’s capabilities output.
– wfs_maxfeatures (optional): Limit the number of GML features to return.
– wfs_request_method (optional): Can be set to “GET” to do a Get request to WFS servers that do not
support Post requests. The default method in MapServer is Post.
...
METADATA
"wfs_filter" "GET"
END
...

– wfs_typename (required): the <Name> of the layer found in the GetCapabil-

ities. An example GetCapabilities request is: http://demo.mapserver.org/cgi-
bin/wfs?SERVICE=WFS&VERSION=1.0.0&REQUEST=GetCapabilities
– wfs_version (required): WFS version, currently “1.0.0”
Note: Each of the above metadata can also be referred to as ‘ows_*’ instead of ‘wfs_*’. MapServer tries the ‘wfs_*’
metadata first, and if not found it tries the corresponding ‘ows_*’ name. Using this reduces the amount of duplication
in mapfiles that support multiple OGC interfaces since “ows_*” metadata can be used almost everywhere for common
metadata items shared by multiple OGC interfaces.

Example WFS Layer

LAYER
NAME "continents"
TYPE POLYGON
STATUS ON
CONNECTION "http://demo.mapserver.org/cgi-bin/wfs?"
CONNECTIONTYPE WFS
METADATA
"wfs_typename" "continents"
"wfs_version" "1.0.0"
"wfs_connectiontimeout" "60"
"wfs_maxfeatures" "10"
END
PROJECTION
"init=epsg:4326"

9.6. WFS Client 445

MapServer Documentation, Release 6.0

END
CLASS
NAME "Continents"
STYLE
COLOR 255 128 128
OUTLINECOLOR 96 96 96
END
END
END # Layer

Connection - deprecated

As of MapServer v4.4 the method of specifying all of the connection information in the CONNECTION parameter has
beendeprecated. The preferred method is mentioned above. If the metadata is not provided, VERSION, SERVICE,
and TYPENAME will be fetched from the CONNECTION, as shown below
CONNECTION "http://demo.mapserver.org/cgi-bin/wfs?SERVICE=WFS&VERSION=1.0.0&TYPENAME=continents"

9.6.3 TODO / Known Limitations

1. Temporary WFS (gml) files are written to the IMAGEPATH directory, but this could become a security concern
since it makes the raw GML data downloadable by someone who could guess the gml filename. We should
consider having a “wfs_cache_dir” metadata that, if it is set will define a directory where temp files should be
written. The default would still be to use the value of IMAGEPATH if “wfs_tmpdir” is not set.

9.7 WFS Filter Encoding

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Author Yewondwossen Assefa
Contact assefa at dmsolutions.ca
Last Updated 2010-10-07

Table of Contents
• WFS Filter Encoding
– Introduction
– Currently Supported Features
– Get and Post Requests
– Use of Filter Encoding in MapServer
– Limitations
– Tests

9.7.1 Introduction

This document describes the procedures for taking advantage of the Filter Encoding (FE) support in WFS GetFeature
requests, which was added to MapServer in version 4.2.

446 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

This document assumes that you are already familiar with the following aspects of MapServer:
• MapServer application development and setting up .map files.
• Familiarity with the WFS specification would be an asset. Links to the MapServer WFS documents are included
in the next section.

Links to SLD-related Information

• Filter Encoding Implementation Specification.

• MapServer WFS Client Howto.
• MapServer WFS Server Howto.
• MapServer OGC Web Services Workshop.
• Open GIS Consortium (OGC) home page.

9.7.2 Currently Supported Features

The following table lists the currently supported features for FE.
Table 1. Currently Supported Features
Feature Set Feature
Spatial Capabilities
Equals
Disjoint
Touches
Within
Overlaps
Crosses
Intersects
Contains
DWithin
BBOX
Scalar Capabilities

Logical Operators
And
Or
Not
Comparison Operators
PropertyIsEqualTo (=)
PropertyIsNotEqualTo (<>)
PropertyIsLessThan (<)
PropertyIsGreaterThan (>)
PropertyIsLessThanOrEqualTo (<=)
PropertyIsGreaterThanOrEqualTo (>=)
PropertyIsLike
PropertyIsBetween (range)

9.7. WFS Filter Encoding 447

 MapServer Documentation, Release 6.0

9.7.3 Get and Post Requests

MapServer already has the capability to receive and parse Get requests and URL-encoded Post requests. The ability
for MapServer to be able to receive Post requests with XML-encoded information sent in the body of the request has
been added. Also, the ability to generate XML-encoded Post requests for WFS layers has been added.
Both Get and Post request are now supported for all WFS requests:
• GetCapabilities
• GetFeatures
• DescribeFeatureType
Supporting these WFS requests in Post was implemented to keep consistency between all supported WFS requests.
When sending requests, the default request method used is Post. To change this behavior, we have introduced a layer
level meta data, wfs_request_method, which can be set to “GET”.

9.7.4 Use of Filter Encoding in MapServer

This section describes how to use FE on both the server and client sides.

Server Side

To be able to use Filter Encoding, you need to create a valid WFS server using MapServer. Please refer to the WFS
Server HOWTO for specifics.
There is nothing special that should be added to a WFS server for Filter Encoding, but you should note that, when
requesting the capabilities of your WFS server, the document returned should contain the supported filters. Here is
part of a Capabilities document (note the “Filter_Capabilities” section):
1 <?xml version="1.0" encoding="ISO-8859-1"?>
2 <WFS_Capabilities version="1.0.0" updateSequence="0" xmlns="http://www.opengis.net/wfs" xmlns:ogc
3

4 <!-- MapServer version 5.6.5 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=SWF OUTPUT=SVG
5

6 <Service>
7 <Name>MapServer WFS</Name>
8 <Title>WMS Demo Server for MapServer</Title>
9 <Abstract>This demonstration server showcases MapServer (www.mapserver.org) and its OGC support
10 <OnlineResource>http://demo.mapserver.org/cgi-bin/wfs?</OnlineResource>
11 </Service>
12

13 <Capability>
14 <Request>
15 <GetCapabilities>
16 <DCPType>
17 <HTTP>
18 <Get onlineResource="http://demo.mapserver.org/cgi-bin/wfs?"/>
19 </HTTP>
20 </DCPType>
21 <DCPType>
22 <HTTP>
23 <Post onlineResource="http://demo.mapserver.org/cgi-bin/wfs?"/>
24 </HTTP>
25 </DCPType>
26 </GetCapabilities>

448 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

27 ...
28 </Request>
29 </Capability>
30 ...
31 <ogc:Filter_Capabilities>
32 <ogc:Spatial_Capabilities>
33 <ogc:Spatial_Operators>
34 <ogc:Equals/>
35 <ogc:Disjoint/>
36 <ogc:Touches/>
37 <ogc:Within/>
38 <ogc:Overlaps/>
39 <ogc:Crosses/>
40 <ogc:Intersect/>
41 <ogc:Contains/>
42 <ogc:DWithin/>
43 <ogc:BBOX/>
44 </ogc:Spatial_Operators>
45 </ogc:Spatial_Capabilities>
46 <ogc:Scalar_Capabilities>
47 <ogc:Logical_Operators/>
48 <ogc:Comparison_Operators>
49 <ogc:Simple_Comparisons/>
50 <ogc:Like/>
51 <ogc:Between/>
52 </ogc:Comparison_Operators>
53 </ogc:Scalar_Capabilities>
54 </ogc:Filter_Capabilities>
55

56 </WFS_Capabilities>

Client Side

To be able to generate a Filter to a WFS server, a layer level metadata called wfs_filter has been added, which should
contain the filter to be sent to the server. Following is an example of a valid WFS client layer with a filter:
LAYER
NAME "cities"
TYPE POINT
STATUS ON
CONNECTION "http://demo.mapserver.org/cgi-bin/wfs?"
CONNECTIONTYPE WFS
METADATA
"wfs_typename" "cities"
"wfs_version" "1.0.0"
"wfs_connectiontimeout" "60"
"wfs_maxfeatures" "100"
"wfs_filter" "<PropertyIsGreaterThan><PropertyName>POPULATION</PropertyName>
< L i t e r a l >10000000</ L i t e r a l ></ P r o p e r t y I
END
PROJECTION
"init=epsg:4326"
END
LABELITEM ’NAME’
DUMP TRUE
CLASS
NAME ’World Cities’

9.7. WFS Filter Encoding 449

MapServer Documentation, Release 6.0

STYLE
COLOR 255 128 128
OUTLINECOLOR 128 0 0
SYMBOL ’circle’
SIZE 9
END
LABEL
COLOR 0 0 0
OUTLINECOLOR 255 255 255
TYPE TRUETYPE
FONT s a n s
SIZE 7
POSITION UC
PARTIALS FALSE
END
END
END

Note:
• The filter given as a value of the wfs_filter metadata should not contain <Filter> start and end tags.
• The CONNECTION points to a valid WFS server supporting filters
• The returned shapes will be drawn using the class defined in the layer.

9.7.5 Limitations

• A limited set of spatial operators are supported.

9.7.6 Tests

Here are some test URLs for the different supported filters:
• PropertyIsEqualTo
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsEqualTo><PropertyName>NAME</PropertyName>
<Literal>Halifax</Literal></PropertyIsEqualTo></Filter>

• PropertyIsNotEqualTo
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsNotEqualTo><PropertyName>NAME</PropertyName>
<Literal>Halifax</Literal></PropertyIsNotEqualTo></Filter>

• PropertyIsLessThan
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsLessThan><PropertyName>POPULATION</PropertyName>
<Literal>1000</Literal></PropertyIsLessThan></Filter>

• PropertyIsGreaterThan
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsGreaterThan><PropertyName>POPULATION</PropertyName>
<Literal>10000000</Literal></PropertyIsGreaterThan></Filter>

• PropertyIsLessThanOrEqualTo

450 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsLessThanOrEqualTo><PropertyName>POPULATION</PropertyName>
<Literal>499</Literal></PropertyIsLessThanOrEqualTo></Filter>

• PropertyIsGreaterThanOrEqualTo
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsGreaterThanOrEqualTo><PropertyName>POPULATION</PropertyName
<Literal>10194978</Literal></PropertyIsGreaterThanOrEqualTo></Filter>

• PropertyIsBetween
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsBetween><PropertyName>POPULATION</PropertyName>
<LowerBoundary>10194978</LowerBoundary><UpperBoundary>12116379</UpperBoundary></PropertyIsBetween></F

• PropertyIsLike
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><PropertyIsLike wildcard=’*’ singleChar=’.’ escape=’!’>
<PropertyName>NAME</PropertyName><Literal>Syd*</Literal></PropertyIsLike></Filter>

• Logical operator OR
http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><OR><PropertyIsEqualTo><PropertyName>NAME</PropertyName>
<Literal>Sydney</Literal></PropertyIsEqualTo><PropertyIsEqualTo><PropertyName>NAME</PropertyName>
<Literal>Halifax</Literal></PropertyIsEqualTo></OR></Filter>

• Logical operator AND

http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><AND><PropertyIsLike wildcard=’*’ singleChar=’.’ escape=’!’>
<PropertyName>NAME</PropertyName><Literal>Syd*</Literal></PropertyIsLike><PropertyIsEqualTo>
<PropertyName>POPULATION</PropertyName><Literal>4250065</Literal></PropertyIsEqualTo></AND></Filter>

• Logical operator NOT

http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&
TYPENAME=cities&Filter=<Filter><AND><NOT><PropertyIsEqualTo><PropertyName>POPULATION</PropertyName>
<Literal>0</Literal></PropertyIsEqualTo></NOT><NOT><PropertyIsEqualTo><PropertyName>POPULATION</Prope
<Literal>12116379</Literal></PropertyIsEqualTo></NOT></AND></Filter>

• Spatial operator BBOX

http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&TYPENAME=cities&
Filter=<Filter><BBOX><PropertyName>Name>NAME</PropertyName><Box%20srsName=’EPSG:42304’>
<coordinates>135.2239,34.4879 135.8578,34.8471</coordinates></Box></BBOX></Filter>

• Spatial operator Dwithin

http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&TYPENAME=cities&F
<Filter><DWithin><PropertyName>Geometry</PropertyName><gml:Point><gml:coordinates>135.500000,34.66666
</gml:Point><Distance units=’m’>10000</Distance></DWithin></Filter>

• Spatial operator Intersects

http://demo.mapserver.org/cgi-bin/wfs?&VERSION=1.0.0&SERVICE=WFS&REQUEST=GetFeature&TYPENAME=cities&F
<Intersect><PropertyName>Geometry</PropertyName><gml:Polygon><gml:outerBoundaryIs><gml:LinearRing>
<gml:coordinates>135.5329,34.6624 135.4921,34.8153 135.3673,34.7815 135.3800,34.6216 135.5361,34.6210
</gml:LinearRing></gml:outerBoundaryIs></gml:Polygon></Intersect></Filter>

9.7. WFS Filter Encoding 451

MapServer Documentation, Release 6.0

• The OGC conformance tests (http://cite.opengeospatial.org/test_engine) have been run on the FE support. The
following table and notes reflect the current status.
Table 2. WFS OGC test suite (over the HTTP Get and Post method)
Test # Description # of Tests # of Failed Tests
1 Basic WFS tests over the HTTP Get and Post method 402 281
1.1 GetCapabilities 16 0
1.2 DescribeFeatureType 18 0
1.3 GetFeature 368 281
1.3.1 Basic WFS tests 20 1
1.3.2 Complex WFS tests 18 18
1.3.3 Arithmetic filter WFS tests 8 8
1.3.4 Comparison WFS tests 50 26
1.3.4.1 GetFeature PropertyIsGreaterThanOrEqualTo filter 2 0
1.3.4.2 GetFeature PropertyIsBetween filter 6 2
1.3.4.3 GetFeature PropertyIsEqualTo filter 4 0
1.3.4.4 GetFeature PropertyIsGreaterThan filter 4 2
1.3.4.5 GetFeature PropertyIsGreaterThanOrEqualTo filter 6 6
1.3.4.6 GetFeature PropertyIsLessThan filter 6 4
1.3.4.7 GetFeature PropertyIsLessThanOrEqualTo filter 6 4
1.3.4.8 GetFeature PropertyIsLike filter 2 0
1.3.4.9 GetFeature PropertyIsNotEqualTo filter 6 0
1.3.4.10 GetFeature PropertyIsNull filter 8 8
1.3.5 Logical WFS test 20 0
1.3.5.1 GetFeature AND PropertyIsEqualTo PropertyIsEqualTo filter 8 0
1.3.5.2 GetFeature OR PropertyIsEqualTo PropertyIsEqualTo filter 8 0
1.3.5.3 GetFeature NOT PropertyIsNotEqualTo filter 4 0
1.3.6 Spatial operator WFS test 252 228
1.3.6.1 GetFeature BBOX filter 36 12
1.3.6.2 GetFeature with other filter types 216 216
2 Transactional WFS test 69 69
The OGC Cite WFS test suite can be found on the OGC Cite portal.
Following are some MapServer specific notes on this test suite:
1. Test number 1.3.1:
• There is a contradiction between the wfs/1.0.0/basic/getfeature/post/3 assertion and the XPath expected
value of the test. The assertion says: “Test that a GetFeature request with no output format de-
fined returns a wfs:FeatureCollection with GML data.” and the expected XPath value for this request:
“boolean(/ogc:ServiceExceptionReport)” is supposed to be true. So, the assertion means that when a WFS
server receives a request which contains an undefined output format or no output format at all, the WFS
server must return a WFS collection containing GML data. The XPath expected value means that when a
WFS server receives a request with an undefined output format or no output format at all, the WFS server
must return a service exception report.
2. Tests number 1.3.2 and 1.3.3:
• Not supported.
3. Tests number 1.3.4.2, 1.3.4.4 to 1.3.4.7:
• The string comparison is not supported using >, <, >=, <=.
• The date comparison is not supported.
See Also:

452 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

bug 461
4. Test number 1.3.4.10:
• This property is not supported in MapServer.
5. Test number 1.3.6.1:
• The returned feature xml won’t validate because the validation is done against a specific xsd (geoma-
try.xsd).
• The data conversion on multipoints and multilayers are not supported within GDAL.
See Also:
bug 461
6. Test number 2:
• The transaction requests are not supported.

9.8 SLD

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Author Yewondwossen Assefa
Contact assefa at dmsolutions.ca
Last Updated 2011-01-14

Contents
• SLD
– Introduction
– Server Side Support
– Client Side Support
– Named Styles support
– Other Items Implemented
– Issues Found During Implementation

9.8.1 Introduction

This document describes the procedures for taking advantage of the Styled Layer Descriptor (SLD) support in WMS
GetMap requests with MapServer. SLD support exists for the server side (ability to read an SLD and apply it with a
GetMap request) and for the client side (includes sending SLD requests to server and generate SLD files on the fly
from MapServer map file). SLD support was added to MapServer in version 4.2.
This document assumes that you are already familiar with the following aspects of MapServer:
• MapServer application development and setting up .map files.
• Familiarity with the WMS specification would be an asset. Links to the MapServer WMS documents are in-
cluded in the next section.

9.8. SLD 453

MapServer Documentation, Release 6.0

Links to SLD-related Information

• Styled Layer Descriptor Implementation Specification.

• MapServer WMS Client HowTo.
• MapServer WMS Server HowTo.
• MapServer OGC Web Services Workshop.
• Open GIS Consortium (OGC) home page.

9.8.2 Server Side Support

General Information

There are two ways a WMS request can pass an SLD document with a GetMap request to MapServer:
• SLD parameter pointing to remote SLD (SLD=http://URL_TO_SLD).
• SLD_BODY parameter to send the SLD definition in the URL.
These two methods are both available through MapServer. An example of a request would be:
http://demo.mapserver.org/cgi-bin/wms?SERVICE=wms&VERSION=1.1.1&REQUEST=GetMap
&LAYERS=country_bounds&SLD=http://demo.mapserver.org/ogc-demos/map/sld/sld_line_simple.xml

Test the remote SLD request.

The SLD in the above request follows:
<?xml version="1.0" encoding="UTF-8"?>
<StyledLayerDescriptor version="1.0.0"
xmlns="http://www.opengis.net/sld"
xmlns:ogc="http://www.opengis.net/ogc"
xmlns:xlink="http://www.w3.org/1999/xlink"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.opengis.net/sld
http://schemas.opengis.net/sld/1.0.0/StyledLayerDescriptor.xsd">
<NamedLayer>
<Name>country_bounds</Name>
<UserStyle>
<Title>xxx</Title>
<FeatureTypeStyle>
<Rule>
<LineSymbolizer>
<Geometry>
<ogc:PropertyName>center-line</ogc:PropertyName>
</Geometry>
<Stroke>
<CssParameter name="stroke">#0000ff</CssParameter>
</Stroke>
</LineSymbolizer>
</Rule>
</FeatureTypeStyle>
</UserStyle>
</NamedLayer>
</StyledLayerDescriptor>

Version 1.1.0 of the same SLD

454 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

<?xml version="1.0" encoding="UTF-8"?>

<StyledLayerDescriptor version="1.1.0"
xmlns="http://www.opengis.net/sld"
xmlns:se="http://www.opengis.net/se"
xmlns:ogc="http://www.opengis.net/ogc"
xmlns:xlink="http://www.w3.org/1999/xlink"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.opengis.net/sld
http://schemas.opengis.net/sld/1.1.0/StyledLayerDescriptor.xsd">
<NamedLayer>
<se:Name>country_bounds</se:Name>
<UserStyle>
<se:Name>xxx</se:Name>
<se:FeatureTypeStyle>
<se:Rule>
<se:LineSymbolizer>
<se:Geometry>
<ogc:PropertyName>center-line</ogc:PropertyName>
</se:Geometry>
<se:Stroke>
<se:SvgParameter name="stroke">#0000ff</se:SvgParameter>
</se:Stroke>
</se:LineSymbolizer>
</se:Rule>
</se:FeatureTypeStyle>
</UserStyle>
</NamedLayer>
</StyledLayerDescriptor>

When MapServer gets a valid SLD through a request, it parses this SLD to extract all the styles attached to the
NamedLayers, and then it applies these styles to the map before it is returned to the client. When applying the SLD,
MapServer compares the <Name> parameter of the NamedLayers in the SLD document to the WMS layer names
(WMS layer names are available in a GetCapabilities request).
Note: All the examples given in this document are live uses of valid SLDs and a MapServer installation with SLD
support.
Additional WMS features related to SLDs have also been developed:
Table1. Additional WMS Features
Features Supported Notes
Method GET : SLD URL Yes
Method GET : SLD_BODY Yes Additional item
Describer Layer Yes
GetLegendGraphic Yes
GetStyles Yes Uses MapScript to get the SLD
Note: As of MapServer version 4.2.3, the GetLegendGraphic request (see section 12 of the Styled Layer Descriptor
Implementation Specification) works as follows: if the RULE keyword is absent from the request, an image containing
the entire legend for the specified layer will be returned. This image consists of the layer name and a symbolization
graphic and label for each class.

Specific SLD Elements Supported

The following tables give a lot of additional details about SLD support in MapServer.
Table2. Named Layers and User Layers

9.8. SLD 455

MapServer Documentation, Release 6.0

Features Supported Notes

Named Layers Yes
User Layers No
Table3. Named Styles and User Styles
Features Supported Notes
Named Styles Yes
User Styles Yes
Table 4. User Styles
Features Sup- Notes
ported
Name No This was removed at implementation time, since it does not fit with MapServer
Title No No use in the MapServer environment
Abstract No No use in the MapServer environment
IsDefault No Only one style is available per layer
FeatureType- Yes MapServer has a concept of one feature type style per layer (either point, line,
Style polygon, or raster)
Table 5. FeatureTypeStyle
Features Supported Notes
Name No No use in the MapServer environment
Title No No use in the MapServer environment
Abstract No No use in the MapServer environment
FeatureTypeName No No use in the MapServer environment
SemanticTypeIdentifier No Still an experimental element in the SLD specifications
Rule Yes
Table 6. Rule
Features Supported Notes
Name Yes
Title Yes
Abstract No No use in the MapServer environment
LegendGraphic Yes
Filter Yes
ElseFilter Yes
MinScaleDenominator Yes
MaxScaleDenominator Yes
LineSymbolizer Yes
PolygonSymbolizer Yes
PointSymbolizer Yes
TextSymbolizer Yes
RasterSymbolizer Yes Applies for 8-bit rasters
• Filter and ElseFilter
For each rule containing a filter, there is a class created with the class expression set to reflect that filter. Avail-
able filters that can be used are Comparison Filters and Logical Filters (see the Filter Encoding HowTo). The
ElseFilter parameters are converted into a class in MapServer and placed at the end of the class list with no
expression set. They are used to render elements that did not fit into any other classes.
• MinScaleDenomibator and MaxScaleDenominator are translated in minscale and maxscale in MapServer.
The following are examples of valid requests using the Filters:
• line with one filter: sld 6a / full request 6a

456 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• line with multiple filters: sld 6b / full request 6b

• line with one filter and an else filter: sld 6c / full request 6c
• spatial filter using BBOX: sld 6d/ full request 6d
• The above example enables spatial filtering using the BBOX parameter as a Filter for a selected area (Africa).
Note that an ElseFilter will not work with a spatial filter.
Table 7. LineSymbolizer
Features Supported Notes
Geometry No MapServer uses the data geometry to do the rendering
Stroke: GraphicFill No Solid color is used
Stroke: GraphicStroke Yes Draws the symbol along the line
Stroke (CssParameter): stroke Yes RGB colors are supported
Stroke (CssParameter): width Yes
Stroke (CssParameter): opacity Yes Only available for AGG driver and mapserver version >=5.2
Stroke (CssParameter): linejoin and linecap No Not supported in MapServer
Stroke (CssParameter): dasharray Yes
Stroke (CssParameter): dashoffset No
PerperdicularOffset (only in SLD 1.1.0) Yes Offset values of the style object will be set
InitialGap(GraphicStroke SLD 1.1.0) No
Gap (GraphicStroke parameter SLD 1.1.0) No
Note that SvgParameter instead of CssParameter are required for SLD 1.1.0.
The following are examples of valid requests using the LineSymbolizer:
• simple line: sld 7a / full request 7a
• line with width: sld 7b / full request 7b
• dashed line: sld 7c / full request 7c
Table 8. PolygonSymbolizer
Features Supported Notes
Geometry No
Stroke Yes Strokes are the same as for the LineSymbolizer
Fill Yes Was developed to support symbol fill polygons in addition to solid fill
Fill-opacity Yes Only available for AGG driver and mapserver version >=5.2
PerperdicularOffset No SLD 1.1.0 parameter
Displacement Yes SLD 1.1.0 parameter. Sets offsetx/y in MapServer
A Fill can be a solid fill or be a Graphic Fill, which is either a well-known Mark symbol (e.g., square, circle, triangle,
star, cross, x) or an ExternalGraphic element (e.g., gif, png) available through a URL. When a Mark symbol is used
in an SLD, MapServer creates a corresponding symbol in the map file and uses it to render the symbols. When a
ExternalGraphic is used, the file is saved locally and a pixmap symbol is created in the mapfile referring to the this
file. Note that the Web object IMAGEPATH is used to save the file.
The following are examples of valid requests using the PolygonSymbolizer:
• simple solid fill: sld 8a / full request 8a
• solid fill with outline: sld 8b / full request 8b
• fill with mark symbol: sld 8c / full request 8c
• fill with external symbol: sld 8d/ full request 8d
Table 9. PointSymbolizer

9.8. SLD 457

MapServer Documentation, Release 6.0

Features Supported Notes

Geometry No
Graphic: Mark symbol Yes Well-known names (square, circle, triangle, star, cross, X) are supported
Graphic: ExternalGraphic Yes Was developed to support symbol fill polygons in addition to solid fill
Opacity Yes Support added in Mapserver 5.4
Size Yes
Rotation Yes Support added in MapServer 5.4
Displacement Yes SLD 1.1.0 Paramater. Support added in MapServer 5.4
AnchorPoint No
Note: refer to the PolygonSymbolizer notes for how the Mark and ExternalGraphic symbols are applied in MapServer.
The following are examples of valid requests using the PointSymbolizer:
• filled mark symbol: sld 9a / full request 9a
• default settings (square, size 6, color 128/128/128): sld 9b / full request 9b
• external symbol: sld 9c / full request 9c
Table 10. TextSymbolizer
Features Sup- Notes
ported
Geometry No
Label Yes
Font(font- Yes Font names used are those available in MapServer font file. If no fonts are available
family) there, default bitmap fonts are used
Font-style Yes
(Italic, ...)
Font- Yes
weight
Font-size Yes If true-type fonts are not used, default bitmap sizes are given
LabelPlace- Yes PointPlacement is supported. LinePlacement is supported for versions >=5.2.1. Only
ment PerperdicularOffset and IsAligned are suported for LinePlacement.
Halo Yes Supported (fill converted to outlinecolor, and radius is converted to outlinewidth. Note
that outlinewidth is only available for AGG in >=5.2)
Fill Yes Only solid color is available
Notes on the TextSymbolizer:
• Font names: when converting Font parameters to MapServer, the following rule is applied to get the font name:
FontFamily-FontStyle-FontWeight. For example, if there is an SLD with a Font Family of arial, a Font Style
of italic, and a Font weight equal to bold, the resulting MapServer font name is arial-bold-italic. Font Style
and Weight are not mandatory and, if not available, they are not used in building the font name. When a Font
Style or a Font Weight is set to normal in an SLD, it is also ignored in building the name. For example, if there
is an SLD with a Font Family of arial, a Font Style of normal and a Font weight equals to bold, the resulting
MapServer font name is arial-bold.
• A TextSymbolizer can be used in MapServer either on an Annotation layer or on a Point, Line, or Polygon layer
- in addition to other symbolizers used for these layers.
• PointPacement: a point placement includes AnchorPoint (which is translated to Position in MapServer) Dis-
placement (which is translated to Offset) and Angle (which is translated to Angle).
• Angle setting (MapServer version >=5.4): by default the angle parameter is set to AUTO. For point features,
users can use the PointPlacement to alter the value. For line features, the user can add a LinePlacement: If
an ‘empty’ LinePlacement is part of the SLD, the angle will be set to FOLLOW, If a LinePlacement contains
the PerpendicualarOffset parameter, the angle will be set to 0 and the PerpendicualarOffset will be used to set

458 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

the offset values in the label object. SLD 1.1.0 introduces the IsAligned parameter for LinePlacement: if this
parameter is set to false, the angle will be set to 0.
The following are examples of valid requests using the TextSymbolizer:
• annotation layer : test for label, font, point placement, color, angle: sld 10a / full request 10a
• annotation layer with text and symbols using 2 symbolizers: sld 10b / full request 10b
Table 11. RasterSymbolizer
Features Supported Notes
Geometry No
Opacity Yes
ChannelSelection No
OverlapBehaviour No
ColorMap Yes
ContrastEnhancement No
ShadedRelief No
ImageOutline No
The current support in MapServer includes only ColorMap parameter support. It can be used to classify 8-bit rasters.
Inside the ColorMap parameters, the color and quantity parameters are extracted and used to do the classification.
Table 12. ColorMap
The following Features are available in SLD 1.0
Features Supported Notes
Color Yes
Opacity No
Quantity Yes
Label No
The following is an example of ColorMap usage for SLD 1.0.
If we have following ColorMap in an SLD:
<ColorMap>
<ColorMapEntry color="#00ff00" quantity="22"/>
<ColorMapEntry color="#00bf3f" quantity="30"/>
<ColorMapEntry color="#007f7f" quantity="37"/>
<ColorMapEntry color="#003fbf" quantity="45"/>
<ColorMapEntry color="#0000ff" quantity="52"/>
<ColorMapEntry color="#000000" quantity="60"/>
</ColorMap>

The six classes that are created are:

class 1: [pixel] >= 22 AND [pixel] < 30 with color 00ff00
class 2: [pixel] >= 30 AND [pixel] < 37 with color 00bf3f
class 3: [pixel] >= 37 AND [pixel] < 45 with color 007f7f
class 4: [pixel] >= 45 AND [pixel] < 52 with color 003fbf
class 5: [pixel] >= 52 AND [pixel] < 60 with color 0000ff
class 6: [pixel] = 60 with color 000000

Note that the ColorMapEntry quantity parameters should be in increasing order.

The following Features are available in SLD 1.1
Features Supported Notes
Categorize Yes

9.8. SLD 459

MapServer Documentation, Release 6.0

The following is an example of and SLD 1.1.0 with a raster symbolizer

<StyledLayerDescriptor version="1.1.0" xsi:schemaLocation="http://www.opengis.net/sld http://schemas.
<NamedLayer>
<se:Name>landsat</se:Name>
<UserStyle>
<se:Name>xxx</se:Name>
<se:FeatureTypeStyle>
<se:Rule>
<se:RasterSymbolizer>
<se:Opacity>0.7</se:Opacity>
<se:ColorMap>
<se:Categorize fallbackValue="#78c818">
<se:LookupValue>Rasterdata</se:LookupValue>
<se:Value>#ffffff</se:Value>
<se:Threshold>22</se:Threshold>
<se:Value>#00ff00</se:Value>
<se:Threshold>30</se:Threshold>
<se:Value>#00bf3f</se:Value>
<se:Threshold>37</se:Threshold>
<se:Value>#007f7f</se:Value>
<se:Threshold>45</se:Threshold>
<se:Value>#003fbf</se:Value>
<se:Threshold>52</se:Threshold>
<se:Value>#0000ff</se:Value>
<se:Threshold>60</se:Threshold>
<se:Value>#000000</se:Value>
</se:Categorize>
</se:ColorMap>
</se:RasterSymbolizer>
</se:Rule>
</se:FeatureTypeStyle>
</UserStyle>
</NamedLayer>
</StyledLayerDescriptor>

The classes that are created are:

class 1: [pixel] < 22 with color ffffff
class 2: [pixel] >= 22 AND [pixel] < 30 with color 00ff00
class 3: [pixel] >= 30 AND [pixel] < 37 with color 00bf3f
class 4: [pixel] >= 37 AND [pixel] < 45 with color 007f7f
class 5: [pixel] >= 45 AND [pixel] < 52 with color 003fbf
class 6: [pixel] >= 52 AND [pixel] < 60 with color 0000ff
class 7: [pixel] >= 60 with color 000000

Examples using 8 bits and 16 bits rasters can be seen at:

• example 1
• example 2

9.8.3 Client Side Support

Client side support of the SLD consists of two parts:

• The first part is using MapServer as a WMS client to send a GetMap request with an SLD. This is done using
two metadata that can be placed at a layer level in a MapServer mapfile. These two metadata are:
– wms_sld_url, which takes a valid URL as a value and appends SLD=xxx to the GetMap request.

460 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

– wms_sld_body, which takes a valid SLD string and appends SLD_BODY=xxx to the GetMap request.
If the value of wms_sld_body is set to AUTO, MapServer generates an SLD based on the classes found in
the layer and send this SLD as the value of the SLD_BODY parameter in the GetMap request.
• The other major item is the generation of an SLD document from MapServer classes. These functions are
currently available through MapServer/MapScript interface. Here are the functions available:
– on a map object: generatesld
– on a layer object: generatesld
Additional MapScript functions have been added or will be added to complement these functions:
– on a map object: applysld
– on a layer object: applysld
Note: When generating an SLD from MapServer classes, if there is a pixmap symbol you need to have this symbol
available through a URL so it can be converted as an ExternalGraphic symbol in the SLD. To do this, you need to
define the URL through a web object level metadata called WMS_SLD_SYMBOL_URL in your map file. The SLD
generated uses this URL and concatenates the name of the pixmap symbol file to get the value that is generated as the
ExternaGraphic URL.

PHP/MapScript Example that Generates an SLD from a Mapfile

The following is a small script that calls the generateSLD() function to create an SLD for a specific layer in a mapfile:
1 <?php
2

3 // define variables
4 define( "MAPFILE", "D:/ms4w/apps/cadastra/map/cadastra.map" );
5 define( "MODULE", "php_mapscript.dll" );
6

7 // load the mapscript module

8 if (!extension_loaded("MapScript")) dl(MODULE);
9

10 // open map
11 $oMap = ms_newMapObj( MAPFILE );
12

13 // get the parcel layer

14 $oLayer = $oMap->getLayerByName("parcel");
15

16 // generate the sld for that layer

17 $SLD = $oLayer->generateSLD();
18

19 // save sld to a file

20 $fp = fopen("parcel-sld.xml", "a");
21 fputs( $fp, $SLD );
22 fclose($fp);
23

24 ?>

9.8.4 Named Styles support

Named styles support are introduced in MapServer 5.2. The support is base on MS RFC 39: Support of WMS/SLD
Named Styles
MapServer 5.2 introduces the possibility to assign a group to a series of classes defined on a layer object using two
new non-mandatory keywords CLASSGROUP (at the layer level) and GROUP at the class level:

9.8. SLD 461

MapServer Documentation, Release 6.0

LAYER
...
CLASSGROUP "group1"
...
CLASS
NAME "name1"
GROUP "group1"
...
END
CLASS
NAME "name2"
GROUP "group2"
...
END
CLASS
NAME "name3"
GROUP "group1"
...
END
...

At rendering time, if the CLASSGROUP is defined, only classes that have the same group name would be used.
Based on this concept, WMS/SLD support uses the class groups as named styles. Each group of classes is considered
equivalent to a named style:
• The GetCapbilities request will output all the styles that are available
• The GetMap request can use the STYLES parameter to specify a named style
• The GetLegendGraphic can use the STYLES parameter to specify a named style

9.8.5 Other Items Implemented

• Support of filled polygons with Mark and ExternalGraphic symbols.

• MapScript functions to parse and apply SLD.
• SLD_BODY request support on client and server side.

9.8.6 Issues Found During Implementation

• Limitation of the FilterEncoding to comparison and logical filters. The spatial filters were not made available
since it required major changes in MapServer WMS support.

9.9 WCS Server

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Author Stephan Meissl
Contact stephan.meissl at eox.at
Last Updated 2011-03-30

462 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Table of Contents
• WCS Server
– Introduction
– Configuring Your Mapfile to Serve WCS Layers
– Test Your WCS 1.0 Server
– WCS 1.1.0+ Issues
– WCS 2.0
– HTTP-POST support
– Reference Section
– Rules for handling SRS in a MapServer WCS
– Spatio/Temporal Indexes
– WCS 2.0 Application Profile - Earth Observation (EO-WCS)
– To-do Items and Known Limitations

9.9.1 Introduction

A WCS (or Web Coverage Service) allows for the publication of “coverages”- digital geospatial information repre-
senting space-varying phenomena. In the MapServer world it allows for unfiltered access to raster data. Conceptually
it is easy think of WCS as a raster equivalent of WFS. The following documentation is based on the Open Geospatial
Consortium’s (OGC) Web Coverage Service Interfaces Implementation Specification version 1.0.0.

Links to WCS-Related Information

• OGC’c WCS Standard page

• WCS 1.0.0 specification
• WCS 1.1.1c1 specification
• WCS 2.0
– GML 3.2.1 Application Schema Coverages
– WCS 2.0 Specification - Core
– WCS 2.0 Specification - KVP Protocol Binding Extension
– WCS 2.0 Specification - XML/POST Protocol Binding Extension
• WMS Server HowTo

Software Requirements

In order to enable MapServer to serve WCS data, it MUST be compiled against certain libraries:
• PROJ.4: The reprojection library. Version 4.4.3 or greater is required.
• GDAL: raster support library.
• MapServer: version >= 4.4 (tested with 5.0.2 while updating this document)
For WCS 1.1.x (MapServer 5.2) and WCS 2.0 (MapServer 6.0) support there is an additional requirement:
• libxml2: An xml parser and generation library.

9.9. WCS Server 463

MapServer Documentation, Release 6.0

Please see the MapServer UNIX Compilation and Installation HowTo for detailed instructions on compiling MapServer
with support for these libraries and features. For Windows users, MapServer for Windows (MS4W) comes with WCS
Server support.

9.9.2 Configuring Your Mapfile to Serve WCS Layers

Much as in the WMS and WFS support, WCS publishing is enabled by adding certain magic METADATA key-
word/value pairs to a .map file.
MapServer will serve and include in its WCS capabilities only the layers that meet the following conditions:
• Data source is a raster, which is processed using GDAL (e.g GeoTIFF, Erdas Imagine, ...)
• LAYER NAME must be set
• LAYER TYPE is set to RASTER
• LAYER DUMP parameter set to TRUE
• WEB metadata “wcs_label” must be set
• LAYER metadata “wcs_label” must be set
• LAYER metadata “wcs_rangeset_name” must be set
• LAYER metadata “wcs_rangeset_label” must be set
• LAYER is enabled to be served via WCS (see MS RFC 67)

Example WCS Server Mapfile

The following is an example of a simple WCS Server mapfile. Note the comments for the required parameters.

NAME W C S _ s e r v e r
STATUS ON
SIZE 400 300
SYMBOLSET "../etc/symbols.txt"
EXTENT -2200000 -712631 3072800 3840000
UNITS METERS
SHAPEPATH "../data"
IMAGECOLOR 255 255 255
FONTSET "../etc/fonts.txt"

#
# Start of web interface definition
#
WEB
IMAGEPATH "/ms4w/tmp/ms_tmp/"
IMAGEURL "/ms_tmp/"
METADATA
"wcs_label" "GMap WCS Demo Server" ### required
"wcs_description" "Some text description of the service"
"wcs_onlineresource" "http://127.0.0.1/cgi-bin/mapserv.exe?" ### recommended
"wcs_fees" "none"
"wcs_accessconstraints" "none"
"wcs_keywordlist" "wcs,test"
"wcs_metadatalink_type" "TC211"
"wcs_metadatalink_format" "text/plain"

464 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

"wcs_metadatalink_href" "http://someurl.com"
"wcs_address" "124 Gilmour Street"
"wcs_city" "Ottawa"
"wcs_stateorprovince" "ON"
"wcs_postcode" "90210"
"wcs_country" "Canada"
"wcs_contactelectronicmailaddress" "blah@blah"
"wcs_contactperson" "me"
"wcs_contactorganization" "unemployed"
"wcs_contactposition" "manager"
"wcs_contactvoicetelephone" "613-555-1234"
"wcs_contactfacimiletelephone" "613-555-1235"
"wcs_service_onlineresource" "http://127.0.0.1/cgi-bin/mapserv.exe?"
END
END

PROJECTION
"init=epsg:42304"
END

LAYER
NAME b a t h y m e t r y
METADATA
"wcs_label" "Elevation/Bathymetry" ### required
"wcs_rangeset_name" "Range 1" ### required to support DescribeCoverage request
"wcs_rangeset_label" "My Label" ### required to support DescribeCoverage request
END
TYPE RASTER ### required
STATUS ON
DATA b a t h _ m a p s e r v e r . t i f
PROJECTION
"init=epsg:42304"
END
DUMP TRUE ### required
END

END # Map File

Output Formats

The raster formats supported by MapServer WCS are determined by the wcs_formats metadata item on the LAYER.
This should contain a space separated list of OUTPUTFORMAT driver names separated by spaces. If absent, all raster
OUTPUTFORMATs are allowed.
WCS is a “raw data” oriented format. So it often most suitable to use it with format using the BYTE, INT16 and
FLOAT32 IMAGEMODEs with GDAL related output formats rather than the built in “rendering oriented” output
formats. By default the only GDAL format driver defined is the GTiff driver. The following are example output format
declarations utilizing the raw image modes:
OUTPUTFORMAT
NAME GEOTIFF_16
DRIVER "GDAL/GTiff"
MIMETYPE "image/tiff"
IMAGEMODE FLOAT32
EXTENSION "tif"

9.9. WCS Server 465

MapServer Documentation, Release 6.0

END

OUTPUTFORMAT
NAME AAIGRID
DRIVER "GDAL/AAIGRID"
MIMETYPE "image/x-aaigrid"
IMAGEMODE INT16
EXTENSION "grd"
FORMATOPTION "FILENAME=result.grd"
END

The FORMATOPTION FILENAME defines the preferred name of the result file when returned WCS GetCoverage
results.

9.9.3 Test Your WCS 1.0 Server

Validate the Capabilities Metadata

OK, now that we’ve got a mapfile, we have to check the XML capabilities returned by our server to make sure nothing
is missing.
Using a web browser, access your server’s online resource URL to which you add the parameters “SER-
VICE=WCS&VERSION=1.0.0&REQUEST=GetCapabilities” to the end, e.g.
http://my.host.com/cgi-bin/mapserv?map=mywcs.map&SERVICE=WCS&VERSION=1.0.0&REQUEST=GetCapabilities

If you get an error message in the XML output then take necessary actions. Common problems and solutions are listed
in the FAQ at the end of this document.
If everything went well, you should have a complete XML capabilities document. Search it for the word “WARN-
ING”... MapServer inserts XML comments starting with “<!–WARNING: ” in the XML output if it detects missing
mapfile parameters or metadata items.
Note that when a request happens, it is passed through WMS, WFS, and WCS in MapServer (in that order) until one
of the services respond to it.
Here is a working example of a GetCapabilities request:
http://demo.mapserver.org/cgi-bin/wcs?SERVICE=wcs&VERSION=1.0.0&REQUEST=GetCapabilities

Test With a DescribeCoverage Request

OK, now that we know that our server can produce a valid XML GetCapabilities response we should test the De-
scribeCoverage request. The DescribeCoverage request lists more information about specific coverage offerings.
Using a web browser, access your server’s online resource URL to which you add the parameters “SER-
VICE=WCS&VERSION=1.0.0&REQUEST=DescribeCoverage&COVERAGE=layername” to the end, e.g.
http://my.host.com/cgi-bin/mapserv?map=mywcs.map&SERVICE=WCS&VERSION=1.0.0&REQUEST=DescribeCoverage&C

Here is a working example of a DescribeCoverage request:

http://demo.mapserver.org/cgi-bin/wcs?SERVICE=wcs&VERSION=1.0.0&REQUEST=DescribeCoverage&COVERAGE=modis

466 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

Test With a GetCoverage Request

The GetCoverage request allows for the retrieval of coverages in a specified output format to the client.
The following is a list of the required GetCoverage parameters according to the WCS spec:
VERSION=version: Request version
REQUEST=GetCoverage: Request name
COVERAGE=coverage_name: Name of an available coverage, as stated in the GetCapabilities
CRS=epsg_code: Coordinate Reference System in which the request is expressed.
BBOX=minx,miny,maxx,maxy: Bounding box corners (lower left, upper right) in CRS units. One of
BBOX or TIME is required.
TIME=time1,time2: Request a subset corresponding to a time. One of BBOX or TIME is required..
WIDTH=output_width: Width in pixels of map picture. One of WIDTH/HEIGHT or RESX/Y is re-
quired.
HEIGHT=output_height: Height in pixels of map picture. One of WIDTH/HEIGHT or RESX/Y is
required.
RESX=x: When requesting a georectified grid coverage, this requests a subset with a specific spatial
resolution. One of WIDTH/HEIGHT or RESX/Y is required.
RESY=y: When requesting a georectified grid coverage, this requests a subset with a specific spatial
resolution. One of WIDTH/HEIGHT or RESX/Y is required.
FORMAT=output_format: Output format of map, as stated in the DescribeCoverage response.
The following are optional GetCoverage parameters according to the WCS spec:
RESPONSE_CRS=epsg_code: Coordinate Reference System in which to express coverage responses.
So to follow our above examples, a valid DescribeCoverage request would look like:
http://my.host.com/cgi-bin/mapserv?map=mywcs.map&SERVICE=wcs&VERSION=1.0.0&REQUEST=GetCoverage
&coverage=bathymetry&CRS=EPSG:42304&BBOX=-2200000,-712631,3072800,3840000
&WIDTH=3199&HEIGHT=2833&FORMAT=GTiff

Here is a working example of a GetCoverage request (note that a 350KB tif is being requested, so this may take a
second):
http://demo.mapserver.org/cgi-bin/wcs?SERVICE=wcs&VERSION=1.0.0&REQUEST=GetCoverage&COVERAGE=modis-
001&CRS=EPSG:26915&BBOX=159707,4597395,1400707,5501395&WIDTH=400&HEIGHT=300&FORMAT=GEOTIFF_RGB

9.9.4 WCS 1.1.0+ Issues

WCS 1.1.0 and later versions of the WCS protocol are supported by MapServer 5.2. For the most part the map file
setup for WCS 1.1.0 is similar to WCS 1.0.0, but the actual protocol is substantially changed.

GetCapabilities

The GetCapabilities request is the same as WCS 1.0 but with a different VERSION value:
SERVICE=WCS&VERSION=1.1.0&REQUEST=GetCapabilities

The format of the returned capabilities document is substantially altered from WCS 1.0, and makes use of OWS
Common for service descriptions.

9.9. WCS Server 467

MapServer Documentation, Release 6.0

DescribeCoverage

The DescribeCoverage request is similar to WCS 1.0, but the IDENTIFIER keyword is used instead of COVERAGE
to name the coverage being requested:
SERVICE=WCS&VERSION=1.1.0&REQUEST=DescribeCoverage&IDENTIFIER=spaceimaging

GetCoverage

The format for GetCoverage is substantially changed from 1.0. The following is a list of GetCoverage required
parameters:
VERSION=version: Request version
REQUEST=GetCoverage: Request name
IDENTIFIER=coverage_name: Name of an available coverage, as stated in the GetCapabilities
BOUNDINGBOX=minx,miny,maxx,maxy,crs: Bounding box corners (lower left, upper right), and the
CRS they are in. The CRS is described using a URN.
FORMAT=output_format: Output format (mime type) of grid product, as stated in the GetCapabilities.
If an alternate spatial resolution is desired, then the following set of keywords must be used to specify the sample
origin and step size of the output grid to be produced. The produced grid will be of a number of pixels and lines as
can be fit in the BOUNDINGBOX starting at GridOrigin, at GridOffsets resolution.
GRIDBASECRS=crs: The grid base CRS (URN).
GRIDCS=crs: The grid CRS (URN).
GridType=urn:ogc:def:method:WCS:1.1:2dGridIn2dCrs: This is the only supported value for
MapServer.
GridOrigin=x_origin,y_origin: The sample point for the top left pixel.
GridOffsets=xstep,ystep: The x and y step size for grid sampling (resolution). Both are positive.
As well, the following optional parameters are available.
RangeSubset=selection: Selects a range subset, and interpolation method. Currently only subsetting on
bands are allowed. Depending on rangeset names, this might take the form “BandsName[bands[1]]” to
select band 1, or “BandsName:bilinear[bands[1]]” to select band 1 with bilinear interpolation.
So a simple GetCoverage might look like:
SERVICE=WCS&VERSION=1.1.0&REQUEST=GetCoverage&IDENTIFIER=dem&FORMAT=image/tiff
&BOUNDINGBOX=43,33,44,34,urn:ogc:def:crs:EPSG::4326

A more complex request might look like:

SERVICE=WCS&VERSION=1.1.0&REQUEST=GetCoverage&IDENTIFIER=dem&FORMAT=image/tiff
&BOUNDINGBOX=33,43,34,44,urn:ogc:def:crs:EPSG::4326
&GridBaseCRS=urn:ogc:def:crs:EPSG::4326&GridCS=urn:ogc:def:crs:EPSG::4326
&GridType=urn:ogc:def:method:WCS:1.1:2dGridIn2dCrs
&GridOrigin=33,44&GridOffsets=0.01,0.01
&RangeSubset=BandsName:bilinear[bands[1]]

It should also be noted that return results from WCS 1.1 GetCoverage requests are in multi-part mime format. Typically
this consists of a first part with an xml document referencing the other parts of the message, and an image file part.
However, for output formats that return multiple files, each will be a separate part. For instance, this means it is
possible to return a jpeg file with a world file, the OUTPUTFORMAT is appropriately configured.

468 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

URNs

In WCS 1.1 protocol coordinate systems are referenced by URN. Some typical URNs are:
urn:ogc:def:crs:EPSG::4326
urn:ogc:def:crs:EPSG:27700
urn:ogc:def:crs:OGC::CRS84

The first two are roughly equivalent to EPSG:4326, and EPSG:27700 while the third is a CRS defined by OGC (es-
sentially WGS84). One critical thing to note is that WCS 1.1 follows EPSG defined axis/tuple ordering for geographic
coordinate systems. This means that coordinates reported, or provided in urn:ogc:def:EPSG::4326 (WGS84) are ac-
tually handled as lat, long, not long,lat. So, for instance the BOUNDINGBOX for an area in California might look
like:
BOUNDINGBOX=34,-117,35,-116,urn:ogc:def:crs:EPSG::4326

And, likewise the bounds reported by GetCapabilities, and DescribeCoverage will be in this ordering as appropriate.

9.9.5 WCS 2.0

Overview

Version 6.0 introduces support for the new version 2.0 of the WCS specification. This section documents the usage of
the new WCS version.

Web Coverage Service (WCS) 2.0 Interface Standard

This specification adopts the new OGC Core and Extension model and at the moment the following documents are
available from the OGC’c WCS Standard page:
• GML 3.2.1 Application Schema Coverages
• WCS 2.0 Specification - Core
• WCS 2.0 Specification - KVP Protocol Binding Extension
• WCS 2.0 Specification - XML/POST Protocol Binding Extension
Technical changes from WCS version 1.1.2 include entirely building on the GML 3.2.1 Application Schema Coverages
and adoption of OWS Common 2.0. Another major change is the introduction of trim and slice concepts which is
explained in more detail below.
There are WCS 2.0 Schemas defined against which all requests and responses should validate.

WCS 2.0 KVP request parameters

The following KVP request parameters are available in WCS 2.0:

COVERAGEID=id: This parameter is technically the same as the COVERAGE parameter for WCS
1.0 or the IDENTIFIER parameter for WCS 1.1. In DescribeCoverage requests, multiple IDs can be
requested by concatenating them with commas.
SUBSET=axis[,crs](low,high): This parameter subsets the coverage on the given axis. This parameter
can be given multiple times, once for each axis. The optional sub-parameter can either be an EPSG

9.9. WCS Server 469

MapServer Documentation, Release 6.0

definition (like EPSG:4326), an URN or an URI or ‘imageCRS’ (which is the default). All crs sub-
parameters from all SUBSET parameters must be equal. (e.g: you cannot subset one axis in imageCRS
and another in EPSG:4326).
SIZE=axis(value): This parameter sets the size of the desired axis to the desired value (pixels).
RESOLUTION=axis(value): This parameter sets the resolution of the desired axis to the desired value
(pixels/unit).
Note: The SIZE and RESOLUTION are mutually exclusive on one axis, but can be mixed on different axes (e.g:
SIZE on x-axis and RESOLUTION on y-axis). Also axis names in SUBSET, SIZE and RESOLUTION parameters
cannot be mixed. E.g: ...&SUBSET=x(0,100)&SIZE=lon(200)&... is not legal although the axis names logically refer
to the same axis.
Note: Recognized values for the axis sub-parameter are: “x”, “xaxis”, “x-axis”, “x_axis”, “long”, “long_axis”, “long-
axis”, “lon”, “lon_axis”, “lon-axis”, “y”, “yaxis”, “y-axis”, “y_axis”, “lat”, “lat_axis” and “lat-axis”.
OUTPUTCRS=crs: This parameter defines in which crs the output image should be expressed in.
MEDIATYPE=mediatype: This parameter is relevant to GetCoverage requests, when multipart
XML/image output is desired. It should be set to ‘multipart/mixed’ (which is currently the only pos-
sible value for this parameter).
INTERPOLATION=intperolation_method: This defines the interpolation method used, for rescaled
images. Possible values are “NEAREST”, “BILINEAR” and “AVERAGE”.
RANGESUBSET=band1[,band2[,...]]: With this parameter, a selection of the bands can be made. Also
the bands can be reordered. The bands can be referred to either by name (which can be retrieved using
the DescribeCoverage request) or by index (starting with ‘1’ for the first band).

Unchanged KVP parameters The following parameters have not (or just slightly) changed since the last version of
the WCS standard.
VERSION=version: For WCS 2.0, this should be set to ‘2.0.0’.
SERVICE=service
REQUEST=request
ACCEPTVERSIONS=versions
SECTIONS=sections
UPDATESEQUENCE=updatesequence
ACCEPTFORMATS=formats: This parameter is currently ignored.
ACCEPTLANGUAGES=languages: This parameter is currently ignored.
FORMAT=format: The desired format can now also be set with the name of the outputformat object
defined in the mapfile.
MAP=mapfile
All other parameter keys are invalid in WCS 2.0 and will therefore produce an exception message.

KVP request examples The below sample request outline the new KVP request syntax:
# GetCapabilities
http://www.yourserver.com/wcs?SERVICE=wcs&VERSION=2.0.0&REQUEST=GetCapabilities
# DescribeCoverage 2.0
http://www.yourserver.com/wcs?SERVICE=wcs&VERSION=2.0.0&REQUEST=DescribeCoverage&COVERAGEID=grey
# GetCoverage 2.0 image/tiff full

470 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

http://www.yourserver.com/wcs?SERVICE=wcs&VERSION=2.0.0&REQUEST=GetCoverage&COVERAGEID=grey&FORMAT=im
# GetCoverage 2.0 multipart/mixed (GML header & image/tiff) full
http://www.yourserver.com/wcs?SERVICE=wcs&VERSION=2.0.0&REQUEST=GetCoverage&COVERAGEID=grey&FORMAT=im
# GetCoverage 2.0 image/tiff trim x y both
http://www.yourserver.com/wcs?SERVICE=wcs&VERSION=2.0.0&REQUEST=GetCoverage&COVERAGEID=grey&FORMAT=im
# GetCoverage 2.0 reproject to EPSG 4326
http://www.yourserver.com/wcs?SERVICE=wcs&VERSION=2.0.0&REQUEST=GetCoverage&COVERAGEID=grey&FORMAT=im

Please refer to the WCS 2.0 tests in msautotest for further sample requests.

Changes to previous versions

The layer name must be a valid NCName, i.e: must not start with a number and can only contain alphanumerical
characters. This constraint derives of the gml:id property which has to be a NCName, that relates to the coverage ID
which is itself taken from the layers name.

New band related metadata entries

In this section new WCS 2.0.0 specific layer metadata entries are discussed.
The following layer metadata fields can be used to return a more detailed description for the range type of a “virtual
dataset” coverage. A coverage is considered as a “virtual dataset” if the (wcs|ows)_extent metadata entry and one of
the (wcs|ows)_size or (wcs|ows)_resolution metadata entries are set.
First of all, the used version of metadata has to be identified. To identify the bands of a coverage, one of the following
fields must be present:
• (wcs|ows)_band_names (corresponding to WCS 2.0)
• (wcs|ows)_rangeset_axes (corresponding to WCS 1.1)
The type of these fields is a space delimited list of names, whereas the count of the names has to match the “bandcount”
metadata field. These names are then used as a prefix for other metadata fields only concerning this band. The possible
metadata keys are the following:
• WCS 2.0:
– {band_name}_band_interpretation
– {band_name}_band_uom
– {band_name}_band_definition
– {band_name}_band_description
– {band_name}_interval
• WCS 1.1
– {band_name}_semantic
– {band_name}_values_types
– {band_name}_values_semantic
– {band_name}_description
– {band_name}_interval

9.9. WCS Server 471

MapServer Documentation, Release 6.0

All values are interpreted as strings, only “interval” is interpreted as 2 double precision float values separated with a
space.
Also default values can be configured for every key. These have the same suffix as the band specific keys but start with
(wcs|ows) instead of the bands name:
• WCS 2.0:
– (wcs|ows)_band_interpretation
– (wcs|ows)_band_uom
– (wcs|ows)_band_definition
– (wcs|ows)_band_description
– (wcs|ows)_interval
• WCS 1.1
– (wcs|ows)_semantic
– (wcs|ows)_values_types
– (wcs|ows)_values_semantic
– (wcs|ows)_description
– (wcs|ows)_interval
If no specific or default value is given, the output is dependant on the metadata key. The UOM, for example will be
set to ‘W.m-2.Sr-1’, interval and significant figures will be determined according to the image type and definition,
description, and interpretation will not be visible in the output at all.
This example demonstrates the use of the band-specific metadata fields with their default values:
METADATA
"ows_srs" "EPSG:4326"
"wcs_extent" "47.5070762077246 16.038578977182 49.0103258976982 17.2500586851354"

"wcs_size" "1200 1100"

"wcs_imagemode" "BYTE"

"wcs_bandcount" "3"
"wcs_band_names" "BandA BandB BandC"

#default values
"wcs_band_interpretation" "This is default interpretation"
"wcs_band_uom" "DefaultUOM"
"wcs_band_definition" "DefaultDefinition"
"wcs_band_description" "This is default description"
"wcs_interval" "0 125"
"wcs_significant_figures" "3"

#specific band values

"BandA_band_interpretation" "This is a specific interpretation"
"BandA_band_uom" "SpecificUOM"
"BandA_band_definition" "SpecificDefinition"
"BandA_band_description" "This is a specific description"
"BandA_interval" "0 255"
END

472 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

The above example would result in having BandA a more specific description, and BandB and BandC having the
default description. It would also be possible to only use some of the specific values for BandA and others from the
default.
If no default and specific values are given for the interval or significant figures metadata field, the a default is generated
from the “imagemode” field, which itself defaults to FLOAT32.
The new metadata fields also contain the (wcs|ows)_nilvalues and (wcs|ows)_nilvalues_reasons
• (wcs|ows)_nilvalues
With this field, specific nilvalues can be set. The values have to be delimited by a space.
• (wcs|ows)_nilvalues_reasons
This field defines the reasons for the specific nilvalues. The reasons are also space delimited and reference the nilvalue
with the same index. The values for the reasons should be URIs or URNs.
The following example demonstrates the use of both metadata fields:
METADATA
"ows_srs" "EPSG:4326"
"wcs_extent" "47.5070762077246 16.038578977182 49.0103258976982 17.2500586851354"

"wcs_size" "1200 1100"

"wcs_imagemode" "BYTE"
"wcs_bandcount" "3"

"wcs_nilvalues" "0 255"

"wcs_nilvalues_reasons" "urn:ogc:def:nil:OGC::BelowDetectionLimit urn:ogc:def:nil:OGC::AboveDetecti
END

9.9.6 HTTP-POST support

Since version 6.0 MapServer also supports HTTP-POST XML requests. All requests possible via HTTP GET can also
be sent via POST. POST requests are possible for WCS 1.1 or WCS 2.0 which adhere to the according standard.
This is an example GetCapabilities request:
<?xml version="1.0" encoding="UTF-8"?>
<ows:GetCapabilities service="WCS">
<ows:AcceptVersions>
<ows:Version>2.0.0</ows:Version>
</ows:AcceptVersions>
<ows:updatesequence>2010-12-21</ows:updatesequence>
<ows:Sections>
<ows:Section>OperationsMetadata</ows:Section>
<ows:Section>ServiceIdentification</ows:Section>
</ows:Sections>
</ows:GetCapabilities>

This is an example DescribeCoverage request, which is only valid for WCS 2.0:
<wcs:DescribeCoverage
xmlns:xsi=’http://www.w3.org/2001/XMLSchema-instance’
xsi:schemaLocation="http://www.opengis.net/wcs/2.0 http://schemas.opengis.net/wcs/2.0/wcsAll.xsd"
xmlns="http://www.opengis.net/wcs/2.0"
xmlns:wcs="http://www.opengis.net/wcs/2.0"
service="WCS"
version="2.0.0">

9.9. WCS Server 473

MapServer Documentation, Release 6.0

<wcs:CoverageID>SOME_ID</wcs:CoverageID>
</wcs:DescribeCoverage>

This example demonstrates the usage of a WCS 2.0 POST-XML GetCoverage request:
<?xml version="1.0" encoding="UTF-8"?>
<wcs:GetCoverage
xmlns:xsi=’http://www.w3.org/2001/XMLSchema-instance’
xsi:schemaLocation="http://www.opengis.net/wcs/2.0 http://schemas.opengis.net/wcs/2.0/wcsAll.xsd"
xmlns="http://www.opengis.net/wcs/2.0"
xmlns:wcs="http://www.opengis.net/wcs/2.0"
service="WCS"
version="2.0.0">
<wcs:Format>image/tiff</wcs:Format>
<wcs:Mediatype>multipart/mixed</wcs:Mediatype>
<wcs:Resolution dimension="x">0.01</wcs:Resolution>
<wcs:Size dimension="y">50</wcs:Size>
<wcs:CoverageID>SOME_ID</wcs:CoverageID>
<wcs:DimensionTrim>
<wcs:Dimension crs="http://www.opengis.net/def/crs/EPSG/0/4326">x</wcs:Dimension>
<wcs:TrimHigh>17.25</wcs:TrimHigh>
<wcs:TrimLow>16.5</wcs:TrimLow>
</wcs:DimensionTrim>
<wcs:DimensionTrim>
<wcs:Dimension crs="http://www.opengis.net/def/crs/EPSG/0/4326">y</wcs:Dimension>
<wcs:TrimLow>47.9</wcs:TrimLow>
</wcs:DimensionTrim>
</wcs:GetCoverage>

Please refer to the WCS 2.0 Specification - XML/POST Protocol Binding Extension and the WCS 2.0 Schemas for
further information on POST request in WCS 2.0.

9.9.7 Reference Section

To avoid confusion only “wcs_*” and “ows_*” prefixed metadata entries are evaluated in OGC WCS services. Pre-
vious versions used “wms_*” prefixed entries as fallback which is dropped in version 6.0 in favor of forcing explicit
decisions. The module will look for the “wcs_*” and “ows_*” metadata prefixes in this order.
The following metadata are available in the setup of the mapfile:

Web Object Metadata

wcs_abstract
• Description: (Optional) A brief description of the service, maps to ows:Abstract (WCS 1.1+ only).
wcs_accessconstraints
• Description: (Optional) A list of codes describing any access constraints imposed by the service provider. The
keyword NONE is reserved to mean no access constraints are imposed.
wcs_address, wcs_city, wcs_contactelectronicmailaddress, wcs_contactfacimiletelephone,
wcs_contactorganization, wcs_contactperson, wcs_contactposition, wcs_contactvoicetelephone, wcs_country,
wcs_postcode, wcs_service_onlineresource, wcs_stateorprovince
• Description: (Optional) Contact address information. If provided then all twelve metadata items are required.
You can also use the responsibleparty* metadata instead.

474 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

wcs_description
• Description: (Optional) A description of the server.
wcs_fees
• Description: (Optional) A text string indicating any fees imposed by the service provider.
wcs_keywords
• Description: (Optional) Short words for catalog searching.
wcs_label
• Description: (Required) A human-readable label for the server.
wcs_metadatalink_format
• Description: (Optional) The file format MIME type of the metadata record (e.g. “text/plain”). The web metadata
wcs_metadatalink_type and wcs_metadatalink_href must also be specified.
wcs_metadatalink_href
• Description: (Optional) The URL to the server’s metadata. The web metadata wcs_metadatalink_format and
wcs_metadatalink_type must also be specified.
wcs_metadatalink_type
• Description: (Optional) The standard to which the metadata complies. Currently only two types are valid:
“TC211” which refers to [ISO 19115], and “FGDC” which refers to [FGDC-STD-001-1988]. The web metadata
wcs_metadatalink_format and wcs_metadatalink_href must also be specified.
wcs_name
• Description: (Optional) A name for the server.
wcs_responsibleparty_address_administrativearea, wcs_responsibleparty_address_city,
wcs_responsibleparty_address_country, wcs_responsibleparty_address_deliverypoint,
wcs_responsibleparty_address_electronicmailaddress, wcs_responsibleparty_address_postalcode,
wcs_responsibleparty_individualname, wcs_responsibleparty_onlineresource, wcs_responsibleparty_organizationname,
wcs_responsibleparty_phone_facsimile, wcs_responsibleparty_phone_voice, wcs_responsibleparty_postionname
• Description: (Optional) Contact address information. If provided then all twelve metadata items are required.
You can also use the address* metadata instead.

Layer Object Metadata

wcs_abstract
• Description: (Optional) A brief description of the service, maps to ows:Abstract (WCS 1.1+ only).
wcs_description
• Description: (Optional) A description of the layer.
wcs_extent
• Description: (Optional) Bounding box of layer, which must be provided for tiled data. Comma-delimited, in
the format of: minx,miny,maxx,maxy
wcs_formats
• Description: (Optional) The formats which may be requested for this layer, separated by a space. (e.g. “GTiff
MrSID”)
wcs_keywords

9.9. WCS Server 475

MapServer Documentation, Release 6.0

• Description: (Optional) Short words for catalog searching.

wcs_label
• Description: (Required) A human-readable label for the layer.
wcs_metadatalink_format
• Description: (Optional) The file format MIME type of the metadata record (e.g. “text/plain”). The web metadata
wcs_metadatalink_type and wcs_metadatalink_href must also be specified.
wcs_metadatalink_href
• Description: (Optional) The URL to the layer’s metadata. The web metadata wcs_metadatalink_format and
wcs_metadatalink_type must also be specified.
wcs_metadatalink_type
• Description: (Optional) The standard to which the metadata complies. Currently only two types are valid:
“TC211” which refers to [ISO 19115], and “FGDC” which refers to [FGDC-STD-001-1988]. The web metadata
wcs_metadatalink_format and wcs_metadatalink_href must also be specified.
wcs_name
• Description: (Optional) A name for the layer.
wcs_nativeformat
• Description: (Optional) The current format of the served raster layer. (e.g. “GTiff”)
Axes Descriptions
MapServer allows you define a number of these for a layer. Individual axis are identified by name when defining
specific metadata (e.g. description). All defined axes must be listed in the rangeset_axes metadata tag so MapServer
knows in advance what to expect. A special rangeset for multiband date is automatically generated by adding the name
“bands” to the rangeset_axes list. If found MapServer will automatically generate metadata for the image bands. You
may of course extend that basic support using the naming conventions below.
wcs_rangeset_axes
• Description: (Optional) Delimited list of defined range sets. If defined, you can also use the following nine
metadata items, where rangeset axis matches the axis name provided in this wcs_rangeset_axes metadata:
{rangeset axis}_semantic
{rangeset axis}_refsys
{rangeset axis}_refsyslabel
{rangeset axis}_description
{rangeset axis}_label
{rangeset axis}_values
{rangeset axis}_values_semantic
{rangeset axis}_values_type
{rangeset axis}_interval
wcs_rangeset_label
• Description: (Required for DescribeCoverage request)
wcs_rangeset_name
• Description: (Required for DescribeCoverage request)

476 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

wcs_srs
• Description: (Optional) Spatial reference system of the layer, in the form of: EPSG:code (e.g. EPSG:42304)
wcs_timeitem
• Description: (Optional) The attribute in the spatio/temporal index that contains time values.
wcs_timeposition
• Description: (Optional) A list of the start and end time of a given coverage (i.e. “2000-11-11T11:11:11Z,2001-
11-11T11:11:11Z”), used when advertising GetCapabilities.

9.9.8 Rules for handling SRS in a MapServer WCS

TODO!

9.9.9 Spatio/Temporal Indexes

MapServer has long supported a method of breaking a dataset into smaller, more manageable pieces or tiles. In this
case a shapefile is used to store the boundary of each tile, and an attribute holds the location of the actual data. Within
a MapServer mapfile the layer keywords TILEINDEX and TILEITEM are used to activate tiling.
Consider the example where an organization wants to serve hundreds or even thousands of MODIS scenes. Five
images cover the spatial extent and each group of five varies by date of acquisition. This turns out to be a fairly
common scenario for organizations interested in WCS, one that the existing tiling support does not adequately address.
In previous versions of MapServer a developer would have to create one tile index and one layer definition for each
group of five images. This could result in configuration files that are prohibitively long and difficult to manage.
In order to more efficiently support the WCS specification a new tiling scheme has been implemented within
MapServer. One that supports spatial sub-setting, but also ad hoc sub-setting based on any attributes found within
tile index. In many cases a temporal attribute could be used, but sub-setting is not limited to that case. The new
scheme introduces the concept of tile index layers, that is, a separate layer definition is used to describe the tile index
dataset. With this we get all the benefits of any MapServer layer, most importantly we can apply MapServer filters
to the data. Filters can be defined at runtime using MapServer CGI, MapScript or via the WCS server interface. The
syntax for the layer using the index remains unchanged except that the value for Tile Indexes refers to the index layer
instead of an external shapefile.
So, looking at the example above again we can reduce our MapServer configuration to two layer definitions, one for
the tile index and one for the imagery itself. Extracting a single dates worth of imagery is now a matter of setting the
appropriate filter within the tile index layer.

Building Spatio-Temporal Tile Indexes

Developing these tile indexes is more difficult than basic indexes simply because there are no ready-made tools to do
so. Fortunately we can leverage existing tool available within MapServer or supporting libraries such as GDAL by
post processing their output.
Taking the above example, building an index is relatively simple task if you are willing to roll up your sleeves and
write a bit of code. First, the basic spatial index needs to be built. The GDAL utility gdaltindex already does this.
Simply point gdaltindex at the directory containing the collection of MODIS images and it will build a shapefile index
suitable for use with MapServer. The next step would be to add the temporal information. The pseudo code would
look something like:
• open the index .dbf file for reading
• create a new column to hold the image acquisition date

9.9. WCS Server 477

MapServer Documentation, Release 6.0

• for each image; 1) extract the image acquisition date and 2) insert it into the new column
• close the index .dbf file
This general approach could be used for many cases. A scripting language such as Perl, PHP or Python works well
since they all have readily available modules for manipulating .dbf files. A worst case would involve hand editing the
resulting .dbf file using a desktop tool such as Mircosoft Access or ESRI Arcview.

9.9.10 WCS 2.0 Application Profile - Earth Observation (EO-WCS)

OGC is currently discussing the adoption of an Earth Observation (EO) Application Profile for WCS 2.0 (EO-WCS).
A draft will be available soon for a public RFC period. For an implementation please refer to the Open Source project
EOxServer which already implements this proposed EO-WCS based on MapServer.

9.9.11 To-do Items and Known Limitations

• MapServer does not derive all of the metadata it could from a given dataset. For example, you must explicitly
list time periods covered by a layer. This should get better with time.
• Only spatial, simple temporal and radiometric band subsetting is possible with the current implementation.
Furture enhancements should allow for arbitrary subsets based on pixel values or tile/image attributes.
• The available set of WCS 2.0 specification documents is not yet complete. Thus, for some implementation
details, the content of some forthcoming extensions had to be anticipated based on the approaches taken for WCS
1.1 and 1.0. The implementation will be adjusted as soon as new specification documents become available.
• If you want to use libxml2 or its derived tools (like xmllint) for validation be aware that there is a currently bug
in libxml2 that breaks the validation of GML 3.2.1.

9.10 WCS Use Cases

Author Norman Barker

Contact nbarker at ittvis.com
Author Gail Millin
Contact nbarker at ittvis.com
Revision $Revision: 8295 $
Date $Date: 2008-12-26 21:08:04 -0800 (Fri, 26 Dec 2008) $
Last Updated 2005/12/12

Contents
• WCS Use Cases
– Landsat
– SPOT
– DEM
– NetCDF

This document explains how to use MapServer to deliver Landsat, SPOT, DEM, and NetCDF temporal/banded data
through the MapServer WCS interface. Thanks go to Steve Lime and Frank Warmerdam for their assistance with these
projects

478 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

9.10.1 Landsat

To serve Landsat imagery through the MapServer Web Coverage Service specify the OUTPUTFORMAT object. For
format support install the GDAL library and from the command prompt and cd to where GDAL is installed and use
the command, gdalinfo –formats. A list of all supported formats will appear and will specify if the format is read only
<ro> or read and write <rw> for WCS the format needs to be supported for read and write (except for GDAL’s own
WCS format, however).
For the example below the Landsat 7 15m resolution mosaic is in a Enhanced Compressed Wavelets format (ECW).
By running the gdalinfo.exe program I could verify that the ECW format has write permissions, therefore the format
can be specified in the MapFile and requested using the GetCoverage request.
OUTPUTFORMAT
NAME "ECW"
DRIVER "GDAL/ECW"
MIMETYPE "image/ecw"
IMAGEMODE "BYTE"
EXTENSION "ecw"
END

The keyword “DUMP” must be set to true to allow the data to download during a GetCoverage request.
LAYER
NAME "Landsat7"
STATUS OFF
TYPE RASTER
PROCESSING "SCALE=AUTO"
UNITS Meters
DUMP TRUE
TILEINDEX "MapServer/wcs/landsat7/l7mosaic15m.shp"
TILEITEM "Location"
METADATA
"wcs_description" "Landsat 7 15m resolution mosaic"
"wcs_name" "Landsat7"
"wcs_label" "Landsat 7 15m resolution mosaic"
"ows_srs" "EPSG:27700"
"ows_extent" "0 0 700005 1050000"
"wcs_resolution" "75 75"
"wcs_bandcount" "3"
"wcs_formats" "ECW"
END
END

A GetCoverage request can then be requested (using the parameters set in the MapFile) by creating a URL with the
elements: - Your Server, MapServer Program, Location of MapFile, Type of Service (WCS), Request (GetCoverage),
Coverage (Landsat7), BBOX (0,0,700005,1050000), CRS (EPSG:27700), ResX (75) ResY (75), Format (ECW).

9.10.2 SPOT

SPOT imagery can be delivered through MapServer Web Coverage Service similarly to the Landsat example above.
The main difference is that as SPOT is a greyscale image the wcs_bandcount = 1 rather than a Landsat image which
consists of 3 bands. For this example the well known GeoTiff format will be used to demonstrate what to specify in a
MapFile for SPOT data.
OUTPUTFORMAT
NAME "GEOTIFF"
DRIVER "GDAL/GTiff"
MIMETYPE "image/tiff"

9.10. WCS Use Cases 479

MapServer Documentation, Release 6.0

IMAGEMODE "BYTE"
EXTENSION "tif"
END
LAYER
NAME "SPOT"
STATUS OFF
TYPE RASTER
PROCESSING "SCALE=AUTO"
UNITS Meters
DUMP TRUE
TILEINDEX "MapServer/wcs/orthospot/spot.shp"
TILEITEM "Location"
METADATA
"wcs_description" "Orthospot mosaic"
"wcs_name" "SPOT"
"wcs_label" "Orthospot mosaic"
"ows_srs" "EPSG:27700"
"ows_extent" "375960 64480 497410 200590"
"wcs_resolution" "100 100"
"wcs_bandcount" "1"
"wcs_formats" "GEOTIFF"
"wcs_nativeformat" "8-bit GeoTIF"
END
END

The key parameters to specify in the WCS MapFile for any data layer and format are:
- Layer Name = Create a short name for the data
- Layer Type = Raster
- Layer Dump = True

The following examples further demonstrate how WCS can be implemented and also how to create WCS containing
layers with a temporal dimension (see NetCDF example).

9.10.3 DEM

It is possible to deliver 16 bit DEM data through the MapServer Web Coverage Service.
Firstly it is necessary to specify the output format in the map file
OUTPUTFORMAT
NAME "GEOTIFFINT16"
DRIVER "GDAL/GTiff"
MIMETYPE "image/tiff"
IMAGEMODE "INT16"
EXTENSION "tif"
END

and in the corresponding layer to specify the keyword “DUMP” to be true

LAYER
NAME "srtm"
STATUS OFF
TYPE RASTER
DUMP TRUE
DATA "srtm.tif"
PROJECTION
"init=epsg:4326"

480 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

END
METADATA
"wcs_label" "SRTM WCS TIF Server"
"ows_extent" "-180 -90 180 90"
"wcs_resolution" "0.00083 -0.00083"
"ows_srs" "EPSG:4326"
"wcs_formats" "GEOTIFFINT16"
"wcs_nativeformat" "geotiff"
END
END

Performance gains can be made by using the gdaladdo utility described at

http://www.gdal.org/gdal_utilities.html#gdaladdo

9.10.4 NetCDF

Firstly GDAL doesn’t support all versions of netCDF (there are a lot, it is a generic format), so for stability it
may be necessary to convert the files into GeoTiff format first. This can be achieved using the netCDF libraries
here http://my.unidata.ucar.edu/content/software/netcdf/index.html. Denis Nadeau and Frank Warmerdam have added
netCDF CF as a read only format within GDAL, so it now possible to read the CF convention netCDF files directly
from disk.
We placed the Z-levels in the bands of the GDAL data file (either GeoTiff or netCDF), and created a shape index for
the time levels. GDAL data is a 2-D format (x,y) and bands. netCDF is an N-D file format, supporting time, x,y,z, and
experiment parameters. By using a set of GDAL netCDF / geoTiff files it is possible to represent this, and to store the
z-level (height) as bands within the data file. Although a hack, it is possible for a custom client to receive important
metadata from the describeCoverage operation of a WCS about the which z-level a band of a geotiff represents by
encoding this in the returned axes description tag.
To create the shape file for the temporal dimension we had to do some hacking with Java code, but we also got it to
work with Steve Lime’ s perl script in the MODIS MapServer demo download (which doesn’t seem to be available
now).
The perl script used in Modis demo by Steve Lime is as follows, and I have placed inline comments below. The script
assumes that gdaltindex has already been run in this directory to create a tile index shape and dbf file. It assumes that
the filenames of your data files have the date in the filename, for example myfileYYYYMMDDHH.tif
1 #!/usr/bin/perl
2 use XBase;
3 opendir(DIR, ’.’); # open the current directory
4 foreach $file (readdir(DIR)) {
5 next if !($file =~ /\.dbf$/); # read the dbf file in this directory created by gdaltindex
6 print "Working on $file...\n";
7 $tfile = ’temporary.dbf’;
8 system("mv $file $tfile");
9 $oldtable = new XBase $tfile or die XBase->errstr;
10 print join("\t", $oldtable->field_names) ."\n";
11 print join("\t", $oldtable->field_types) ."\n";
12 print join("\t", $oldtable->field_lengths) ."\n";
13 print join("\t", $oldtable->field_decimals) ."\n";
14 $newtable = XBase->create("name" => $file,
15 "field_names" => [$oldtable->field_names, "IMGDATE"], # this is the FILTERITEM in
16 "field_types" => [$oldtable->field_types, "C"], # character column type
17 "field_lengths" => [$oldtable->field_lengths, 13], # length of the date string
18 "field_decimals" => [$oldtable->field_decimals, undef]) or die "Error creating new
19 foreach (0 .. $oldtable->last_record) {
20 ($deleted, @data) = $oldtable->get_record($_);

9.10. WCS Use Cases 481

 MapServer Documentation, Release 6.0

21 print " ...record $data[0]\n";

22 # extract the date
23 $year = substr $data[0], 8, 4; # year is at position 8 in the filename string
24 $month = substr $data[0], 12, 2; # month is at position 12 in the filename string
25 $day = substr $data[0], 14, 2; # day is at position 14 in the filename string
26 $hour = substr $data[0], 16, 2; # hour is at position 16 in the filename string
27 $date = "$year-$month-$day" . "T" . "$hour\n"; # format is YYYY-MM-DDTHH, or any ISO format
28 print "$date";
29 push @data, "$date";
30 $newtable->set_record($_, @data);
31 }
32 $newtable->close();
33 $oldtable->close();
34 unlink($tfile);
35 }

If have used the perl script then skip to the layer definitions below, if you wish to code your own the description is
here.
The DBF file has to have the column ‘location’ that indicates the location of the data file (either absolute path or
relative to the map file location, and the second column that can be called whatever you want but indexes time. In our
case we called it ‘time’ :-)
The corresponding shapefile then has to contain Polygons with the bounding boxes of the tif file for each time. So
OGRInfo timeIndex.shp looks something like:
OGRFeature(timeIndex):116
location(String) = mytime.tif
time(String) = 2001-01-31T18:00:00
POLYGON ((xxx,xxxx,.......))

Define your output format as

OUTPUTFORMAT
NAME "GEOTIFF_FLOAT"
DRIVER ’GDAL/GTiff’
MIMETYPE ’image/tiff’
IMAGEMODE FLOAT32
EXTENSION ’tif’
END

Then you need to define your tile index within the map file
LAYER
NAME ’time_idx’
TYPE TILEINDEX
DATA ’timeIndex’
FILTERITEM ’time’
FILTER ’%time%’
END

and the actual layer

LAYER
NAME ’TempData’
STATUS OFF
TYPE RASTER
DUMP TRUE
TILEINDEX ’time_idx’
PROJECTION

482 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

"init=epsg:4326"
END
METADATA
"wcs_label" ’Temperature data’
"ows_extent" ’-180 -90 180 90’
"wcs_resolution" ’1.125 -1.125’
"ows_srs" ’EPSG:4326’
"wcs_formats" ’GEOTIFF_FLOAT’
"wcs_nativeformat" ’netCDF’
"wcs_bandcount" ’27’
"wcs_rangeset_axes" ’bands’
"wcs_rangeset_label" ’Pressure (hPa units) Levels’
"wcs_rangeset_name" ’bands’
"wcs_rangeset_description" ’Z levels ’
"wcs_timeposition" ’2001-01-01T06:00:00,2001-01-01T12:00:00,2001-01-01T18:00:00,2001-01-02T00:00:
"wcs_timeitem" ’time’
END
END

The TempData coverage layer will now let you subset with the &bands=... &time=... subset parameters!
To do a coordinate reprojection specify in the request &Response_CRS=ESPG:xxxx
When you start doing temporal subsetting with WCS and MapServer you can see the need for an automatic way of
generating map files such as using an XSL stylesheet!
For a tile-index layer you need to provide the following extra metadata in order to use it for WCS:
"OWS_EXTENT" "10050 299950 280050 619650"
"WCS_RESOLUTION" "100 100"
"WCS_SIZE" "2700 3197"
"WCS_BANDCOUNT" "3"

If your image has a colortable and only one band, it will come out greyscale unless you set the IMAGEMODE to
PC256 instead of BYTE.

9.11 SOS Server

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Revision $Revision: 9024 $
Date $Date: 2009-05-19 07:14:40 -0700 (Tue, 19 May 2009) $
Last Updated 2007/12/06

Table of Contents
• SOS Server
– Introduction
– Setting Up an SOS Server Using MapServer
– Limitations / TODO
– Reference Section
– Use of sos_procedure and sos_procedure_item

9.11. SOS Server 483

MapServer Documentation, Release 6.0

9.11.1 Introduction

SOS (Sensor Observation Service), currently an OGC discussion paper, is part of of the OGC’s SensorWeb Enablement
(SWE) group of specifications. These specifications describe how applications and services will be able to access
sensors of all types over the Web. Specifically, SOS provides an API for managing deployed sensors and retrieving
sensor data.
SOS support is available in MapServer 4.10.0 or more recent. Note that no client tools currently exist in MapServer
for SOS. More SWE based software is available at http://www.52north.org/
SOS support was implemented in MapServer to the guidelines of MapServer MS RFC 13: Support of Sensor Obser-
vation Service in MapServer.
This document assumes that you are already familiar with certain aspects of MapServer:
• MapServer application development and setting up .map files.

Links to SOS-Related Information

• SOS discussion paper

• Sensor Web Enablement and OpenGIS SensorWeb

Relevant Definitions

The following is taken from the SOS discussion paper:

Observation An observation is an event with a result which has a value describing some phenomenon.
Observation Offering An observation offering is a logical grouping of observations offered by a service that are
related in some way.
Observed Value A value describing a natural phenomenon, which may use one of a variety of scales including nom-
inal, ordinal, ratio and interval.
Sensor An entity capable of observing a phenomenon and returning an observed value. A sensor can be an instrument
or a living organism (e.g. a person).

9.11.2 Setting Up an SOS Server Using MapServer

Install the Required Software

SOS requests are handled by the “mapserv” CGI program. The first step is to check that your mapserv exe-
cutable includes SOS support. One way to verify this is to use the “-v” command-line switch and look for “SUP-
PORTS=SOS_SERVER”.
Example 1. On Unix:
$ ./mapserv -v
MapServer version 4.9 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=SVG S
UPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT SUPPORTS=
WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER SUPPORTS=SOS_SERVER SUPPORTS=
THREADS INPUT=JPEG INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE DEBUG=MSDEBUG

Example 2. On Windows:

484 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

C:\Apache\cgi-bin> mapserv -v
MapServer version 4.9 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP OUTPUT=SVG S
UPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT SUPPORTS=
WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER SUPPORTS=SOS_SERVER SUPPORTS=
THREADS INPUT=JPEG INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE DEBUG=MSDEBUG

If you don’t have SOS support in your MapServer build, then you must compile MapServer with the following in
mind:
• flag -DUSE_SOS_SVR is required
• requires either -DUSE_WMS_SVR or -DUSE_WFS_SVR flags to be enabled
• requires libxml2 and proj libraries
• requires ICONV support (-DUSE_ICONV) on Windows
For more help with MapServer compilation see the appropriate HowTo: Unix / Windows

Configure a Mapfile For SOS

Each instance of SOS server that you setup needs to have its own mapfile. It is just a regular MapServer mapfile in
which some parameters and some metadata entries are mandatory. Most of the metadata is required in order to produce
a valid GetCapabilites output.
Here is the list of parameters and metadata items that usually optional with MapServer, but are required (or strongly
recommended) for a SOS configuration:
MAP level:
• Map NAME
• Map PROJECTION
• Map Metadata (in the WEB Object):
– sos_title
– sos_onlineresource
– sos_srs
– see the Reference Section of this document for a full list of metadata and descriptions
LAYER level:
• Layer NAME
• Layer PROJECTION
• Layer METADATA
– sos_offering_id
– sos_observedproperty_id
– sos_observedproperty_id
– sos_describesensor_url
– see the Reference Section of this document for a full list of metadata and descriptions

9.11. SOS Server 485

MapServer Documentation, Release 6.0

Onlineresource URL

The sos_onlineresource metadata is set in the map’s web object metadata and specifies the URL that should be used
to access your server. This is required for the GetCapabilities output. If sos_onlineresource is not provided then
MapServer will try to provide a default one using the script name and hostname, but you shouldn’t count on that too
much. It is strongly recommended that you provide the wfs_onlineresource metadata.
Here is a valid online resource URL:
http://my.host.com/cgi-bin/mapserv?map=mysos.map&

By creating a wrapper script on the server it is possible to hide the “map=” parameter from the URL and then your
server’s online resource URL could be something like:
http://my.host.com/cgi-bin/mapserv?

This is covered in more detail in the “More About the Online Resource URL” section of the WMS Server document.

Example SOS Server Mapfile

The following is an example of a bare minimum SOS Server mapfile. Note the comments for the required parameters.
NAME "SOS_DEMO"
STATUS ON
SIZE 300 300
EXTENT -66 44 -62 45
UNITS METERS
SHAPEPATH "./data/"
IMAGECOLOR 255 255 0
SYMBOLSET "./etc/symbols.sym"

IMAGETYPE png

WEB
IMAGEPATH "/ms4w/tmp/ms_tmp/"
IMAGEURL "/ms_tmp/"

METADATA
"sos_onlineresource" "http://127.0.0.1/cgi-bin/mapserv.exe?map=D:/ms4w/apps/sos/sos_test.map" ## R
"sos_title" "My SOS Demo Server" ## Recommended
"sos_srs" "EPSG:4326" ## REQUIRED
END
END

PROJECTION
"init=epsg:4326"
END

LAYER
NAME "test_sos_layer"
METADATA
"sos_procedure" "NS01EE0014" ## REQUIRED
"sos_offering_id" "WQ1289" ## REQUIRED
"sos_observedproperty_id" "Water Quality" ## REQUIRED
"sos_describesensor_url" "http://some/url/NS01EE0014.xml" ## REQUIRED
END
TYPE POINT
STATUS ON

486 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

DATA "sos_test"

PROJECTION
"init=epsg:4326"
END

CLASS
NAME "water quality"
STYLE
COLOR 255 0 0
SYMBOL "circle"
SIZE 8
END
END
END

END #map

Test Your SOS Server

GetCapabilities Request

The GetCapabilities request allows the clients to retrieve service metadata about a specific service instance. For an
SOS service, it allows to identify such things as offerings and observed property available, as well as information on
sensors that are used.
Using a web browser, access your server’s online resource URL to which you add the parameters “SER-
VICE=SOS&REQUEST=GetCapabilities” to the end, e.g.
http://my.host.com/cgi-bin/mapserv?MAP=mysos.map&SERVICE=SOS&REQUEST=GetCapabilities

If everything went well, you should have a complete XML capabilities document. Search it for the word “WARN-
ING”... MapServer inserts XML comments starting with “<!–WARNING: ” in the XML output if it detects missing
mapfile parameters or metadata items. If you notice any warning in your XML output then you have to fix all of them
before you can try your server with an SOS client, otherwise things are likely not going to work.
Note: The SERVICE parameter is required for all SOS requests.

GetObservation Request

The GetObservation request is designed to query sensor systems to retrieve observation data in the form defined in
the Observation and Measurement specification (O&M), and more information on this O&M spec can be found at
http://www.opengeospatial.org/functional/?page=swe. Upon receiving a GetObservation request, a SOS shall either
satisfy the request or return an exception report.
The following is a list of the possible parameters for a GetObservation request:
request: (Required) value must be “GetObservation”.
service: (Required) value must be “SOS”.
version: (Required) value must be “1.0.0”.
offering: (Required) The Offering identified in the capabilities document.
observedProperty: (Required) The property identified in the capabilities document.

9.11. SOS Server 487

MapServer Documentation, Release 6.0

responseFormat: (Required) The format / encoding to be returned by the response.

eventTime (Optional) Specifies the time period for which observations are requested.
procedure: (Optional) The procedure specifies the sensor system used. In this implementation, the pro-
cedure is equivalent to be the sensor id that will be used when doing a DescribeSensor request.
featureOfInterest: (Optional) In this implementation, this will be represented by a gml envelope defining
the lower and upper corners.
Result: (Optional) The Result parameter provides a place to put OGC filter expressions based on property
values.
resultModel: (Optional) Identifier of the result model to be used for the requested data. The resultModel
values supported by a SOS server are listed in the contents section of the service metadata (GetCapabil-
ities). MapServer currently supports om:Observation and om:Measurement. om:Measurement provides
a flat model of the geometry and attributes, similar to WFS GetFeature output. om:Observations pro-
vides a more compact definition which includes an XML header of the field names and defintions, fol-
lowed by a “DataBlock” of delimited records (default is CSV delimited output). The default output is
om:Measurement.
srsName: (Optional) srs (EPSG code) of the output response.
Here are some valid examples:
Example 1:
http://127.0.0.1/cgi-bin/mapserv.exe?map=D:/ms4w/apps/sos/sos_test.map&Request=GetObservation&service
Offering=WQ1289&observedproperty=Water Quality&version=1.0.0&responseFormat=text/xml; subtype="om/1.0

Example 2:
http://127.0.0.1/cgi-bin/mapserv.exe?map=D:/ms4w/apps/sos/sos_test.map&Request=GetObservation&service
Offering=WQ1289&observedproperty=Water Quality&eventtime=<ogc:TM_Equals><gml:TimePeriod><gml:beginPos
<PropertyName>COLOUR</PropertyName><Literal>180</Literal></PropertyIsEqualTo><PropertyIsEqualTo>
<PropertyName>COLOUR</PropertyName><Literal>200</Literal></PropertyIsEqualTo></or></Filter>&version=1

Example 3:
http://127.0.0.1/cgi-bin/mapserv.exe?map=D:/ms4w/apps/sos/sos_test.map&Request=GetObservation&service
Offering=WQ1289&observedproperty=Water Quality&featureofinterest=
<gml:Envelope><gml:lowerCorner srsName=’EPSG:4326’>-66 43</gml:lowerCorner><gml:upperCorner srsName=’
-64 45</gml:upperCorner></gml:Envelope>&version=1.0.0&responseFormat=text/xml; subtype="om/1.0.0"

Example 4:
http://127.0.0.1/cgi-bin/mapserv.exe?map=D:/ms4w/apps/sos/sos_test.map&Request=GetObservation&service
Offering=WQ1289&observedproperty=Water Quality&version=1.0.0&responseFormat=text/xml; subtype="om/1.0

DescribeSensor Request

The DescribeSensor request gives the client the ability to retrieve the characteristics of a particular sensor and return
the information in a SensorML xml document. In this implementation, MapServer does not generate the SensorML
document but only redirect the request to an existing SensorML document.
The following is a list of the possible parameters for a DescribeSensor request:
request: (Required) value must be “DescribeSensor”
service: (Required) value must be “SOS”.

488 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

version: (Required) value must be “1.0.0”.

procedure: (Required) This is the sensor id, which was specified in the “sos_procedure” metadata.
outputFormat: (Required) The format encoding to be returned by the response.
Here is a valid example:
http://127.0.0.1/cgi-bin/mapserv.exe?map=D:/ms4w/apps/sos/sos_test.map&Request=DescribeSensor&
procedure=urn:ogc:def:procedure:NS01EE0014&service=SOS&version=1.0.0&outputFormat=text/xml; subtype="

9.11.3 Limitations / TODO

1. Have MapServer generate the SensorML document, instead of redirecting the request to an existing SensorML
document.

9.11.4 Reference Section

The following metadata are available in the setup of the SOS Server mapfile:
Note: Each of the metadata below can also be referred to as ‘ows_*’ instead of ‘sos_*’. MapServer tries the ‘sos_*’
metadata first, and if not found it tries the corresponding ‘ows_*’ name. Using this reduces the amount of duplication
in mapfiles that support multiple OGC interfaces since “ows_*” metadata can be used almost everywhere for common
metadata items shared by multiple OGC interfaces.

Web Object Metadata

ows_updatesequence
• Description: (Optional) The updateSequence parameter can be used for maintaining the consistency of a client
cache of the contents of a service metadata document. The parameter value can be an integer, a timestamp in
[ISO 8601:2000] format, or any other number or string.
ows_language
• Description: (Optional) Descriptive narrative for more information about the server. Identifier of the language
used by all included exception text values. These language identifiers shall be as specified in IETF RFC 1766.
When this attribute is omitted, the language used is not identified. Examples: “en-CA”, “fr-CA”, “en-US”.
Default is “en-US”.
ows_schemas_location
• Description: (Optional) (Note the name ows_schemas_location and not sos/_... this is because all OGC Web
Services (OWS) use the same metadata) Root of the web tree where the family of OGC SOS XMLSchema files
are located. This must be a valid URL where the actual .xsd files are located if you want your SOS output to val-
idate in a validating XML parser. Default is http://www.opengeospatial.net/sos. See http://ogc.dmsolutions.ca
for an example of a valid schema tree.
sos_abstract
• Description: (Optional) Descriptive narrative for more information about the server.
sos_keywordlist
• Description: (Optional) A comma-separated list of keywords or keyword phrases to help catalog searching.
sos_accessconstraints

9.11. SOS Server 489

MapServer Documentation, Release 6.0

• Description: (Optional) Text describing any access constraints imposed by the service provider on the SOS or
data retrieved from this service.
sos_addresstype, sos_address, sos_city, sos_stateorprovince, sos_postcode, sos_country
• Description: Optional contact address information. If provided then all six metadata items are required.
sos_contactelectronicmailaddress
• Description: Optional contact Email address.
sos_contactperson, sos_contactposition, sos_contactorganization
• Description: Optional contact information. If provided then all three metadata items are required.
sos_contactvoicetelephone
• Description: Optional contact voice telephone number.
sos_contactfacsimiletelephone - * Description: Optional contact facsimile telephone number.
sos_fees
• Description: (Optional) Fees information. Use the reserved word “none” if there are no fees.
sos_onlineresource
• Description: (Required) The URL that will be used to access this OGC server. This value is used in the
GetCapabilities response.
• See the section “Onlineresource URL” above for more information.
sos_service_onlineresource
• Description: (Optional) Top-level onlineresource URL.
sos_srs
• Description: (Required) Contains a list of EPSG projection codes that should be advertized as being available
for all layers in this server. The value can contain one or more EPSG:<code> pairs separated by spaces (e.g.
“EPSG:4269 EPSG:4326”) This value should be upper case (EPSG:42304.....not epsg:42304) to avoid problems
with case sensitive platforms.
sos_title
• Description: (Recommended) A human-readable name for this Layer.
sos_hoursofservice
• Description: (Optional) Time period (including time zone) when individuals can contact the organization or
individual.
sos_contactinstructions
• Description: (Optional) Supplemental instructions on how or when to contact the individual or organization.
sos_role
• Description: (Optional) Function performed by the responsible party. Possible values of this Role shall include
the values and the meanings listed in Subclause B.5.5 of ISO 19115:2003.
sos_maxfeatures
• Description: (Optional) The number of elements to be returned by the WFS server. If the not set all observations
are returned
sos_encoding_blockSeparator

490 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

• Description: (Optional) For GetObservation requests using resultModel=om:Observation (SWE DataBlock en-
coding). Record separator to be used. Default is ‘\n’
sos_encoding_tokenSeparator
• Description: (Optional) For GetObservation requests using resultModel=om:Observation (SWE DataBlock en-
coding). Token (field) separator to be used. Default is ‘,’

Layer Object Metadata

sos_describesensor_url
• Description: (Required) This metadata item is only a temporary measure until the describe sensor is generated
from MapServer. Right now when a DescribeSensor request is sent with a procedure (sensorid), it will redirect
it to the url defined by this metadata item.
• In MapServer 5.0, it is possible to use variable substituion on the url. For example “sos_describesensor_url”
“http://foo/foo?mysensor=%procedure%” will substitute the %procedure% in the metadata with the procedure
value coming from the request.
"sos_describesensor_url" "http://some/url/NS01EE0014.xml"

sos_[item name]_alias
• Description: (Optional) An alias for an attribute’s name that will be returned when executing a GetObservation
request.
sos_[item name]_definition
• Description: (Optional) An associated definition (usually a URN) for a component, that will be returned when
executing a GetObservation request. Default is “urn:ogc:object:definition”
sos_[item name]_uom
• Description: (Optional) An associated unit of measure URN) for a component, that will be returned when
executing a GetObservation request. Default is “urn:ogc:object:uom”
sos_observedproperty_id
• Description: (Required) ID of observed property, possibly in number format.
sos_observedproperty_name
• Description: (Optional) Name of observed property, possibly in string format.
sos_observedproperty_authority
• Description: (Optional) An associated authority for a given component of an observed property
sos_observedproperty_version
• Description: (Optional) An associated version for a given component of an observed property
sos_offering_description
• Description: (Optional) Description of offering.
sos_offering_extent
• Description: (Optional) Spatial extents of offering, in minx, miny, maxx, maxy format:
"sos_offering_extent" "-66, 43, -62, 45"

The logic for the bounding box returned as part of the offering is the following:

9.11. SOS Server 491

MapServer Documentation, Release 6.0

– note that it is a mandatory element that needs an espg code and lower/upper corner coordinates
– looks for the espg parameter in the first layer of the offering (this could be an ows/sos_srs or a projection
object with the epsg code (mandatory)
– looks for sos_offering_extent. If the metadata is not available, the extents of all layers in the offering will
be used to compute it.
Here is an example result from a GetCapabilities request:
<gml:boundedBy>
<gml:Envelope>
<gml:lowerCorner srsName="EPSG:4326">-66 43</gml:lowerCorner>
<gml:upperCorner srsName="EPSG:4326">-62 45</gml:upperCorner>
</gml:Envelope>
</gml:boundedBy>

sos_offering_id
• Description: (Required) ID of offering, possibly in number format.
sos_offering_intendedapplication
• Description: (Optional) The intended category of use for this offering.
sos_offering_name
• Description: (Optional) Name of offering, possibly in string format.
sos_offering_timeextent
• Description: (Optional) Time extent of offering, in the format of “begin/end”. Here is an example:
"sos_offering_timeextent" "1990/2006"

If end is not specified it will be set to now. Here is an example result from a GetCapabilities request:
<sos:eventTime>
<gml:TimePeriod>
<gml:beginPosition>1990</gml:beginPosition>
<gml:endPosition>2006</gml:endPosition>
</gml:TimePeriod>
</sos:eventTime>

sos_procedure
• Description: (Required) Normally a sensor unique id. One per layer:
"sos_procedure" "NS01EE0014"

Note: sos_procedure can also be a list, separated by spaces, i.e.:

"sos_procedure" "35 2147 604"

All sos_procedure links from layers in the offerings will be outputed together, such as the following taken from
a GetCapabilities response:
<procedure xlink:href="urn:ogc:object:feature:Sensor:3eTI:csi-sensor-1"/>
<procedure xlink:href="urn:ogc:object:feature:Sensor:3eTI:csi-sensor-2"/>

sos_procedure_item
• Description: (Required if sos_procedure is not present): See section 5 for more details

492 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

"sos_procedure_item" "attribute_field_name"

sos_timeitem
• Description: (Optional) Name of the time field. It will be used for queries when a GetObservation request is
called with an EVENTTIME parameter. It is layer specific and should be set on all layers.
"sos_timeitem" "TIME"

9.11.5 Use of sos_procedure and sos_procedure_item

In MapServer 5.0 SOS support has been upgraded to use a new metadata called sos_procedure_item. The value for
sos_procedure_item is the field/attribute name containing the procedure values. The use of this metadata as well as
the sos_procedure is described here per type of request (refer to http://trac.osgeo.org/mapserver/ticket/2050 for more
description):
It should be noted that, for very large datasets defined only with sos_procedure_item, this may result in costly process-
ing, because MapServer has to process attribute data. It is advised to setup and manage datasets accordingly if dealing
with large observation collections.

GetCapabilities

• if sos_procedure is defined, use it

• if not look for sos_procedure_item : procedure values are extracted from the layer’s attribute specified by this
metadata. Not that this can be time consuming for layers with a large number of features.
• if none is defined return an exception

DescribeSensor

• if sos_procedure is defined, use it

• if not look for sos_procedure_item : procedure values are extracted from the layer’s attribute specified by this
metadata
• if none is defined return an exception

GetObservation

Both sos_procedure and sos_procedure_item can be define. Here are the cases:
• case 1 [only sos_procedure is defined.]
– Use this metadata to match the layer with the procedure value sent in the request
– When outputing the <member/procedure> output the value of the metadata
Note: If more than one procedure is defined per LAYER object, output observations will have incorrect sos:procedure
values, because there is no way to map procedures to observations. This is where sos_procedure_item should be used
(i.e. when more than one procedure makes up a LAYER object).
• case 2: only procedure_item is defined.
– Use the sos_procedure_item and do a query on the layer to match the procedure with the layer.

9.11. SOS Server 493

MapServer Documentation, Release 6.0

– When outputting the <member/procedure> use the procedure_item as a way to only output the at-
tribute value corresponding to the feature.
• case 3: both are defined.
– check in sos_procedure to match the procedure with the layer.
– When outputting the <member/procedure> use the procedure_item as a way to only output the at-
tribute value corresponding to the feature.

9.12 MapScript Wrappers for WxS Services

Author Frank Warmerdam

Contact warmerdam at pobox.com
Revision $Revision: 8278 $
Date $Date: 2008-12-23 13:34:31 -0800 (Tue, 23 Dec 2008) $

Contents
• MapScript Wrappers for WxS Services
– Introduction
– Python Examples
– Perl Example
– Java Example
– PHP Example
– Use in Non-CGI Environments (mod_php, etc)
– Post Processing Capabilities

9.12.1 Introduction

With the implementation of MapServer MS RFC 16: MapScript WxS Services in MapServer 4.9, MapScript now has
the ability to invoke MapServer’s ability to execute OGC Web Service requests such as WMS, WCS, and WFS as well
as capturing the results of processing the requests.
This makes it possible to dynamically configure a map object based on information in the original request, and to
capture the output of processing requests for further post-processing.

9.12.2 Python Examples

The following trivial example, in Python, demonstrates a script that internally provides the map name, but otherwise
uses normal mapserver processing.
import mapscript

req = mapscript.OWSRequest()
req.loadParams()

map = mapscript.mapObj( ’/u/www/maps/ukpoly/ukpoly.map’ )

map.OWSDispatch( req )

494 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

The OWSRequest object is used to manage a parsed list of OWS processing options. In the above example they are
loaded from the environment using the loadParams() call which fetches and parses them from QUERY_STRING in
the same way the mapserv executable would.
Then we load a map, and invoke OWSDispatch with the given arguments on that map. By default the results of the
dispatched request are written to stdout which returns them back to the client.
The following example ignores all passed in arguments, and manually constructs a request argument by argument. It
is likely more useful for testing purposes than for deploying WxS services, but demonstrates direct manipulation of
the request object.
import mapscript

req = mapscript.OWSRequest()
req.setParameter( ’SERVICE’, ’WMS’ )
req.setParameter( ’VERSION’, ’1.1.0’ )
req.setParameter( ’REQUEST’, ’GetCapabilities’ )

map = mapscript.mapObj( ’/u/www/maps/ukpoly/ukpoly.map’ )

map.OWSDispatch( req )

The previous example have all let results be returned directly to the client. But in some cases we want to be able to
capture, and perhaps modify the results of our requests in some custom way. In the following example we force the
hated OGC required mime type for errors to simple text/xml (warning - non-standard!)
import mapscript

req = mapscript.OWSRequest()
req.loadParams()

map = mapscript.mapObj( ’/u/www/maps/ukpoly/ukpoly.map’ )

mapscript.msIO_installStdoutToBuffer()
map.OWSDispatch( req )

content_type = mapscript.msIO_stripStdoutBufferContentType()
content = mapscript.msIO_getStdoutBufferBytes()

if content_type == ’vnd.ogc.se_xml’:
content_type = ’text/xml’

print ’Content-type: ’ + content_type

print
print content

This example demonstrates capture capturing output of OWSRequest to a buffer, capturing the “Content-type:” header
value, and capturing the actual content as binary data. The msIO_getStdoutBufferBytes() function returns the stdout
buffer as a byte array. If the result was known to be text, the msIO_getStdoutBufferString() function could have been
used to fetch it as a string instead, for easier text manipulation.

9.12.3 Perl Example

Most of the same capabilities are accessable in all SWIG based mapscript languages. In perl, we could script creation
of a request like this:
#!/usr/bin/perl

use mapscript;

9.12. MapScript Wrappers for WxS Services 495

MapServer Documentation, Release 6.0

$req = new mapscript::OWSRequest();

$req->setParameter( "SERVICE", "WMS" );
$req->setParameter( "VERSION", "1.1.0" );
$req->setParameter( "REQUEST", "GetCapabilities" );

$map = new mapscript::mapObj( "/u/www/maps/ukpoly/ukpoly.map" );

mapscript::msIO_installStdoutToBuffer();

$dispatch_out = $map->OWSDispatch( $req );

printf "%s\n", mapscript::msIO_getStdoutBufferString();

One issue in Perl is that there is currently no wrapping for binary buffers so you cannot call
msIO_getStdoutBufferBytes(), and so cannot manipulate binary results.

More Perl example code

#!/usr/bin/perl
############################################################################
#
# Name: wxs.pl
# Project: MapServer
# Purpose: MapScript WxS example
#
# Author: Tom Kralidis
#
##############################################################################
#
# Copyright (c) 2007, Tom Kralidis
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies of this Software or works derived from this Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
############################################################################/

use CGI::Carp qw(fatalsToBrowser);

use mapscript;
use strict;
use warnings;
use XML::LibXSLT;
use XML::LibXML;

496 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

my $dispatch;

# uber-trivial XSLT document, as a file

my $xsltfile = "/tmp/foo.xslt";

# here’s the actual document inline for

# testing save and alter $xsltFile above

=comment
<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:wfs="http://www.opengis.net/wfs">
<xsl:output method="xml" indent="yes"/>
<xsl:template match="/">
<WFSLayers>
<xsl:for-each select="//wfs:FeatureType">
<wfs_layer>
<name><xsl:value-of select="wfs:Name"/></name>
<title><xsl:value-of select="wfs:Title"/></title>
</wfs_layer>
</xsl:for-each>
</WFSLayers>
</xsl:template>
</xsl:stylesheet>
=cut

my $mapfile = "/tmp/config.map";
# init OWSRequest object
my $req = new mapscript::OWSRequest();

# pick up CGI paramters passed

$req->loadParams();

# init mapfile
my $map = new mapscript::mapObj($mapfile);

# if this is a WFS GetCapabilities request, then intercept

# what is normally returned, process with an XSLT document
# and then return that to the client
if ($req->getValueByName(’REQUEST’) eq "GetCapabilities" && $req->getValueByName(’SERVICE’) eq "WFS")

# push STDOUT to a buffer and run the incoming request

my $io = mapscript::msIO_installStdoutToBuffer();
$dispatch = $map->OWSDispatch($req);

# at this point, the client’s request is sent

# pull out the HTTP headers

my $ct = mapscript::msIO_stripStdoutBufferContentType();

# and then pick up the actual content of the response

my $content = mapscript::msIO_getStdoutBufferString();

my $xml = XML::LibXML->new();
my $xslt = XML::LibXSLT->new();

# load XML content

9.12. MapScript Wrappers for WxS Services 497

MapServer Documentation, Release 6.0

my $source = $xml->parse_string($content);

# load XSLT document

my $style_doc = $xml->parse_file($xsltfile);
my $stylesheet = $xslt->parse_stylesheet($style_doc);

# invoke the XSLT transformation

my $results = $stylesheet->transform($source);
# print out the result (header + content)
print "Content-type: $ct\n\n";
print $stylesheet->output_string($results);
}

# else process as normal

else {
$dispatch = $map->OWSDispatch($req);
}

9.12.4 Java Example

One benefit of redirection of output to a buffer is that it is thread-safe. Several threads in the same process can
be actively processing requests and writing their results to distinct output buffers. This Java example, used to test
multi-threaded access demonstrates that.
import edu.umn.gis.mapscript.mapObj;
import edu.umn.gis.mapscript.OWSRequest;
import edu.umn.gis.mapscript.mapscript;

class WxSTest_thread extends Thread {

public String mapName;

public byte[] resultBytes;

public void run() {

mapObj map = new mapObj(mapName);

map.setMetaData( "ows_onlineresource", "http://dummy.org/" );

OWSRequest req = new OWSRequest();

req.setParameter( "SERVICE", "WMS" );

req.setParameter( "VERSION", "1.1.0" );
req.setParameter( "REQUEST", "GetCapabilities" );

mapscript.msIO_installStdoutToBuffer();

int owsResult = map.OWSDispatch( req );

if( owsResult != 0 )
System.out.println( "OWSDispatch Result (expect 0): " + owsResult );

resultBytes = mapscript.msIO_getStdoutBufferBytes();
}
}

public class WxSTest {

public static void main(String[] args) {

498 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

try {
WxSTest_thread tt[] = new WxSTest_thread[100];
int i;
int expectedLength=0, success = 0, failure=0;

for( i = 0; i < tt.length; i++ )

{
tt[i] = new WxSTest_thread();
tt[i].mapName = args[0];
}

for( i = 0; i < tt.length; i++ )

tt[i].start();

for( i = 0; i < tt.length; i++ )

{
tt[i].join();
if( i == 0 )
{
expectedLength = tt[i].resultBytes.length;
System.out.println( "Document Length: " + expectedLength + ", expecting somewhere a
}
else if( expectedLength != tt[i].resultBytes.length )
{
System.out.println( "Document Length:" + tt[i].resultBytes.length + " Expected:" +
failure++;
}
else
success++;
}

System.out.println( "Successes: " + success );

System.out.println( "Failures: " + failure );

} catch( Exception e ) {
e.printStackTrace();
}
}
}

9.12.5 PHP Example

Most of the same capabilities are accessible in php mapscript. Here is an example displaying a wms capabilities.
Example1 : get the capabilities
This is for example what a url could look like :
http://.../php/ows.php?service=WMS&version=1.1.1&Request=GetCapabilities

<?php

dl("php_mapscript_4.10.0.dll");

$request = ms_newowsrequestobj();

$request->loadparams();

9.12. MapScript Wrappers for WxS Services 499

MapServer Documentation, Release 6.0

/*exampple on how to modify the parameters :

forcing the version from 1.1.1 to 1.1.0 */
$request->setParameter("VeRsIoN","1.1.0");

ms_ioinstallstdouttobuffer();

$oMap = ms_newMapobj("../../service/wms.map");

$oMap->owsdispatch($request);

$contenttype = ms_iostripstdoutbuffercontenttype();

$buffer = ms_iogetstdoutbufferstring();

header(’Content-type: application/xml’);
echo $buffer;

ms_ioresethandlers();

?>

Example2 : get the map

This is for example what a url could look like :

http://.../php/ows.php?SERVICE=WMS&VeRsIoN=1.1.1&Request=GetMap&
LAYERS=WorldGen_Outline

<?php

dl("php_mapscript_4.10.0.dll");

$request = ms_newowsrequestobj();

$request->loadparams();

ms_ioinstallstdouttobuffer();

$oMap = ms_newMapobj("../../service/wms.map");

$oMap->owsdispatch($request);

$contenttype = ms_iostripstdoutbuffercontenttype();

if ($contenttype == ’image/png’)
header(’Content-type: image/png’);

ms_iogetStdoutBufferBytes();

ms_ioresethandlers();

?>

500 Chapter 9. OGC Support and Configuration

 MapServer Documentation, Release 6.0

9.12.6 Use in Non-CGI Environments (mod_php, etc)

The loadParams() call establish parses the cgi environment varabiables (QUERY_STRING, and RE-
QUEST_METHOD) into parameters in the OWSRequest object. In non-cgi environments, such as when php, python
and perl are used as “loaded modules” in Apache, or Java with Tomcat, the loadParams() call will not work - in fact in
4.10.x it will terminate the web server instance.
It is necessary in these circumstances for the calling script/application to parse the request url into keyword/value pairs
and assign to the OWSRequest object by other means, as shown in some of the above examples explicitly setting the
request parameters.

9.12.7 Post Processing Capabilities

In the following python example, we process any incoming WxS request, but if it is a GetCapabilities request we
replace the Service section in the capabilities with a section read from a file, that is carefully tailored the way we want.
#!/usr/bin/env python

import sys
import elementtree.ElementTree as ET

import mapscript

req = mapscript.OWSRequest()
req.loadParams()

map = mapscript.mapObj( ’/u/www/maps/ukpoly/ukpoly.map’ )

#
# Handle anything but a GetCapabilities call normally.
#
if req.getValueByName(’REQUEST’) <> ’GetCapabilities’:

map.OWSDispatch( req )

#
# Do special processing for GetCapabilities
#
else:
mapscript.msIO_installStdoutToBuffer()

map.OWSDispatch( req )

ct = mapscript.msIO_stripStdoutBufferContentType()
content = mapscript.msIO_getStdoutBufferString()
mapscript.msIO_resetHandlers()

# Parse the capabilities.

tree = ET.fromstring(content)

# Strip out ordinary Service section, and replace from custom file.

tree.remove(tree.find(’Service’))
tree.insert(0,ET.parse(’./Service.xml’).getroot())

# Stream out adjusted capabilities.

9.12. MapScript Wrappers for WxS Services 501

MapServer Documentation, Release 6.0

print ’Content-type: ’ + ct
print
print ET.tostring(tree)

502 Chapter 9. OGC Support and Configuration

 CHAPTER 10

Optimization

10.1 Debugging MapServer

Author Jeff McKenna

Contact jmckenna at gatewaygeomatics.com
Last Updated 2010-05-07

Table of Contents
• Debugging MapServer
– Introduction
* Links to Related Information
– Steps to Enable MapServer Debugging
* Step 1: Set the MS_ERRORFILE Variable
* Step 2: Set the DEBUG Level
* Step 3: Turn on CPL_DEBUG (optional)
* Step 4: Turn on PROJ_DEBUG (optional)
* Step 5: Test your Mapfile
* Step 6: Check your Web Server Logs
* Step 7: Verify your Application Settings
– Debugging MapServer using Compiler Debugging Tools
* Running MapServer in GDB (Linux/Unix)
– Debugging Older Versions of MapServer (before 5.0)

10.1.1 Introduction

When developing an application for the Internet, you will inevitably across problems many problems in your environ-
ment. The goal of this guide is to assist you with locating the problem with your MapServer application.

Links to Related Information

• RFC 28: Redesign of LOG/DEBUG output mechanisms

503
MapServer Documentation, Release 6.0

• MapServer Errors

10.1.2 Steps to Enable MapServer Debugging

Starting with MapServer 5.0, you are able to control the levels of debugging/logging information returned to you by
MapServer, and also control the location of the output log file.
In technical terms, there are msDebug() calls in various areas of the MapServer code that generate information that
may be useful in tuning and troubleshooting applications.

Step 1: Set the MS_ERRORFILE Variable

The MS_ERRORFILE variable is used to specify the output of debug messages from MapServer. You can pass the
following values to MS_ERRORFILE:
[filename] Full path and filename of a log file, to contain MapServer’s debug messages. Any file extension can be
used, but .log or .txt is recommended. The file will be created, if it does not already exist.
Starting with MapServer 6.0, a filename with relative path can be passed via the CONFIG MS_ERRORFILE
directive, in which case the filename is relative to the mapfile location. Note that setting MS_ERRORFILE via
an environment variable always requires an absolute path since there would be no mapfile to make the path
relative to.
stderr Use this to send MapServer’s debug messages to the Web server’s log file (i.e. “standard error”). If you are
using Apache, your debug messages will be placed in the Apache error_log file. If you are using Microsoft
IIS, your debug messages will be sent to stdout (i.e. the browser), so its use is discouraged. With IIS it is
recommended to direct output to a file instead.
stdout Use this to send MapServer’s debug messages to the standard output (i.e. the browser), combined with the rest
of MapServer’s output.
windowsdebug Use this to send MapServer’s debug messages to the Windows OutputDebugString API, allowing the
use of external programs like SysInternals debugview to display the debug output.

Through the Mapfile

The recommended way to set the MS_ERRORFILE variable is in your mapfile, within the MAP object, such as:
MAP
...
CONFIG "MS_ERRORFILE" "/ms4w/tmp/ms_error.txt"
...
LAYER
...
END
END

Through an Environment Variable

You can also set the MS_ERRORFILE variable as an environment variable on your system. Apache users can set the
environment variable in Apache’s httpd.conf file, such as:
SetEnv MS_ERRORFILE "/ms4w/tmp/ms_error.txt"

504 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

Windows users can alternatively set the environment variable through the Windows System Properties; but make sure
to set a SYSTEM environment variable.
Note: If both the MS_ERRORFILE environment variable is set and a CONFIG MS_ERRORFILE is also set, then the
CONFIG directive takes precedence.

Step 2: Set the DEBUG Level

You can retrieve varying types of debug messages by setting the DEBUG parameter in the Mapfile. You can place the
DEBUG parameter in any LAYER in the mapfile, or instead, set it once in the MAP object so that it applies to each
layer. Use the value of the DEBUG parameter to set the type of information returned, as follows:

DEBUG Levels

Level 0 Errors only (DEBUG OFF, or DEBUG 0)

In level 0, only msSetError() calls are logged to MS_ERORFILE. No msDebug() output at all. This is the default
and corresponds to the original behavior of MS_ERRORFILE in MapServer 4.x
Level 1 Errors and Notices (DEBUG ON, or DEBUG 1)
Level 1 includes all output from Level 0 plus msDebug() warnings about common pitfalls, failed assertions or
non-fatal error situations (e.g. missing or invalid values for some parameters, missing shapefiles in tileindex,
timeout error from remote WMS/WFS servers, etc.)
Level 2 Map Tuning (DEBUG 2)
Level 2 includes all output from Level 1 plus notices and timing information useful for tuning mapfiles and
applications. this is the recommended minimal debugging level
Level 3 Verbose Debug (DEBUG 3)
All of Level 2 plus some debug output useful in troubleshooting problems such as WMS connection URLs being
called, database connection calls, etc.
Level 4 Very Verbose Debug (DEBUG 4)
Level 3 plus even more details...
Level 5 Very Very Verbose Debug (DEBUG 5)
Level 4 plus any msDebug() output that might be more useful to developers than to users.

Mapfile Example

The following example is the recommended method to set the DEBUG parameter:
MAP
...
CONFIG "MS_ERRORFILE" "/ms4w/tmp/ms_error.txt"
DEBUG 5
...
LAYER
...
END
END

10.1. Debugging MapServer 505

MapServer Documentation, Release 6.0

The MS_DEBUGLEVEL Environment Variable

Instead of setting the DEBUG Debug level in each of your mapfiles, you can also be set the level globally by using the
MS_DEBUGLEVEL environment variable.
When set, this value is used as the default debug level value for all map and layer objects as they are loaded by the
mapfile parser. This option also sets the debug level for any msDebug() call located outside of the context of a map or
layer object, for instance for debug statements relating to initialization before a map is loaded. If a DEBUG value is
also specified in the mapfile in some map or layer objects then the local value (in the mapfile) takes precedence over
the value of the environment variable.
Apache users can set the environment variable in Apache’s httpd.conf file, such as:
SetEnv MS_DEBUGLEVEL 5

Windows users can alternatively set the environment variable through the Windows System Properties; but make sure
to set a SYSTEM environment variable.

Step 3: Turn on CPL_DEBUG (optional)

MapServer relies on the GDAL library to access most data layers, so you may wish to turn on GDAL debugging, to
hopefully get more information on how GDAL is accessing your data file. This could be very helpful for problems
with accessing raster files and PostGIS tables. You can trigger this GDAL output by setting the CPL_DEBUG variable
in your mapfile, within the MAP object, such as:
MAP
...
CONFIG "CPL_DEBUG" "ON"
...
LAYER
...
END
END

Step 4: Turn on PROJ_DEBUG (optional)

MapServer relies on the PROJ.4 library to handle data projections, so you may wish to turn on PROJ debugging,
to hopefully get more information back from the PROJ library. You can trigger this PROJ output by setting the
PROJ_DEBUG variable in your mapfile, within the MAP object, such as:
MAP
...
CONFIG "PROJ_DEBUG" "ON"
...
LAYER
...
END
END

Step 5: Test your Mapfile

Once you have set the MS_ERRORFILE and DEBUG level in your mapfile, you should now test your mapfile and read
your generated log file.

506 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

Using shp2img

The recommended way to test your mapfile is to use the MapServer commandline utility shp2img, to verify that your
mapfile creates a valid map image. shp2img should be included in your MapServer installation (MS4W users need to
execute setenv.bat before using the utility).
You can set the DEBUG level by passing the shp2img following parameters to your commandline call:
Note: If you have already set MS_ERRORFILE in your mapfile, you must comment this out in order to use these
shp2img options
Note: When using shp2img to debug, your layer’s STATUS should be set to ON or DEFAULT. If the layer’s STATUS
is set to OFF, you must additionally pass the layer name to shp2img by using the “-l layername” syntax

-all_debug Use this setting to set the debug level for the MAP object and all layers. this is the recommended switch
to use
shp2img -m spain.map -o test.png -all_debug 5

msLoadMap(): 0.002s
msDrawMap(): Layer 0 (spain provinces), 0.012s
msDrawRasterLayerLow(orthophoto): entering.
msDrawGDAL(): src=0,0,3540,2430, dst=188,48,1,1
source raster PL (-793.394,-1712.627) for dst PL (188,48).
msDrawGDAL(): red,green,blue,alpha bands = 1,2,3,0
msDrawMap(): Layer 1 (orthophoto), 0.150s
msDrawMap(): Layer 2 (urban areas), 0.004s
msDrawMap(): Layer 3 (species at risk), 0.008s
msDrawMap(): Layer 4 (populated places), 1.319s
msDrawMap(): Drawing Label Cache, 0.014s
msDrawMap() total time: 1.513s
msSaveImage() total time: 0.039s
msFreeMap(): freeing map at 0218C1A8.
freeLayer(): freeing layer at 0218F5E0.
freeLayer(): freeing layer at 030C33A0.
freeLayer(): freeing layer at 030C3BC8.
freeLayer(): freeing layer at 030C4948.
freeLayer(): freeing layer at 030C7678.
shp2img total time: 1.567s

-map_debug Use this setting to set the debug level for the MAP object only.
shp2img -m spain.map -o test.png -map_debug 5

msDrawMap(): Layer 0 (spain provinces), 0.012s

msDrawRasterLayerLow(orthophoto): entering.
msDrawMap(): Layer 1 (orthophoto), 0.144s
msDrawMap(): Layer 2 (urban areas), 0.004s
msDrawMap(): Layer 3 (species at risk), 0.008s
msDrawMap(): Layer 4 (populated places), 1.323s
msDrawMap(): Drawing Label Cache, 0.013s
msDrawMap() total time: 1.511s
msSaveImage() total time: 0.039s
msFreeMap(): freeing map at 0205C1A8.

-layer_debug Use this setting to set the debug level for one layer object only.

10.1. Debugging MapServer 507

MapServer Documentation, Release 6.0

shp2img -m spain.map -o test.png -layer_debug orthophoto 5

msDrawRasterLayerLow(orthophoto): entering.
msDrawGDAL(): src=0,0,3540,2430, dst=188,48,1,1
source raster PL (-793.394,-1712.627) for dst PL (188,48).
msDrawGDAL(): red,green,blue,alpha bands = 1,2,3,0
msDrawMap(): Layer 1 (orthophoto), 0.151s
freeLayer(): freeing layer at 02F23390.

Set CPL_DEBUG At the commandline execute the following:

set CPL_DEBUG=ON

shp2img -m spain.map -o test.png -layer_debug orthophoto 5

msDrawRasterLayerLow(orthophoto): entering.
GDAL: GDALOpen(D:\ms4w\apps\spain\map/.\../data/ov172068_200904_c100u50x75c24n.jpg, this=0
4059840) succeeds as JPEG.
msDrawGDAL(): src=0,0,3540,2430, dst=188,48,1,1
source raster PL (-793.394,-1712.627) for dst PL (188,48).
msDrawGDAL(): red,green,blue,alpha bands = 1,2,3,0
GDAL: GDALDefaultOverviews::OverviewScan()
msDrawMap(): Layer 1 (orthophoto), 0.155s
freeLayer(): freeing layer at 03113390.
GDAL: GDALDeregister_GTiff() called.

Reading Errors Returned by shp2img If there is a problem with your mapfile, shp2img should output the line
number in your mapfile that is causing the trouble. The following tells us that there is a problem on line 85 of my
mapfile:
getSymbol(): Symbol definition error. Parsing error near (truetype2):(line 85)

If you are using mapfile INCLUDEs, it may be tricky to track down this line number, but most of the time the line
number is useful.

Using mapserv CGI

Another handy way to test your mapfile is to call the mapserv CGI executable at the commandline, such as the follow-
ing:
mapserv -nh "QUERY_STRING=map=/ms4w/apps/spain/map/spain.map&mode=map"

ON_MISSING_DATA

If you are using tile indexes to access your data, you should also be aware of the configuration settings added in
MapServer 5.4 that allow you to tell MapServer how to handle missing data in tile indexes. Please see the CONFIG
parameter’s ON_MISSING_DATA setting in the MAP object for more information.
Hint: You can check the attributes in the tileindex by executing “ogrinfo -al” on your data file

508 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

Step 6: Check your Web Server Logs

Once you have verified that there are no problems with you mapfile, next you should check your Web server log files,
for any related information that may help you narrow down your problem.

Apache

Unix users will usually find Apache’s error_log file in a path similar to:
/var/log/apache2/

Windows users will usually find Apache’s log files in a path similar to:
C:\Program Files\Apache Group\Apache2\logs

MapServer for Windows (MS4W) users will find Apache’s log files at:
\ms4w\Apache\logs

Microsoft IIS

IIS log files can be located by:

1. Go to Start -> Control Panel -> Administrative Tools
2. Open the Internet Information Services (IIS) Manager.
3. Find your Web site under the tree on the left.
4. Right-click on it and choose Properties.
5. On the Web site tab, you will see an option near the bottom that says “Active Log Format.” Click on the
Properties button.

10.1. Debugging MapServer 509

MapServer Documentation, Release 6.0

6. At the bottom of the General Properties tab, you will see a box that contains the log file directory and the log file
name. The full log path is comprised of the log file directory plus the first part of the log file name, for example:
C:\WINDOWS\system32\LogFiles\W3SVC1\ex100507.log

You may also want to check the Windows Event Viewer logs, which is located at:
1. Go to Start -> Control Panel -> Administrative Tools
2. Computer Management
3. Event Viewer

Warning: As mentioned previously, in IIS the MapServer stderr debug output is returned to the client instead of
routed to the Web Server logs, so be sure to log the output to a file, using:

CONFIG "MS_ERRORFILE" "/ms4w/tmp/ms_error.txt"

510 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

CGI Error - The specified CGI application misbehaved by not returning a complete
set of HTTP headers

This error is often caused by missing DLL files. You should try to execute “mapserv -v at the commandline, to make
sure that MapServer loads properly.

Step 7: Verify your Application Settings

If you have verified that MapServer creates a valid map image through shp2img, you’ve checked your MapServer
log files, and there are no problems noted in your Web server logs, then you should focus your attention on possible
application configuration problems. “Application” here means how you are displaying your map images on the Web
page, such as with OpenLayers.

PHP MapScript

If you are using PHP MapScript in your application, here are some important notes for debugging:
1. Make sure your php.ini file is configured to show all errors, by setting:
display_errors = On

2. To enable debugging in PHP MapScript, if you are using MapServer 5.6.0 or more recent, make sure to define
ZEND_DEBUG in the PHP source.
If you are using MapServer < 5.6.0, then:
• open the file /mapscript/php3/php_mapscript.c
• change the following:
#define ZEND_DEBUG 0

to

#define ZEND_DEBUG 1

10.1.3 Debugging MapServer using Compiler Debugging Tools

Running MapServer in GDB (Linux/Unix)

Section author: Frank Warmerdam

Building with Symbolic Debug Info

It is not strictly necessary to build MapServer with debugging enabled in order to use GDB on linux, but it does ensure
that more meaningful information is reported within GDB. To enable full symbolic information use the –enable-
debug configure switch. Note that use of this switch disables optimization and so it should not normally be used for
production builds where performance is important.
./configure --enable-debug <other switches>
make clean
make

10.1. Debugging MapServer 511

MapServer Documentation, Release 6.0

Running in the Debugger

To run either mapserv or shp2img, give the name of the executable as an argument to the “gdb” command. If it is not
in the path, you will need to provide the full path to the executable.
gdb shp2img
GNU gdb (GDB) 7.0-ubuntu
Copyright (C) 2009 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /wrk/home/warmerda/mapserver/shp2img...done.
(gdb)

Once you are at the “(gdb)” prompt you can use the run command with the arguments you would normally have passed
to the mapserv or shp2img executable.
(gdb) run -m test.map -o out.png
Starting program: /wrk/home/warmerda/mapserver/shp2img -m test.map -o out.png
[Thread debugging using libthread_db enabled]

Program received signal SIGSEGV, Segmentation fault.

0x00007ffff67594a2 in JP2KAKDataset::Identify (poOpenInfo=0x0)
at jp2kakdataset.cpp:962
962 if( poOpenInfo->nHeaderBytes < (int) sizeof(jp2_header) )
Current language: auto
The current source language is "auto; currently c++".
(gdb)

If the program is crashing, you will generally get a report like the above indicating the function the crash occurred
in, and some minimal information on why. It is often useful to request a traceback to see what functions led to the
function that crashed. For this use the “where” command.
(gdb) where
#0 0x00007ffff67594a2 in JP2KAKDataset::Identify (poOpenInfo=0x0)
at jp2kakdataset.cpp:962
#1 0x00007ffff67596d2 in JP2KAKDataset::Open (poOpenInfo=0x7fffffffb6f0)
at jp2kakdataset.cpp:1025
#2 0x00007ffff6913339 in GDALOpen (
pszFilename=0x83aa60 "/home/warmerda/data/jpeg2000/spaceimaging_16bit_rgb.jp
2", eAccess=GA_ReadOnly) at gdaldataset.cpp:2170
#3 0x00007ffff69136bf in GDALOpenShared (
pszFilename=0x83aa60 "/home/warmerda/data/jpeg2000/spaceimaging_16bit_rgb.jp
2", eAccess=GA_ReadOnly) at gdaldataset.cpp:2282
#4 0x0000000000563c2d in msDrawRasterLayerLow (map=0x81e450, layer=0x839140,
image=0x83af90, rb=0x0) at mapraster.c:566
#5 0x000000000048928f in msDrawRasterLayer (map=0x81e450, layer=0x839140,
image=0x83af90) at mapdraw.c:1390
#6 0x0000000000486a48 in msDrawLayer (map=0x81e450, layer=0x839140,
image=0x83af90) at mapdraw.c:806
#7 0x00000000004858fd in msDrawMap (map=0x81e450, querymap=0) at mapdraw.c:459
#8 0x0000000000446410 in main (argc=5, argv=0x7fffffffd918) at shp2img.c:300
(gdb)

It may also be helpful to examine variables used in the line where the crash occured. Use the print command for this.

512 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

(gdb) print poOpenInfo

$1 = (GDALOpenInfo *) 0x0

In this case we see that the program crashed because poOpenInfo was NULL (zero). Including a traceback like the
above in bug report can help the developers narrow down a problem more quickly, especially if it is one that is difficult
for the developers to reproduce themselves.

10.1.4 Debugging Older Versions of MapServer (before 5.0)

1. Make sure that MapServer is compiled in debug mode (on unix this is enabled through ./configure –enable-
debug).
You can verify that your build was compiled in debug mode, by executing the following at the commandline
(look for “DEBUG=MSDEBUG”):
./mapserv -v

MapServer version 4.10.2 OUTPUT=GIF OUTPUT=PNG OUTPUT=WBMP

OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER
SUPPORTS=WMS_CLIENT SUPPORTS=WCS_SERVER SUPPORTS=THREADS SUPPORTS=GEOS
INPUT=EPPL7 INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE
DEBUG=MSDEBUG

2. Set the MS_ERRORFILE variable is in your mapfile, within the MAP object, such as:
MAP
...
CONFIG "MS_ERRORFILE" "/ms4w/tmp/ms_error.txt"
...
LAYER
...
END
END

3. If you don’t use the MS_ERRORFILE variable, you can use the LOG parameter in your WEB object of the
mapfile, such as:
MAP
...
WEB
LOG "mapserver.log"
END
...
LAYER
...
END
END

4. Specify DEBUG ON in your MAP object, or in your LAYER objects, such as:
MAP
...
WEB
LOG "mapserver.log"
END
DEBUG ON
...
LAYER
...

10.1. Debugging MapServer 513

MapServer Documentation, Release 6.0

END
END

5. Note that only errors will be written to the log file; all DEBUG output goes to stderr, in the case of Apache that
is Apache’s error_log file. If you are using Microsoft IIS, debug output is routed to stdout (i.e. the browser), so
be sure to remove DEBUG ON statements if using IIS on a production server.
.

10.2 FastCGI

Author Frank Warmerdam

Contact warmerdam at pobox.com
Author Howard Butler
Contact hobu.inc at gmail.com
Revision $Revision: 10014 $
Date $Date: 2010-03-25 08:22:08 -0700 (Thu, 25 Mar 2010) $
Last Updated 2008/07/15

Table of Contents
• FastCGI
– Introduction
– Obtaining the necessary software
– Configuration
– Common Problems
– FastCGI on Win32

10.2.1 Introduction

FastCGI is a protocol for keeping cgi-bin style web applications running as a daemon to take advantage of preserving
memory caches, and amortizing other high startup costs (like heavy database connections) over many requests.

10.2.2 Obtaining the necessary software

1. There are three pieces to the MapServer FastCGI puzzle. First, you need the actual FastCGI library. This can
be downloaded from http://www.fastcgi.com/. This library does the usual configure, make, make install dance.
One added complication is that it installs by default in /usr/local, and you might give the configure command a
–prefix=/usr to put it in a location that is already visible to ldconfig.
2. Assuming you are running Apache, the next piece you need is mod_fcgi. Mod_fcgi depends on the version of
Apache you are running, so make sure to download the correct fork (Apache 1.3 vs. Apache 2).
3. The third and final piece is to compile MapServer with FastCGI support. This is pretty straightforward, and all
you need to do is tell configure where the FastCGI library is installed. If you changed the prefix variable as
described above, this would be:

514 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

./configure [other options] --with-fastcgi=/usr

With those pieces in place, the MapServer CGI (mapserv) should now be FastCGI-enabled. You can verify this by
testing it on with the command line:
[hobu@kenyon mapserver-4.4.2]# ./mapserv -v
MapServer version 4.4.2 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
SUPPORTS=PROJ SUPPORTS=FREETYPE SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT
SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=FASTCGI INPUT=EPPL7
INPUT=SDE INPUT=ORACLESPATIAL INPUT=OGR INPUT=GDAL
INPUT=SHAPEFILE DEBUG=MSDEBUG

10.2.3 Configuration

1. Modify http.conf to load the FastCGI module.

LoadModule fastcgi_module /usr/lib/apache/1.3/mod_fastcgi.so

2. Add an Apache handler for FastCGI applications.

AddHandler fastcgi-script fcgi

3. Set FastCGI processing information

FastCgiConfig -initial-env PROJ_LIB=/usr/local/share/proj
-initial-env LD_LIBRARY_PATH=/usr/local/lib:/usr/local/pgsql/lib:/usr3/pkg3/oracle9/lib
-appConnTimeout 60 -idle-timeout 60 -init-start-delay 1
-minProcesses 2 -maxClassProcesses 20 -startDelay 5

4. Install a copy of the mapserv executable into the cgi-bin directory with the extension .fcgi (ie. mapserv.fcgi).
Use this executable when you want to utilize fastcgi support.
5. In your mapfile, set a PROCESSING directive to tell FastCGI to cache the connections and layer information
on all layers for which connection caching is desired - ie. all slow layers.
PROCESSING "CLOSE_CONNECTION=DEFER"

10.2.4 Common Problems

File permissions

Fedora Core 3 doesn’t allow FastCGI to write to the process logs (when you use RedHat’s Apache 2 rather than your
own). This is described here.
Also, FastCGI needs to write its socket information somewhere. This can be directed with the FastCgiIpcDir directive.

10.2.5 FastCGI on Win32

MS4W Users

MS4W (MapServer for Windows) >= version 2.2.2 contains MapServer compiled with FastCGI support. MS4W ver-
sion >= 2.2.8 also contains the required Apache module (mod_fcgid), and users must follow the README instructions
to setup FastCGI with their application.

10.2. FastCGI 515

MapServer Documentation, Release 6.0

Building fcgi-2.4.0

I used libfcgi-2.4.0 for use with Apache2 from http://www.fastcgi.com.

Binary IO Patch

It is critical that stdio be in binary mode when PNG and other binary images are written to it. To accomplish this for
stdio handled through the FastCGI library, I had to do the following hack to libfcgi/fcgi_stdio.c within the fcgi-2.4.0
distribution.
In FCGI_Accept() made he following change
if(isCGI) {
FCGI_stdin->stdio_stream = stdin;
FCGI_stdin->fcgx_stream = NULL;
FCGI_stdout->stdio_stream = stdout;
FCGI_stdout->fcgx_stream = NULL;
FCGI_stderr->stdio_stream = stderr;
FCGI_stderr->fcgx_stream = NULL;

/* FrankWarmerdam: added so that returning PNG and other binary data

will still work */
#ifdef _WIN32
_setmode( _fileno(stdout), _O_BINARY);
_setmode( _fileno(stdin), _O_BINARY);
#endif

} else {

Also, add the following just before the FCGI_Accept() function

#ifdef _WIN32
#include <fcntl.h>
#include <io.h>
#endif

I’m sure there is a better way of accomplishing this. If you know how, please let me know!

Building libfcgi

The makefile.nt should be fine. Just ensure you have run VCVARS32.BAT (as is needed for building MapServer) and
then issue the command:
nmake /f makefile.nt

Then the .lib and .dll will be in libfcgi/Debug?. Make sure you copy the DLL somewhere appropriate (such as your
cgi-bin directory).

Other Issues

1) FastCGI’s receive a very limited environment on win32, seemingly even more restricted than normal cgi’s started
by apache. Make sure that all DLLs required are either in the fastcgi directory or in windowssystem32. Any missing
DLLs will result in very cryptic errors in the error_log, including stuff about Overlapping read requests failing or
something like that.

516 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

2) Make sure you use a libfcgi.dll built against the same runtime library as your mapserv.exe (and possibly
libmap_fcgi.dll) or you will suffer a world of pain! Different runtime libraries have different “environ” vari-
ables (as well as their own stdio and heaps). You can check that everything is using MSVCRT.DLL (or all using
MSVCRTD.DLL) using the MS SDK Dependency Walker.

10.3 Mapfile

Author David Fawcett

Contact david.fawcett at gmail.com
Revision $Revision: 8365 $
Date $Date: 2008-12-31 07:49:02 -0800 (Wed, 31 Dec 2008) $
Last Updated 2007/08/01

Table of Contents
• Mapfile
– Introduction

10.3.1 Introduction

The contents of a Map File are used by MapServer for configuration, data access, projection, and more. Because the
Map File is parsed every time a map image is requested, it is important to think about what you include in the file
in order to optimize performance. The optimal Map File is one that doesn’t include or reference anything that isn’t
needed.

1. Projections

There are two ways to define projections in a Map File. You can either use inline projection parameters or specify an
EPSG code for that projection. If you use the EPSG code method, Proj.4 looks up the projection parameters in the
Proj4 database using the EPSG code as an ID. This database lookup takes significantly more resources than when the
projection parameters are defined inline. This lookup takes place for each projection definition using EPSG codes in a
Map File.
Projection defined using inline projection parameters
PROJECTION
"proj=utm"
"ellps=GRS80"
"datum=NAD83"
"zone=15"
"units=m"
"north"
"no_defs"
END

Projection defined using EPSG Code

PROJECTION
"init=epsg:26915"
END

10.3. Mapfile 517

MapServer Documentation, Release 6.0

Optimization Suggestions
• Use inline projection parameter definitions in place of EPSG codes.
• If you want to use EPSG codes, remove all unneeded projection definition records from the Proj.4 EPSG
database.

2. Layers

For every layer in a Map File that has a status of ON or DEFAULT, MapServer will load that layer and prepare it for
display, even if that layer never gets displayed.
Optimization Suggestions
• Build lean Map Files, only include layers that you plan to use.
• Turn off unnecessary layers; the more layers MapServer is displaying, the more time it takes. Have your opening
map view show only the minimum necessary to orient the user, and allow them to turn on additional layers as
needed. This is particularly true of remote WMS or very large rasters.
• Related to turning off layers, is turning them on but using MINSCALEDENOM and MAXSCALEDENOM to
determine at what zoomlevels the layer is available. If a map’s display is outside of the layer’s MINSCALE-
DENOM and MAXSCALEDENOM range, then MapServer can skip processing that layer. It also makes for a
really cool effect, that the national boundaries magically change to state boundaries.
• If you have a complex application, consider using multiple simple and specific Map Files in place of one large
‘do everything’ Map File.
• In a similar vein, each class also supports MINSCALEDENOM and MAXSCALEDENOM. If your dataset
has data that are relevant at different zoomlevels, then you may find this a very handy trick. For example,
give a MINSCALEDENOM of 1:1000000, county roads a MINSCALEDENOM of 1:100000, and streets a
MAXSCALEDENOM of 1:50000. You get the cool effect of new data magically appearing, but you don’t have
MapServer trying to draw the nation’s roads when the entire nation is in view!
• Classes are processed in order, and a feature is assigned to the first class that matches. So try placing the most
commonly-used classes at the top of the class list, so MapServer doesn’t have to try as many classes before
finding a match. For example, if you wanted to highlight the single state of Wyoming, you would probably do
this:
CLASS
EXPRESSION (’[NAME]’ eq ’WY’])
STYLE
COLOR 255 0 0
END
END
CLASS
STYLE
COLOR 128 128 128
END
END

But it would be a lot more efficient to do this, since 98% of cases will be matched on the first try:
CLASS
EXPRESSION (’[NAME]’ ne ’WY’])
STYLE
COLOR 128 128 128
END
END
CLASS

518 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

STYLE
COLOR 255 0 0
END
END

• Use tile indexes instead of multiple layers.

3. Symbols

When the Map File is loaded, each raster symbol listed in the symbols file is located on the filesystem and loaded.
Optimization Suggestions
• Only include raster symbols in your symbols file if you know that they will be used by your application.

4. Fonts

To load a font, MapServer opens up the fonts.list FONTSET file which contains an alias for the font and the path for
that font file. If you have a fonts.list file with a long list of fonts, it will take more time for MapServer to locate and
load the font that you need.
Optimization Suggestions
• Limit the entries in fonts.list to fonts that you actually use.

10.4 Raster

Author HostGIS
Revision $Revision: 8365 $
Date $Date: 2008-12-31 07:49:02 -0800 (Wed, 31 Dec 2008) $
Last Updated 2008/08/08

Table of Contents
• Raster
– Overviews
– Tileindexes and Internal Tiling
– Image formats
– Remote WMS

10.4.1 Overviews

TIFF supports the creation of “overviews” within the file, which is basically a downsampled version of the raster
data suitable for use at lower resolutions. Use the “gdaladdo” program to add overviews to a TIFF, and MapServer
(via GDAL) will automagically choose which downsampled layer to use. Note that overviews significantly increase
the disk space required by a TIFF, and in some cases the extra disk reading may offset the performance gained by
MapServer not having to resample the image. You’ll just have to try it for yourself and see how it works.

10.4. Raster 519

MapServer Documentation, Release 6.0

10.4.2 Tileindexes and Internal Tiling

Tiling is mostly effective for cases where one commonly requests only a very small area of the image.
A tileindex is how one creates an on-the-fly mosaic from many rasters. This is described in the Tile Indexes. That
document describes common cases where a tileindex makes sense. In particular, if you have a very large raster and
most requests are for a very small spatial area within it, you may want t consider slicing it and tileindexing it.
As an alternative to slicing and mosaicing, TIFFs do support a concept of internal tiling. Like a tileindex, this is mostly
effective when the requests are for a small portion of the raster. Internal tiling is done by adding “-co TILED=YES”
to gdal_translate, e.g.:
gdal_translate -co TILED=YES original.tif tiled.tif

10.4.3 Image formats

The TIFF image format is the fastest to “decipher”, but once you get beyond a certain point, the disk reading (since
TIFF is very large) may become slow enough to make it worthwhile to consider other image formats.
For TIFFs larger than 1 GB, ECW images tend to render faster than TIFFs, since decompressing the data (CPU and
RAM) is faster than reading the uncompressed data (disk). The downside is that ECW is not open-source, and the
licensing is often prohibitive.
JPEG2000 is a very slow image format, as is JPEG.

10.4.4 Remote WMS

Remote WMS servers are often slow, especially the public ones such as TerraServer or NASA’s Landsat server. There’s
nothing you can do about that, except to reconsider when the remote WMS layer should be used.
For example, there may be a different WMS server (or a different set of imagery, or even vector outline maps) suitable
for drawing the countries or states to orient the user. You could then have the WMS layer come on at a certain scale,
or have the layer always available but turned off so the user can choose when to turn it on.
See Also:
Raster Data

10.5 Tile Indexes

Author HostGIS
Revision $Revision: 8365 $
Date $Date: 2008-12-31 07:49:02 -0800 (Wed, 31 Dec 2008) $
Last Updated 2007/08/03

Table of Contents
• Tile Indexes
– Introduction
– What is a tileindex and how do I make one?
– Using the tileindex in your mapfile
– Tileindexes may make your map faster

520 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

10.5.1 Introduction

An introduction to tileindexes, MapServer’s method for doing on-the-fly mosaicing.

10.5.2 What is a tileindex and how do I make one?

A tileindex is a shapefile that ties together several datasets into a single layer. Therefore, you don’t need to create
separate layers for each piece of imagery or each county’s road data; make a tileindex and let MapServer piece the
mosaic together on the fly.
Making a tileindex is easy using ‘gdaltindex’ for GDAL data sources (rasters) and ‘ogrtindex’ for OGR data sources
(vectors). You just run them, specifying the index file to create and the list of data sources to add to the index.
For example, to make a mosaic of several TIFFs:
gdaltindex imagery.shp imagery/*.tif

And to make a mosaic of vectors:

ogrtindex strees.shp tiger/CA/*.shp tiger/NV/*.shp

Note: ogrtindex and gdaltindex add the specified files to the index. Sometimes you’ll have to delete the index file to
avoid creating duplicate entries.

10.5.3 Using the tileindex in your mapfile

Using a tileindex as a layer is best explained by an example:

LAYER
NAME "Roads"
STATUS ON
TYPE LINE
TILEINDEX "tiger/index.shp"
TILEITEM "LOCATION"
END

There are two items of note here: TILEINDEX and TILEITEM. TILEINDEX is simply the path to the index file, and
TILEITEM specifies the field in the shapefile which contains the filenames referenced by the index. The TILEITEM
will usually be “LOCATION” unless you specified the -tileindex option when running gdaltindex or ogrtindex.
Two important notes about the pathnames:
• The path to TILEINDEX follows the same conventions as for any other data source, e.g. using the SHAPEPATH
or else being relative to the location of the mapfile.
• The filenames specified on the command line to gdaltindex or ogrtindex will be used with the same conventions
as well, following the SHAPEPATH or else being relative to the mapfile’s location. I find it very useful to
change into the SHAPEPATH directory and then run ogrtindex/gdaltindex from there; this ensures that I specify
the correct pathnames.

10.5.4 Tileindexes may make your map faster

A tileindex is often a performance boost for two reasons:

• It’s more efficient than having several separate layers.

10.5. Tile Indexes 521

MapServer Documentation, Release 6.0

• MapServer will examine the tileindex to determine which datasets fall into the map’s view, and will open only
those datasets. This can result in a great savings for a large dataset, especially for use cases where most of the
time only a very small spatial region of the data is being used. (for example, citywide maps with nationwide
street imagery)
A tileindex will not help in the case where all/most of the data sources will usually be opened anyway (e.g. street data
by county, showing states or larger regions). In that case, it may even result in a decrease in performance, since it may
be slower to open 100 files than to open one giant file.
The ideal case for a tileindex is when the most typically requested map areas fall into very few tiles. For example,
if you’re showing state and larger regions, try fitting your data into state-sized blocks instead of county-sized blocks;
and if you’re showing cities and counties, go for county-sized blocks.
You’ll just have to experiment with it and see what works best for your use cases.

10.6 Vector

Author HostGIS
Revision $Revision: 8365 $
Date $Date: 2008-12-31 07:49:02 -0800 (Wed, 31 Dec 2008) $
Last Updated 2008/08/08

Table of Contents
• Vector
– Splitting your data
– Shapefiles
– PostGIS
– Databases in General (PostGIS, Oracle, MySQL)

10.6.1 Splitting your data

If you find yourself making several layers, all of them using the same dataset but filtering to only use some of the
records, you could probably do it better. If the criteria are static, one approach is to pre-split the data.
The ogr2ogr utility can select on certain features from a datasource, and save them to a new data source. Thus, you
can split your dataset into several smaller ones that are already effectively filtered, and remove the FILTER statement.

10.6.2 Shapefiles

Use shptree to generate a spatial index on your shapefile. This is quick and easy (“shptree foo.shp”) and generates a
.qix file. MapServer will automagically detect an index and use it.
MapServer also comes with the sortshp utility. This reorganizes a shapefile, sorting it according to the values in one
of its columns. If you’re commonly filtering by criteria and it’s almost always by a specific column, this can make the
process slightly more efficient.
Although shapefiles are a very fast data format, PostGIS is pretty speedy as well, especially if you use indexes well
and have memory to throw at caching.

522 Chapter 10. Optimization

 MapServer Documentation, Release 6.0

10.6.3 PostGIS

The single biggest boost to performance is indexing. Make sure that there’s a GIST index on the geometry column,
and each record should also have an indexed primary key. If you used shp2pgsql, then these statements should create
the necessary indexes:
ALTER TABLE table ADD PRIMARY KEY (gid);
CREATE INDEX table_the_geom ON table (the_geom) USING GIST;

PostgreSQL also supports reorganizing the data in a table, such that it’s physically sorted by the index. This allows
PostgreSQL to be much more efficient in reading the indexed data. Use the CLUSTER command, e.g.
CLUSTER the_geom ON table;

Then there are numerous optimizations one can perform on the database server itself, aside from the geospatial compo-
nent. The easiest is to increase max_buffers in the postgresql.conf file, which allows PostgreSQL to use more memory
for caching. More information can be found at the PostgreSQL website.

10.6.4 Databases in General (PostGIS, Oracle, MySQL)

By default, MapServer opens and closes a new database connection for each database-driven layer in the mapfile. If
you have several layers reading from the same database, this doesn’t make a lot of sense. And with some databases
(Oracle) establishing connections takes enough time that it can become significant.
Try adding this line to your database layers:
PROCESSING "CLOSE_CONNECTION=DEFER"

This causes MapServer to not close the database connection for each layer until after it has finished processing the
mapfile and this may shave a few seconds off of map generation times.

10.6. Vector 523

MapServer Documentation, Release 6.0

524 Chapter 10. Optimization

 CHAPTER 11

Utilities

11.1 legend

11.1.1 Purpose

Creates a legend from a mapfile. Output is either PNG or GIF depending on what version of the GD library used.

11.1.2 Syntax

legend [mapfile] [output image]