Cognos 8 Planning

ANALYST USER GUIDE

Product Information
This document applies to Cognos 8 Planning Version 8.3 and may also apply to subsequent releases. To check for newer versions of this document, visit the Cognos Global Customer Services Web site (http://support.cognos.com).

Copyright
Copyright 2007 Cognos Incorporated. Portions of Cognos software products are protected by one or more of the following U.S. Patents: 6,609,123 B1; 6,611,838 B1; 6,662,188 B1; 6,728,697 B2; 6,741,982 B2; 6,763,520 B1; 6,768,995 B2; 6,782,378 B2; 6,847,973 B2; 6,907,428 B2; 6,853,375 B2; 6,986,135 B2; 6,995,768 B2; 7,062,479 B2; 7,072,822 B2; 7,111,007 B2; 7,130,822 B1; 7,155,398 B2; 7,171,425 B2; 7,185,016 B1;7,213,199 B2. Cognos and the Cognos logo are trademarks of Cognos Incorporated in the United States and/or other countries. All other names are trademarks or registered trademarks of their respective companies. While every attempt has been made to ensure that the information in this document is accurate and complete, some typographical errors or technical inaccuracies may exist. Cognos does not accept responsibility for any kind of loss resulting from the use of information contained in this document. This document shows the publication date. The information contained in this document is subject to change without notice. Any improvements or changes to either the product or the document will be documented in subsequent editions. U.S. Government Restricted Rights. The software and accompanying materials are provided with Restricted Rights. Use, duplication, or disclosure by the Government is subject to the restrictions in subparagraph (C)(1)(ii) of the Rights in Technical Data and Computer Software clause at DFARS 252.227-7013, or subparagraphs (C)(1) and (2) of the Commercial Computer Software - Restricted Rights at 48CFR52.227-19, as applicable. The Contractor is Cognos Corporation, 15 Wayside Road, Burlington, MA 01803. This software/documentation contains proprietary information of Cognos Incorporated. All rights are reserved. Reverse engineering of this software is prohibited. No part of this software/documentation may be copied, photocopied, reproduced, stored in a retrieval system, transmitted in any form or by any means, or translated into another language without the prior written consent of Cognos Incorporated.

Table of Contents
Introduction

19

What's New? 23 New Features in Version 8.3 23 Microsoft Excel 2007 23 Select Folders in Cognos Connection 23 Select Framework Manager Package for D-List Import 23 Chapter 1: Analyst Overview

25

Saving Analyst Data 25 Registry Settings 25 Restart Analyst 26 Understanding Analyst 26 Formulas in Analyst and Excel Spreadsheets 26 Using LIB and LIBNO Parameters 27 Other Differences 27 Multi-User Object Access 28 Analyst Samples 28 Tutorialgo 29 Great Outdoors Analyst 29 The BiF Library 31 The Slice Update Sample 31 Customizing the Analyst Toolbar 34 Use Custom Menus 35 Create Custom Toolbar Buttons 36 Chapter 2: Administration

37

Viewing and Editing Analyst Workspace Settings 37 Creating Planning Tables 38 Set Filesys.ini 38 Rebuild Index Files 39 Refresh References 39 Validate D-Lists 39 DOS File Names 39 Close ODBC Links 40 Configuration Settings 40 Chapter 3: Objects

43

Add a Description to an Object 43 Reveal Object Name from DOS Filename 43 Chapter 4: Security

45

Cognos Namespace 45 Authentication Providers 46 Deleting or Restoring Unconfigured Namespaces 47

User Guide 3

Table of Contents Users, Groups, and Roles 47 Users 48 Groups and Roles 48 Setting up Security for a Cognos 8 Planning Installation 50 Configure Cognos 8 to Use an Authentication Provider 51 Add or Remove Members From Planning Rights Administrators and Planning Contributor Users Roles 52 Enabling Planning Roles in Cognos 8 53 Restricting Access to the Everyone Group 53 Recommendation - Creating Additional Roles or Groups for Contributor 54 Configuring Access to Analyst 54 Configure Integrated Windows Authentication 56 Specify a Default Library 57 Restrict D-Cube Access 57 Assign Access at the Library Level 57 Assigning Access at the Object Level 58 Assigning Access at the Item Level 60 Chapter 5: Integration

63

Financial Planning with Cognos Finance and Cognos Planning Products 63 Prerequisites and Required Components 64 Understanding the Process 64 Using Cognos Finance Input Forms or Report Files 65 Importing From Cognos Finance 65 Financial Planning with Cognos Performance Applications and Analyst 67 IQD Files and the Import from IQD Wizard 68 Creating Planning Models and Data from Cognos Performance Applications Data 72 Determine Plan Granularity 74 Create a Planning Model in Analyst 74 Set up a Contributor Application 76 Populate the Planning Application in Contributor 77 Publish Planning Data 79 Automation 81 Command Line 81 Chapter 6: Importing Data from Cognos 8 Data Sources

83

Create a Data Source Connection 84 Create a Framework Manager Project and Import Metadata 86 Create and Publish the Cognos Package 87 Working with SAP BW Data 88 Create a Detailed Fact Query Subject 88 Recommendation - Query Items 89 Recommendation - Hierarchy 89 Recommendation - Hiding the Dimension Key Field 90 Working with Packages 90 Chapter 7: D-Lists

91

Open a D-List 91 Create a D-List 91 Formulas 93

4 Analyst

Table of Contents Import a Formula (CalcTexts) 93 Preview or Print the Formulas 95 Create a Subtotal 95 Enter a Formula into a D-List 95 Edit a Formula 97 Copy a Formula 97 Remove a Formula 98 Troubleshooting Formula Errors 98 View Formulas 100 D-List Conditional Formulas 101 Formula Priority 104 Apply a Time Average 106 Weighted Averages 106 Choosing Which Item to Weight By 107 Apply a Weighted Average 107 Override a Weighted Average 108 Force to Zero 108 Remove Averages 108 Applying Local Formats to D-Lists 109 Types of Formats 109 Assign Local Formats 110 Save a Local Format 110 Load a Local Format 110 Formulas and D-List Formatted Items 111 Format a Specific Row or Column 111 Numeric Formats 112 Date and Time Formats 114 Apply D-List Formats 117 Apply Free-Text Format to a D-List Item 118 Timescale D-Lists 118 Create a Timescale D-List 119 Create a Custom Timescale 120 Timescales and BiFs 120 Common Errors in Timescales 120 Set Periods of Uneven Lengths 120 Copy an Existing D-List 121 Edit a D-List 121 Edit D-List Item Names 123 Insert items from a D-List 123 Create an Import Link into a D-List 124 Run an Import Link into a D-List 124 Paste Items from a Spreadsheet, Database, or Other Text Source 125 Import D-List Items from Another D-List 125 Import D-List Items from Unmapped ASCII Files 126 Import D-List Data from Mapped ASCII Files 127 Import D-List Items from a D-Cube 128 Import D-List Items Using ODBC 129 Import D-List Items From a Cognos Package 130 Export a D-List as an e-List 131

User Guide 5

Table of Contents Implement Changes 131 Delete items from a D-List 131 Import Mode 132 Example - Import Modes 132 Copying from Another D-List 133 Importing from ASCII files, ODBC sources, Cognos Packages, or D-Cube data 134 Where Drop-Box 135 Subtotals Drop-Box 135 Maintaining Hierarchies 135 Manage D-Lists 139 Rename/Move a D-List 139 << Move 139 Delete a D-List 139 Search for a D-List 140 Show a D-List's Dependants 140 Show a D-List's Precedents 140 Upgrade Pipe Symbols in D-List Item Names 141 Unique Names 141 Edit Unique Names 141 Define the Unique Part of a D-List Item 141 Manually Reorder D-List Items 142 Sort D-List Items 143 View/Edit Summary Information on a D-List 144 Add Sub Headings to Reports 145 Set D-List Colors 146 Chapter 8: D-Cubes

147

Interrupt a Calculation 148 Set or Clear Audit Trails 148 View the Audit Trail of a Cell Within a D-Cube 149 Breakback 149 Default Rules for Breakback 149 Use Breakback to Set Global Targets 151 Holds and Breakback 151 Using Breakback with Integer Arithmetic 152 Set Breakback to Integer or Decimal Mode 153 Set a Target Using Breakback 153 Create D-Cubes 154 Open D-Cubes 156 Open Multiple D-Cubes 157 Expand a Subtotal 157 View Multiple Slices of a D-Cube in Separate Windows 157 D-Cube Data Allocations 158 Enter Data 159 Enter Data into Individual Cells of a D-Cube 159 Color Conventions for Data 160 View a Formula 161 View the Origin of a Detail Cell 161 Edit D-Cubes 161

6 Analyst

Table of Contents Copy a Range on the Same Page 161 Copy Ranges in a D-Cube from Page to Page 162 Copy Data Using Operators 163 Copy from a Spreadsheet to Analyst 163 Insert Lines to Separate Totals from Detail Items 164 Edit Undo and Redo 165 Suppress Zero Rows, Columns, or Pages 165 Reveal All Zero Suppressed Rows and Columns 166 Change the Column Width or Row Label Width 166 Annotate a Cell 167 Edit Data 168 Edit Data on the Current Page of a D-Cube 168 Edit the Data in Individual Cells Using Operators 169 Select a Range of Cells in a D-Cube 170 Reset Data 171 Recover from Errors 172 D-Cube Commands 173 Apply Commands 175 Edit a Range of Data on the Current Page 176 Delete the Data from an Entire D-Cube 176 Set a Column or Row to Zero 177 Change Ranges of Data Using Menu Commands 178 Locks, Protects, and Holds 180 Hold Data 181 Protect Data 183 Lock Data 183 Special Copy and Paste 185 Random Number D-Cube Command 185 Round Command 185 Export Data 186 D-Cube Export 186 Format Prior to Export 189 Export to a Spreadsheet 190 AutoSum 190 Find Text and Character Matches 191 Formats 191 Local vs Global Formats 191 Load or Remove a Global Format 191 Save a Global Format 192 Enter Prefixes and Suffixes 192 Access Blank if Zero 193 D-Cube Selections 193 Expanded Selections 193 Creating a Selection 194 Facilitate Selection Using the Selection Dialog Box 195 Saved Selections and D-Links 197 Save a Selection 197 Load a Saved Selection 198 Load a Saved Selection on Opening a D-Cube 198

User Guide 7

Table of Contents Edit the Selection on Opening a D-Cube 199 Edit a Saved Selection 200 Manage D-Cubes 202 Memory Management 202 Split Column Headings onto Two Lines 203 Show Details or Formulas Only 203 Sort Rows, Columns, and Pages 204 Manipulate D-Cube Structure 205 Work with Dimensions 207 Navigate Around a D-Cube 212 View a Different Slice 212 View a Different Page 213 Save D-Cubes 213 Chapter 9: D-Links

215

The D-Link Dialog Box 216 Dimensions and D-Lists 216 Create D-Links 217 Use D-Cube as Source and Target 217 Use Cognos Package as Source in a D-Link 218 Pair Source and Target Dimensions 219 Select Required Items from Unpaired Dimensions 219 Change Optional Settings in D-Link 220 Name and Save the D-Link 220 Open D-Links 220 Open More than One D-Link 221 Open a D-Link that Targets an Open D-Cube 221 Open a D-Link that Uses an Open D-Cube as its Source 221 Open D-Links Associated with Selected D-Cubes 222 Run D-Links 222 Run an Open D-Link 223 Run a D-Link Using a Specific D-Cube as its Source or Target 223 Run a D-Link Using an Open D-Cube as its Target 223 Run D-Links with the Source D-Cube Open 224 Run Update D-Links in a Single D-Cube (Manually) 225 Run Batches of D-Links Using Library Functions 225 Memory Considerations 226 Run Batches of D-Links using Macros 227 Tun an Inverse D-Link 227 Dimensions 230 Virtual Dimensions 230 Dimensions and D-Lists 230 Unvisited Dimensions 230 Unpaired Dimensions 231 Match Descriptions 233 How Match Descriptions Pairs Data 233 Create a Match Descriptions Pairing 233 Case Sensitivity 234 Match Calculated Target Items 235

8 Analyst

Table of Contents Allocation 235 Maintain Allocation Tables 236 Allocation Table Menu Options 236 How Allocation Tables Assign Data 237 Navigate Around an Allocation Table 238 Load an A-Table into a D-Link 240 Change to Matched Descriptions 240 Change to Allocation 241 Change Dimension Items in a D-Link 241 Target Formula Items 241 Local Allocation Tables (A-Tables) 242 Create a Local Allocation Table Pairing 242 Delete Entries in a Local Allocation Table 244 Use Wildcard Characters in Local Allocation Tables 244 Edit Local Allocation Table Entries 246 Reorder Lines in a Local Allocation Table 246 Add Entries to a Local A-Table 246 Loaded Allocation Tables 248 Saved Allocation Tables 248 Copy and Paste Allocation Table Entries 248 D-Cube Allocations 249 Example 249 Use D-Cube Data in Allocation 250 Select and Slice an Allocation D-Cube 250 Execution Modes 252 Run D-Links inversely 252 Fill Mode 253 Substitute Mode 253 Add Mode 253 Subtract Mode 253 The Target Area 253 Dump Options 254 When will records be unassigned? 255 Edit 255 Print 255 File 255 Dump Item 256 Drill Down on Data Assigned to Dump Items 257 Dump Items Used with Dump Options 257 Scale and Round Data within a D-Link 257 Scaling Factors 257 Rounding Factors 257 Set Scaling and Rounding 258 Subcolumns 259 Cut a Subcolumn 259 Change the Position of an Existing Subcolumn 260 Clear a Subcolumn 261 Duplicate Target Items 261 Lookup and Accumulation D-Links 262

User Guide 9

Table of Contents Database D-Cube 262 Sparse D-Cube 262 Lookup and Accumulation D-Link Restrictions 264 Lookup D-Links 264 Accumulation D-Links 270 Analyst<>Contributor Links 276 When to use Analyst<>Contributor Links 276 Installation 277 Security 277 How Analyst<>Contributor and Contributor<>Contributor Links Work 278 Copying Analyst<>Contributor Links 280 Library Method 280 Library Copy Wizard Method 281 Factors That Can Affect Memory Usage 281 Opening a Link From a Computer that Does not Have Access to the Original Datastore 282 Running Batches of D-Links using the @DLinkExecuteList macro 282 Running D-Links While Making Model Changes 283 Effect of Access Tables in Contributor 284 Select Contributor Application 284 Analyst<>Cognos Finance Links 285 When to Use Analyst <> Cognos Finance Links 286 Installation 286 How Analyst <> Cognos Finance Links Work 286 D-Link Options 286 One-off and Internal D-Links 286 Fill a D-Cube with Data Using a One-Off Internal D-Link 287 Date Allocations 287 Copy data in a D-Cube from Page to Page 288 Update Models Using D-Cube Update List 288 Order of Running D-Links 289 D-Cube Update List Dialog Box 289 Drill Down 290 How Drill Down Works 291 Drill Down on a Cell Using a Source or Mapped ASCII File 291 Drill Down on One Cell Using a D-Cube as a Source 291 Drill Down on a Range of Cells 292 Drill Down on Break Back Allocations 293 Troubleshoot D-Links 294 The "Nothing to Transfer" Message 294 Nothing Happens When You Drill Down 294 Error Messages Appear When You Drill Down 295 Chapter 10: ODBC Links

297

Creating ODBC Links 297 Install an ODBC Driver 297 Set Up an ODBC Data Source 298 Import Using ODBC 298 Import D-List Items into a D-List Using ODBC 298 Import Data into a D-Cube Using an ODBC Link 299

10 Analyst

Table of Contents Chapter 11: Libraries

301

Work with Libraries 301 Library Administration 302 Create a Library 302 Delete a Library 303 Display Library Name 303 Change Library Details 303 The Library Window 304 Rename an Object 304 Delete an Object 304 Move an Object to a Different Library 305 Show the Precedents of an Object 305 Show the Dependants of an Object 306 Highlight Unused Objects in the Library 307 Reveal the DOS File Name 307 Show the Description of an Object 307 Copy Objects 307 Remap Objects 308 Check Integrity 309 Open Multiple Objects 310 Copy Libraries or Objects Using the Library Copy Wizard 310 Select Libraries 311 Select Objects 312 Chapter 12: Built in Functions (BiFs) BiF Library 315 Work with BiFs 316 Steps to Create a BiF 316 Steps to Edit a BiF 316 Steps to Delete a BiF 317 BiF Results 317 BiF Outputs 317 Priority of Calculations 319 Circularity in BiFs 319 Nesting in BiFs 319 Breakback in BiFs 320 BiF Input Parameters 320 Show Calculation Errors 320 BiF Examples 320 Overview of Examples 320 @Cumul 324 @Cycles 324 @Days 327 @DaysOutstanding 328 @DCF 331 @Decum 336 @Delay 337 @DelayDebt 340 @DelayStock 343

315

User Guide 11

Table of Contents @DepnAnnual 346 @DepnDB 352 @DepnSLN 355 @DepnSYD 356 @Deytd 359 @Differ 360 @Drive 362 @Drive1 365 @Drive2 368 @ErlangDelayAgents 371 @ErlangDelayFull 373 @ErlangDelayLite 375 @ErlangLossLite 375 Understanding the Erlang BiFs Equations 376 Erlang Built-in Functions Glossary 378 @Feed 379 @FeedParam 380 @Forecast 383 @Funds 394 @FV 395 @Grow 406 @ICF 408 @IRR 411 @Lag 418 @Last 421 @Lease 423 @LeaseVariable 443 @Linavg 472 @Mix 473 @Movavg & @Movsum 475 MoveMed 481 @Nper 484 @NPV 493 @Outlook 498 @PMT 501 @PV 509 @Proportion 517 @Rate 518 @Repeat 529 @SeasonLite 529 @SeasonPro 533 @Simul 568 @StockFlow 571 @StockFlowAF 577 @StockflowBQ 582 @Tier 585 @Time 586 @TimeSum 593 @TMax 601

12 Analyst

Table of Contents @TMin 602 @Transform 603 @TRound 605 @Ytd 607 Switchover Dates 609 Examples: 610 Set up a Switchover Date in a Timescale D-List 610 Set up a Switchover Date in a BiF Formula 610 Print or Preview BiF Specifications 611 Chapter 13: Macros

613

Creating and Running Macros 613 Create a Macro using the Wizard 614 Record a Macro 615 Run a Macro 615 Macro Editor 615 Editing Macro Code 615 Editing Macro Variables 617 Start a Macro With Batch Utility Wizard 619 Configure Analyst Security 619 Run the Batch Utility Wizard 619 Using the Command Line to Run a Macro or Batch Job 621 Command Line Options 621 Command Line Examples 622 D-List Macros 627 @DListNew 627 @DListOpen 628 @DListUpdate 629 @ExportToEList 629 @ItemDelete 630 Item Import Macros 631 @DListItemImportCognosPackage 632 @DListItemCopyFromDList 635 @DListItemImportDelimitedText 637 @DListItemImportFileMap 640 @DListItemImportFinance 642 @DListItemImportDCube 644 @DListItemImportIQD 645 @DListItemImportOdbc 646 @RefreshDataWarehouse 649 ODBC Macros 650 Control Macros 650 @Activate 651 @AddLocalPreSelection 651 @CheckAccess 652 @CheckAccessLevel 653 @Close 653 @Delay 653 @FileTranslate 653

User Guide 13

Table of Contents @LibCopy 654 @MacroExecute 655 @Message 656 @PackDir 656 @PackDirSel 657 @Rem 658 @Reset 658 @Run 658 @Save 660 @ShutDown 660 @TestData 660 @UnPackDir 661 D-Link Macros 662 @DLinkActivateQueue 663 @DLinkExecuteInv 663 @DLinkExecSel 664 @DLinkExecute 665 @DLinkExecuteList 666 @DLinkNew 667 @DLinkOpen 668 @DLinkSelectList 668 D-Cube Macros 669 @DCubeCalculate 670 @DCubeClearMask 671 @DCubeCommand 672 @DCubeCreateDSels 673 @DCubeCreateTSels 676 @DCubeDeleteSels 678 @DCubeDeselect 679 @DCubeExport 680 @DCubeIncreaseSelect 683 @DCubeInput 684 @DCubeNew 685 @DCubeOpen 686 @DCubeOpenChooseSel 687 @DCubeOpenNamedSel 688 @DCubeOpenSelect 689 @DCubePage 689 @DCubePageId 690 @DCubePrint 691 @DCubeReselect 693 @DCubeSort 694 @DCubeTranspose 695 @DCubeUpdate 696 @GenerateTransformerModel 697 @Publish 698 @SliceCommand 699 @SliceUpdate 700 File Map Macros 704

14 Analyst

Table of Contents @FMapNew 705 @FMapOpen 705 A-Table Macros 706 @ATabOpen 706 @ATabRefresh 707 @ATabImportCognosPackage 708 @ATabImportDelimitedText 709 @ATabImportFileMap 710 @ATabImportOdbc 710 Chapter 14: A-Tables (Allocation Tables)

713

Example 713 Creating an A-Table 714 Selecting Source and Target for an A-Table 714 Attaching a D-List to an A-Table 718 Add A-Table Entries 719 Add Single Allocation Table Entries 719 Add Multiple Allocation Table Entries 720 Add New Entries for New Dimension Items 721 Add One-to-Many Entries 721 Allocate Entries Using Matching Descriptions 722 One-Sided Allocation Table Entries 723 Insert Items Buttons 723 Select Source and Target Dimension Items 724 Change the Source or Target for an A-Table 725 Managing A-Tables 725 Reorder Allocation Table Entries 725 Show and Hide Dimension Items 726 Change Entry Signs in an Allocation Table 727 How the A-Table Adapts when Dimension Item Lists Change 727 Refresh a Dimension Item List from a Mapped ASCII File 728 Delete A-Table Entries 729 Chapter 15: File Maps

731

Creating a File Map 731 Create a File Map 731 Map Editor Page 1 731 Map Editor Page 2 732 Map Editor Page 3 732 Using the LIB parameter 733 Delimited and Fixed Width ASCII Files 734 Define Columns in an ASCII File 734 Text Qualifiers 737 Special Cases for text qualifiers 738 Date and Text Data in File Maps 738 Import Date Data from a File Map 738 Import Text Data from a File Map into D-List Formatted Cells 740 Follow On 741 What does Follow On do? 742 An alternative view 743 User Guide 15

Table of Contents Which rows will a subheading apply to? 743 Use Follow On and Overlapping Subheadings 745 Drill Down and Follow On 747 Dummy Maps 747 ASCII Files 749 Effects of Changing an ASCII File 749 Effects of Changing Delimited ASCII Files 750 Effects of Changing Fixed Width ASCII Files 752 Effects of Changing the Source ASCII File for a D-Link 752 Chapter 16: Analyst Publish

755

Dimensions for Publish 755 Selecting a Dimension for Publish for Reporting 756 Understanding the Publish Process 757 Table-only Publish Layout 757 Database Object Names 758 Items Tables 758 Hierarchies 759 Export Tables 762 Annotations Tables 763 Metadata Tables 763 Creating a Table-only Publish Layout 766 View Publish Layout 768 Database object names 768 D-Lists 769 D-Cube Data and Export Tables 769 Annotations 769 Metadata 770 Views 771 Creating a View Publish Layout 771 Publishing Using the Command Line 773 Create a DSN for the Database Server 773 Set up Microsoft SQL Server 2000 Desktop Engine as a Database Server 775 Reporting Directly From Publish Tables 775 Generate Framework Manager Model Wizard 776 Configuring Your Environment 777 Create a Set of Framework Manager Models 779 Update a Framework Manager User Model 780 Generate Transformer Model Wizard 780 Configuring Your Environment 781 Generate a Transformer Model 781 Creating PowerCube(s) 782 Model Changes that Impact the Publish Tables 783 Chapter 17: Printing and Previewing

785

Print Setup 785 Previewing 786 Preview Formulas and Details of a D-List 786 Preview a D-Cube 786 Preview D-Cube Summary Information 787 16 Analyst

Table of Contents Printing 788 Print a D-List and Formulas 788 Print a D-Cube 788 Print Nested Macros 789 Print a List of Objects in a Library 789 Print to .csv Files 790 Print Annotations 790 Glossary Index

791

799

User Guide 17

Table of Contents

18 Analyst

Introduction
This document is intended for use with Cognos 8 Planning - Analyst. It helps you understand how to use all of the functionality in Analyst. Cognos 8 Planning provides the ability to plan, budget, and forecast in a collaborative, secure manner. The major components are Analyst and Contributor.

Cognos 8 Planning - Analyst

Analyst is a flexible tool used by financial specialists to define their business models. These models include the drivers and content required for planning, budgeting, and forecasting. The models can then be distributed to managers using the Web-based architecture of Cognos 8 Planning Contributor.

Cognos 8 Planning - Contributor

Contributor streamlines data collection and workflow management. It eliminates the problems of errors, version control, and timeliness that are characteristic of a planning system solely based on spreadsheets. Users have the option to submit information simultaneously through a simple Web or Microsoft Excel interface. Using an intranet or secure Internet connection, users review only what they need to review and enter data where they are authorized. For more information about using this product, visit the Cognos Global Customer Services Web site (http://support.cognos.com).

Best Practices for Cognos 8 Planning

The Cognos Innovation Center for Performance Management provides a forum and Performance Blueprints which you can use to discover new ideas and solutions for finance and performance management issues. Blueprints are pre-defined data, process, and policy models that incorporate best practice knowledge from Cognos customers and the Cognos Innovation Center. These Blueprints are free of charge to existing customers or Platinum and Gold partners. For more information about the Cognos Innovation Center or the Performance Blueprints, visit http://www.cognos.com/ innovationcenter.

Audience
This guide is for both new and experienced Analyst users. Familiarity with financial data is helpful, but not required.

Related Documentation
Our documentation includes user guides, getting started guides, new features guides, readmes, and other materials to meet the needs of our varied audience. The following documents contain related information and may be referred to in this document.

User Guide 19

Introduction Note: For online users of this document, a Web page such as The page cannot be found may appear when clicking individual links in the following table. Documents are made available for your particular installation and translation configuration. If a link is unavailable, you can access the document on the Cognos Global Customer Services Web site (http://support.cognos.com). Logon credentials are available either from your administrator or by request from support.america@cognos. com.

Document
Contributor Administration Guide Manager User Guide Analyst for Microsoft Excel Tutorial Cognos 8 Planning Installation and Configuration Guide

Description
Creating and administering Cognos 8 Planning - Contributor Applications. Using Cognos 8 Planning - Manager Getting to know Cognos 8 Planning - Analyst for Microsoft Excel

Installing and Configuring Cognos 8 Planning Products

Finding Information
To find the most current product documentation, including all localized documentation, access the Cognos Global Customer Services Web site (http://support.cognos.com). Click the Documentation link to access documentation guides. Click the Knowledge Base link to access all documentation, technical papers, and multimedia materials. Product documentation is available in online help from the Help menu or button in Cognos products. You can also download documentation in PDF format from the Cognos Global Customer Services Web site. You can also read PDF versions of the product readme files and installation guides directly from Cognos product CDs.

Getting Help
For more information about using this product or for technical assistance, visit the Cognos Global Customer Services Web site (http://support.cognos.com). This site provides product information, services, user forums, and a knowledge base of documentation and multimedia materials. To create a case, contact a support person, or to provide feedback, click the Contact Us link. For information about education and training, click the Training link.

Printing Copyright Material

You can print selected pages, a section, or the whole book. Cognos grants you a non-exclusive, non-transferable license to use, copy, and reproduce the copyright materials, in printed or electronic

20 Analyst

Introduction format, solely for the purpose of operating, maintaining, and providing internal training on Cognos software.

User Guide 21

Introduction

22 Analyst

What's New?
This section contains a list of new features for this release. It will help you plan your upgrade and application deployment strategies and the training requirements for your users. For information about upgrading, see the Cognos 8 Planning Installation and Configuration Guide. To review an up-to-date list of environments supported by Cognos products, such as operating systems, patches, browsers, Web servers, directory servers, database servers, and application servers, visit the Cognos Global Customer Services Web site (http://support.cognos.com).

New Features in Version 8.3

Listed below are new features since the last release.

Microsoft Excel 2007

This release supports Contributor and Analyst for Excel using Microsoft Excel 2007. For more information about using Analyst or Contributor with Excel, see the Analyst for Microsoft Excel User Guide and Contributor for Microsoft Excel User Guide.

Select Folders in Cognos Connection

This release supports folder selection in Cognos Connection when generating Framework Manager and Transformer models. For more information see the following sections. "Create a Set of Framework Manager Models" (p. 779) "Generate a Transformer Model" (p. 781)

Select Framework Manager Package for D-List Import

This release supports browsing in Cognos Connection folders to select a Framework Manager package for a D-List import. For more information, see "Import D-List Items From a Cognos Package" (p. 130).

User Guide 23

What's New?

24 Analyst

Chapter 1: Analyst Overview

When you open Analyst, you are asked to select a namespace and log on with your user id and (optional) password. Passwords are not case sensitive. If you decide not to use a password, leave the Password box blank. When you close Analyst, if you have any open objects, you are prompted to save them.

Steps
1. Click Start, Programs, Cognos 8, Cognos Planning - Analyst. 2. Choose a namespace and click OK. 3. Type your User ID and [optional] password, and click OK. 4. Select a default library to use for the session. Click OK. 5. To close Analyst, from the File menu, click Exit.

Saving Analyst Data

The data directories are where users store and access data. All users must have full network access rights to their own as well as all other users' data directories. Users store their D-Cubes and other data in directories.You can restrict user access internally within Analyst. We recommend that you do not store your data in the directory where Analyst is installed. Save your data in another directory, preferably on another drive.

Registry Settings
When the application is installed, the registry keys created by the install are written to <HKEY_LOCAL_MACHINE> (HKLM). This allows several users to use the same PC for running Analyst. Writing to HKLM requires power-user rights and access rights to the registry. All registry settings created or changed at runtime are written to <HKEY_CURRENT_USER> (HKCU). Analyst usually writes to HKCU. When trying to read from the registry, it will first look in HKCU to find a specific value. If it doesn't find the value it needs, it defaults to the value in HKLM. Consequently, multiple users can use the same machine but have different registry settings in the HKU/SID, which will keep customized settings from being overwritten by other users. Optional: Changes can be made to the keyboard layout. Only users with power user rights can change these settings. You can also change the Maximum Workspace Size (kb) setting in Cognos Configuration.

User Guide 25

Chapter 1: Analyst Overview

Restart Analyst
Restarting Analyst allows you to select a different default library, or to return to Cognos Planning - Manager if you opened Analyst through Manager.

Steps
From the File menu, click Restart. If you have open objects, you are prompted to save them.

Understanding Analyst
Analyst is very different from Excel. If you have an Excel background, it is important to understanding how Analyst differs from Excel in the key areas.

Formulas in Analyst and Excel Spreadsheets

In a spreadsheet, formulas generally use cell references demonstrated in the following: A4 = B2 + C3 In Analyst, the formulas use D-List item names as demonstrated in the following: Profit = Sales - Costs In a spreadsheet, you can set up formulas that can refer to different rows and different columns (and different pages). In the example above, the cells B2 and C3 are in different rows and columns from the result in cell A4. However, formulas set up in Analyst refer to other items in the same D-List, so formulas always span across the rows or down the columns. This means that a different approach to formulas must be adopted to make complex systems more structured and easier to maintain. When you edit a formula in a spreadsheet, you edit only a single cell at a time. You then must copy the formula so that the relative addresses are updated. In a spreadsheet, you cannot simply add a column or row and automatically expect existing formulas to be copied. However, when you edit a formula in a D-List, that formula affects every cell that uses the D-List and implements the changes without your needing to copy the amended formula. So, for example, a four-dimensional model with twenty items in each D-List and a subtotal at the bottom of each list would contain only four formulas in 400 pages of data. The same model in a series of flat spreadsheets would contain over 20,000 formulas; thus, it would take major effort and discipline to maintain these formulas even with aids such as copy commands and macros. A summary of the major differences between Analyst and spreadsheets is given below.

Spreadsheet Formulas
Cell specific

D-List Formulas
Global

26 Analyst

Chapter 1: Analyst Overview

Spreadsheet Formulas
Spreadsheet specific Use cell references or range names Change by editing a cell then copying

D-List Formulas
Can be used in multiple D-Cubes Use item names Change by editing a single formula: copied automatically

Flexible: can cross-reference different rows, Structured across rows or down columns or pages columns and worksheets Use formulas to refer to a different worksheet Thousands in a large spreadsheet @function ( ) Use D-Links to refer to a different D-Cube

Few are needed Built in Functions (BiFs)

Using LIB and LIBNO Parameters

System parameters are used to insert information into object names, descriptions, paths. System parameters can be used when creating file maps and D-Links. The table below outlines the system parameters used in Analyst.

System Parameter
{LIB}

Description
Inserts the current library path of the object where {LIB} is used. Can be used in file maps, import D-Links, and the @DCubeExport macro.

{LIBNO}

Inserts the current library number of the object where {LIBNO} is used. Can be used in file maps.

Other Differences
Multi-page spreadsheets generally are only three-dimensional, whereas D-Cubes can contain four, five, or even six dimensions. Some spreadsheets now can be pivoted so page labels become column headings and vice versa. The options for viewing D-Cubes go further than this. The rows, columns, and pages can be interchanged. Spreadsheets generally use cell references for formulas and store the calculations in the worksheet. D-Cubes, however, do not store the formulas. Rather, the formulas are stored in D-Lists as separate entities. The practical effect of this is that one set of formulas can be used in many User Guide 27

Chapter 1: Analyst Overview different D-Cubes. D-List formulas use the same names throughout, so a formula is not dependent on a relative cell reference or the position of a cell. In spreadsheets, the arithmetic is generally one way. Data is typed in designated cells and the results are shown in other cells. Any attempt to type data in formula cells results in an error. In D-Cubes, the arithmetic is two-way: Targets can be set by typing data in a formula cell and splitting it to its component parts according to predefined rules. For example, you could enter a budget for the full year, and the program would split the annual figure into months by allocating it pro rata across a given seasonality profile. This two-way arithmetic is known as breakback. To refer to a different worksheet, spreadsheets generally use formulas. To refer to a different D-Cube, D-Cubes use D-Links rather than formulas.

Multi-User Object Access

At any one time, only one person is allowed to write to an object. Suppose two people have been assigned write-access to the same D-Cube. The first person who opens the D-Cube will open it in write mode and can save any changes to the original. The second person to open the D-Cube will be told that the D-Cube is currently being updated. They will be permitted to open the D-Cube in read-only mode. Any changes that are made by the second person will not be saved although they are permitted to save a copy with a different name. This can only be used later if a D-Link is set up to overwrite the original with the copy. Suppose the second person wants to make changes to the original. He or she must contact the first person and request that they save what they are doing and switch to read-only mode. This will allow the second person to open (or reopen) the D-Cube in write-mode. He or she can then make changes to the original D-Cube and save it. A similar situation exists for any object where two or more people want to make changes to an object simultaneously. Any object can be opened in read-only mode to allow another user to update it elsewhere. However, after you have switched to read-only mode, you can not revert to write-mode without closing and reopening the object. This ensures that you are always working on the latest version.

Analyst Samples
Sample models are provided with Cognos Planning. Samples illustrate key features in these products. Samples are installed to the following location for English installation _location\samples\en\Planning. For French and German language installs, the path would be ...\samples\fr\Planning and ...\samples\ de\Planning. They can be accessed directly from Cognos Planning - Analyst without any further configuration if the system administrator has installed them. The following table lists the samples available in Cognos Planning. This is not necessarily an exhaustive list. Visit the Cognos Global Customer Service Web site (http://support.cognos.com) for more information. 28 Analyst

Chapter 1: Analyst Overview We recommend that you do not edit and save any of these samples directly. Instead, copy the required D-Lists, including any formats such as D-List formats, into another library, and then rebuild the D-Cubes. You may want to use other dimensions such as Time, as appropriate.

Name
tutorialgo

Planning Component
Analyst

Description
A lightweight model used with the Analyst Tutorial and Analyst for ExcelTutorial to demonstrate key Analyst functionality Used to demonstrate and test much of the Analyst functionality. Provides working examples of all Built in Functions. Contains a macro which runs D-Links to D-Cubes in either Analyst or Contributor in a series of steps, each targeting a small slice or chunk of the target D-Cube, and thus requiring far less memory to execute the link.

great outdoors analyst

Analyst

bif

Analyst

slice_update

Analyst

Tutorialgo
The Analyst Tutorial demonstrates how to create a simple Analyst model. The tutorialgo library is the finished model. It is designed to introduce users to the concepts of Analyst. In particular, the principles of the D-List, D-Cube and D-Link. See the Analyst Tutorial for more information. Analyst for Excel tutorial also uses this library, see the Analyst for Microsoft Excel Tutorial for more information. The tutorialgo model does not make full use of the functionality available in Analyst. For more functionality, use the great outdoors analyst model.

Great Outdoors Analyst

The great outdoors analyst sample is designed to illustrate many of the features and functionality available in Analyst. It is an example of a plan of expenditure of a fictional business. The structure of the great outdoors analyst model is described in the following sections.

Depreciation
The Depn policy D-Cube holds the asset life and depreciation methods for all the different asset types.

User Guide 29

Chapter 1: Analyst Overview The Asset purchases D-Cube allows entry by division, by budget version of the cost, asset type, month of purchase, and month to start depreciation. The Depreciation D-Cube allows calculation of depreciation on new asset purchases, and combines this with depreciation on existing items. This is typically imported from a fixed asset register to determine appreciation charge.

Sales
The Price and cost D-Cube holds sales price and unit cost by product and by version. The Sales plan D-Cube determines the calculation of a growth margin from its elements by product, channel, division, month, and version.

Salaries
The Base Salaries D-Cube makes the assumption that individual salaries are based on grade. It holds the base salary by grade, division, and, version. The Salary plan D-Cube determines the salary by individual, version, and by month based on their division, grade, and month they are due an increase. The start date has been included for information purposes. It is not part of the calculation of the salary, although this functionality could be included if required.

Expenses
The Expenses D-Cube has links from Salary Plan and Commissions to import the appropriate data. Additionally the Expenses D-Cube holds details of the overhead expenditure. The overhead expenditure is planned at annual level and breakback functionality is then used to apportion it to months appropriately. The OH adjust D-Cube holds the annual base data by overhead item, division, and month. It allows a percent change to be applied to the base amount to provide an adjusted amount. The OH profiles and Profile types D-Cubes enable the adjusted overhead values to be apportioned into the appropriate months depending on the profile to be used and the monthly structure of that profile. The Commissions D-Cube takes the growth sales revenue and applies the appropriate commission percentage to determine the commission charge by division, month, and version.

Income Statement
The Franchise rev D-Cube holds the imported franchise revenue by division, month, and version. The Inc Statement D-Cube is a summary D-Cube combining the summary level from the other D-Cubes. In addition, it allows for below the line expenditure to be entered, by division.

Miscellaneous
Additionally, this sample includes the following: Csv files that contain the base data D-Links to import the data

30 Analyst

Chapter 1: Analyst Overview Macros to help with model maintenance Manager screens have been provided to aid the understanding and usability of the sample. They consist of a front page, two flowcharts, and some sample screens to manage and maintain the model.

The BiF Library

The BiF library provides working examples of all Built in Functions. It is optionally installed with the software. This self-contained BiF library may be accessed using Cognos Planning - Analyst, however a Cognos Planning - Manager front end has been provided to simplify navigation. There are four reports included in the BiF library. They are linked together and provide links to the appropriate D-Cubes. Where appropriate, graphs are provided to further explain the impact of the different methods available. The BiF library contains a D-Cube for each BiF. Each D-Cube in the BiF library consists of a minimum number of dimensions, enough to demonstrate functionality only. When working with your own D-Cubes, it is likely that they will contain more dimensions than those demonstrated in the library. Where ever possible, the D-Cube data supplied in the BiF library mirrors the Online Help D-Cube data examples to further integrate the BiF library and the Online Help. D-Cubes in the BiF library should therefore not be saved following any changes. Where appropriate, annotations have been provided to further explain the data provided.

The Slice Update Sample

The Slice Update sample is supplied as an optional download from the Cognos Global Customer Services Web site (http://support.cognos.com). The slice_update sample contains a macro which runs D-Links to D-Cubes in either Analyst or Contributor, in a series of steps each targeting a small slice or chunk of the target D-Cube. This means that far less memory is required to execute the link. To ensure the whole D-Cube is updated, a series of slices is required, instead of setting these up manually (create a looping macro to run the D-Link to each defined slice sequentially). This sample model contains examples of macros which allow the target D-Cube to be sliced on either one, or two dimensions in any number of steps.

Differences between Analyst and Contributor

When using slice_update, consider the differences between Analyst D-Cubes and Contributor cubes as targets. If Analyst is your target, then the D-Cube must be updated using a @DCubeUpdate command. The slice_update does not operate if a @DLinkExecute command is used. If Contributor is the target, then the commands @DLinkExecute or @DLinkExecuteList can be used in the macro. Where Contributor is the target, you must store the elements making up the looping macro in a separate library from the library in which the Contributor application is based. This is because the

User Guide 31

Chapter 1: Analyst Overview Steps cube does not need to contain the e.List, but must have an internal D-Link. Contributor does not allow internal D-Links if the cube does not contain the e.List. This sample has been set up so that it will work in Analyst. For information on using it with Contributor, see (p. 34). There are three parameters in the @SliceUpdate macro:

Lists
A list of the D-Lists which are to be used as slice dimensions.

Lookup
A slice of a D-Cube which stores data about the dimensions being sliced on. When Analyst is the target, the rows D-List of this D-Cube must contain items that have the same names as some or all of the items in the D-Lists in the target D-Cubes on which you are slicing. When Contributor is the target, the rows dimension of this cube must contain the same items as the dimension being sliced on in the Contributor cube, and it must also have exactly the same name.

Criteria
The basis for deciding which elements of the target D-Cube should be targeted when the D-Link is run.

The D-Cubes
The Target Cube Analyst1 D-Cube has the dimensions: Single Item, Target, sources, Divisions, and Versions. The Target Cube Analyst2 D-Cube has the dimensions: Single Item, Target, sources, European Areas, and Versions. The European Areas D-List contains some of the items in the Divisions D-List but not all of them. There is one link targeting each D-Cube. In each case it is included in the @DCubeUpdate list. When the D-Cubes are updated, either the Versions dimension, the Divisions/European Areas dimension, or both could be used to define the slices.

Example 1 - A One Dimensional Slice on the Versions D-List

The Steps slice dimension 1 D-Cube was updated to contain the same items as the Versions D-List. The update steps have been defined as follows 1st step 2nd step 3rd step 4th step Version 4 and Version 5 Version 2 Empty Version 6.

32 Analyst

Chapter 1: Analyst Overview The empty step is left in to illustrate that the macro will not stop if one step is empty. Versions 1, 3, and 7 are not set to be updated. The 'Target Cube Analyst 1' and 'Target Cube Analyst 2' D-Cubes are all zero. The 'Update macro target analyst' macro runs the D-Cube update steps for both D-Cubes. The single dimension slice update is defined in the 'Template Macro 1' macro. The Versions dimension is defined as the dimension on which to slice. Any item from the Versions D-List is targeted when the D-Link is run, if it is defined as the 1st step in the Current Step column of the Slice Dimension 1 D-Cube. The update steps were defined in a separate column of this same D-Cube.

Template Macro Loop 1

When the 'Template Macro Loop 1' macro is run, the following process takes place. The data input in the Specify Step column of the 'Slice Dimension 1' D-Cube specifying which version is updated in each step is transferred by D-Link to the Current Step column. The macro that updates the two D-Cubes runs under the control of the slice update mechanism so that only Versions defined as 1st Step are targeted, in this case Version 4 and Version 5. A command in the macro deducts 1 from each item in the Current Step column of the 'Steps slice dimension 1' D-Cube. Any item which was previously 2nd Step now becomes 1st Step. The macro tests the data in the Current Step column. If any items greater than 1 remain it restarts itself from the beginning, otherwise it stops.

When the macro is complete, in the Target D-Cubes, Versions 2, 4, 5 and 6 now contain a number 1 as they have been targeted by the link.

Example 2 - A two dimensional slice on Versions and Divisions/European Areas.

If the D-Cubes you are targeting are very large, you may want to slice them in two dimensions. When using this macro with the sample model, ensure that both target D-Cubes are set to zero. The two dimensional looping macro illustrates this process. It uses the same slices for the Versions D-List as defined in example 1. Additionally, in the 'Steps slice dimension 2' D-Cube, steps for the Divisions/European Areas are defined. Americas and Central Europe are to be updated as Step 1. Southern Europe is to be updated as Step 2. The slice update is defined in the 'Template Macro 2' macro. Note that both Divisions and European Areas D-Lists are included in the List of Lists. Because the 'Template Macro 2' macro has been set up as a nested macro, the update will take place as follows. All of the processes defined in the first example will take place for Americas and Central Europe first. These processes are repeated for Southern Europe. User Guide 33

Chapter 1: Analyst Overview When the macro has finished, only cells in Americas, Central Europe, or Southern Europe and Version 2, 3 5 or 6 will contain a 1. This means that only very small slices of the D-Cube have been updated in each round.

Steps to Adapt the Template for a User Model

Adapt the template for a user model to slice a D-Cube into two dimensions. 1. Decide which target dimensions in your model are to be used for slicing. 2. Update the D-Lists 'Slice Dimension 1' and 'Slice Dimension 2' (only for a two-dimensional slice) with items from your D-Lists. 3. Open the 'Template Macro 1' macro and replace the second command line with your own update macro. 4. Edit the SliceUpdate parameters so that the List of Lists contains your D-Lists. 5. Set the update steps in the 'Steps slice dimension 1' and 'Steps slice dimension 2' D-Cubes, if applicable. Running the macro DCO Steps 1 or DCO Steps 2 will open the correct slice. 6. Run either 'Template macro loop 1' or 'Template macro loop 2', depending on whether you set up a one or two dimensional slice.

Using the Slice Update Macro with Contributor

When using a Contributor target, it is only possible to have one D-List in the List of Lists. This could be your Contributor e.List but does not have to be. If you have several dimensions containing similar items on which you wish to slice, as in the case of Divisions/European Areas, you must set up separate slice update Lookup cubes for each of them.

Steps
1. Rename the 'Steps slice Dimension 1' D-Cube to exactly the same name as your Contributor dimension. 2. Update the D-List with all the items contained in the Contributor dimension. 3. Edit the SliceUpdate parameter in 'Template macro 1' so that the list of lists contains only this D-List. 4. Replace the second command line with your own update macro. 5. Set your update steps in the 'Steps slice Dimension 1 cube. Running the 'DCO Steps 1' macro opens the correct slice. 6. Run the 'Template macro 1' macro.

Customizing the Analyst Toolbar

Customize the Analyst toolbar to add user-created menus and buttons.

34 Analyst

Chapter 1: Analyst Overview

Use Custom Menus

User defined menus are added before the Window and Help menus on the menu bar. When using custom menus, you must create a .txt file.

Steps
1. From the Tools menu, click Options. 2. Click the Custom tab. 3. In the Custom Menu File text box, type the path of the custom menu, or browse for the file. 4. Click OK. 5. You are prompted to restart Analyst. 6. For the changes to take effect, click Yes. Analyst closes and re-opens.

Example of a custom menu file

;Custom menu for Cognos 1Cognos &Incorporated 2&Lists 3&Timescale::M|600102 DLO-periods 3&Versions::M|600102 DLO-Budget Versions 3&Profit and Loss Accounts::M|600102 DLO-Accounts 2&Update 3&Profit and Loss::M|600102 DCU-Accounts When creating custom menus, only macros (p. 613) can be used. For example, M|600102 DLO-Accounts runs an @OpenDList macro named DLO-Accounts from library number 600102. When a line begins with a semicolon (;), it is a comment. Otherwise, the first character must be a digit specifying the Level of the menu (item). The level may be 1, 2, or 3. Level 1 is a menu on the main menu bar, level 2 is a submenu of level 1 items, and level 3 is a submenu of level 2. You must start with a level 1 item, that will appear on the analyst toolbar. Following this, you use a level 2 item to define the first sub-menu step. If this level 2 item has a sub-menu, you follow it with level 3 steps until the sub-menu is fully defined. You then proceed to use another level 2 item, because it is the second step of the main menu. This can again be followed by level 3 items if needed. If you include another level 1 item, it will also appear on the toolbar and can be followed by sub-menus of its own. Continue until your whole menu is defined. Immediately following the digit defining the level up to the first colon (:), is the caption to go on the menu (item). If one character is preceded by an ampersand (&), this character is underlined and can be used to select the item when the menu is active. From the first to the second colon is the tip text for the menu (item); currently tip texts are not displayed for the menus and are used only as comments in the menu definition file.

User Guide 35

Chapter 1: Analyst Overview For a menu item, the action code follows the second colon. The only documented action code is "M|" for a macro call. Other action codes exist but currently are for internal use only. Macros are defined by a numeric library number followed by the full macro name. Note: Only use letters after the ampersand (&) or Analyst will not function. The character following the ampersand (&) must be a letter of the English alphabet. If you use other characters, Analyst will not work on start-up.

Create Custom Toolbar Buttons

You can create custom toolbar buttons in Analyst that, when clicked, will run macros (p. 613). The user-defined toolbar buttons display on the toolbar before the execute button (which appears only when there are unprocessed entries in a D-Cube). When you create the custom toolbar file, you must save it as a text (.txt) file. For example, a sample toolbar .txt file is C:\Program Files\Cognos\YourDirectory\CustomToolbar. txt. The layout of a custom toolbar text file to add a button which opens a D-List called 'Accounts' in library number 600102 would appear as follows: ;custom toolbar file Open Accounts D-List,C:\Program Files\Cognos\YourDirectory\DLOPEN. bmp,10,M|600102 DLO-Accounts If a line begins with a semicolon (;), it is a comment. Otherwise, each record must contain the following four fields, separated by commas. A description, which is also used as a tip on the button. The full path of the bitmap file to use for the button, for example, C:\Program Files\Cognos\ YourDirectory\DLOPEN.bmp. Offset in pixels from the previous button, 10 in this example. An action code to run a named macro. For example M|600102 DLO-Accounts runs an @OpenDList macro named DLO-Accounts from library number 600102.

Note: When creating the Custom Toolbar text (.txt) file, ensure that there are no spaces between the commas separating the four fields.

Steps
1. Create the custom toolbar text (.txt) file. 2. Create (or use an existing) a toolbar button saved as a bitmap (.bmp) file. 3. From the Tools menu, click Options. 4. Click the Custom tab. 5. In the Custom Toolbar File text box, type the path of the custom toolbar, or browse for the file, and then click OK. 6. Click Yes to restart Analyst.

36 Analyst

Chapter 2: Administration
There are a number of administrative tasks that must be performed on a regular basis. A necessary part of application administration is the care and maintenance of the Analyst database. As a regular part of these efforts, you must: view and edit Analyst workspace settings(p. 37) set FileSys.ini file (p. 38) rebuild index files (p. 39) refresh references to objects (p. 39) validate D-Lists (p. 39) close ODBC links (p. 40) manage memory (p. 40) manage masks (p. 40) administer users (p. 40) set configuration settings (p. 40) administer groups (p. 41)

Viewing and Editing Analyst Workspace Settings

You can view and edit your Analyst workspace settings. Their is a separate Cognos configuration for every computer where Analyst is installed that controls the settings on that computer. Use Cognos Configuration to change the Analyst Workspace settings. Edit and view the workspace setting from within Cognos Configuration. You can also view the workspace settings from the Analyst Tools, Options menu.

Steps to Edit and View Workspace Settings

1. Start Cognos Configuration. 2. In the Explorer window, under Environment, click Planning. 3. Under Planning - Component Properties, Analyst maximum workspace size in KB, Value, enter a workspace value in KB. 4. From the File menu, click Save.

User Guide 37

Chapter 2: Administration The amount of memory you specify can be a value between 64000 and 2000000, and the memory is allocated as it is needed up to the limit you set.

Steps to View the Workspace Setting From Analyst

1. In Analyst, from the Tools menu, click Options. 2. Click the General tab. 3. In the Maximum Workspace Size (kb) box, you can view the current maximum workspace setting.

Creating Planning Tables

The first time Planning is run, a check is run to see if any Planning tables exist in the Planning content store. If not, you are prompted to create them. You can either create them directly or select the option to create a script which can be modified and run by a Database Administrator (DBA). You also have the option to specify the path to the filesys.ini file. You change this setting if the default path is not used. The filesys.ini file is a control file that contains file paths for the Libs.tab, Users.tab, and Groups.tab that control the specific library and user setup. You can edit the filesys. ini path by selecting Tools, Edit FileSys.Ini path.

Set Filesys.ini
The filesys.ini file is the main control file that is referred to when the program starts. You can specify the pathnames of the Libs.tab, Users.tab, and Groups.tab that control your specific library and user set-ups. You can edit the location of the filesys.ini file from Analyst, and from the Contributor Administration Console.

Steps in Analyst
1. From the Tools menu, click Options. 2. Click the General tab. 3. In the Active Filesys.ini file box, click type the path of the filesys.ini file or click Browse to locate the file manually.

Steps in the Contributor Administration Console

1. From the Tools menu, click Edit FileSys.Ini path. 2. Type or browse to the path of the filesys.ini.

38 Analyst

Chapter 2: Administration

Rebuild Index Files

When copying or moving large models using Windows Explorer, the index files can become corrupt. Index files are system-generated tables that list all objects in a library. Their DOS names are given the suffixes .ho, .d1 and .d2, and, on rare occasions, they may need to be rebuilt and then refreshed.

Step
From the File menu, click Administration, Rebuild Index Files. The index files will be rebuilt from DOS pathnames of objects in the current library.

Refresh References
Because objects have references to other objects - for example, a D-Cube refers to its D-Lists, a D-Link refers to its source and target D-Cubes, and a selection refers to its D-Cube and D-Lists, and because models can become very large and complicated, references may become corrupt. On rare occasions, you may need to refresh these references.

Step
From the File menu, click Administration, Refresh References. The references to other objects in the current library are refreshed. If any libraries on the system are ever offline, and work is done on the remaining libraries, it will be necessary to refresh references on these libraries when they come back online. Also, if work has been done on the libraries while offline, all remaining libraries must be refreshed when it all comes online again.

Validate D-Lists
This function enables you to check the syntax of every formula in every D-List in the current library.

Step
From the File menu, click Administration, Validate D-Lists. Cognos Planning - Analyst checks the validation of each calculation contained in every D-List.

DOS File Names

The program automatically generates a DOS file name for each object you create. Each file will have a suffix indicating which type of object it is. Never attempt to copy or move files in DOS or Windows Explorer because it can create problems with the references between objects and the system.

User Guide 39

Chapter 2: Administration

Close ODBC Links

If you select the option to keep an ODBC link open at the time of setting it up, you can go back and close the online links with a database.

Step
From the File menu, click Close ODBC.

Configuration Settings
You can alter the configuration settings. These control the path to the start-up control file, the maximum memory usage, the undo/redo facility, and the customized menu options.

Steps to use a custom menu file

1. From the Tools menu, click Options. 2. Click the Custom tab. 3. In the Custom Menu File text box, type the path of the custom menu. 4. Click OK. 5. For changes to take effect, restart Analyst.

Steps to change the keyboard layout

1. From the Tools menu, click Options. 2. Click the Language tab. 3. In the Keyboard Layout box, select a language. 4. Click OK.

Steps to change the number of undos and redos

1. From the Tools menu, click Options. 2. Click the Undo tab. 3. Select or clear the Enable Undo/Redo check box to enable or disable the feature. Note: By default, this is disabled because it takes up valuable memory. 4. Type the stack size of undo in the Undo Stack Size (excluding D-Cube data) box. StackBytes is set to a default of 1024 KB (1 MB). This controls the amount of memory allocated to undoing any operation apart from data entry in cells in a D-Cube. Note: This should not be set higher than the workspace setting. Because this is used for small objects like D-Lists, macros, and D-Links, it does not require a large stack size.

40 Analyst

Chapter 2: Administration 5. Type the stack size of D-Cube data undo in the D-Cube Data Undo Stack Size box. CubeStackBytes in the CP section of the registry or in Cognos.ini is the amount of memory that Analyst uses for storing the cells that are to be undone/redone. These are measured in kilobytes (that is, 1024 KB = 1 MB). The rule of thumb is that 8 bytes is needed for each cell that is to be undone/redone. For example if the D-Cube Data Undo Stack Size = 1024, 128,000 cells can be undone: 1024000 / 8 = 128,000. Thus a 10,000 cell D-Cube selection could be undone 12 times, whereas a 128,000 cell D-Cube selection can be undone only once. This assumes that the Maximum Undoable View Size is increased to allow it. When calculating the number of levels of undo, it is the total number of cells in the selection that counts, not the number of cells that actually have changed. Note: This does not use workspace. It uses a machine's extra PC RAM, and will therefore be restricted by the amount of available RAM and the number of applications that are running. 6. Type a value for the maximum undoable size in the Maximum Undoable View Size (MaxViewCells*8) box. This is the setting that is used to filter what goes into the D-Cube Data Undo Stack Size. By default, this is set to 128, which means a D-Cube selection of more than 16,000 cells cannot be undone (128,000 bytes/8 bytes per cell = 16,000 cells). For example. If you have a large selection from one D-Cube and many small selections from other D-Cubes, you may not be interested in undoing operations on the large D-Cube. You could set the limit to 10000 cells, for instance, so the undo/redo engine will ignore selections larger than 10000 cells (avoiding taking snapshots of them). Note: For very large D-Cubes, if memory is limited, you may want to disable the undo/redo function.

Step to change MAXWS and keyboard settings

Modify the MAXWS and keyboard layout settings are stored in the HKEY_Local_Machine part of the registry. Note: Contact your system administrator if you do not have access.

Steps to locate ODBC sources

1. From the File menu, click Administration, Locate ODBC-Sources. 2. Select the ODBC source required from the drop down box, or click All to search for all ODBC sources. 3. Select the library in which to search, or click All to search in all libraries. 4. Select the check boxes to search for D-lists, D-Links or both. 5. Click Search Now to display a table of objects using the source(s). You may filter on the object name or the table name to limit the number of objects displayed. 6. Click Highlight and specify only the objects that you want to appear.

User Guide 41

Chapter 2: Administration 7. From the table, right-click to open an object, or print or preview the list of objects.

Steps to locate Built in Functions

1. From the File menu, click Administration, Locate Built in Function. 2. Select a BiF from the drop-down box, or click All to search for all BiFs. 3. Select the library in which to search, or click All to search in all libraries. 4. Click Search Now to display a table of objects using the source(s). You may filter on the object name or the table name to limit the number of objects displayed. 5. Click Highlight and specify only the objects that you want to appear. 6. From the table, right-click to open an object, or print or preview the list of objects.

42 Analyst

Chapter 3: Objects
Objects are the basic building blocks used to create models in Cognos Planning - Analyst.

Objects include:
D-Lists (p. 215) D-Cubes (p. 147) D-Links (p. 215) A-Tables (Allocation Tables) (p. 713) File Maps (p. 731) Macros (p. 613) Formats (p. 109) Selections (p. 197) Libraries (p. 301)

Add a Description to an Object

Descriptions and owner's notes can be attached to various objects.

Steps
1. Open the object. 2. From the File menu, click Summary Info. 3. Click the General tab and type an owner's note or description in the Notes box. 4. Click OK. 5. Save and close the object to ensure the notes are kept.

Reveal Object Name from DOS Filename

You can show the Analyst object name from the DOS filename and vice-versa.

Step
Select File, Administration, File to Object then select a file name.

User Guide 43

Chapter 3: Objects The object name and the Windows path are revealed.

44 Analyst

Chapter 4: Security
Cognos 8 security is designed to meet the need for security in various situations. You can use it in everything from a proof of concept application where security is rarely enabled to a large scale enterprise deployment. The security model can be easily integrated with the existing security infrastructure in your organization. It is built on top of one or more third-party authentication providers. You use the providers to define and maintain users, groups, and roles, and to control the authentication process. Each authentication provider known to Cognos 8 is referred to as a namespace. In addition to the namespaces that represent the third-party authentication providers, Cognos 8 has its own namespace named Cognos. The Cognos namespace makes it easier to manage security policies and deploy applications. For more information, see the Cognos 8 Security and Administration Guide.

Cognos Namespace
The Cognos namespace is the Cognos 8 built-in namespace. It contains the Cognos objects, such as groups, roles, data sources, distribution lists, and contacts. During the content store initialization, built-in and predefined security entries are created in this namespace. You must modify the initial security settings for those entries and for the Cognos namespace immediately after installing and configuring Cognos 8. You can rename the Cognos namespace using Cognos Configuration, but you cannot delete it. The namespace is always active. When you set security in Cognos 8, you may want to use the Cognos namespace to create groups and roles that are specific to Cognos 8. In this namespace, you can also create security policies that indirectly reference the third-party security entries so that Cognos 8 can be more easily deployed from one installation to another. The Cognos namespace always exists in Cognos 8, but the use of the Cognos groups and roles it contains is optional. The groups and roles created in the Cognos namespace repackage the users, groups, and roles that exist in the authentication providers to optimize their use in the Cognos 8 environment. For example, in the Cognos namespace, you can create a group named HR Managers and add to it specific users and groups from your corporate IT and HR organizations defined in your authentication provider. Later, you can set access permissions for the HR Managers group to entries in Cognos 8.

User Guide 45

Chapter 4: Security

Authentication Providers
User authentication in Cognos 8 is managed by third-party authentication providers. Authentication providers define users, groups, and roles used for authentication. User names, IDs, passwords, regional settings, personal preferences are some examples of information stored in the providers. If you set up authentication for Cognos 8, users must provide valid credentials, such as user ID and password, at logon time. In Cognos 8 environment, authentication providers are also referred to as namespaces, and they are represented by namespace entries in the user interface.

Cognos 8 does not replicate the users, groups, and roles defined in your authentication provider. However, you can reference them in Cognos 8 when you set access permissions to reports and other content. They can also become members of Cognos groups and roles. The following authentication providers are supported in this release: Active Directory Server Cognos Series 7 eTrust SiteMinder LDAP NTLM SAP

You configure authentication providers using Cognos Configuration. For more information, see the Installation and Configuration Guide.

Multiple Namespaces
If multiple namespaces are configured for your system, at the start of a session you must select one namespace that you want to use. However, this does not prevent you from logging on to other namespaces later in the session. For example, if you set access permissions, you may want to reference entries from different namespaces. To log on to a different namespace, you do not have to log out of the namespace you are currently using. You can be logged on to multiple namespaces simultaneously. Your primary logon is the namespace and the credentials that you use to log on at the beginning of the session. The namespaces that you log on to later in the session and the credentials that you use become your secondary logons. When you delete one of the namespaces, you can log on using another namespace. If you delete all namespaces except for the Cognos namespace, you are not prompted to log on. If anonymous access is enabled, you are automatically logged on as an anonymous user. If anonymous access is not enabled, you cannot access the Cognos Connection logon page. In this situation, use Cognos Configuration to enable anonymous access.

46 Analyst

Chapter 4: Security

Deleting or Restoring Unconfigured Namespaces

You can preserve namespaces and all their contents in the content store even if they are no longer configured for use in Cognos 8. When a namespace is not configured, it is listed as inactive in the directory tool. An inactive namespace is one that was configured, but later deleted in Cognos Configuration. The namespace can be deleted from the content store by members of the System Administrators role. You cannot log on to an inactive namespace. If a new version of Cognos 8 detects a previously configured namespace that is no longer used, the namespace appears in the directory tool as inactive. You can configure the namespace again if you still require the data. If the namespace is not required, you can delete it. When you delete a namespace, you also delete all entries in My Folders that are associated with that namespace, and their contents. An active namespace cannot be deleted, but can be updated. To recreate a namespace in Cognos Configuration, you must use the original ID of the namespace. For information about configuring and recreating namespaces, see the Installation and Configuration Guide.

Delete an Inactive Namespace

If a namespace was removed from Cognos Configuration and is no longer required, a member of the System Administrators role can delete it permanently in the directory tool. Deleting a namespace also deletes all the entries in My Folders that are associated with the namespace. To access the directory administration tool, you must have execute permissions for the directory secured feature and traverse permissions for the administration secured function.

Steps
1. In Cognos Connection, in the upper-right corner, click Launch, Cognos Administration. 2. On the Security tab, click Users, Groups, and Roles. If the namespace you want to delete does not have a check mark in the Active column, it is inactive and can be deleted. 3. In the Actions column, click the delete button. If the namespace is active, the delete button is not available. The namespace is permanently deleted. To use the namespace again in Cognos 8, you must add it using Cognos Configuration.

Users, Groups, and Roles

Users, groups, and roles are created for authentication and authorization purposes. In Cognos 8, you can use users, groups, and roles created in third-party authentication providers, and groups

User Guide 47

Chapter 4: Security and roles created in Cognos 8. The groups and roles created in Cognos 8 are referred to as Cognos groups and Cognos roles.

Users
A user entry is created and maintained in a third-party authentication provider to uniquely identify a human or a computer account. You cannot create user entries in Cognos 8.

Information about users, such as first and last names, passwords, IDs, locales, and email addresses, is stored in the authentication providers. However, this may not be all the information required by Cognos 8. For example, it does not specify the location of the users' personal folders, or format preferences for viewing reports. This additional information about users is stored in Cognos 8, but when addressed in Cognos 8, the information appears as part of the external namespace.

Access Permissions for Users

Users must have at least traverse permissions for the parent entries of the entries they want to access. The parent entries include container objects such as folders, packages, groups, roles, and namespaces. Permissions for users are based on permissions set for individual user accounts and for the namespaces, groups, and roles to which the users belong. Permissions are also affected by the membership and ownership properties of the entry. Cognos 8 supports combined access permissions. When users who belong to more than one group log on, they have the combined permissions of all the groups to which they belong. This is important to remember, especially when you are denying access. Tip: To ensure that a user or group can run reports from a package, but not open the package in a Cognos studio, grant the user or group execute and traverse permissions on the package.

Groups and Roles

Users can become members of groups and roles defined in third-party authentication providers, and groups and roles defined in Cognos 8. A user can belong to one or more groups or roles. If users are members of more than one group, their access permissions are merged. Groups and roles represent collections of users that perform similar functions, or have a similar status in an organization. Examples of groups are Employees, Developers, or Sales Personnel. Members of groups can be users and other groups. When users log on, they cannot select a group they want to use for a session. They always log on with all the permissions associated with the groups to which they belong. Roles in Cognos 8 have a similar function as groups. Members of roles can be users, groups, and other roles. The following diagram shows the structure of groups and roles.

48 Analyst

Chapter 4: Security

Group

Role

User

Group

User

Group

Role

You create Cognos groups and roles when you cannot create groups or roles in your authentication provider groups or roles are required that span multiple namespaces portable groups and roles that can be deployed are required In this case, it is best to populate groups and roles in the third-party provider, and then add those groups and roles to the Cognos groups and roles to which they belong. Otherwise, you may have trouble managing large lists of users in a group in the Cognos namespace. you want to address specific needs of Cognos 8 administration you want to avoid cluttering your organization security systems with information used only in Cognos 8

Cognos 8 Planning Roles

There are two predefined roles for Cognos 8 Planning: Planning Rights Administrators This role enables you to access Contributor Administration Console, Analyst, and all associated objects in the application for the first time following installation. You can then change the roles, groups, and users who can access the Contributor Administration Console and to Analyst. Planning Contributor Users This is the default role for users who want to access the Contributor Web client, Contributor for Excel, or Analyst. However, anyone can be assigned rights to use the Contributor Web client, or Contributor for Excel regardless of whether they are a member of the Planning Contributor Users role. Analyst users must be members of the Planning Contributor User role. Note: You do not have to use these roles, they can be deleted or renamed. If you decide not to use the predefined roles, you must assign the access permissions and capabilities required by Cognos 8 Planning to other groups, roles, or users.

Capabilities
Capabilities are secured functions and features. If you are an administrator, you set access to the secured functions and features by granting execute permissions for specified users, groups, or roles. Users must have at least one capability to be accepted through the Cognos Application Firewall.

User Guide 49

Chapter 4: Security The Planning Contributor Users role has the Planning Contributor capability by default. If you do not want to use this role, you can assign the capability to any groups, users, or roles that you create to replace this role by giving execute permissions to the appropriate members. The Planning Rights Administrators role has the Planning Rights Administration capability by default. To assign this capability to groups, users, or roles, you must give execute permissions to the appropriate members. You must also give members permissions to traverse the Administration folder. Tip You change capabilities through Cognos Administration, by clicking the Security tab. For more information, see "Securing Functions and Features" in the Administration and Security Guide.

Capabilities Needed to Create Cognos 8 Planning Packages

You can create a Planning Package during the Go to Production process, giving users access to Cognos 8 studios from the Contributor application and enabling users to report against live Contributor data using the Planning Data Service. To do this, the Planning Rights Administrators role must be granted the Directory capability. Members of the System Administrator role are automatically granted this capability, but Planning Rights Administrator members are not.

Setting up Security for a Cognos 8 Planning Installation

You must set up security for a Cognos 8 Planning installation. To configure security for Cognos 8 Planning, do the following: Using Cognos Configuration, configure Cognos 8 to use an authentication provider Using Cognos Administration add Contributor Administration Console and Analyst administrators to the Planning Rights Administrators role add Contributor application members and Analyst users to the Planning Contributor Users role enable Planning Roles in Cognos 8 restrict access to the everyone group create additional roles or groups for Cognos 8 Planning (optional) Note: We recommend that you add groups of users as defined in your authentication provider to the roles in Cognos 8 Planning, rather than individual users. This means that changes in group membership are reflected immediately in the roles without having to make changes in Cognos 8 To configure Analyst security: configure integrated Windows authentication if you want to execute macros without interaction specify a default library

50 Analyst

Chapter 4: Security assign access at object, library, or item level

Using the Contributor Administration Console set access rights for Contributor administrators to Contributor administration functions set rights for Contributor application users for Contributor applications

Configure Cognos 8 to Use an Authentication Provider

Cognos 8 components can run with two types of access: anonymous and authenticated. By default, anonymous access is enabled. To use Cognos 8 Planning, you must disable anonymous access so that users are required to log on. Only authenticated users can access your Cognos 8 Planning applications. For authenticated access, you must configure Cognos 8 components to use a namespace associated with an authentication provider used by your organization. You can also configure multiple namespaces. At run time, users can choose which namespace they want to use. Note: If you are using the Generate Transformer Model extension, you must add the Cognos Series 7 namespace. Local authentication export (LAE) files cannot be used.

Steps to Disable Anonymous Access

1. On each Content Manager computer, start Cognos Configuration. 2. In the Explorer window, under Security, Authentication, click Cognos. The Cognos resource represents the Cognos namespace. For more information, see the Administration and Security Guide. 3. In the Properties window, ensure that Allow Anonymous Access is set to False. 4. From the File menu, click Save.

Steps to Configure Authentication Providers

1. On each Content Manager computer, start Cognos Configuration. 2. In the Explorer window, under Security, right-click Authentication, and then click New resource, Namespace. 3. In the Name box, type a name for your authentication namespace. 4. In the Type list, click the appropriate namespace and then click OK. The new authentication provider resource appears in the Explorer window, under the Authentication component. 5. In the Properties window, for the Namespace ID property, specify a unique identifier for the namespace. 6. In the Properties window for Authentication, for the Allow session information to be shared between client applications, set the value to True.

User Guide 51

Chapter 4: Security This enables you to have single signon between multiple clients on the same computer. Note that you cannot have single signon between a Windows application, and a Web client application, for example, Contributor administration and Cognos 8. 7. Specify the values for all other required properties to ensure that Cognos 8 components can locate and use your existing authentication provider. 8. Test the connection to a new namespace. In the Explorer window, under Authentication, right-click the new authentication resource and click Test. 9. From the File menu, click Save. Cognos 8 loads, initializes, and configures the provider libraries for the namespace. For specific information about configuring each kind of authentication provider, see the Cognos 8 Planning Installation and Configuration Guide.

Add or Remove Members From Planning Rights Administrators and Planning Contributor Users Roles
Using Cognos Administration, add Contributor Administration Console administrators and Analyst administrators to the Planning Rights Administrators role. Add Contributor application and Analyst users to the Planning Contributor Users role.

Steps
1. In Cognos Connection, in the upper-right corner, click Cognos Administration. 2. On the Security tab, click Users, Groups, and Roles. 3. Click on the Cognos namespace. 4. In the Actions column, click the properties button for the Planning Rights Administrators or Planning Contributor Users role. 5. Click the Members tab. 6. To add members, click Add and do the following: To choose from listed entries, click the appropriate namespace. To search for entries, click the appropriate namespace and then click Search. In the Search string box, type the phrase you want to search for. For search options, click Edit. Find and click the entry you want. To type the name of entries you want to add, click Type and type the names of groups, roles, or users using the following format, where a semicolon (;) separates each entry: namespace/group_name;namespace/role_name;namespace/user_name; Here is an example: Cognos/Authors;LDAP/scarter;

52 Analyst

Chapter 4: Security 7. Click the right-arrow button, and when the entries you want appear in the Selected entries box, click OK. Tip: To remove entries from the Selected entries list, select them and click Remove. To select all entries in a list, click the check box in the upper-left corner of the list. To make the user entries visible, click Show users in the list. 8. Click OK. For more information, see the Cognos 8 Administration and Security Guide.

Enabling Planning Roles in Cognos 8

Planning tasks that require access to the Cognos 8 data store, such as running macros, go to production, and adding job servers, require additional security configuration. Access to the data store is restricted to certain groups through Cognos Administration. You must have a system administrator or a user from one of the following groups perform tasks that require the Cognos 8 data store. Optionally, you can add users to the required groups to perform the tasks.

Group
Data Manager Authors

Tool
Framework Manager

Task
Only members of the Data Manager Authors group can import from a Framework Manager data source. You must have a Data Manager Authors group member perform this task.

Directory Administrators

Cognos Administration You must have a Directory Administrator create Configuration, Data a data source named Cognos Planning Source Connections Contributor with a connection of type Cognos Planning - Contributor before performing go to production.

Report Administrators or Server Administrators

Cognos Administration A Report Administrators or Server Administrators Configuration, Content group member must publish and run macros in Administration Content Administration.

Restricting Access to the Everyone Group

The Everyone group represents all authenticated users and the Anonymous user account. The membership of this group is maintained by the product and cannot be viewed or altered. By default, the Everyone group belongs to several built-in groups and roles in the Cognos namespace. To restrict access, remove the Everyone group the System Administrators role and replace it with authorized groups, roles, or users. Optionally, remove the Everyone group from the Planning Contributor Users role to restrict access to Contributor plans. User Guide 53

Chapter 4: Security For more information about the Everyone group, and System Administrators role, see "Initial Security" in the Administration and Security Guide.

Recommendation - Creating Additional Roles or Groups for Contributor

To secure your Contributor applications, you may want to create roles or groups for the following users: Contributor client extensions Contributor work offline users for example, create one work offline role per Contributor application and assign the offline users to the relevant role. The application administrator must also belong to this role. system links translated applications

Planning Contributor User Roles

When you assign user rights to Contributor applications, the first time you click User, Group, Role in the Rights window, a list of all the Users, Groups, and Roles that are members of the Planning Contributor Users role is displayed. If the Planning Contributor Users role contains a large number of members directly below, you can improve performance by creating a smaller number of groups or roles below the Planning Contributor Users role to act as filters. Note: Members of roles can be users, groups, and other roles. Groups can contain users and other groups, but not roles.

Planning Rights Administrator Roles

If you have a large number of administrators, you may wish to create a roles, or groups for specific tasks, and then add the individual users to this role or group. For example, a role Allow System Links can be used for this task, and any user added to that role is assigned that right. For more information about creating groups or roles, see the Administration and Security Guide.

Configuring Access to Analyst

You must configure access to Analyst. Members of the Planning Rights Administrators group can set access rights at the library, object, or item level in Analyst. The type of access can be read, write, or control access. Usually users are allowed read, write, and control access rights for their own libraries, and read access to other libraries. However, you can change this to allow two or more users, groups, and roles, read and write access to the same objects, although not at the same time. The No Access setting is available only at the object level. It is used to override the library settings so that other users cannot view, read, write, or control specified objects.

54 Analyst

Chapter 4: Security Removing access rights at library level is achieved in a different manner. To remove access rights at library level, remove a user from the access list.

Filesys.ini Recommendations
We recommend using NT security on the filesys.ini file, and giving only the Planning Rights Administrators write access to the filesys.ini file. Every other user should have read rights.

Access Right Description

Read Access Read access allows objects to be viewed and copied but not changed. Read access is needed to view another user's data or to set up D-Links from another user's D-Cube. Read access allows a user to open other user's D-Cubes to view the data. Although changes can be entered, the data cannot be saved in the original D-Cube. If you try to save a D-Cube without write-access, you are not allowed to save the original but are prompted to save a copy with a new name. Write Access Write access allows a user to make changes to an object and save it. You can delete or rename the object. You can be given write access to another library. This allows you to open someone else's D-Cube, make changes, and save it. If you have write-access to a D-List, you can change formulas and formats and save it. The access is cumulative, so anyone with write access has all the privileges of read-access as well. Control Access Control access allows a user to set the security access in addition to having full read and write access. You can view, change, delete, rename, and set access to the object. If you have control access to a library, you can govern who can and who cannot access data in the library.

User Guide 55

Chapter 4: Security

Access Right Description

Default Access By default, libraries are set to read access for others and write access for your Settings own library. Objects that do not have an access setting of their own assume the access status of the library to which they belong. By default, objects have read-only access (the default setting for the library), but you can restrict this to no access if you want. Members of the Planning Rights Administrators group can set access to users, groups, or roles to determine who can and who cannot have access to objects. The access can be set at library level or object level. The type of access can be read, write, or control access. If access is given to a group, or role, users in that group or role inherit the maximum access rights available. To configure Analyst security, there are a number of steps you can take. Configure integrated Windows authentication if you want to execute macros without interaction, see (p. 56). Specify a default library, see (p. 57). Restrict D-Cube Access, see (p. 57). Assign access at the library level, see (p. 57). Assign access at the object level, see (p. 58). Assign access at the item level, see (p. 60).

Configure Integrated Windows Authentication

If you are going to run the Analyst Batch Scheduler wizard, you must configure integrated Windows authentication to secure jobs and macros in Analyst. To use Integrated Windows Authentication, you must use an authentication provider that supports it. All authentication providers supported by Cognos 8 can be configured to use Integrated Windows authentication, except for the SAP provider.

Steps
1. In Internet Information Services (IIS) Manager, right-click the Cognos8 virtual directory, and select Properties. 2. Click the Directory Security tab, and click Edit next to Enable anonymous access and edit the authentication methods for this resource. 3. Clear Anonymous access, and select Integrated Windows authentication. Click OK, and close IIS. 56 Analyst

Chapter 4: Security

Specify a Default Library

After you log on to Analyst, you must choose a default library. Otherwise, you are prompted to select from a list of available libraries based on your access rights, to use in the session.

Steps
1. Log on to Analyst. 2. Select a library from the list to use as a default library. 3. Click OK. Tips: To prevent you from being asked to select a library every time you open Analyst, from the Tools menu, click Options. Click the View tab, and select the Do not Prompt for default library check box, and click OK. You can select default libraries for Users, Groups, and Roles. From the File menu, select Administration, Maintain Libraries and Users. Click the name of the user, group, or role, and then click Properties. Select the default library.

Restrict D-Cube Access

You can select the Restricted D-Cube Access check box to prevent users from setting hold, lock, and protect patterns. The users will also be unable to remove the hold, lock, and protect patterns settings in D-Cubes.

Steps
1. In Analyst, from the File menu, click Administration, Maintain Libraries and Users. 2. Click the Users, Groups and Roles tab. 3. Select the users, groups, or roles that you want to restrict access to and click Properties. 4. Select the Restricted D-Cube Access check box. 5. Click OK.

Assign Access at the Library Level

Library level access is commonly used when a library contains confidential data such as salary information and expense accounts. Setting access rights at the library level means you allow or restrict access to a library for a user, group, or role.

Steps
1. In Analyst, from the File menu, select Administration, and click Maintain Libraries and Users.

User Guide 57

Chapter 4: Security The Table maintenance dialog box appears. 2. Click the library you want to set access rights for and click Properties. The Properties dialog box appears showing access rights for the library. 3. Click the Access tab, then click the name whose access you want to modify. 4. Click Read Access, Write Access, or Control Access. The name appears in the right-hand column with the appropriate access assigned. Tip: To remove access rights, click Remove. Note: Only names that have been added to Analyst are available, and by default, users have Control access over libraries they own. 5. Click OK. 6. In the Table Maintenance dialog box, click Close to return to Analyst.

Assigning Access at the Object Level

Security can be assigned to an object or group of objects. By default, no object-level security is specified, in which case objects inherit the security of the library to which they belong. The following table describes object -level access.

Access
Leave Blank No Access

Description
Objects inherit the library security settings Removes all access rights. You cannot view, read, write, or control objects. Can view and copy, but not change an object. Can change and save an object. It also confers full read-access. Can set security on an object. It also confers full read and write-access.

Read Access Write Access

Control Access

Set Access Rights at the Object Level

Security can be restricted for selected objects. If an object has access levels set at both the object and library level, the object assumes the more secure of the access levels. For example, if a D-Cube has read-only object-level security, but the library it belongs to has write-access, the D-Cube assumes read-only access, overriding the library security setting. However,

58 Analyst

Chapter 4: Security any attempt to give the D-Cube control access will be overruled by the write-access settings for the library. Access rights can be set or changed only if you have control access to the library and objects you are working on.

Steps
1. From the File menu, click Library, Objects. 2. Select the objects for which you want to set security. 3. Right-click anywhere in the list and choose Define Access. 4. Click Add to add a list of users who you want to give access to these objects. Only users, groups, and roles that have been added to Analyst are available. 5. Click the name of the user, group, or role. 6. Select the access level from the drop-down menu. 7. Click OK.

Set Read-Only Access to an Active Object

You can assign read-only access to the current active object.

Step
From the File menu, click Read Only Mode, Current Object.

[Read Only] appears in the title bar of the current object. Note: After you have switched to read-only mode, you cannot revert to write-mode without closing and re-opening the object.

Set Access Rights for a Library

Access rights for a library can be set up, changed, or removed for each user, group, or role. For each library, the type of access - Read, Write, or Control must be assigned to the user, group, or role. Access rights can be altered only by Planning Rights Administrators and users that have been given control-access.

Steps
1. From the File menu, click Administration, Maintain Libraries and Users. 2. In the Administration dialog box, click the Libraries tab and then double-click the library you want to change. 3. Click Access, and do one of the following:

User Guide 59

Chapter 4: Security If there are no users, groups, or roles listed, click Add, select the name in the Give Access to dialog box, select the access level and click OK. If users are already listed, click the user, and select the appropriate level of access from the Change Access to box. Note: Everyone is automatically a member of All Users, which by default has Read Access to all libraries. If you want to make a library secure, All Users access must be removed. Optionally, you can repeat this process to add access rights for other users. 4. After the access rights for users are set, click OK to return to the Users page and then click Close.

Set Read-Only Access Globally

You can assign read-only access to all objects globally so all objects are opened in read-only mode.

Step
From the File menu, click Read Only Mode, On/Off.

[Read-Only] appears in the program title bar.

Assigning Access at the Item Level

Security can be assigned to an item by using masks. A mask is used to hide individual items within a D-List or to prevent someone from writing over specified items. This is useful when you are on a network and want to allow limited access to large D-Cubes. You can be specific about the level of security applied to each item by applying an item named a mask. A mask contains a security pattern with a list of users and their access rights. It can be attached to one or more D-List items.

Managing Masks
You can customize the mask to give each user a key to unlock the restrictions. The choices are Read, Write, or Invisible. Items that are set to Read appear on the window but cannot be typed over or changed except by an indirect method such as a breakback. Items that are set to Write can be changed as usual. Items that are set to Invisible do not appear on the selection window. If a user has access at one level and a group to which the user belongs is restricted at a different level, the system takes on the highest access level of the two settings.

Example
If users have Invisible access but are members of All Users, which has Write access, they are able to see and change items because the program grants the highest level of access it can find. The only exception is that a user cannot have higher access to a D-List item than the object or library it belongs to. Anyone with the status of Planning Rights Administrator can override any security settings. If you are not on a network, it is typical for all standalone users to be Planning rights administrators, meaning they can access all models that are given to them. 60 Analyst

Chapter 4: Security Only users with Control access to a D-List can apply or remove a mask. It is possible to lock yourself out of certain items by attaching a mask that makes the items invisible. However, it is not as serious as it sounds because if you had control access to attach the mask, you also have the right to remove it. If you do not have Control access to a D-List, you cannot add or remove a mask. Usually, you only apply a mask to items from a single D-List within a D-Cube. In the unlikely event that item level security is applied to items from two D-Lists in the same D-Cube, there is a potential conflict of access rights at the intersection. In this case, the most secure of the two settings is applied. Thus, if you have Write access to UK and Read access to Sales, the UK Sales cell will give you Read access, the more secure of the two settings.

Create a Mask
Create a mask to hide individual items within a D-List or to prevent someone from writing over specified items. This is useful when you are on a network and want to allow limited access to large D-Cubes. When you open a D-Cube, items assigned to masks that are set to Invisible do not appear on the selection window. Items marked Read appear and can be changed temporarily but cannot be saved. If you have items that do not have full write-access in the current selection, you cannot save the D-Cube. This applies even if you have altered only items with Write access. Also, you cannot save the data in the D-Cube if a total can breakback over an item to which you do not have write access.

Steps
1. From the File menu, click Administration, Maintain Libraries and Users. 2. Click the Mask tab. 3. Click Add to create a new mask. 4. In the Name box, type a name for the mask. By default, a new mask has All Users defined as Invisible. This is the most secure setting possible and applying such a mask to a D-List item renders it invisible to everyone. 5. To customize the access levels for each individual or group to allow limited access: In the Create a new Mask dialog box, click Add and select the user and then click OK. 6. Change access to Read, Write, or Invisible. 7. Repeat for other users and groups until you have built up the required security pattern. If you are giving users write access at item level, ensure that they have write access at both object level and library level. Otherwise they cannot save the data. The default is for users to have read access to libraries unless they are given write access by the Planning Rights Administrator. If the selection opened has any items with read access, the whole D-Cube opens in read-only mode and you cannot save the D-Cube. To open with Write access, you must open a selection or slice of the D-Cube that has Write access. You can use Saved Selections to define Write or Read Only slices of the D-Cube. 8. Optionally, click Remove to withdraw the access privileges for a user or a group.

User Guide 61

Chapter 4: Security 9. Click OK to save the masks; then click Close to return to the main window.

Apply a Mask to Each D-List Item

Apply a mask to a D-List item to hide or protect items. A default mask can be set for a D-List. The default mask automatically applies to any items that are added after you set the default mask. If you want the default mask to apply to existing items, you must assign it to them when you set the default mask. Planning Rights Administrators can make a mask Public. This allows other users to use the same security pattern by attaching a public mask onto D-List items of their own. Changing a mask from public to not public makes the Planning Rights Administrator the new owner.

Steps
1. Open the D-List. If masks already are applied to a D-List, a small yellow padlock appears in the top left corner of the D-List. 2. From the D-List menu, click Options and then click the Security tab. 3. Click the item you want to mask, and then select a mask from the menu. 4. If you want to view or change the security pattern defined by a mask, click Edit Masks. 5. Optionally, you can right-click to get a menu allowing you to create a New mask, Edit or Clear an existing one, or Assign a mask to other items. 6. To assign a mask to other items, select a mask in the Multiple Assignment box and then click Assign. Select items that you want to apply the mask to and then click Move>>. 7. Click OK. Note: You can set a default mask that will apply to any new D-List items that are added later. 8. Click OK and save the D-List. To enter data in a D-Cube that contains masked D-Lists, from the File menu, click Open, D-Cubes, Edit Selection. Select only those items to which you have full write access. Do not select any sub total if you do not have write access to all the components of that subtotal. Items with Write access can be changed but only if the user also has write-access at object and library level.

62 Analyst

Chapter 5: Integration
Cognos Planning - Analyst can be used with other products. You can to take actuals from an Enterprise Resource Planning (ERP) system and combine them with planning information from Analyst and Contributor to perform comparative analysis using Cognos Performance Applications. The following diagram illustrates the various integration points between Cognos Planning and other Cognos products.

Published Data Real Time Data

Cognos 8 Business Intelligence

Cognos 8 Metrics Manager

Planning Cognos Connection

OLAP

Relational

Cognos 8 Business Intelligence Transformer 7.4

Cognos 8 Planning - Contributor

Cognos 8 Planning - Analyst

Performance Applications

Cognos 8 Controller

Cognos Finance

Financial Planning with Cognos Finance and Cognos Planning Products

This section describes what you need to know about using Cognos Finance with Cognos Planning. Cognos Finance, Cognos Planning - Analyst and Cognos Planning - Contributor create a highly complementary financial planning solution. Cognos Finance provides the consolidation, reporting, and schedule-based budgets. Analyst and Contributor provide planning, modeling, and driver-based

User Guide 63

Chapter 5: Integration forecasting and budgeting. Users can take historical actuals gathered in Cognos Finance and use them in Analyst and Contributor to forecast for the future. The Import From Cognos Finance wizard in Analyst facilitates the import of metadata and data from Cognos Finance into Analyst and the subsequent movement of data back and forth between the two applications.

Prerequisites and Required Components

To ensure seamless integration and effective use of Cognos Finance and Cognos Planning, you must: install both Cognos Finance and Cognos Planning in the same installation folder share a common authentication source (directory server namespace) have the Full Cognos Finance Client if using Network Installs. have at least one library in Analyst with control or write access.

Understanding the Process

Before you begin your planning exercise, you should understand the behavior of the new integration features and how they can be combined with existing features to represent the overall flow: The process starts in Cognos Finance with the creation of a Cognos Finance input form (LIF) or report (LRF). The Cognos Finance user defines the scope of the Planning problem and generates a report or input form representation, which contains the metadata and references to data. In Analyst, the Analyst modeler processes the Cognos Finance input form (LIF) or report (LRF) by using the Import From Cognos Finance wizard which: Generates the related D-Lists with an import link. Optionally, generates a D-Cube and D-Links to facilitate the data movement. Optionally, generates an e.List text file for later import into Contributor.

Once the modeling is completed in Analyst, the Contributor administrator: creates an application from the Analyst model. imports the e.List from the text file created in Analyst during the Import From Cognos Finance step. imports Cognos Finance data into Contributor using existing import mechanisms. The application goes to production and the plans are delivered and updated.

In Analyst, the modeler user defines a D-Link, using Contributor as a Source and Cognos Finance as the target to move the latest D-Cube data from Contributor to Cognos Finance. Financial reporting can then be done via the Cognos Finance Reporting Module.

64 Analyst

Chapter 5: Integration

Using Cognos Finance Input Forms or Report Files

Before you create your Input Form or Report in Cognos Finance, you need to understand how information is shared between Cognos Finance and Analyst. The following rules describe how the structural, financial, and formatting information in Cognos Finance is interpreted in Analyst. Analyst respects the switchable accounts options in the Cognos Finance report or input form, bringing data over as period activity or YTD. Note: All forms must have the Account dimension. The Import From Cognos Finance process respects the color attribute of the input form or report. Account dimension summaries are calculated accordingly to take account of Debit/ Credit formatting. The Import From Cognos Finance process automatically creates the summary total (sum of children) for each parent member in a hierarchical dimension (parent-child) relationship. If your input form or report includes a total but only some of the members that make up that total, the Import process brings in the subset of members and recalculates the summary total based on the subset. This may result in a different total from the parent total in Cognos Finance. The Import From Cognos Finance process recreates summary totals automatically (parent/child) . It does not transfer other non-aggregate formulas from Cognos Finance. You must recreate the formulas that apply in Analyst. Locked items and dimensions in Cognos Finance are not automatically locked in Analyst. Select the appropriate member for each filter dimension on the report or input form. The Import From Cognos Finance process honors this filter when creating D-Links. The filter dimensions will not appear in the Cognos Finance D-Cube created in Analyst. A new calculation status, External, is now available in Cognos Finance for accounts and components. It enables you to keep the values brought in from an external source for calculated and aggregated items of these dimension types. If the calculation status is not set to External, the value is changed the next time a Submission Calculation is done in Cognos Finance. For more information, see the Cognos Finance Application Administration Guide.

Importing From Cognos Finance

The Analyst modeler can import metadata and data from Cognos Finance using the Import From Cognos Finance wizard.

Steps
1. From the File menu in Analyst, click Import From Cognos Finance. This menu is only enabled if Planning and Cognos Finance are installed in the same installation folder. 2. Click the Cognos Finance system. 3. Click or type the location of the Cognos Finance report or input form. 4. Click a destination library to create the D-Lists, D-Links, and D-Cube. User Guide 65

Chapter 5: Integration 5. Click an item format. 6. Select the Define an e.List box if there is a future need to import e.List data into the Contributor Administration console. An e.List is a dimension with a hierarchical structure that typically reflects the structure of the organization. It determines how the application is distributed to end users. It can also be used to establish rules of security. The e.List data is typically imported into the Contributor Administration Console as a text file. Selecting the Define an e.List box causes an e.List specification page to be displayed. The specifications are used to automatically generate the necessary Contributor e.List import text file. 7. If you want to allow for the creation of a D-Cube, click Create a D-Cube. To create a D-Link that can be used to refresh a D-Cube from a Cognos Finance system, click Create a Cognos Finance to Analyst D-Link. To create a D-Link that can be used to update a Cognos Finance system, click Create an Analyst to Cognos Finance D-Link. When the D-Link name exceeds 31 characters, the D-Cube, the LRF and the LIF names are truncated to 31 characters. 8. Click the Cognos Finance dimensions that you want to create during import. The list of Cognos Finance dimensions is based on the definition of the LRF and LIF files selected in Cognos Finance. 9. If you are defining an e.List, click a dimension as an e.List. 10. If you are defining an e.List, click either Create D-List with all items or Create D-List with one item. If you choose Create D-List with one item, the first item is used. 11. If you are defining an e.List, type the location of the e.List import text file to be created. 12. When you define a D-Cube, select a filter, then double-click one or more items in the Source list to move the item to the Target list. When the options for creating a D-Link are selected and the target list contains existing D-Lists from Analyst, the D-Link mapping definition is partially created for the existing D-Lists. You must manually map the existing D-List to Cognos Finance dimensions to complete the mapping definition. 13. You can rename the D-Link, D-Cube, and D-List, which are created in Analyst. Click the object, press F2, and type the new name. 14. If you want to run the Cognos Finance to Analyst D-Link right away, click Run Cognos Finance to Analyst D-Link. If you do not click this option, an empty D-Cube is created. 15. Click Finish to proceed with import process. Once the process is completed, the newly created D-Cube is automatically opened you clicked Open D-Cube on Finish.

66 Analyst

Chapter 5: Integration

Financial Planning with Cognos Performance Applications and Analyst

You can use Cognos Planning and Cognos Performance Applications together to extract actuals from the data warehouse of an Enterprise Resource Planning (ERP) system and bring them into Cognos Planning, as structural data for the Analyst model and as the initial figures from the previous planning cycle for Contributor return completed planning data to the data warehouse using an ETL tool such as Data Manager for comparative analysis monitor live or published planning data during the planning cycle against current operational data in the Performance Applications warehouse

The data warehouse extracts, and changes that occur during the planning cycle, are managed using the Import from IQD wizard. Monitoring is done directly against Contributor data using the appropriate extensions. Steps Preparing Cognos Performance Applications Data for Planning In the Performance Application, identify the key performance indicators (KPIs) to plan by, monitor, and report on. Because planning is often performed at a different level from actuals, you may need to add to the dimensions from the data warehouse. Cognos consultants can help you in this identification and analysis. The Import from IQD wizard expects each dimension to have both an ID field and a description field, each of which must be unique across the dimension. Preparing for the Model in Analyst After the planning measures and dimensions that are available from Cognos Performance Applications have been identified, the Analyst user designs a model, and identifies any alternate data sources that are needed for the dimensions and measures. Because Performance Applications use multiple currencies for reporting, the Analyst user should determine what currency to use when data is published back into the Performance Applications warehouse. Note: If you create a D-List using the Import from IQD wizard, you should not add any items manually. If you do add items manually, these items will be removed every time you refresh the D-List. After planning models are designed and sourcing is identified, the solution to integrate the actuals information with planning information can be implemented using either the mapping table that is generated during the IQD import, or if the mapping tables are not required, you can use a Cognos package as a source to populate D-Lists in Analyst. Preparing e.Lists for Contributor Data As well as importing D-List data for the Analyst model, you can choose to generate e.Lists using data from IQD files, or if the data is modeled in Framework Manager and published as a package, you can also use Contributor Administration Links. User Guide 67

Chapter 5: Integration

IQD Files and the Import from IQD Wizard

The Import from IQD wizard in Cognos Planning works with any Cognos product that can generate Impromptu Query Definition (IQD) files, such as Impromptu and Framework Manager see "Generate Framework Manager Model Wizard" (p. 776). An Impromptu Query Definition is a transparent file that holds the SQL statement from which a hierarchy can be built. One IQD file is needed per dimension (D-List or e.List). The following diagram illustrates how the IQD wizard works in Cognos Planning and Performance Applications.
IQD Files

(2)

Analyst
Import from Impromptu Query Definition (IQD) Refresh from IQD Update from Data Warehouse D-Lists D-Cubes IQD SQL

(3)

e.List Text files

(4) (1)
PQD

Impromptu/ Framework Manager Contributor Application

Data Warehouse

(7) Performance Applications (or other data warehouses)

Contributor Publish Container

Cognos Planning

You can follow the workflow in the diagram from numbers 1 through 7 in sequence. 1. You model the data in Framework Manager or Impromptu that needs to be imported into Planning and save this as an IQD file. 2. Using the Analyst Import from IQD feature, browse to the IQD file and use the contents of the file to populate the various wizard steps. After the IQD file has been used once, it is not required by Analyst since Analyst stores the IQD details internally. The IQD file itself is only required if it is necessary to re-run the import wizard, as opposed to simply refreshing the D-List. Also as part of the import wizard, a mapping table is created that holds the relationship between the source business keys and the IDs generated by Planning. You can choose where this mapping table is stored, for example the source database. See Step 7 for more details. 3. When running the Import from IQD feature to create an e.List, the wizard generates the D-List in Analyst, either fully populated or with just a placeholder item, and also creates a text file that holds the e.List in the correct format for import into Contributor.

68 Analyst

Chapter 5: Integration 4. As part of the import wizard, you have the option to create a PQD (Planning Query Definition) file. This stores the import configuration for reuse. You can choose where this file is stored. The IQD file is not needed unless the import wizard needs to be re-run, and the essential elements of the IQD file, such as the SQL statement, are stored in the library folder in Analyst. 5. Since all the essential elements for the IQD import are stored in Analyst, after the initial import, all updates and refreshes are performed directly with the database. 6. Once the D-Lists in Analyst have been populated, the Contributor application is created/updated. As part of this step, the e.List file that was created as part of the import is imported into the Contributor Administration Console. Contributor end users enter their numbers into the created application. 7. When the Publish process in Contributor runs, the Planning IDs are output with the Planning data. To get the published data back into the source system, the published data is moved to the source system and the mapping table generated during the Import from IQD wizard is used to map Planning IDs to source business keys.

IQD File
The IQD file holds a definition of the data that will be imported into Cognos Planning, very similar to a SQL statement. You specify the order of the D-List or e.List items within the IQD, or the order of the underlying table or view is used by default. Alternative ordering can be applied manually to the generated D-List or e.List within Analyst, but this is lost if the D-List is updated again from the IQD. The wizard imports from flattened hierarchy structures only. Ragged hierarchies are not supported. In a flattened hierarchy, all level items have the same number of parent levels. In a parent-child (ragged) hierarchy, not all level items are at the same hierarchy level.) To import a ragged hierarchy, use the standard Analyst D-List import. The Import from IQD wizard expects each level to have an ID field as well as a description field, which must both be unique across the dimension. All settings are retained in PQDs (Planning Query Definitions) after the initial configuration. You can use PQDs to speed up the repeated import of the same IQD file. If you use the PQD to import another IQD file, ensure they both generate similar result columns.

Mapping Table
The Import from IQD wizard generates a mapping table in a database selected by the user to map the IDs of the data warehouse table to the Planning IDs. This is necessary so that Planning data can be returned to the Performance Applications data warehouse even if the warehouse has changed. Use this table to map the published data from Contributor into the source database (or other database that has the same source dimension data). You can specify an alternative target database for the mapping table, as long as the database connection and user details are configured in Cognos Connection or Framework Manager.

User Guide 69

Chapter 5: Integration

Connections and Security

The Import from IQD wizard uses any connection that has been set up in Cognos Connection or Framework Manager. If a connection that requires a client installation, such as Oracle, DB2 or ODBC, is used when setting up a database connection for the IQD, the appropriate client configuration must also exist on the machine where the IQD import will be executed. This is a limitation of such connections. Configure the security to the chosen data source in Cognos Connection or Framework Manager. When the wizard is run, it checks to confirm that you have access to the data source used in the IQD. If the data source connection does not exist in Cognos Connection for Framework Manager, the wizard will stop the process and display a message stating that security details must be configured before the wizard can proceed. You can tell whether a D-List is sourced from an IQD file by opening the D-List, and checking to see if the Refresh option is disabled, which indicates the D-List is not sourced from an IQD file.

Standard D-Lists
For standard D-Lists, specify the leaf-level of the hierarchy (Item Name), and the leaf-level business ID and the table-unique identifier (Dimension ID) the other fields from the SQL statement in the IQD file that are to be used as levels how to configure the relationship between levels, starting from the top level Any additional, non-hierarchy fields that are required in the mapping table, such as background information about products or customers for additional reporting, are added as pass through fields. A Pass-Through field is one that appears in the Mapping table but not in Analyst or Contributor.

Time D-Lists
For time D-Lists, specify the same hierarchical information as a standard D-List. In addition, select the date format, and the period start and end date fields that will be used in Analyst to define the 'from' and 'to' columns for the timescale options of the D-List.

e.Lists
For e.Lists, specify the file path for each e.List file that will be generated by the process. Also, specify whether the entire e.List should be imported into Analyst or just a single item. Regardless of choice, the entire e.List is created in the e.List file that should be imported in Contributor via the Administration Console. Remember the D-List generated in Analyst is only the placeholder D-List that represents the e.List in Contributor. There are scenarios where it makes sense to have the entire e.List in Analyst, but often the e.List is too large to hold in Analyst, so only a single placeholder item is required in Analyst. After this step, the same hierarchical information will be specified that is applicable to the standard D-List. The e.List file that is generated will contain IDs that match those in the mapping table.

70 Analyst

Chapter 5: Integration

Managing Changes from the Data Warehouse

There may be changes to the data warehouse during the Planning cycle. If no items have been added manually to the Analyst D-List, use the Import from IQD wizard to manage updates, inserts, and deletes of the D-List or e.List items automatically. Refreshing from the IQD overwrites any manual changes you have made so you cannot manage updates, inserts, and deletes outside of the IQD. You can use the Import from IQD wizard to update any D-List created by the wizard in previous releases. Note: If you add items to the D-List manually in Analyst, they are removed when you refresh the D-List from the warehouse. If you run the Import from IQD wizard, you can monitor changes. If you use a macro, you do not receive any warnings. For all existing D-Lists, if a new ID field is not unique, or if the truncated description field (the first 50 characters) is not unique, the import from IQD process stops and the D-List is returned to its original state before the process began. In the wizard, you can choose to generate new IDs for all D-List items. This refreshes the entire dimension, including the IID, GUIDs, and mapping table, however changing internal IDs could lead to a loss of data in Contributor.

Update
If only the level description changes (not the field marked as ID) then the dimension is updated and retains the original IDs and mappings. The rest of the dimension remains untouched. Any pass-through columns should also be updated in the mapping table in the source database, even if no change has occurred within the D-List or e.List. The position of the items within the hierarchy should also be retained or updated as appropriate. If an item moves from one position to another within the hierarchy, then all item IDs should remain the same.

Insert
If a new item appears in the source table, or an existing item is given a new ID, then it is inserted as a new item into the D-List or e.List in the same position as it is found in the IQD SQL statement (according to the IQD ordering or according to the SQL table order). The other items of the dimension remain untouched. The mapping table is updated with the new item, but the existing mapping details remain untouched.

Delete
If an item is removed from the source table, then it is removed from the D-List or e.List and marked as deleted in the mapping table, although the entry still remains in the mapping table. The other items remain untouched.

Deployment Options for IQD

You can move the IQD-sourced D-Lists between different environments, like Development, Test, Production.

User Guide 71

Chapter 5: Integration

Unchanged IQD Setup

If the IQDs that have been set up in Development do not need to be changed in the other environments, for example if the SQL statement in the IQD is correct regardless of environment, then you can move the Analyst libraries through the environment like normal. You do not need to the move the IQD file itself because the import definition is stored in the Analyst library after the initial run-through of the IQD wizard. The only things you need to create in the target environment are: The appropriate database connections. Depending on how you have configured the IQD import, you will need connections for the source and mapping table databases (which could be the same). The names of the connections should be the same in the target environment as in the source environment. The mapping table will need to be moved to the target environment. If the mapping table cannot be found, the update will fail.

Changed IQD Setup

If the IQDs need to be updated in the target environment, for example to change the SQL statement, then these too will need to be moved to the target environment. For each D-List that relies on an IQD file that needs to be updated, do the following: Edit or recreate the IQD file to incorporate the changes. If the file path to the IQD is not the same as in the previous environment, you will receive a warning and then be able to browse to the new location. Run the IQD Wizard and run the PQD file. This will step through the wizard with the settings that were used for the original configuration. This will also read the IQD file into Analyst again, so any changes you have made will be picked up. If you have made changes like adding a WHERE clause to the SQL statement in the IQD file, then no alterations to the settings will be required. If you have removed fields from the SQL statement that were previously used in the IQD Wizard, or have added fields that need to be incorporated, you can add them at the appropriate wizard step. At the end of the wizard, click Overwrite Existing D-List and also Keep IDs for Matching D-List Items so that the existing D-List is correctly updated. After this process runs, the settings are incorporated into Analyst and you no longer need the IQD files.

Creating Planning Models and Data from Cognos Performance Applications Data
When you create planning models from Performance Application data, you use: the Cognos Performance Applications data warehouse, to supply data for the planning model and the initial values in Contributor As for any data source, the connection information to the Cognos Performance Applications data warehouse must be in Cognos Connection or Framework Manager. Ensure that the Cognos Performance Applications data warehouse is prepared with actuals. 72 Analyst

Chapter 5: Integration For information about dimensions available, see the Report Administration Guide for your Performance Application. Impromptu or Framework Manager, to generate the IQDs, or if IQDs are not needed for mapping tables, then you can publish models as a package Analyst, to create the planning model Contributor, to publish the model, the initial values, and to collect the planning data Contributor Administration Links to import data that is modeled in Framework Manager and published as a package Cognos Data Manager, to move the data between the Contributor and Performance Applications databases,

To create a Planning application from a Performance Application, do the following: Determine the granularity of the plan (p. 74). Create D-Lists in Analyst from IQD files. Create a placeholder e.List in Analyst from an IQD file. Create D-Cubes in Analyst from D-Lists. Create an application in Contributor by importing the planning models created in Analyst. Populate the e.List that was creating in Analyst as a placeholder. Assign user access rights to each e.List item. User rights to the data in the D-Cube are established using an e.List. An e.List is the D-List that represents the business area responsible for providing the planning information. Populate Contributor application D-Cubes with actuals from the Cognos Performance Applications as seeding plan values. Make Contributor models available to planning users. In the Contributor Administration Console, the administrator runs the Go to Production process and sets up the Contributor Web site so that the planning models become available to all participating planning users. Users can review plans and enter data using their distributed Web browser environment. Publish planning data back to Cognos Performance Applications' data warehouses.

After user input and modifications, the planning data can be published back to the data warehouse where the planning data and actuals can be compared and analyzed. The Contributor administrator performs a Publish from Contributor which exports the D-Lists and the fact data to database tables. From the Publish tables, the administrator needs to ETL (Extract-Transform-Load) the data into the Performance Applications database, using the mapping table generated by the IQD to match Contributor IDs with Business Codes. For more information about publishing planning data, see "Publish Planning Data " (p. 79).

User Guide 73

Chapter 5: Integration

Monitoring Planning Activity using Business Intelligence (BI) and Cognos Metrics Manager
Live and published planning data can be compared with actuals using Cognos PowerPlay cubes, Cognos Impromptu catalogs and reports, Cognos Metrics Manager and Cognos Framework Manager models for reporting as appropriate. All these can be done using Contributor extensions.

Determine Plan Granularity

You must ensure that Cognos Performance Applications has been set up to support planning before creating a planning model in Cognos Planning. Steps Determine the granularity of the plans that will be created in Cognos Planning. Assess the business requirements to determine the planning measures and the dimensional context for the measures before creating the extract queries. The dimensional context identifies the D-List candidates critical to the measures. The dimensions you create should have some relation to the measures. Also, ensure that the measures and dimensional relationships exist in the Cognos Performance Application. You must determine which dimension will control the access to the data (the e.List). Understand the planning requirements and develop an initial draft of the planning model. The time you spend drafting the model will greatly reduce the need to make changes later, of a kind that could affect the effort required to publish data and generate plan data to the warehouse. Determine the level of data that is required for planning, and compare that against the data that is available. Using this information, create and customize the dimension structure to define the business model.

Create a Planning Model in Analyst

D-List information is brought from dimensions in Cognos Performance Applications into Analyst using the Import from Impromptu Query Definition (IQD) wizard.

Mandatory IQD (Impromptu Query Definition) Columns

The following identifies the mandatory columns that must exist in the IQD in order for the D-List/ Planning dimension process to be successful. For normal D-Lists the IQD has the following mandatory columns: Dimension ID. The Dimension ID is reserved for other Cognos reporting tools. It should take the values of item code. Item Code (ID field). This column is the reference key in the source database. Item codes and descriptions must be unique across all imported levels.

74 Analyst

Chapter 5: Integration Item Name (description field). All item names inside a D-List must be unique and is limited to 50 characters. Input item names that are longer than 50 characters are automatically truncated during the D-List creation.

For date based or time scale D-Lists, the IQD must have the following additional mandatory columns: Period start date Period end date

To create a D-List, either a brand new D-List is created or an existing one gets replaced. There are two options for replacing an existing D-List: Replace the existing D-List and retain the existing, corresponding item IDs. This option preserves the data in the D-Cubes that use the D-List. All other attributes of the D-List are replaced. Use this option when it is necessary to track historical changes to cube data. Replace the existing D-List and preserve only the D-List name. All item IDs are new. Even items with the same name receive new IDs.

Both options replace all D-List items. Any new items added to the D-List using Analyst is be lost.

Steps
1. From the File menu, click Import from, Impromptu Query Definition (IQD). 2. Select the type of D-List to create. You can choose to create a normal D-List, a time scale D-List, or an e.List. You can also use PQD files that are saved IQD imports. These files contain all the settings for IQD imports that you might have run previously and are designed to minimize configuration for similar D-Lists. 3. Select the mandatory IQD columns to import. For information about mandatory IQD columns, see "Mandatory IQD (Impromptu Query Definition) Columns" (p. 74). 4. For time scale D-Lists only, select the start date and end date. 5. Optionally, define the D-List hierarchies. 6. Optionally, preview the data source records. 7. Name the D-List and select a library with which to associate it. 8. Select the target for the mapping table. This target database will need to be configured in Cognos Connection/Framework Manager. After you run the wizard, the import process creates a mapping table for each D-List in a user selected database. The mapping tables contain the following critical reference information for the Analyst Planning model and the data source. Reference information for the Analyst item IDs, which are unique for each item. Reference information for the data source Item code, which is also unique for each item.

User Guide 75

Chapter 5: Integration When importing initial values to a D-Cube later, the GUID and item code from the planning dimension tables should be used to build correct insert values. Repeat this process to create all D-Lists required for one or more D-Cubes.

Create an e.List for the Contributor Application

A Contributor application is created from D-Cubes selected from an existing Analyst model. Each D-Cube must have an e.List dimension. An e.List is created in Analyst as a placeholder D-List. This e.List can either be created with one item or with all items. If created with one item only, this item acts as a place holder with one root item. If created with all items, the e.List is created in Analyst with all items. The complete e.List hierarchy and item IDs are saved to a text file that Contributor will use to populate the complete e.List. Note: If you create the e.List with all the items in Analyst, you may run the risk of having a large Analyst model because the whole Contributor model would effectively be in Analyst.

Create D-Cubes
A D-Cube in Analyst is made up of two or more D-Lists. To create a D-Cube based on Cognos Performance Applications, all D-Lists can be created from IQD files.

Set up a Contributor Application

Once the model has been created in Analyst, in the Contributor Administration Console, the administrator creates the Contributor application, grants access to users and imports data. Steps The administrator creates a new application, selecting the library containing the model, and selecting the D-List to use as the e.List. Populates the e.List by importing the text file created earlier. Sets up access to the application through the e.List, users and rights, and through access tables. Configures application settings by defining one application data store and two publish data stores. Imports data into D-Cubes as initial planning values or seeding values. These values are normally from the actuals of the Performance Applications data warehouse. Dimensions in Cognos Performance Applications are often at a lower level of granularity than is required for planning purposes so values will typically be generated from summarizations or aggregations. Runs the Go to Production process to put the application into production state. Sets up the Contributor Web site to make the production application available to planning (contributing) users.

76 Analyst

Chapter 5: Integration

Populate the Planning Application in Contributor

You can import data that has been modeled in Framework Manager and published as a package into Contributor using Administration Links. You can also import historical actual measures from Cognos Performance Applications into Cognos Planning D-Cubes in the format of: Item name of dimension (D-List) 1 Item name of dimension (D-List) 2 Item name of dimension (D-List)... Item name of dimension (D-List) n Measure value

The item names are those seen in Planning. They may not be the same as produced by the original IQD, due to manual modifications after the D-List creation or truncations during the D-List creation. The planning dimension tables must be used to retrieve correct item names used for the value import. The most efficient way to import data into Cognos Planning is through Contributor. This can be done using SQL or Data Manager. The decision whether to use SQL or Data Manager depends on the complexity of the import and your expertise with SQL and Data Manager. If the data is modeled in Framework Manager and published as a package, you can also use Contributor Administration Links. The processes to import the data are: Using SQL for Loads A simple SQL query can be used to load directly from the Cognos Performance Applications into the Contributor import IM table. Execute the Build and Validate the AAA_ Table Data Data Manager is unable to deliver data directly to the IM_ table so this step loads the IM_table from the AAA_table using an INSERT statement. Notes In a multi-measure planning cube, match the Measures D-List column to the Measures Column in the Import Table. If re-loading, TRUNCATE the table first. TRUNCATE TABLE <table_name>, where the <table_name> is table that all rows are to be removed. This is a SQL command in Oracle and SQL Server. There is an option in Contributor Import, Zero Data, which accomplishes the same thing. Zero Data should only be used if intending to replace the target D-Cube's data entirely with the data set being imported. Import the data into Contributor To import the data into the Contributor Production environment, the Go to Production process must be completed.

User Guide 77

Chapter 5: Integration Assign access for the published table Assign synonyms and grants for published table to user schema (Oracle) and assign permissions and access rights for user (SQL Server). Using an appropriate SQL tool: Login to the Contributor data store that is linked by the Contributor application's data store. Build planning views for reporting on published tables with Cognos Performance Applications fact. Build reports on planning views.

Access to Published Cognos Planning - Contributor Tables

In order to access the data in published Contributor tables, you must assign: synonyms and grants for a published table to a user schema (Oracle) permissions and access rights for user (SQL Server)

Steps
1. Using an appropriate SQL tool, log in to the Cognos Planning - Contributor application's data store. Note: The DBA must assign the appropriate login security and privileges. For example in Oracle:
GRANT CONNECT TO <Planning Contributor account>;

2. Grant access to the published tables so that the Cognos Performance Applications account can select the data. For example in Oracle:
CREATE SYNONYM ET_REVENUE_AND_EXP FOR <Planning Contributor account>.ET_REVENUE_AND_EXP; GRANT SELECT ON <Planning Contributor account>..ET_REVENUE_AND_EXP TO <Performance Applications account>;

Tip: The following Oracle script can be used to generate the above scripts:
select 'CREATE SYNONYM ' || TABLE_NAME || ' FOR <Planning Contributor account>.' || TABLE_NAME || ';' from ALL_TABLES WHERE OWNER = '<Planning Contributor account>' AND (TABLE_NAME LIKE 'HY_%' OR TABLE_NAME LIKE 'ET_%') select 'GRANT SELECT ON <Planning Contributor account>.' || TABLE_NAME || ' TO <Performance Applications account>;' from ALL_TABLES WHERE OWNER = '<Planning Contributor account>' AND (TABLE_NAME LIKE 'HY_%' OR TABLE_NAME LIKE 'ET_%')

78 Analyst

Chapter 5: Integration

Publish Planning Data

Users can carry out a plan and actuals comparison, after the required planning is complete and the actuals for the same period are captured. To do this, the plans need to be extracted and compared with the actuals from Cognos Performance Applications.

Publish Layouts
You can choose from these types of publish layouts: table-only, incremental, and view. The table-only layout gives users greater flexibility in reporting on Planning data. The table-only layout can also be used as a data source for other applications. This layout is required by the Generate Framework Manager Model Admin extension and the Generate Transformer Model Admin extension. The incremental publish layout publishes only the e.List items that contain changed data. Users can schedule an incremental publish using a macro or through Cognos Connection and Event Studio. You can achieve near real-time publishing by closely scheduling incremental publishes. The view layout generates views in addition to the export tables. This layout is for historical purposes.

The following types of tables are created when you publish using the table-only layout and the incremental publish layout.

Table type
Items Hierarchy

Description
Describes the D-List items.

Prefix or name
it_

Contains the hierarchy sy_ for the simple hierarchy cy_ information derived from the for the calculated hierarchy. D-list, which is published to two associated tables. Contains published D-Cube data. et_ Contains annotations, if the an_ for cell and audit annotations option to publish annotations is annotationobject for tab (cube) selected. and model annotations Contains metadata about the publish tables. P_APPLICATIONCOLUMN P_APPCOLUMNTYPE P_APPOBJECTTYPE P_APPLICATIONOBJECT

Export Annotation

Metadata

Common

Contains tables used to track when major events occurred in the publish container.

adminevent adminhistory containeroption

User Guide 79

Chapter 5: Integration

Table type
Job

Description

Prefix or name

Contains tables with information job jobitem jobitemstatetype relating to jobs. jobstatetype jobtask jobtype A table used to lock objects in the P_OBJECTLOCK system when they are being processed. publishparameter Publish information about files attached to the Contributor application. ad_cube

Object locking

Publish parameter Attached document

The following types of tables are created when you publish using the view layout.The Contributor administrator publishes from Contributor which exports the D-Lists and the fact data to database tables. Contributor publishes the planning data into a set of tables with standard prefixes.

View Layout Contributor Standard Prefixes Planning Data

an av et ev fv hy it Annotation (for D-Cubes) Annotation view Export/publish (for D-Cubes) Export view Fact view Hierarchy (for D-Lists) Item (in D-Lists)

In order to perform the actuals vs. plan comparison, this data must be made available in such a manner that the Business Intelligence tools can access. The problems are: Planning data is linked to the D-Lists via GUIDs, whereas Performance Applications actuals data is linked via item code of dimensions. The planning dimension tables provide lookup links. Physically, the Planning data may be stored in a different database (or schema in Oracle) than the Performance Applications data and thus with different default security on the tables.

80 Analyst

Chapter 5: Integration In order to accomplish this, views can be created in the Planning database to expose the Planning data as well as provide access via grants.

Automation
You can automate the update of a D-List or recreation of an e.List file using either of these Analyst macros: @DListItemImportIQD (p. 645) @RefreshDataWarehouse (p. 649)

Manual updates to D-Lists and e.Lists will be removed without messages if you update lists using macros.

Command Line
The re-use functionality can be executed using these command line options: Update D-List from IQD Update IQD Mapping Table Upgrade D-List created from an earlier release

Update D-List from IQD

This command line updates a D-List from its IQD source. Any changes done manually in Analyst will be lost. Item names, hierarchies and display orders are synchronized with the IQD source. D-Lists created prior to this release are upgraded before running this command line.

Update IQD Mapping Table

This command line synchronizes the mapping table with the Analyst D-List. It is provided for backward compatibility only, where a D-List can be changed manually (not recommended practice) . D-Lists created prior to this release are upgraded before running this command line.

Upgrade D-List created from an Earlier Release

This command line upgrades a D-List created from a previous release.

User Guide 81

Chapter 5: Integration

82 Analyst

Chapter 6: Importing Data from Cognos 8 Data Sources

You can import data into Cognos 8 Planning - Analyst and Cognos 8 Planning - Contributor from any data source that can be published as a Cognos 8 package. For more information about supported data sources, visit the Cognos Global Customer Services Web site (http://support.cognos.com). There are additional considerations when importing SAP BW data into Cognos 8 Planning. For more information, see "Working with SAP BW Data" (p. 88). For information on Cognos 8 Planning configuration requirements for SAP BW, see the Cognos 8 Planning - Installation and Configuration Guide. You must have Framework Manager installed. If you are working with SAP BW data, you must install the SAP gateway functions. For more information, see the Cognos 8 Planning - Installation and Configuration Guide. Importing data from Cognos 8 data sources involves the following tasks. In Cognos Connection, create a data source connection (p. 84). In Framework Manager, create a new project and import the metadata into the project (p. 86). In Framework Manager, model the source. See the Framework Manager User Guide for more information. Create and publish the Cognos package to Cognos Connection (p. 87). If importing into a Contributor application, in the Contributor Administration Console, create and run an administration link. Tip: You can create and schedule macros that run administration links. If importing into an Analyst model, choose one of the following options: Select a Cognos package as a source in a D-List Import. Select a Cognos package as a source in a D-Link. Select a Cognos package as a source in an A-Table, or import a Cognos package as a Source in an A-Table.

You can also automate the import of Cognos packages using the @DListItemImportCognosPackage macro.

User Guide 83

Chapter 6: Importing Data from Cognos 8 Data Sources

Create a Data Source Connection

When you create a data source, you must provide the information required to connect to the database. This information is provided in the form of a connection string. You can include authentication information for the database in the data source connection by creating a signon. Users need not enter database authentication information each time the connection is used because the authentication information is encrypted and stored on the server. The signon produced when you create a data source is available to the Everyone group. Later, you can modify who can use the signon or create more signons.

New Data Sources

You can create data sources in the portal or in Framework Manager. Because they are stored on the server, data sources appear in both places, regardless of where they were created. Existing data source connections can be edited only in the portal. If you are an administrator, you can set up all required data sources before models are created in Framework Manager so that all connections are available in the Framework Manager Import wizard. Data sources are stored in the Cognos namespace and must have unique names. For example, you cannot use the same name for a data source and a group.

Physical Connections
A data source defines the physical connection to a database. A data source connection specifies the parameters needed to connect to a database, such as the location of the database and the timeout duration. Note: The schema name in the connection string for an Oracle database is case-sensitive. If the schema name is typed incorrectly, you cannot run queries.

Required permissions
Before creating data sources, you must have write permissions to the folder where you want to save the data source and to the Cognos namespace. You must also have execute permissions for the Data Source Connections secured feature.

Steps to Create a Connection

1. Open Cognos Connection. 2. In the upper-right corner, click Launch, Cognos Administration. 3. On the Configuration tab, click Data Source Connections. 4. Click the New Data Source button. 5. In the name and description page, type a unique name for the connection and, if you want, a description and screen tip, and then click Next.

84 Analyst

Chapter 6: Importing Data from Cognos 8 Data Sources 6. In the connection page, click the type of database to which you want to connect, select an isolation level, and then click Next. Note: For SAP BW data sources, the isolation level is read-only. Note: For Cognos Planning - Contributor 7.3 data sources, select Cognos Planning - Series 7. For Cognos 8 Planning - Contributor data sources, select Cognos Planning - Contributor. The connection string page for the selected database appears. 7. Enter any parameters that make up the connection string, and specify any other settings, such as a signon or a timeout. One of the following options may apply depending on the data source to which you are connecting: If you are connecting to a Cognos cube, you must enter the full path and file name for the cube. An example for a local cube is C:\cubes\Great Outdoors Company.mdc. An example for a cube on your network is \\servername\cubes\Great Outdoors Company.mdc. If you are connecting to a password protected PowerCube, click Cube Password, and then type the password in the Password and Confirm Password boxes. If you are connecting to an ODBC data source, the connection string is generated from the name you enter in the ODBC data source box and any signon information. The data source name is an ODBC DSN that has already been set up. You can include additional connection string parameters in the ODBC connect string box. These parameters are appended to the generated connection string. If you are connecting to a Microsoft Analysis Services data source, select an option in the Language box. If you selected Microsoft Analysis Services 2005 you must specify an instance name in the Named instance since you can have more than one instance on each server. If you use a Microsoft Active Directory namespace and you want to support single signon with Microsoft SQL Server or Microsoft Analysis Server, select An External Namespace, and select the Active Directory namespace. For more information about configuring an Active Directory namespace, see the Cognos 8 Planning Installation and Configuration Guide. If you selected Microsoft Analysis Services 2005, you must specify an instance name in the Named instance since you can have more than one instance on each server. If you selected Cognos Planning - Series 7, you must specify the Planning Administration Domain ID and the namespace. If you selected Other Type as the data source type, you must build the connection string manually.

Tip: To test whether parameters are correct, click Test. If prompted, type a user ID and password or select a signon, and then click OK. If you are testing an ODBC connection to a User DSN, you must be logged on as the creator of the DSN for the test to succeed. 8. Click Finish.

User Guide 85

Chapter 6: Importing Data from Cognos 8 Data Sources The data source appears in Data Source Connections on the Configuration tab, and can be selected when using the Import wizard in Framework Manager.

Create a Framework Manager Project and Import Metadata

A project is a set of models, packages, and related information for maintaining and sharing model information.

Steps
1. From the Windows Start menu, click Programs, Cognos 8, Framework Manager. 2. In the Framework Manager Welcome page, click Create a new project, and specify a name and location. You can add the new project to a source control repository, see the Framework Manager Help for more information. 3. In the Select Language page, click the design language for the project. You cannot change the language after you click OK, but you can add other languages. Note: If an SAP BW server does not support the selected language, it uses the content locale mapping in Cognos Configuration. If a mapping is not defined, Framework Manager uses the default language of the SAP BW server. 4. In the metadata source page, select Data Sources. 5. Select a data source connection and click Next. If the data source connection you want is not listed, you must first create it (p. 84). 6. Select the check boxes for the tables and query subjects you want to import. Tip: For usability, create a package that exposes only what is required. 7. Specify how the import should handle duplicate object names. Choose either to import and create a unique name, or not to import. If you choose to create a unique name, the imported object appears with a number. For example, you see QuerySubject and QuerySubject1 in your project. 8. If you want to import system objects, select the Show System Objects check box, and then select the system objects that you want to import. 9. Specify the criteria to use to create relationships and click Import. For more information, see the Framework Manager User Guide. 10. Click Next and then Finish. Note: You save the project file (.cpf) and all related XML files in a single folder. When you save a project with a different name or format, ensure that you save the project in a separate folder.

86 Analyst

Chapter 6: Importing Data from Cognos 8 Data Sources

Create and Publish the Cognos Package

You create and publish a package to make the data available to Cognos 8 Planning.

Steps to Create a Package

1. Click the Packages folder, and from the Actions menu, click Create, Package. 2. In the Provide Name page, type the name for the package and, if you want, a description and screen tip, and click Next. 3. Specify whether you are including objects from existing packages or from the project and then specify which objects you want to include. 4. Choose whether to use the default access permissions for the package: To accept the default access permissions, click Finish. To set the access permissions, click Next, specify who has access to the package, and click Next. You can add users, groups, or roles. See the Framework Manager User Guide for more information. 5. Move the language to be included in the package to the Selected Languages box, and click Next. 6. Move the sets of data source functions you want available in the package to the Selected function sets box. If the function set for your data source vendor is not available, make sure that it was added to the project. 7. Click Finish and choose whether to publish the package.

Steps to Publish a Package

1. Select the package you want to publish. 2. From the Actions menu, click Package, Publish Packages. 3. Choose where to publish the package: To publish the package to the report server, click Cognos 8 Content Store. Click Public Folders to publish the package to the public folder. You can create folders in the public folder also. Click My Folders to create your own folder and publish the package to it. To publish the package to a network location, click Location on the network.

4. To enable model versioning when publishing to the Cognos 8 Content Store, select the Enable model versioning check box and type the number of model versions of the package to retain. Tip: To delete all but the most recently published version on the server, select the Delete all previous model versions check box.

User Guide 87

Chapter 6: Importing Data from Cognos 8 Data Sources 5. If you want to externalize query subjects, select the Generate the files for externalized query subjects check box. 6. By default, the package is verified for errors before it is published. If you do not want to verify your model prior to publishing, clear the Verify the package before publishing check box. 7. Click Publish. If you chose to externalize query subjects, Framework Manager lists which files were created. 8. Click Finish.

Working with SAP BW Data

The SAP BW model is an OLAP source and is optimized for reporting rather than for high volume access that is sometimes required for planning activities. To efficiently access data for Cognos 8 Planning, create a detailed fact query subject that will access fact data at a detail level suitable for use with Cognos 8 Planning. Tip: If you have OpenHub, you can use it to generate a text file or database table from SAP BW. You can then manually create a Framework Manager model and Cognos Package from the tables and then import the package into Planning using an Administration Link, D-Link, or D-List import. For Cognos products to be able to access SAP BW as a data source, the user accounts used to connect to SAP must have specific permissions. These permissions are required for the OLAP interface to SAP BW and are therefore relevant to both reporting and planning activities. For more information about guidelines for working with SAP BW data, see the Framework Manager user Guide. For more information about access permissions for modelling and reporting access, see the Cognos 8 Planning Installation and Configuration Guide. For information about setting up your environment to work with SAP BW and Planning, see the Cognos 8 Planning Installation and Configuration Guide.

Create a Detailed Fact Query Subject

The detailed fact query subject is a model query subject based on database query subjects and calculations. The relational folder is where the SAP star schema is imported to. The detailed fact query subject is the logical representation of the fact table and the query subjects in the relational folder are the physical representation of the SAP fact table. We recommend that you do not modify the contents of the relational folder, unless advised by customer support.

Steps
1. In Framework Manager, click the Key Figures dimension. 2. From the Tools menu, click Create Detailed Fact Query Subject. 3. In the metadata wizard, select the data source you want to use.

88 Analyst

Chapter 6: Importing Data from Cognos 8 Data Sources You can create a new data source by clicking the New button and specifying SAP BW for Planning as the type. 4. Click OK. Framework Manager creates a model query subject named Detailed_Key_Figures and a separate folder containing references to the relational objects. The references to the relational objects are the physical layer. 5. Create the package. Note: Packages that contain the Detailed_Key_Figures query subject are only accessible or supported for the report authoring tools, such as Query Studio and Report Studio if they are hidden by doing the following: In the Define Objects screen click the down arrow and choose Hide Component and Children. Click Detailed_Key_Figures and Relational_Objects.

6. Publish the package.

Recommendation - Query Items

It is a common requirement to concatenate two or more fields from a data source when creating D-Lists in Analyst. When importing D-Lists from a Cognos Package, you perform the concatenation in Framework Manager by creating a new query item. The query item can then be included in the published package and imported into D-Lists and used in D-Links. When working with SAP BW, you can use a concatenated query item to build a D-List in Analyst. However, when you create a link, either in Analyst or Contributor, then the concatenated query item cannot be used. Instead, use one of the underlying query items for the source and use a substring on the target dimension. When applying a filter in Framework Manager, you specify how it is used by selecting a usage value. To see the filtered data when publishing a package in Planning, select Always or Optional. See the Framework Manager User Guide for more information.

Recommendation - Hierarchy
These recommendations will help improve performance when working with the SAP BW import process. Use manageably sized dimensions when importing SAP BW data. This is because Planning relies on lookups against the SAP BW hierarchies during the import process, so larger hierarchies slow down the import process. This may require modelling in SAP BW since it is at a higher level of detail than the Planning process requires. Where possible, take data from the lowest level in the BW hierarchies. This is because data is taken from the fact table level and aggregated to the level selected in the Planning link. The

User Guide 89

Chapter 6: Importing Data from Cognos 8 Data Sources further up the hierarchy that members are mapped into Planning, the more aggregations are needed to be recreated during the import process. This may require modelling in SAP BW since it is at a higher level of detail than the Planning process requires.

Recommendation - Hiding the Dimension Key Field

When working with SAP BW data, the Dimension Key field for any dimension should be hidden in the Model (not the package) - both for the OLAP and Detailed Fact Query Subject access before the package is published. It is not intended for direct use from within Cognos Planning.

Working with Packages

To avoid a high number of Query Subjects and Query Items when working with and creating packages in Planning, you should make them as specific as possible so they contain only objects that are useful to a Planning user. Using a naming convention may also be useful, like using Planning as a prefix for your packages. For advanced users, you could also create a single package that holds all of the source objects.

90 Analyst

Chapter 7: D-Lists
Think of a D-List as a field in a database. The items contained in a D-List should be related to each other so that their relationships can be placed into formulas that proceed down the rows or across the columns. A D-List is a list of related items that can be put into formulas that proceed down the rows or across the columns of a D-Cube. Commonly used D-Lists include profit and loss items, months, divisions, products, customers, and cost centers. In addition to the items that make up the row, column, and page label, a D-List also contains formulas. You can use letters, numbers, punctuation marks, and spaces in D-List item names. However, you should avoid using the following: The semicolon (;) because it is used in special calculation formulas named built-in functions (BiFs). The at sign (@) and braces ({}) because these are used in formulas, built-in functions, and macros. The brackets ([]) because this naming convention is used to mark D-List formatted items, which appear as virtual dimensions in the D-Link editor.

Wildcard characters like * and ? can be used, but caution should be used when importing from a source that has item names without the wildcard characters. Use a local or loaded Allocation Table. You can also make the link using match descriptions to avoid any potential problems. See "Use Wildcard Characters in Local Allocation Tables" (p. 244). D-Lists that contain calculations can have those calculations force to zero. This means that if the calculation result is zero or if the result is nonsensical, the cell displays blank. This option is particularly useful for D-Lists with multiple calculations.

Open a D-List
You can open a D-List directly from the File menu or from a D-Cube.

Steps
1. From the File menu, click Open, D-List. 2. If you want to open a D-List from a D-Cube, right-click a D-List label, and then select Edit D-List from the list.

Create a D-List
Create a D-List to define the items that make up the row, column, and page labels. D-Lists can also contain formulas that proceed down the rows or across the columns of a D-Cube. User Guide 91

Chapter 7: D-Lists Alternatives to typing items manually are available. For more information, click a link below. Create an Import Link into a D-List (p. 124) Import Items from ASCII Files into a D-List (p. 126) Import D-List Items from Mapped ASCII Files into a D-List (p. 127) Import items from D-Cube data (p. 128) Import items from an ODBC source (p. 129) Import D-List items from Another D-List (p. 125) Import D-List items from a Cognos package (p. 130) Paste D-List Items from a Spreadsheet Database or Word Processor (p. 125)

For long lists, we recommend that each item consist of a code followed by a unique name (p. 141) (for example, A1234 salaries). For data imported from other sources, it is easier to match the unique names. D-List item names can be up to 50 characters long using letters, numbers, punctuation marks, and spaces. However, you should avoid the following characters in D-List names: Semicolon (;)The semicolon is used in special calculation formulas named built-in functions (BiFs). Brackets ([ ])These are used to mark D-List formatted items, which appear as a virtual dimension in the D-Link editor. At Sign (@)These are used in formulas, built-in functions (BiFs), and macros. Duplicate names are not allowed and are removed. Braces ({})These are used in formulas, built-in functions (BiFs), and macros. Duplicate names are not allowed and are removed.

Steps
1. From the File menu, click New, D-List. 2. In the Input New Items box, enter the items to include in the D-List. 3. Click OK. 4. To specify the attributes for each D-List item, click the attribute in the column you wish to specify, and then click Change item attributes. FormatOption Specify the data in the D-List as Numeric Format (p. 112), Date or Time Format (p. 114), D-List (p. 117), or Text (p. 118). Calculation. Define the calculations and built-in functions for D-List items (p. 93).

92 Analyst

Chapter 7: D-Lists Calc Option Specify D-List items to be weighted, specify time averages, or choose the calculation to force to zero (p. 106). 5. From the File menu, click Save. 6. Name the D-List. Note: The D-List name is case sensitive and must be unique. You can type up to 31 characters including spaces. 7. Click OK.

Formulas
You can create calculations if you need a comparison or ratio that does not exist in the data source. For example, you can calculate financial ratios, such as liquidity ratios, debt ratios, and activity. A calculation D-List is a generic term for any D-List containing formulas. D-List formulas come in four main categories: Numeric Date/Time D-List Text

These categories display against items in a D-List in the calculation column of the Format Attribute (p. 109) screen. You can arrange a formula however you want. The program ignores spaces, tabs, and carriage returns. For example, the program would consider the following three calculations identical. + Sales - {Rent and Rates} - Electricity + Sales - {Rent and Rates} - Electricity + Sales {Rent and Rates} Electricity

Import a Formula (CalcTexts)

You can import the calculations to apply to D-List items from any of the following sources:

User Guide 93

Chapter 7: D-Lists D-Cube data Mapped Ascii file Unmapped Ascii file ODBC (SQL database) Cognos package Another D-List Cognos Finance Impromptu Query Definition (IQD)

When setting up the import, your source data must contain a column with the names of the items against which the calculations are to be set, and a column containing the calculations themselves. The syntax must avoid spaces and symbols in item names. For example, if you were to use Margin % instead of Margin_percent as an item name, then when shown in the formula, margin % would have to be in braces {Margin %} to indicate to the program that this is an item name. While it is possible to import formulas from spreadsheets, it should be treated with caution. Generally, this facility is more useful for formulas generated from databases consisting of structured hierarchies. If you import formulas from spreadsheets, cell references in the formulas must be displayed and converted to text. First, you must ensure that the formulas all refer to cells in the same column or the same row and use + symbols rather than =SUM( ) in the formulas. Second, you must create range names for each cell based on the labels, apply the names to the formulas, and display the formulas in text format in the cells. Finally, this can be saved as a comma delimited (.csv) ASCII file and edited to remove leading = signs from the formulas. Zeros against detail items should be blanked out.

Steps
1. From the File menu, click New, D-List. 2. Click Import and select the import option you require. 3. At the Import of D-List Items screen, in the Select Attribute box, define the column containing the item names as Item Name and the column with the calculations as CalcTexts. 4. Make any other import settings you require. 5. Click OK. 6. Save the D-List, or use D-List > Implement to apply the calculations.

94 Analyst

Chapter 7: D-Lists

Preview or Print the Formulas

You can use the Print Preview function to view all the formulas before they are printed.

Steps
1. From the File menu, click Print or Print Preview. 2. Click the D-List tab. 3. Select the Calculations check box. 4. Click Preview. 5. Click the right or left arrow to scroll through the pages containing the formulas.

Create a Subtotal
When large amounts of data are shown, you can use subtotals to present data in a way that can be analyzed more conveniently. To add summaries, you need to specify the items for which you want to create a subtotal.

Steps
1. Open an existing D-List or create a new D-List containing the necessary items to subtotal. 2. Click the calculation cell of the item to which you want to add a subtotal. 3. Click Change item attributes. 4. Click Paste to select items specifically in the Selection dialog box. 5. Click Apply to test the syntax. 6. Save and close the D-List.

Enter a Formula into a D-List

To help you make better business decisions, you can summarize data or create conditional and statistical formulas for management analysis. Data is summarized using row or column calculations or by creating calculated D-List items. To select items needed in a formula, you can use the Search function. Items can be selected from either the Items Available list or the Items Included list. To use the Search feature, follow these guidelines: Wildcard characters of * and ? are allowed in the filter. Use a question mark (?) to represent any single character. Use the multiplication symbol (*) to represent any series of characters.

User Guide 95

Chapter 7: D-Lists Click And to show items that satisfy both criteria. Click Or to show items that satisfy either criteria. If Match Case is selected, it will further refine the filter to match on capitals or small letters. The Name box can be set using the equal symbol (=), or the not equal symbol (<>). Use the equal symbol (=) to show items that meet the criterion. Use the not equal symbol (<>) to exclude items that meet the criteria.

Steps
1. Open the D-List. 2. Click the calculation cell that you want to format. Then click Change item attributes. 3. Define the calculation: Type the formula in the Calculation box. You can enter calculations, BiFs, and conditional statements in the Calculation box. Remember that D-List items containing more than one word must be placed in braces to indicate that they are single variables. For example, {Rent & Rates} or {Margin %} If you want to select the items needed in the formula, click Paste. The arithmetic operators, plus {+}, minus {-}, multiply {*}, and divide {/}, should be entered at a later time. Select nonadjacent items by holding CTRL and clicking each item in the Items Available list. To move the selected items to the Items Included list, click Move. When you select single items from the Items Available list, double-clicking the item automatically moves it to the Items Included list. Tips: In the Select items to include list, you can select items using Search. This allows the use of wildcard characters. For example, if you have an item named P01, and you enter P01*, it will search for anything beginning with the characters P01. if you enter P?????????, it means search for any D-List item that starts with P and is ten characters long, including spaces. If you want to move items back to the Items Available list, select the items in the Items Included list and click Move. Expand explodes the constituent parts of a selected subtotal. 4. Click OK. 5. Type the operators. Ensure that D-List items of more than one word are in braces { } to treat them as a single variable. The program ignores spaces, tabs, and line returns. 6. Click Apply. 7. To calculate, from the D-List menu, click Implement.

96 Analyst

Chapter 7: D-Lists Implementing lets you test a formula in an open D-Cube before you save the D-List. Clicking Reset reverts the formula to the last saved version. 8. To save the formula without calculating, from the File menu, click Save. Formulas are shown in the Calculation column of the D-List attribute screen.

Edit a Formula
Depending on modeling requirements, you may need to modify an existing formula. You cannot edit a formula that is an output (p. 317) calculation from a built-in function (BiF). You are equally restricted in what you can do when editing a BiF calculation. You can change the BiF parameters (p. 320) but you cannot insert other items into or operate on the BiF formula.

Steps
1. Open a D-List. 2. Click the calculation cell of the item you want to edit. 3. Click Change item attributes. 4. In the Calculation box, edit the formula. 5. If you want to test the syntax, click Apply. 6. To calculate the formula, from the D-List menu, click Implement. Using the Implement function lets you test a formula before saving the D-List. If the formula is wrong, clicking Reset from the File menu reverts to the last saved version of the D-List. 7. If you want to save the formula without calculating, from the File menu, click Save.

Copy a Formula
You can copy a formula from one D-List to another to facilitate the creation of formulas that are similar among D-Lists.

Steps
1. Open the D-List containing the formula you want to copy (this is the source D-List). 2. In the Calculation box of the item in the source D-List, select the formula (or the section of the formula) from which you want to copy. 3. From the Edit menu, click Copy. 4. Close the D-List. 5. Open the D-List containing the formula to which you want to paste the copied formula (this is the target D-List). 6. Click the calculation cell of the item in the target D-List to which you want to copy the formula.

User Guide 97

Chapter 7: D-Lists 7. With the cursor in the calculation screen, from the Edit menu click Paste. 8. Click Apply. 9. From the File menu, click Save.

Remove a Formula
You can remove formulas that are no longer applicable to the D-Lists or required as part of the planning process. You cannot remove a formula that is an output calculation from a built-in function without removing the BiF formula first.

Steps
1. Open a D-List. 2. Select the calculation cell of the applicable item and click Change Item Attributes. 3. Select the calculation formula you want to remove. Existing formulas contain the following syntaxes in the calculation cell of the item: equal sign (=), Subtotal, Conditional, or BiF. 4. To remove the formula, click Clear. 5. To apply the changes to the item calculation cell, click Apply. 6. From the File menu, click Save.

Troubleshooting Formula Errors

To help with diagnosing problems with calculations, you can perform several checks. For example, you can verify if the expression includes all the details required to perform the calculation, if D-List items were double-counted, or if the formula exhibits circularity.

Check Formulas for Omission

You may want to check the subtotals that make up the grand total or the details that make up a subtotal. Viewing the details allows you to verify that the calculation expression includes the required D-List items. Examine the omitted items from a formula to ensure that the formula is valid and designed as intended.

Steps
1. Open the D-Cube containing the D-List you wish to check. 2. From the D-Cube menu, click Selections, Reselect. 3. Click on the tab for the relevant D-List.

98 Analyst

Chapter 7: D-Lists 4. Select a total and then click Expand(p. 157) to select the subtotals that form a part of the grand total. 5. Click Expand again to select the details that comprise the subtotals. Items that are not selected are omitted from the formula. 6. Scroll through the list and note the omissions. To make it easier, click Move to move the selected details to the Items Included list; this leaves only the omissions in the Items Available list.

Check for Double Counting in Formulas

In a hierarchical D-List, it is possible to check for double counting by using a simple D-Cube. Do not use this method when D-Lists use more complex calculations or there is no overall grand total.

Steps
1. In the D-List, create a check item and set it to be a sub-total of all the detail items. When setting the calculation use the option to show Detail items in the Show drop box. 2. Create a single item D-List. 3. Use the hierarchical D-List and the D-List created in step 2 to make a D-Cube. 4. Use D-Cube commands to set all detail items to 1 in the cube. The Grand Total in your hierarchy and the check item you inserted should now show the same value. If they do not, then some items have been omitted or double counted in setting the sub-totals in your hierarchy.

Check for Circularity

A formula exhibits circularity if x = function(y) and, directly or indirectly, y = function(x). Such circularity is not mathematically permissible and the program takes considerable steps to avoid circular formulas. When you select items to put in a formula, the selection screen hides the items causing circularity. For example, suppose you set up the formula:
Margin = {Margin %} * Sales/100 Costs = Sales - Margin

Then, at a later date, the formula you want to set up is:

Margin %= (Sales - Costs)/ Sales * 100

When you click Paste to select items for the calculation on Margin %, the item Costs would not be available for selection because it is indirectly a function of Margin% from the first formula. Hence, the formula would be circular.

User Guide 99

Chapter 7: D-Lists The program does not allow circularity within a D-List even if the application of a timescale removes the circularity. For example, suppose the built-in function, @Feed, is set up to feed the closing balance of one month to the opening of the next:
Close_n = Open_n + Inflow_n - Outflow_n Open_n = Close_n-1

where: n = period number Inflow is not allowed to be a function of the opening balance despite the fact that the presence of a timescale means that the formula is not strictly circular. The built-in function, @FeedParam, is specifically designed to avoid this.

View Formulas
You can view formulas for an individual cell in a D-Cube or view all formulas in a D-List. You can also view formulas for an individual cell in a D-Cube by pressing F7.

Steps to View a Single Formula

1. Open the desired D-Cube. 2. Click the cell containing the formula. 3. From the D-Cube menu click Show Formulae. A Calculations box appears and displays the formula. 4. When you finish viewing the formula, close the Calculation box. Alternatively, you can view D-List formulas. 5. In the Calculation box of a D-List, select the calculation cell of the D-List item. 6. Click Change item attributes.

Steps to View all Formulas in a D-List

1. Open a D-List. 2. From the File menu, click Print Preview. 3. Click the D-List tab. 4. Click the Calculations box. 5. Click Preview. 6. Click the right or left arrow to scroll through the pages containing the formulas.

100 Analyst

Chapter 7: D-Lists

D-List Conditional Formulas

Conditional or Logical Formulas
Logical functions are used to compare a value or the result of a formula with another expression. The logical operators available for conditional tests are IF, THEN, ELSE, AND, OR. The relational operators used are less than (<), greater than (<), equal to (=), greater than or equal to (>=), less than or equal to (<=), and not equal to (<>). Notes: The logical operators IF, THEN, ELSE, OR, and AND must be in capital letters and must be followed by parentheses ( ), to denote the start and end of the expression. You cannot enter the result of an IF THEN ELSE statement and expect breakback to give the reason for a particular result. Unlike simple mathematical formulas, conditional statements are not reversible.

The common syntax for a conditional formula is expressed as:

IF(test) THEN(expression) ELSE(expression)

AND Syntax
IF(test1 AND test2) THEN(expression) ELSE(expression)

OR Syntax
IF(test1 OR test2) THEN(expression) ELSE(expression)

The calculations included in a conditional statement can include further IF THEN ELSE calculations:
IF(Profit<=3525) THEN 0 IF (Profit<=6725) THEN 1 ELSE 2

Include the conditional statements in the required order because the calculation will return the result from the first condition satisfied. In the example above, if the profit is 3000 then this calculation will return the answer 0 as the first condition is satisfied.

User Guide 101

Chapter 7: D-Lists However, if the calculation cell were to read

IF (Profit<=6725) THEN 1 IF(Profit<=3525) THEN 0 ELSE 2

Then a result of 1 would be returned as, having found the first condition satisfied, the second condition is not considered. There are various other abbreviations For example, you can exclude the final ELSE statement entirely and it assumes ELSE 0. IF(Profit>3525) THEN 1 means:
IF(Profit>3525) THEN 1 ELSE 0

You can test the truth of a statement 1=TRUE 0=FALSE without putting in the IF THEN ELSE: (Item1=Item2) means:
IF(Item1=Item2) THEN 1 ELSE 0

IF(Item1) THEN 10 means:

IF (Item1<>0) THEN 10 ELSE 0

Example - Set Up a Conditional Formula

Suppose you want to group people by tax band using a conditional statement. The syntax would be:
IF(Profit<=3525) THEN 1 IF(Profit>3525 AND Profit<=6725) THEN 2 IF(Profit>6275 AND Profit<27825) THEN 3 ELSE 4

The result uses the conditional statement to group people by tax band: A more complicated example of a conditional formula is used to work out the tax amount:
IF(Profit<=3525) THEN (0) IF(Profit>3525 AND Profit<=6725) THEN ((Profit-3525)*.2) IF(Profit>6275 AND Profit<27825) THEN ((6725-3525)*.2 + (Profit-6725)*.23) ELSE ((6725-3525)*.2 + (Profit-6725)*.23)+(Profit-27825)*.40)

102 Analyst

Chapter 7: D-Lists

D-List Formatted Items in Formulas

Typically, text is used in formulas to display messages as a result of a conditional test such as OK or ERROR. However, the text must be D-List formatted text. You cannot type the conditional result text (such as OK or ERROR) directly into the formula as you can in a spreadsheet. The program stores text in a D-List formatted column as a number (named the item identification {IID} number). The formula must include a number as a result that refers to the IID number of a D-List item from another D-List. To view an IID, from the File menu click Print Preview, Preview. To find an IID, click the Summary Info button in the Analyst toolbar. In the example that follows, a flag has been set up to select names when the Grade or Division has been excluded accidentally. The item, Error-Flag, is a formula:
IF(Division=0) OR (Grade=0)) THEN 1 ELSE 2

However, rather than returning the number 1 or 2 according to the result of the conditional formula, the item Error-Flag has been formatted to accept text from another D-List. The other D-List contains only two items: ERROR and OK, which have identification numbers of 1 and 2, respectively. So when the result of the formula gives the answer 1, it looks up the identification number 1 and displays ERROR. Similarly, when the result of the formula gives the answer 2, it looks up the identification number 2 and displays OK.

Create a Conditional Formula Using D-List Formatted Text

The procedure for setting up a formula using D-List formatted text is the same as for setting up any other formula except that the numeric result is converted to text by applying a D-List format to the item containing the result. To apply the D-List format to the conditional formula, you must: assign a conditional formula to a D-List item create a separate D-List with a list of the text you want to display apply a D-List format

Steps
1. Open or create the D-List that you want to format to D-List formatted text. 2. Click the calculation cell of the item you want to format. 3. Click Change item attributes. 4. Create the formula in the D-List as a normal logical formula. 5. To test the syntax, click Apply. If the syntax is correct, the word Conditional appears in the calculation cell of the D-List item. 6. From the File menu, click Save. User Guide 103

Chapter 7: D-Lists Note: You may have to set conditional formulas to Low in the Priority box so that the totals add up correctly. The priority sorts conflicting formulas near the edges and corners of D-Cubes where calculations must be carried out in a specific order. You must consider this issue more than you would in a two-dimensional spreadsheet. In the event of a conflict, the higher priority formula is calculated last; therefore, its result is used. 7. To create a separate D-List with a list of the text you want to display, from the File menu, click New, D-List. 8. Type the item names, and click OK. 9. From the File menu, click Save As. 10. Name the D-List. 11. From the File menu, click Close. 12. In the D-List that requires formatting, click the Format cell of the item you want to format (the same item that contains the conditional formula), and then click Change item attributes. 13. In the Attribute box, click D-List and then select the library containing the desired D-List. 14. In the Available D-Lists box, select the D-List, and click Apply. The word D-List appears in the format cell of the item. 15. From the File menu, click Save. The formatted column converts the numeric result from the chosen D-List into text according to the identification number of the items contained therein.

Formula Priority
Priority conflicts can occur when a particular cell has a list of several formulas from which it must choose. Typically, this can occur when using logical IF THEN ELSE operators and percentages. The program can perform the calculation of the total either across the rows or down the columns, which can produce different results. For example, the percentage of the total is not the same as the total of the percentages. Setting priorities tells the program which formula to use in the event of a conflict. The program always uses the formula with the higher priority. Formula priority can be set to High, Medium, or Low.

View the Formula Priorities

View formula priorities to verify that the correct priority is set. The priority sorts conflicting formulas near the edges and corners of D-Cubes where calculations must be carried out in a specific order. You must consider this issue more than you would in a two-dimensional spreadsheet. In the event of a conflict, the higher priority formula is calculated last; therefore, its result is used.

Steps
1. Open the appropriate D-Cube. 104 Analyst

Chapter 7: D-Lists 2. Click the D-Cube calculation cell of an item containing a formula, and then press F7. A text box appears displaying the formula used, together with a list of the formulas it has overridden. 3. Close the text box.

View Priorities in the Priority Box Steps

1. Open the D-List. 2. Click the calculation cell of the item containing the formula. 3. Click Change item attributes. The priority is designated in the Priority box.

View Single Formulas

You can use this method to view all formulas, priorities, and other attributes.

Steps
1. Open the D-List. 2. From the File menu, click Print Preview, Preview.

Set Formula Priorities

The default priority for formulas is medium priority. If a conflict between two or more formulas with the same priority occurs, the program uses the formula contained in the earlier D-List. The order of the D-Lists is the order in which the D-Lists were chosen when the D-Cube was created. Calculations in the first D-List are calculated first and therefore are overridden by those in the later D-Lists, which are calculated last.

Steps
1. Open the D-List. 2. Click the calculation cell of the item for which you want to set the priority. 3. Click Change item attributes. 4. In the Priority box, change the priority of the formula to Low, Medium, or High. 5. Click Apply. 6. Save the D-List.

User Guide 105

Chapter 7: D-Lists

Apply a Time Average

You must define certain items in the calculation D-List as time averages so they give sensible results when calculating year-end totals or other subtotals. There are four kinds of time averages: Time Average = sum / number of periods. First period = use the data from the first period in the time formula. Last period = use the data from the last period in the time formula. Zero = displays as zero. Do not try to put a time average on a timescale D-List. Put the time average on a calculation D-List instead. The first and last time averages refer to the first and last items in the D-List formulas, not necessarily the chronological order. For example, if the formula is Q1=Feb+Jan+Mar, the first period is Feb not Jan.

Steps
1. Open or create a D-List containing formulas (not a timescale D-List) 2. Click the Calc. Option cell of the item to which you want to apply the time average. 3. Click Change item attributes. 4. In the Calc. Option list, click Time averages. 5. In the Available functions box, select one of the following options: Time average First period Last period Zero

6. Click Apply. 7. Save the D-List.

Weighted Averages
All items that are percentages, ratios, prices, or performance measures or items that can be expressed on a per unit basis must be set as weighted averages. This ensures that an item is correctly subtotaled and lets you breakback over subtotals. Weighted averages affect how an item is summed across any other dimension.

106 Analyst

Chapter 7: D-Lists

Quarter 1
Units Sold Price Sales Value 50 5 250

Quarter 2
50 6 300

Quarter 3
100 7 700

Quarter 4
50 6 300

Full Year
250 ? 1550

The Price for the full year is clearly not 5 + 6 + 7 +6 (=24). To calculate it we need ((5 x 50) + (6 x 50) + (7 x 100) + (6 x 50) )/ (50 +50 +100+50) =1550/250 = 6.2 The item Price must be set as a weighted average, weighted by Units Sold.

Choosing Which Item to Weight By

Generally, if D-List item A is a formula: A = B / C. Then item A should be weighted always by item C, the denominator, even if the relationship between the items is expressed differently. For example, the calculation above could be arranged such that B is the formula result: B = A * C. Item A still should be weighted by item C. A common example is a D-List containing the items Sales, Price, and Units:Price = Sales/Units. Price must be a weighted average weighted by Units.

Apply a Weighted Average

Weighted averages are applied by setting the D-List items to weighted averages.

Steps
1. Open or create a D-List containing formulas (not a timescale D-List) 2. Click the Calc. Option cell of the item to which you want to apply the weighted average. 3. Click Change item attributes. 4. In the Calc. Option list, click Weighted Averages. 5. In the Items to weight by box, select an item to weight by. 6. Click Apply. 7. Save the D-List.

User Guide 107

Chapter 7: D-Lists

Override a Weighted Average

It is possible to override a weighted average so that a total across another dimension uses straight addition instead. In the example shown below, a zero was added to the calculation of Total 1. Even though Unit Price has been set up as a weighted average weighted by Units, Total1 adds the prices together: 10 + 7 + 8 + 12 = 37. The annual total (Year) uses the more common weighted average: Year = Q1 + Q2 + Q3 + Q4 Total1 = Q1 + Q2 + Q3 + Q4 + 0

Q1
Unit Price Units Sold 10 50

Q2
7 100

Q3
8 200

Q4
12 50

Year
8.5 400

Total 1
37 400

In this example the weighted average is set on the item Unit Price in the rows D-List. The overridden total, Total 1, is in the timescale D-List forming the columns.

Steps
1. Open the D-List containing the total where you wish to override the weighted average. This might be a total in a timescale D-List for example. 2. Click the calculation cell of the item with the calculation. 3. Click Change item attributes. 4. Add a zero to the calculation. Adding a zero uses a straight total rather than a weightedaverage.

Force to Zero
Applying this average to any D-List item will cause any calculation on it in another D-Cube dimension to give a result of zero. It is particularly useful where the item applies in the detail only and is irrelevant on a total. This setting is applied automatically to D-List and Date formatted items but may be removed manually if desired by changing the setting to None.

Remove Averages
Steps
1. Open the D-List containing at least one formula (not a timescale D-List) with an average. 2. Click the Calc. Option cell of the item of which you want to remove the average.

108 Analyst

Chapter 7: D-Lists 3. Click Change item attributes. 4. In the Calc. Option list, click None. 5. Click Apply. 6. Save the D-List.

Applying Local Formats to D-Lists

A format sets the form in which data displays and can be in numeric, date/time, or D-List (text) format. Use numeric formats (p. 112) to set decimal places, insert thousand delimiters, set braces for negative numbers, and set prefixes and suffixes. Use date and time formats to display dates and times (such as 6-Jul-99) as data in a D-Cube. Use D-List Formats (p. 117) to enter text, restricting the text to the codes and names displayed in a selected D-List. Free-text formats (p. 118) are allowed. You can name and save formats for repeated use.

Local vs. Global Formats

A local format (p. 110) involves formatting D-Lists only and applies to an individual row or column. Generally, you apply local formats in the Format Attribute screen. The Format Attribute screen lets you control the appearance of items in a D-List. The content of the Format Attribute screen depends on the format type selected in the Format type list. Some elements of the Format Attribute screen are common to all format types: Attribute list, Format Type list, Prev. and Next, Assign, Apply, Reset, Load, Save, and Save As. A global format involves formatting an entire D-Cube. Generally, you apply global formats by clicking Format on the D-Cube menu. Although the two screens are almost identical, a D-Cube format is a general format applied to all data in the entire D-Cube, not just a specific item from a D-List. However, a local format applied to an individual row or column has priority over the global D-Cube format.

Types of Formats
The format affects how the data appears. A Numeric format determines the number of decimal places, the appearance of negative numbers, commas separating thousands, blank cells rather than zeroes, and showing percentage signs or currency symbols before and after the number. A Date/Time format determines whether the date appears in hours, months, years, or any of the standard formats such as DD/MM/YY. A D-List format lets you type text contained in another D-List into the formatted row or column. A Text format lets you type any text.

User Guide 109

Chapter 7: D-Lists

Assign Local Formats

This feature is useful when you would like to assign one format to several items within a D-List rather than having to define and then apply the same format to each item.

Steps
1. Create a format. 2. Click Assign to assign a format to other D-List items in the Selection dialog box. 3. Select the items to assign the format to in the Items Available list. 4. Click Move to move the items to the Items Included list. 5. Click OK to return to the D-List edit screen. 6. Repeat as necessary.

Save a Local Format

This feature is useful when you want to save a format within a D-List to be used several times in other D-Lists rather than having to define and then apply the same format repeatedly.

Steps
1. Open or Create an appropriate D-List 2. Create a format (numeric (p. 112),date or time (p. 117), or D-List (p. 117)) or edit an existing format 3. In the D-List editor, click Save. Click Save As to save the format with a new name. 4. Select a library to which you will save the local format from the Libraries list. 5. Type a name in the Saved Format Name box. 6. Click OK to return to the D-List edit screen.

Load a Local Format

This feature is useful when you want to apply a saved format.

Steps
1. Open the appropriate D-List. 2. Click the format cell of the item you want to format. 3. Click Load. 4. Select a library from the Libraries list.

110 Analyst

Chapter 7: D-Lists 5. Select a name in the Saved Format Name box. 6. Click OK. 7. Click Apply. 8. Save the D-List.

Formulas and D-List Formatted Items

Although the D-Cube displays an item name from the format D-List in a D-List formatted cell, it actually stores a value. The underlying number in a D-List formatted cell is the item identification number (IID) of the item from the format D-List. Each D-List item is assigned an IID when it is first inputted and no amount of reordering or deleting items can change it (short of substituting with a new D-List). New items are given new IID numbers - not reassigned old unused numbers. A D-List formatted cell whose underlying value is zero, or a number that is not a valid ID in the format D-List, displays as a blank cell in the D-Cube. Both detail and formula items can be D-List formatted. If a formula returns a number that is not a valid ID in the format D-List, the formula cell is blank in the D-Cube. To find an IID number, click the Summary Info button in the Analyst toolbar. You can display the item identification number of a D-List item by clicking the File menu, pointing to Print Preview, and then clicking the Preview button. The IID displays before the item name.

Format a Specific Row or Column

To format an individual row, column or page, open the relevant D-List and apply a format to a specific D-List item. This local format of a D-List item will always have priority over the global D-Cube format.

Steps
1. Open the D-List. 2. Click Change item attributes. 3. Select a format from the Attribute box as follows: A numeric format (p. 112) determines the number of decimal places, the appearance of negative numbers, commas separating thousands, blank cells instead of zeroes, and showing percentage signs or currency symbols before and after the number. A date or time format (p. 114) determines whether the date is displayed in hours, months, years, or any of the standard formats such as DD/MM/YY. A D-List format (p. 117) allows text from another D-List to be typed into the formatted row or column. A free-text format (p. 118) lets you type any string of characters or symbols.

4. Format the items accordingly.

User Guide 111

Chapter 7: D-Lists 5. Click Apply. 6. Repeat for other D-List items you wish to format. 7. Optional: Click Assign to assign a format to other D-List items. For example, the same numeric format applied to one specific item also can be assigned to another item. Click Assign, and then select the items to assign the format to. Select the items then click Move to move from the Items Available list to the Items Included list. Click OK to return to the D-List edit screen.

8. Apply the global format to the D-Cube. 9. From the File menu, click Save.

Numeric Formats
The numeric format lets you change how data displays. You can change the decimal places, assign currency such as $, , or FF, display negative numbers in brackets, hide zero cells, insert comma separators for thousands, or apply a scaling factor. Suffixes or prefixes are allowed - characters, numbers, punctuation marks, or percent signs (%) up to a recommended maximum of 10 characters. The default format is zero decimal places, comma delimiter active for thousands, brackets for negative numbers, a scaling factor of one, and zeros displayed when they display in a cell (Blank if zero is cleared). This displays the number 1234.0 as 1,234 and the negative number -1234.0 as (1,234). When you change the numeric format options, the Sample field updates showing how two sample numbers (1234 and -1234) will be displayed in a D-Cube. All values with a numeric format are right justified in the D-Cube. Where an item in a D-List has a scaling format applied, the behavior will be different in Analyst and Contributor. In Analyst the numbers are entered, ignoring any scaling applied in the formatting. So where an item is scaled to 1000s, if you type in 22k, it shows as 22 (i.e. 22k), but if you type in 22, it shows as 0 (0.022k), assuming that less than 2 decimal places are showing. In Contributor the numbers are entered as displayed - if the cell shows 1000 for an underlying value of 10, and you type in 1200, the new value shows as 1200 with the underlying value of 12.

Apply a Numeric Format

By applying different numeric formats, you can change the appearance of numbers. A numeric format does not affect the actual cell value that Analyst uses to perform calculations.

Steps
1. Ensure that the D-List is open and active. 2. Click the format cell of the item you want to format. 112 Analyst

Chapter 7: D-Lists 3. Click Change item attributes. An alternative is to double-click the format cell of the item. 4. Select Format from the Attribute list box, then click Numeric. 5. Set the decimal places (p. 113) and type the prefix and suffix (p. 192) for both positive and negative numbers. 6. Select Blank if zero if you want cells containing zero to be blanked out. 7. Select Use thousand delimiter to use a delimiter to show the number 1000000 as 1,000,000. 8. Type a numeral in the Scaling Factor(p. 257) box to display data in thousands, millions, and so on. Note: A scaling factor of 1000 will store the number 3333 as 3333.000 but display it as 3.333. If you set the decimal places to zero, it will display it as 3. This can cause confusion when adding three numbers X=A+B+C where A, B and C are all equal to 3333.000. With a scaling factor applied, this will be displayed as 3+3+3 =10 where the underlying calculation is correct as 3333+3333+3333=9999. 9. (Optional) Click Save and name the format. This allows you to apply the same format elsewhere. To load an existing format, choose Load and then select the name of the format. The Save As button allows you to edit an existing format and save it under a new name. 10. (Optional) The Assign button allows you to assign the current format to several other D-List items. 11. From the File menu, click Save.

Set the Decimal Places

Set how many decimal places to display. The default setting is zero.

Steps
1. Open a D-List. 2. Click the Format cell of the item you want to set decimal places to. 3. Select Numeric from the Attribute list. 4. In the Decimal Places scroll box, set the decimal places.

Set Scaling Factor within a D-List

Setting a Scaling Factor allows you to display numbers in round thousands, millions, and so on. The stored number is divided by the scaling factor to give the number displayed on the screen. The default scaling factor is 1, meaning that numbers are stored as they are displayed. For example, a scaling factor of 1000 applied to the number 1234.00 would display it as 1.23. The number that is stored by the program would always remain 1234.00.

User Guide 113

Chapter 7: D-Lists A scaling factor of 0.01 applied to the number 1234.00 would display the number as 123400.00. Again, the number that is stored by the program would always remain 1234.00. You must be careful with scaling factors. The program keeps the original number stored to many decimal places so there are no significant rounding errors. However, adding the numbers 1333 + 1333 +1333 = 3999 would give 1 + 1 + 1 = 4 when scaled by a factor of 1000. Applying the scaling factor is not the same as rounding to integers. Note: Use caution when applying scaling factors to individual D-List items because there is no indication in a D-Cube that scaling factors are set, and data can seem to be incorrectly calculated. Remember that you can set a scaling factor for an individual D-Cube. When you type a number into a cell formatted with a scaling factor, type the stored number, not the displayed number. The same rule applies when using D-Links to copy numbers into cells formatted with a scaling factor. You can also set scaling factors in D-Links.

Steps
1. Open a D-List. 2. Select the Format cell of the item you want to format. 3. Click Change item attributes. 4. From the Attribute list, click Numeric. 5. From the Scaling Factor box, type a scaling factor. Choose a factor between 1,000,000 (one million) and .000001 (one millionth). The default setting is 1 (no scaling). Normally, you want a D-Cube to display the actual value stored in a cell (the underlying number). You can use the scaling factor if you want the numbers displayed in the D-Cube to be scaled up or down from the stored value. The stored value is divided by the scaling factor to arrive at the displayed number.

Date and Time Formats

All values with a date or time format are left-justified in the D-Cube. There are three categories of date and time format. All date formats convert an underlying number (the date serial number) in the D-Cube grid into a date. Data entry in date-formatted cells in the D-Cube grid is restricted to valid dates (a date in the appropriate format).

Dates in Calculations
You can create calculations referring to date formatted items - they will calculate using the serial number. For example, you have a D-List containing two detail items, Start Date and Duration, and a formula item, End Date: {End Date} = {Start Date} + Duration Start Date is date formatted (DD/MM/YY).

114 Analyst

Chapter 7: D-Lists If Start Date displays 01/01/97 (serial number is 35,429), and Duration displays 2, then End Date will display 35,431. If End Date is date formatted (DD/MM/YY), it will display 03/01/97.

Date Formats
Thirty-two date format options are available, each of which is represented using a format code. The serial number for date formats containing a day, a month, and a year code (for example, YY-MM-DD, DD/MM/YY) is calculated from the number of days between January 1st, 1900 and the start of the period. For example, an item has been given the date format DD-Mon-YY. In the D-Cube, the date serial number and the displayed cell contents are related.

Serial Number
0 1 35,429 35,429.5

Cell Display
01-Jan-00 02-Jan-00 01-Jan-97 01-Jan-97

Other serial numbers are calculated differently. The day format is an example.

Serial Number
0 1 6 14 14.5

Cell Display
Mon Tue Sun Mon Mon

The following date formats show 4 digit years: DD-MM-YYYY MM-DD-YYYY DD/MM/YYYY MM/DD/YYYY DD.MM.YYYY

User Guide 115

Chapter 7: D-Lists MM.DD.YYYY DDMMYYYY MMDDYYYY

Time Formats
There is one time format, represented by the format code HH:mm. The serial number stored for a time-formatted cell is a fraction of a day. For Example:

Serial Number
0 0.5 0.75 1 1.25

Cell Display
00:00 12:00 18:00 00:00 03:00

Specific
The specific format provides the same twenty-four date formats provided by the date format, each supplemented with the time format. Serial numbers are calculated the same as the date. For Example: (Date/Time -Specific DD/MM/YY):

Serial Number
0 1 35,429 35,429.5

Cell Display
01/01/00-00:00 02/01/00-00:00 01/01/97-00:00 01/01/97-12:00

116 Analyst

Chapter 7: D-Lists

Apply a Date or Time Format

The date or time format shows date and time serial numbers as date and time values, according to the type and locale that you specify.

Steps
1. With the D-List active, click the Format column of the item you want to format. 2. Click Change item attributes. 3. Select Format from the Attribute box and then click Date/Time. 4. Select Date, Time, Date/Time, and/or Blank if Zero from the Format section. Note: To display un-entered dates as blank (rather than 1st January 1900, which is the base date), select Blank if Zero. This option is available in Analyst and Manager, but is unavailable in Analyst for Excel. 5. Select a date or time format from the Available formats box. Dates may be added or subtracted in formulas. The calculation uses the stored number as the number of days or fractions of a day since January 1st, 1900 (Jan 1st, 1900 is stored as 0, Jan 2nd, 1900 has a stored number of 1, and so on). 6. (Optional) If you want the count to start a day earlier (so that January 1st, 1900 is stored as 1, January 2nd, 1900 as 2, and so on) click Date/Time in the Format section, then click End instead of Start. Then choose the format as usual. 7. From the File menu, click Save. The dates and times can now be typed directly into the formatted cells in the D-Cube.

Apply D-List Formats

A D-List format lets you type text from another D-List in a row or a column. The D-List format is a very powerful feature of Analyst. The format is used in database type functions to consolidate data in a similar manner to query-style reports. The format also can be used to provide look-up tables. When typing in text, the first character or two of a text entry is sufficient as long as it is not ambiguous. If you are unsure of what to type, type a question mark (?) and a drop down menu appears showing all the available items from the D-List. Free text format is not allowed because the text typed must correspond to an item in a D-List.

Steps
1. Open the D-List that contains the items you want to format to accept text. 2. Click the Format column of the item you want to format and then click Change item attributes. 3. Select D-List from the Attribute box. 4. Select a library from the Libraries box. User Guide 117

Chapter 7: D-Lists 5. Select a D-List from the Available D-Lists box. Only those items appearing in this D-List will be allowed in the formatted column. 6. Click Apply. 7. From the File menu, click Save. Tip: When typing in text, the first character or two of a text entry is sufficient as long as it is not ambiguous. If you are unsure of what to type, type a question mark (?) and a drop down menu appears showing all the available items from the D-List.

Apply Free-Text Format to a D-List Item

You can format a D-List item to accept free text of 50 characters or less. Text format has priority over time and numeric formats, but not over D-List formats. Text may be transferred to another D-Cube using a D-Link as long as the target D-List item is also a text format. Formula cells cannot accept free text and will always display blank.

Steps
1. Open or create a D-List. 2. Click the format cell of the item you wish to format. 3. Click Change item attributes. 4. Select text from the Attribute list. 5. Save and close the D-List.

Timescale D-Lists
D-Lists containing items such as days, weeks, months, quarters, or years need to be defined as timescales. These are necessary for special built in functions (BiFs) that use time to apply calculations such as debtor days, stock-turn days, and creditor days to determine closing balances. Timescale D-Lists also are used for importing data containing dates into the correct period. They also can be used by the Grow command to grow a base value by a certain percentage per period (linear or compound). Typing Jan01 in a Period column automatically generates dates. Dates also may be imported from spreadsheets using CTRL+C to copy a selected range and CTRL+V to paste into the period column. Dates in Analyst run from 1950 to 2049. So if you type the two digit 49, it means 2049. But if you type 50, it means 1950.

118 Analyst

Chapter 7: D-Lists

Create a Timescale D-List

Steps
1. From the File menu, click New, D-List. 2. Type the date items (for example, Jan 00, Feb 00, Mar 00, and so on). 3. (Optional) Type the names of any subtotals required. 4. Click OK. 5. Set the calculations on any calculation item. 6. From the D-List menu, click Options. 7. Click the Timescale tab, and set the following options: For certain standard formulas, you need dates. Either set the dates by defining the start and end of each period in the format dd/mm/yy or set the months in the period column. Combinations of from and to dates and months in the period column are not permitted. Note: After you create your timescale D-List, check the timescale output. Ensure that the dates are correct according to what you set. For example, if you have the format as mm/dd/yyyy, some of the entries may switch to dd/mm/yyyy. If this occurs, manually change the From and To fields. For the built-in functions (BiFs), ytd (year to date) and de-ytd, you need a fiscal year start date. Type the start date and month in the Start of fiscal year box. For the built-in functions Drive, Drive1, Drive2, Outlook, DCF, ICF (discounted and inflated cashflow), Grow, Mix, and Stockflow; and for some of the options in @Time, you need a switchover date denoting the dividing point between past and present. Select the Use Switchover box and then type the date if you require the built in function to read the switchover setting from the D-List. You can use from, to, and period dates to define a time period. 8. If you want to use a generic monthly timescale (Jan, Feb, Mar), click Use as Timescale, and click OK. 9. From the File menu, click Save. 10. Name the D-List. Note: The D-List name is case sensitive and must be unique. You can type up to 31 characters including spaces. 11. Click OK.

User Guide 119

Chapter 7: D-Lists

Create a Custom Timescale

You can define the timescale by specifying the start date and the length of the time period in days. This method is much quicker if you have a weekly or a mixed timescale.

Steps
1. Open a D-List and choose D-List>Options. Click the Timescale tab and select the Use as timescale option. The timescale dates can now be set in two modes: Normal and Custom. 2. Choose whether you want to set the Normal or Custom mode: If you want normal mode, set the From & To dates manually or pick up on the D-List item name and the program fills in sensible From and To dates. If you want custom mode, set the length in days and the first From date.

Items of zero length are treated as non-time items and do not have a From and To date attached to them. This allows you to start the timescale at an item other than the first D-List item. You can also allow for non-time items by setting the length to zero. You can switch between the Normal mode and Custom mode.

Timescales and BiFs

All Built-in Functions (BiFs) are time-dependent. However, you should not set up BiFs in the timescale D-List. You should set up the BiF in the main calculation D-List. The BiF becomes active only when set up in a D-Cube against a timescale D-List. Usually, you should set up the D-Cube with the calculation D-List first and the timescale D-List last. This ensures that the timescale D-List has priority in the event of conflicting calculation formulas. If the BiF formula accidentally is given a higher priority, the result is zero in time totals such as quarterly or annual totals.

Common Errors in Timescales

The most common error in defining timescale D-Lists is failing to define the periods for items that require them. This gives the BiF an ambiguous calculation when setting results using inputs such as debtor days, switchover dates, and so on. Another common error is to define the timescale D-List as a mixture of generic months (Jul, Aug, Sep, and so on) and specific dates (for example, 01/05/02 to 31/05/02). Choose only one date format. Also, remember to use D-List as a parameter in the BiF formula if you want to use the switchover date defined in the D-List.

Set Periods of Uneven Lengths

Occasionally, you must use the number of days in a period to take account of periods of unequal length. For example, in the built-in function, @DelayStock, you should set the indicator to 1 for days if the periods are not equal in length. If you do not define the dates in the timescale D-List, 120 Analyst

Chapter 7: D-Lists the program assumes that all periods are equal. The default days setting for a month uses an average of 30.42 days (365/12), which can lead to inaccuracies.

Steps
1. Open the timescale D-List. 2. From the D-List menu, click Options. 3. Ensure that the From and To dates are defined correctly for each period. 4. Click OK. 5. Save the timescale D-List.

Copy an Existing D-List

You can make an exact copy of an existing D-List so that the formulas and formatting are copied with the item names.

Steps
1. Open the D-List 2. Select the D-List you want to copy. 3. Click OK. 4. From the File menu, click Save As. 5. Type the new name. 6. Click OK.

Edit a D-List
You can edit items names and related attributes. You can edit other features, such as timescales, import links, subheaders, and masks definition. A mask is used to hide individual items within a D-List or to prevent people writing over specified items. This is particularly useful when you are operating on a network and want to allow limited access to large D-Cubes. You can be quite specific about the level of security applied to each item by means of applying an item called a mask. A mask contains a security pattern with a list of users and their access rights. It can be attached to one or more D-List items. You may customize the mask to give each user a key to unlock the restrictions. The choices are Read-only, Read/Write, or Invisible. Items that are set to Invisible will not appear on the selection screen at all, whereas items marked Read-only will appear on the screen but can not be over-typed or changed except by some indirect method such as a breakback. Read/Write access means that the items can be changed as usual.

User Guide 121

Chapter 7: D-Lists These settings can be defined for each user or for groups of users. If a user is also a member of a group for which access rights have been defined, he assumes the highest level of access available. Steps Open the D-List. To edit item names, click on them and overtype with the new names (p. 123). To add items, from the D-List menu, click Add Items (p. 123). It is possible to add items to a D-List from different sources. For more information, click a link below. Create an Import Link into a D-List (p. 124). Paste D-List items from a spreadsheet database or word processor (p. 125). Import D-List items from another D-List (p. 125). Import items from unmapped ASCII files into a D-List (p. 126). Import D-List items from mapped ASCII files into a D-List (p. 127). Import items from D-Cube data (p. 128). Import items from an ODBC source (p. 129). Import D-List items from a Cognos package (p. 130)

To delete items, from the D-List menu, click Delete Items (p. 131). To change the attributes of an item, select the item you want to edit and then click Change item attributes. Edit the attribute of the items you want to change. Formats (p. 109) Averages, such as Time (p. 106) and Weighted (p. 106) Formulas (p. 93)

To edit other features of the D-List, from the D-List menu, click Options, and then the appropriate tab. Timescales (p. 118) Import Links (p. 124) Sub headers (p. 145) Unique names (p. 141) Security (Masks)

Edit the features you want to change and then click OK. Implement the changes and save the D-list (p. 131).

122 Analyst

Chapter 7: D-Lists

Edit D-List Item Names

Change the D-List item names to reflect a name that is more obvious to users. You can use letters, numbers, punctuation marks, and spaces in D-List item names. However, you should avoid using the following: The semicolon (;) because it is used in special calculation formulas named built-in functions (BiFs). The at sign (@) and braces ({}) because these are used in formulas, built-in functions, and macros. The brackets ([]) because this naming convention is used to mark D-List formatted items, which appear as virtual dimensions in the D-Link editor.

Note: If you change D-List item names, you should check any D-Links in which the D-List is used. If the D-List is paired with a different D-List on a match descriptions basis then the item you have renamed will no longer match.

Steps
1. Open the D-List. 2. Click the item you want to edit. 3. Double-click the item name and type the new or edited name. Note: To split long column headings, double-click the item you want to edit in the Item Name column. Use the pipe symbol "|" (shift+\) to denote a line break so that the column heading displays on two lines. You can have more than one line break. The pipe symbol does not display on the screen. 4. From the File menu, click Save.

Insert items from a D-List

You can insert items from a D-List while in a D-List or D-Cube.

Steps
1. Choose to insert items while in a D-List or D-Cube To insert from a D-List, from the D-List menu, click Add Items and select Input. To insert from a D-Cube, right-click your mouse on the D-List dimension, and then click Insert Items. The Input new Items screen allows you to type the new D-List item names and select the options you need, including whether to import items, and if so, what mode to import it by; where to place the new items, and a choice of Subtotals into which the new items may be inserted. 2. Type the D-List item names, or make your selections, and click OK.

User Guide 123

Chapter 7: D-Lists

Create an Import Link into a D-List

Use an import link to update the items in a D-List on a regular basis from a source file. This could be a monthly download of a file containing a list of account codes or a periodic review of product codes. After you create the link, you can run it by clicking the D-List menu and then click Update. This scans the source file and looks for any new items based on the unique part of the D-List, and rejects any duplicates. It then sorts the new items in a predefined manner or inserts them individually as desired.

Steps
1. Open the D-List. 2. From the D-List menu, click Options, Import link tab. 3. Choose the type of source file from the Import From drop down box (click one of the following links for more information). ASCII file (p. 127) (text, .csv, .txt or .prn files, and so on). D-List (p. 125) (to import items from another D-List). ODBC (databases) (p. 129). Cognos package (p. 130) D-Cube (p. 128) (to import items from another D-Cube). Finance. No Import Link (to turn off the default setting by selecting).

4. Choose the import mode (p. 132), from the drop down box. If using Update mode the Remove Obsolete check box will become active. Select it to keep specified items. 5. Choose the position and sort order of the items from the Location/Sort Order drop box (p. 143). 6. Use the Subtotal drop box if you wish to insert the new items into existing subtotals in the D-List (p. 145). Whenever you import items to a D-List, or create a new D-List by importing from an eligible source, you will be offered the option to turn this import into an import link for the D-List. If you answer Yes to this question, it will overwrite any existing import link.

Run an Import Link into a D-List

Running an import link into a D-List will update the D-List with the new items by reading straight from the source file. If the import link is run to a closed D-List by means of a macro, then any

124 Analyst

Chapter 7: D-Lists changes will be saved automatically. Depending on the settings in the import link, the existing items could be renamed or deleted altogether, so use caution when running these links.

Steps
1. With the D-List active, from the D-List menu click Update. 2. Position the new items. Note: Skip this step unless the position is set to Select in the Where box. In the Selection window, select items in the Items Available list and choose the insertion point by clicking the Items Included list. Click Move >> to insert the selected items below the chosen insertion point. If duplicates exist that are solely based on the unique part of the D-List item, the program rejects the duplicate items. 3. Check any formulas. 4. From the File menu, click Save.

Paste Items from a Spreadsheet, Database, or Other Text Source

You can copy text into a D-List from another program by pasting it in using the Microsoft Windows clipboard.

Steps
1. In the source program, select the items you want to copy. 2. Copy the items to the Windows clipboard by pressing Ctrl+C. 3. Return to Analyst. Create a new D-List, or for an existing active D-List, from the D-List menu, click Add Items and then click Input. 4. In the Enter item names text box, paste items from the clipboard by pressing Ctrl+V. 5. From the File menu, click Save. 6. Name the D-List if a new D-List was created. 7. Click OK.

Import D-List Items from Another D-List

You can create a new D-List or insert items into an existing D-List by importing items from another D-List. Optionally you may also import formulas, formats and calculation options (such as weighted averages). The source data for the D-list item must be a single page of the D-Cube. It can consist of several columns of D-List formatted data and may also include the rows D-List.

User Guide 125

Chapter 7: D-Lists The procedure below assumes that you are importing D-List items into a new D-List. If, however, you are importing D-List items into an existing D-List, you must open the D-List, and then from the D-List menu, click Add Items, and then start at step 2.

Steps
1. From the File menu, click New, D-List. 2. Click Import and then click Import From Another D-List. 3. Select the D-List containing the source data. 4. Make a selection from the D-List items. Ctrl-click to select non-adjacent items. A blank selection will import all the items. 5. Click OK. 6. Look at the Import of D-List Items dialog box: Click Import Mode (p. 132), and select either Append or Update. Click Subtotals and specify a subtotal, or choose either <None> or <Allocate>. Click Where to select where you would like the items from the ASCII file placed in the D-List. Click Select Attribute to choose from the following: Skip, Item name, Parent, Parent 1 through Parent 8. 7. In the import option box make selections to determine the items to be imported. If you wish to import formulas and format attributes, you must check the relevant boxes. 8. Click OK. 9. Choose whether you want to turn this import into an import link for the D-List. If you click Yes, it will overwrite any existing import link. 10. Click OK to import the items, and then save and close the D-List.

Import D-List Items from Unmapped ASCII Files

To import from D-List items from an unmapped ASCII file, you must specify the delimiter used in the ASCII file.

Steps
1. From the File menu, click New, D-List. - or From the File menu, click Open, D-List, and select the appropriate D-List. 2. From the D-List menu click Add Items. 3. Click Import and then select Import Unmapped ASCII. 126 Analyst

Chapter 7: D-Lists 4. Browse for the correct file and then click Open. 5. Look at the Apply Structure dialog box and do the following: Select Use Delimiter and then specify comma, semicolon, colon, tab, or space as the delimiter 6. Look at the Import of D-List Items dialog box: Click Import Mode (p. 132), and select either Append or Update. Click Subtotals, and either specify a subtotal or choose either <None> or <Allocate>. Click Where to select where you would like the items from the ASCII file placed in the new D-List. Click Select Attribute to choose from the following: Skip, Item name, Parent, Parent 1 through Parent 8 or Calc Texts. 7. Click OK. 8. Choose whether you want to turn this import into an import link for the D-List. If you click Yes, it will overwrite any existing import link. 9. From the File menu, click Save. 10. Name the D-List if necessary. Note: The D-List name is case sensitive and must be unique. You can type up to 31 characters including spaces. 11. Click OK.

Import D-List Data from Mapped ASCII Files

Before you import D-List data from mapped ASCII files, you must first create the file map from which the data is imported.

Steps
1. From the File menu, click New, D-List. - or From the File menu, click Open, D-List, and select the appropriate D-List. 2. From the D-List menu click Add Items. 3. Click Import and select Import Mapped ASCII File. 4. Select a file map, and then click Open. 5. From the Import of D-List Items dialog box: Click Import Mode(p. 132), and select either Append or Update. Click Subtotals and specify a subtotal, or choose either <None> or <Allocate>.

User Guide 127

Chapter 7: D-Lists Click Where to select where you would like the items from the ASCII file placed in the new D-List. Click Select Attribute to choose from the following: Skip, Item name, Parent, Parent 1 through Parent 8, or Calc Texts. 6. Click OK. 7. Choose whether you want to turn this import into an import link for the D-List. If you click Yes, it will overwrite any existing import link. 8. From the File menu, click Save. 9. Name the D-List if necessary. 10. Click OK.

Import D-List Items from a D-Cube

You can create a new D-List or insert items into an existing D-List by importing items from a D-Cube. The source data for the D-List item must be a single page of the D-Cube. It can consist of several columns of D-List formatted data and may also include the rows D-List. The procedure below assumes that you are importing D-List items into a new D-List. If, however, you are importing D-List items into an existing D-List, you must open the D-List, and then from the D-List menu, click Add Items, and then start at step 2.

Steps
1. From the File menu, click New, D-List. 2. Click Import, and then select Import D-Cube Data. 3. Select the D-Cube containing the source data. 4. Make a selection from the D-Cube dimensions which consists of one page only and columns of formatted data. 5. Click OK. 6. From the Import of D-List Items dialog box: Click Import Mode (p. 132), and select either Append or Update. Click Subtotals, and either specify a subtotal or choose either <None> or <Allocate>. Click Where to select where you would like the items from the ASCII file placed in the new D-List. Click Select Attribute to choose from the following: Skip, Item name, Parent, Parent 1 through Parent 8 or Calc Texts.

128 Analyst

Chapter 7: D-Lists 7. Click OK. 8. Choose whether you want to turn this import into an import link for the D-List. If you click Yes, it will overwrite any existing import link. 9. Click OK to import the items, and then save and close the D-List.

Import D-List Items Using ODBC

You can create a new D-List or insert items into an existing D-List by importing items from a database or spreadsheet using an ODBC link. Before importing D-List items using an ODBC link, you must check that the ODBC driver has been correctly installed and that a data source has been set up for the spreadsheet or database you want to access. The procedure below assumes that you are importing D-List items into a new D-List. If, however, you are importing D-List items into an existing D-List, you must open the D-List, and then from the D-List menu, select Add Items, and then start at step 2.

Steps
1. From the File menu, click New, D-List. 2. Click Import, and then select Import from ODBC (SQL database). 3. Select an ODBC source, then click Connect. If required, you may need to log on with your ID and password. 4. Select the table and column that contain the items to import. You can click Fetch to preview the column. 5. Optional: Click Create SQL to create a SQL statement. Experienced SQL users can edit this statement or type a SQL statement directly into the text box. For example, to combine two columns into a single D-List item, type a SQL expression such as Select ProductID & ProductName from Products. 6. Click OK. 7. From the Import of D-List Items dialog box: Click Import Mode(p. 132), and select either Append or Update. Click Subtotals and specify a subtotal, or choose either <None> or <Allocate>. Click Where to select where you would like the items from the ASCII file placed in the new D-List. Click Select Attribute to choose from the following: Skip, Item name, Parent, Parent 1 through Parent 8, or Calc Texts. 8. Click OK.

User Guide 129

Chapter 7: D-Lists 9. Choose whether you want to turn this import into an import link for the D-List. If you click Yes, it will overwrite any existing import link. 10. Click OK to import the items, and then save and close the D-List.

Import D-List Items From a Cognos Package

You can create and run an import link into a D-List using a Cognos package as a source. After you create the link, you can run it by clicking the D-List menu and then click Update. This scans the source file and looks for any new items based on the unique part of the D-List, and rejects any duplicates. It then sorts the new items in a predefined manner or inserts them individually as desired.

Steps
1. Open the D-List. 2. From the D-List menu, click Options, Import Link tab. 3. Choose Cognos Package as the type of source file from the Import From drop down box. 4. In the Cognos Package section, you can click the ...button to select a Cognos package, or if you already have a Cognos package, you can switch to a different Cognos package and then select new Query Items. You can display a preview of the selected Query Items. Select the Display preview of selected query items check box to preview the Query Items. The Preview option only works with Query Items that have not been selected, and helps you select the correct Query Items.

5. Choose the import mode from the drop down box. If using Update mode, the Remove Obsolete check box will become active. Select it to keep specified items. 6. Choose the position and sort order of the items from the Location/Sort Order drop box. 7. Use the Subtotal drop box if you wish to insert the new items into existing subtotals in the D-List. 8. Select an attribute from the drop box. 9. Optional: If you are editing an existing Import link, click Connect. 10. Click OK. Whenever you import items to a D-List, or create a new D-List by importing from an eligible source, you will be offered the option to turn this import into an import link for the D-List. If you answer Yes to this question, it will overwrite any existing import link.

130 Analyst

Chapter 7: D-Lists

Export a D-List as an e-List

You can export a D-List in a format that the Contributor Administration Console can use to import into an e-List.

Steps
1. Open the D-List. 2. From the D-List menu, click Export as E.List. 3. Choose a location to save the exported D-List. 4. Enter a name for which to save the exported D-List. 5. Click Save to export the D-List as an e.List. Note: The order of the items in the e.List is determined by the order of the items in the D-List item calculations, not the order that the items appear in the D-List itself. A macro called @ExportToEList is also available to automate this export process.

Implement Changes
If you make a change to a D-List, it is necessary to implement the change.

Step
From the D-List menu, click Implement Implement lets you see the effect of your proposed changes in any open D-Cube which uses the D-List, and still revert to the saved version in the event of an error. To revert to the saved version, from the D-List menu click Reset. - or From the File menu, click Save.

Delete items from a D-List

Deleting items from D-Lists should be done carefully because all data relating to each D-List item is deleted with it. If the D-List is used in several D-Cubes, it is deleted from all D-Cubes, not just the one you are working on. If the D-list is used as a dimension in a D-Link, you should check carefully to ensure that the D-link will still function correctly without the item you are deleting. In multi user systems, we recommend you check to see who is using a particular D-List before deleting items from it.

Steps
1. Choose whether to delete items while in a D-List or D-Cube: If from a D-List, from the D-List menu, click Delete Items. If from a D-Cube, right-click your mouse on the D-List dimension, and click Delete Items. User Guide 131

Chapter 7: D-Lists 2. In the Selection dialog box, select the items you want to delete and then click Move >>. 3. Click OK. This message appears: X items to be deleted. Do you want to proceed? 4. Click Yes to delete or No to cancel. 5. Save the D-List or D-Cube to make the deletion permanent. Tip: A shortcut for deleting an individual D-List item is to click the row numbers in the D-List attribute screen and then press Delete.

Import Mode
When importing items into a D-List, the import mode determines how to insert new items which may contain a unique code into a D-List.

Unique Names
A D-list item name may consist of a unique code, a specified number of characters in length followed by a description (p. 141). If this is the case and the source data for the import contains an item with the same unique code as an existing D-List item, but a different description, then Append mode will leave the existing description unchanged. Update mode will change the item to use the description in the source data.

Example - Import Modes

This example illustrates the different behaviors of the four basic import modes Append Update Update+Remove Obsolete Update+Remove Obsolete but Keep

In the example shown, the original D-List prior to copying contains 4 items: A, B and C adding to a subtotal X total. No special formatting or calculation options exist on the original D-List, but the unique code portion has been defined as being the first character.
A B C X total = A+B+C

132 Analyst

Chapter 7: D-Lists The source D-List has 4 items: A product, D product, E product adding into X total. In each case all the items from the source D-List are copied together with their formulas and formats. The sort order is set to 'hierarchical' and no subtotal is chosen. In each case, the numeric formats and the formulas are copied across.
A product D product E product X total= A product+ D product+ E product

When copying from ASCII, ODBC or D-Cube data, the source file has two columns consisting of an item and its parent.
A product X total D product X total E product X total

Copying from Another D-List

When copying from another D-list, formulas are replaced in Update mode, left unchanged in Append mode. The descriptions are always modified in Update mode, but not in Append mode. In Update mode, you are allowed to remove obsolete items, but can prevent the deletion of selected items by pressing the Keep button. Note: The behavior of Keep does indeed keep item B in the D-List, but does not keep it in the formula X total. Even if you had included X total in the selection of items to Keep, its formula and format would still have been updated.

Target D-List Source Target D-List After Copying D-List: Copy before copying all items Append Update Update+Remove Obsolete Update+Remove Obsolete, but KeepB
A B C A Product D product E product A B C D product E product A product D product E product A product D product E product A product

X total =A + B X total = A +C product + D product + E product

X total =A + B X total = A +C product + D product + E product

X total = A product + X total = A product + D product + E product D product + E product

User Guide 133

Chapter 7: D-Lists

Target D-List Source Target D-List After Copying D-List: Copy before copying all items Append Update Update+Remove Obsolete Update+Remove Obsolete, but KeepB
D product E product B C B

Importing from ASCII files, ODBC sources, Cognos Packages, or D-Cube data
When importing from ASCII files, D-Cube data, ODBC sources, or Cognos packages using macros, simple subtotals are combined in both Update and Append mode. The descriptions are always modified in Update mode, but not in Append mode. In Update mode, you are allowed to remove obsolete items, but can prevent the deletion of selected items by pressing the Keep button.

Target D-List

Source File

Target D-List After Running Macros

Item Name Parent

Append

Update

Update+Remove Update+Remove Obsolete Obsolete,but Keep B

A B C X total = A+B+C

A Product D Product E Product

X Total X Total X Total

A B C D Product

A Product B C

A Product D Product E Product

D Product E product A Product X total= A product+ D product+ E product B

D Product X total= A product+ D product+ E product E Product X total = A+B+C+D product+ E product

E Product X total = A+B+C+D product+ E product

134 Analyst

Chapter 7: D-Lists

Where Drop-Box
When moving or importing items, the Where box lets you define where new items are positioned. Top puts all the new items at the top of the list. Bottom puts all the new items at the bottom of the list. Select lets you position items individually using the selection screen. You must move each new item to the Items Included list by clicking Move >>. You can reposition the items with the arrow buttons on the right of the selection screen. The Reset button lets you start over if an error occurs. Alphabetical sorts the entire list alphabetically and numerically, including reordering existing items. Hierarchical automatically creates subtotals, totals, and grand totals according to the position of each item in a hierarchy defined in the source data.

Subtotals Drop-Box
The Subtotals drop-box lets you modify subtotal formulas to incorporate new items. The subtotal option is available only when the formulas consist of simple additions of items in the D-List. <None> does not change subtotal formulas. <Allocate> lets you set up an allocation table to allocate items into one or more subtotals. This is particularly recommended for long lists with multiple formula hierarchies.

A final option is for you to select a single subtotal from the list. This incorporates all new items into that subtotal formula.

Maintaining Hierarchies
You can create source files for hierarchical D-Lists in ASCII files, databases with ODBC connections or D-Cubes. In all import screens there is a sorting option called hierarchical. This groups newly imported D-List items with their subtotals.

Simple Hierarchies
You do not have to have each level of the hierarchy in a separate column. By defining the first column Item name and the second column Parent in the example below, cities add up into countries, add up into continents You can go straight from Item name to Parent 2 without an intermediate Parent being assigned.

User Guide 135

Chapter 7: D-Lists

Item Name
Ottawa Montreal Canada New York Rio de Janeiro Brazil

Parent
Canada Canada Americas Americas Brazil Americas

Create Hierarchies from an ASCII (Text) File

You can import items into a D-List and automatically create subtotals, totals, and grand totals according to the position of each item in a hierarchy defined in an ASCII or text file. Each item should be listed as a subtotal or total that each item belongs to. Using this method, you can create D-Lists as well as set up a D-List import link (p. 124) for ongoing maintenance.

Steps
1. From the File menu, click New, D-List. 2. Click Import and then select Import from ASCII-files. Note: If the ASCII file is a fixed-width text file, you need to set up a file map to define where each column of data starts and ends. In that case, you select Import from Mapped ASCII files. 3. Select the name of the ASCII file or file map to import. Note: If importing from a delimited ASCII file, select Use Delimiter and then select Comma as the delimiter. 4. Select the column containing the lowest level of the hierarchy and select Item name in the Select Attribute box. Then select the column containing the subtotals and select Parent in the Select Attribute box. At the next highest level of the hierarchy, select Parent 2 (grandparent). 5. Repeat until all columns have been assigned to a level. 6. Set the import mode (p. 132). 7. Click OK to import the items and then save and close the D-List.

Multiple Independent Hierarchies

Multiple independent hierarchies can be set up provided the relationship is defined by labeling one column containing the Item name and a second column containing the Parent. Hierarchies do not have to be branches of the same tree; they can be totally independent of each other.

136 Analyst

Chapter 7: D-Lists For example, the ASCII file shown below would have the first column defined as Item name and the second column defined as Parent. One hierarchy adds up the products by sub-totalling into countries. A second independent hierarchy adds up the products into soft and hard cheese types. The item names do not have to be in any specific order.

Camembert
St Paulin Edam Roquefort Camembert St Paulin Edam Roquefort

French cheeses
French cheeses Dutch cheeses French cheeses Soft cheeses Hard Cheeses Hard cheeses Soft cheeses

Multiple Column Source File

As an alternative to maintaining your data in two columns in your source file (which may be an Ascii file, a database or a D-Cube) you can use multiple columns defined as Item Name, Parent, Parent2 etc. There does not need to be an entry in every intermediate parent column.

Item name
Ottawa Montreal New York Rio de Janeiro

Parent
Canada Canada

Parent2
Americas Americas Americas

Brazil

Americas

Import D-Cube Data

You can update a D-List from the hierarchical data contained in a D-Cube. You can use the same data to maintain both the e-list in Contributor and the equivalent D-List in Analyst. By maintaining them from the same source, you ensure that they remain in step when the hierarchy changes.

User Guide 137

Chapter 7: D-Lists

Example: Create and Maintain Subset D-Lists from D-Cube Data

This example shows how to maintain a D-List hierarchy from data in a D-Cube in the format shown below. For example, suppose you have a long list of cities, but you want to create and maintain a subset list that contains just the European cities. After you have set it up, the D-List can be updated very simply from the source D-Cube data by selecting Update from the D-Cube menu.

Steps
1. Open the D-List containing the European cities. Because you cannot have a D-List with no items, you must have at least one item in this list to begin with. 2. Select Options from the D-List menu and click the Import Link tab. 3. In the Import From box, select D-Cube. 4. Select Update from the Import Mode list, then select the Remove Obsolete Items check box. 5. Select Hierarchical from the Location/Sort Option list. 6. Under D-Cube Data Import, click the button to browse for the D-Cube to use as a source table. You must specify a selection and orientation that opens it. 7. At the first selection screen click the slice. Ensure Cities is set as rows, and city hierarchy set as columns. 8. Do not make a specific selection on the Cities dimension (rows). This means that all rows will be looked at, including ones inserted at a future date. 9. Click the second tab and select the column or columns that contain the hierarchy. These are the columns containing the parent or grand-parent names. Within certain limitations, any number of hierarchy levels are allowed. 10. In the Select attribute box, click the first column (E.g. Cities) and select Item name. 11. In the Select attribute box, click the second column and select Parent. If there are more hierarchy levels, repeat this process by highlighting each column in turn and selecting Parent2, Parent3, and so on. 12. To ensure that only the items with a parent are imported, select the Suppress Zero Rows check box, and then click OK. 13. Select Update from the D-List menu. The new items will be imported into the D-List. 14. Save the D-List. The D-List can now be updated very simply at a later date by selecting Update from the D-Cube menu. This will take account of changes to the hierarchical data in the source D-Cube and update the target D-List appropriately.

138 Analyst

Chapter 7: D-Lists

Manage D-Lists
Management of D-Lists occurs in the library. You can copy, rename, print details, see where a D-List is used, and see what other objects the D-List is dependant on.

Rename/Move a D-List
Before you rename a D-List, ensure the list and all D-Cubes using it are closed.

Steps
1. From the File menu, click Library, D-Lists. 2. Select the D-List and move it to the bottom with the down arrow button. 3. Click the Rename/Move button. 4. Type the new name. Note: You also can move D-Lists to a different library. To move the D-List, select a new library from the Target Library list. 5. Click OK.

<< Move
Click << Move to deselect highlighted items in the D-List Selection dialog box. It removes selected items from the Items Included list and moves them to the Items Available list.

Delete a D-List
You can delete D-Lists only if they are not used by another object, and they are not active or open.

Steps
1. From the File menu, click Library, D-Lists. 2. Select the D-List and move it to the bottom with the down arrow button. 3. Click the Delete button. 4. If you are not allowed to delete a D-List, you can check to see its usage by clicking the Show objects that the selected object(s) is used by (p. 140) button. You must delete the D-Cubes and other objects that use the D-List, or amend them to use a different D-List, before you can delete the D-List.

User Guide 139

Chapter 7: D-Lists

Search for a D-List

You can search for specific D-Lists within a Library by using a filter, or by typing the D-List name.

Steps
1. From the File menu, click Library, D-Lists. 2. From the drop-down box, select a library. 3. Click the Filter button to limit the choice of D-Lists shown. For example, Filter = P* shows anything beginning with P. Alternatively, you can search for a specific name by clicking the Binoculars button and typing the name to search for.

Show a D-List's Dependants

Occasionally, a D-List is dependent on other objects. For example, a D-List might use a saved format named Format-A to set the numeric format of one of its items. Format-A would be a dependant of the D-List.

Steps
1. From the File menu, click Library, D-Lists. 2. Select the D-List. 3. Click the Show objects that the selected object(s) is using button.

Show a D-List's Precedents

A D-List can be used in several D-Cubes. It also can be used in D-Links or even be used in another D-List. The objects (such as D-Cubes or D-Links) that use a D-List are known as its precedents. For example, a D-Cube might be a precedent of a D-List.

Steps
1. From the File menu, click Library, D-Lists. 2. Select the desired D-List. 3. Click the Show objects that the selected object(s) is used by button to see where the D-List is used.

140 Analyst

Chapter 7: D-Lists

Upgrade Pipe Symbols in D-List Item Names

D-lists created prior to Cognos Planning 7.3 with item names containing a pipe symbol (|) may not appears correctly in the D-list editor. Pipe symbols in D-lists created in older versions of Analyst will continue to split long column names.

Steps
1. From the File menu, click Library, D-Lists. 2. Select the D-List and move it to the bottom with the down arrow button. 3. Right-click the D-List and select Upgrade Pipe symbols in D-List item names.

Unique Names
A D-List item may consist of a unique code and name, which makes it easier to match items, especially for data imported from other sources.

Edit Unique Names

Steps
1. Open a D-List. 2. From the D-List menu, click Options, Unique names tab. 3. To remove the column breaks defining unique names, right-click both the starting and ending positions. 4. To redefine the starting and ending positions, drag an existing column break. Note: The Reset button resets the settings to the last saved version.

Define the Unique Part of a D-List Item

When importing D-List items into a D-List on a regular basis, it often is useful to define which part of a D-List item contains a unique code. This lets you test whether it is a genuinely new item or only a slightly different spelling of the description. The unique portion of the D-List item is case sensitive and takes into account leading and trailing spaces. After you set the unique part of a D-List item you cannot type, paste, or import any duplicate items using a D-Link beginning with that code. For example, a typical product code consists of a code followed by a description such as P03 Camping Gear (P03 is the code, Camping Gear is the description). If the following month you are updating the list of product codes and an item named P03 Camping Gear displays, the program recognizes the code (P03). If you are updating the D-List, you can use Append mode to leave the existing description intact, or use Update mode to use the new spelling, thereby preventing duplication of the same product.

User Guide 141

Chapter 7: D-Lists

Total and Subtotal

Be cautious of items beginning with the word, Total. When the unique part of the D-List is defined as the first four or five characters, the word Total is treated the same way as any other code consisting of five characters. If more than one item starts with the word Total the program indicates that this is a duplicate code and prompts you to delete duplicates. If you receive a message telling you that if you continue, items will be removed and asks if you want to continue, click No. You can then go back and rename the totals by assigning unique codes to them.

Steps
1. Open the appropriate D-List. 2. From the D-List menu, click Options, Unique names tab. 3. Look at the Unique Names page of the D-List Options dialog box. To mark the start of the unique part, click once in the text box. To mark the end, click again. To remove line breaks, right-click the appropriate break. - or In the Unique Range box, specify the unique range by entering beginning and ending numbers. 4. Usually, the unique part consists of a code, but it could be the entire description. If this is the case, click Select All, OK. 5. From the File menu, click Save.

Manually Reorder D-List Items

You can choose the order for items in a D-List by manually reordering the D-List. Before reordering, you can search for items based on a character match. Click Search and then type the name of the item to select. Wildcard characters are allowed. Use a question mark (?) for a single character and an asterisk (*) for a series of characters. To use the Search feature, follow these guidelines: Wildcard characters of * and ? are allowed in the filter. Use a question mark (?) to represent any single character. Use the multiplication symbol (*) to represent any series of characters. Click And to show items that satisfy both criteria. Click Or to show items that satisfy either criteria. If Match Case is selected, it will further refine the filter to match on capitals or small letters.

142 Analyst

Chapter 7: D-Lists The Name box can be set using the equal symbol (=), or the not equal symbol (<>). Use the equal symbol (=) to show items that meet the criterion. Use the not equal symbol (<>) to exclude items that meet the criteria.

Steps
1. Choose whether you are reordering from a D-List or from a D-Cube: If reordering from a D-List, from the D-List menu, select Reorder, Manual If reordering from a D-Cube, right-click a D-List dimension and choose Re-Order Items from the list.

2. In the Reorder Itemsdialog box, in the Items Included area, select the items you want to reorder and use the arrow keys to move the selected item up or down to move to the top or bottom. Tip: Non adjacent items can be grouped together: Select the items to group. While holding CTRL, click to select non adjacent items. Group all selected items together at the top or bottom using the arrow keys. You now can position the group of items using the arrow keys to move up or down. 3. When you finish reordering items, click OK. 4. From the File menu, click Save.

Sort D-List Items

You can sort D-List items based on one of the hierarchical sort orders available.

Steps
1. Open the D-List you want to sort. 2. From the D-List menu, select Reorder and choose the sort order you want. A summary of the sort orders is given below.

Sort Order
Normal Alphabetical Rev Alphabetical

Explanation
The sort order saved on the D-List. Alphanumeric abc123. Reverse alphanumeric 321cba.

User Guide 143

Chapter 7: D-Lists

Sort Order
Totals After/Calc

Explanation
A hierarchical sort, with subtotals below their children, sorted within each subtotal according to the order found in the calculation formulas. The grand total is at the bottom. In the event of multiple overlapping hierarchies, items outside the main hierarchy appear at the bottom, below the grand total of the main hierarchy. The main hierarchy is chosen as the hierarchy below the first grand total in the list. A hierarchical sort, starting with the grand total and then each subtotal appearing above its children. In the event of multiple overlapping hierarchies, items outside the main hierarchy appear at the bottom, below the grand total of the main hierarchy. Same as Totals Below, but sort alphabetically within each subtotal. Same as Totals Above, but sort alphabetically within each subtotal. Same as Totals Below, but sort reverse alphabetically within each subtotal. Same as Totals Above, but sort reverse alphabetically within each subtotal. Same as Totals Below, but uses the current D-list order within each subtotal. Same as Totals Above, but uses the current D-list order within each subtotal.

Totals Before/Calc

Totals After/A-Z Totals Before/A-Z Totals After/Z-A

Totals Before/Z-A

Totals After/None

Totals Before/None

View/Edit Summary Information on a D-List

Steps
1. Open the D-List. 2. From the File menu, click Summary Info. 3. On the general information page, you can add a description and write notes about the D-List. The DOS file name is available only to the system administrator. 4. To see what objects a D-List is using, click the Objects used tab. 5. To see D-Cubes and other objects that use a D-List, click the Usage page.

144 Analyst

Chapter 7: D-Lists 6. Click OK. 7. If you changed the description or the owner's note, save the D-List.

Add Sub Headings to Reports

You can insert subheadings in reports, but you see them in Manager only, not in Analyst. You can assign subheadings to one or more D-List items such that when the item is selected as a row label in a report, the subheading displays above it.

Steps
1. From the File menu, select Open and then click D-List. 2. From the D-List menu, click Options. 3. Click the Subheaders tab. 4. On the Subheaders page, click New and then type the header text in the Header Text box. 5. In the Alignment box, select the position of the text: Select Left for left-justified text. Select Center for centered text. Select Right for right-justified text.

6. In the Fonts section: Select Common font size for one font size and specify the size in the Font size list. Select Multiple font sizes for more than one font size; then click Edit font size to specify. Click OK.

7. In the Select items to include in current header box, click the items you want to go into the subheading from the Items Available list. Whenever any of these items display as row labels in a report, the subheading displays above the first item. 8. To move the selected items to the Items Included list, click Move>>. 9. Click OK to return to the D-List options screen. 10. Click OK to return to the D-List. 11. From the File menu, click Save. Note: In normal use, the subheader displays in reports but not in the D-Cube.

User Guide 145

Chapter 7: D-Lists

Set D-List Colors

It is sometimes useful to group families of D-Lists by assigning them the same color. This can serve as a memory jogger when working with D-Lists. Most D-Lists belong to one of these categories. In general, when you set up D-Links to copy data between D-Cubes, you should map the links between D-Lists of the same family.

Steps
1. Open a D-List. 2. Click the small color box to the top left of the D-List. 3. Select a color from the palette. The recommended color conventions are listed below:

Types of D-List
Account codes, Profit and Loss items, balance sheet, other calculation D-Lists

Color
White

Divisions, regions, geographical, cost centers, personnel names Green Products, brands, services Customers Days, months, weeks, quarters, years, time Versions, actual/budget/variance Purple Blue Cyan Yellow

146 Analyst

Chapter 8: D-Cubes
A D-Cube is a store of data within a model. It is multi-dimensional and contains rows, columns and any number of pages. The D-Lists form the dimensions of the cube. Unlike a spreadsheet, D-Cubes can be sliced so that any pair of dimensions can comprise the rows and columns while additional dimensions comprise the pages. Cognos Planning - Analyst can handle any number of dimensions, the only practical limitation being the memory in your PC, but typically a D-Cube will contain no more than 5 or 6 dimensions. A D-Cube must contain at least two D-Lists (similar to a flat spreadsheet). Alternatively, a D-Cube can have three D-Lists, in which case it resembles a three-dimensional worksheet consisting of several flat sheets stacked behind one another. A four or five-dimensional D-Cube can be considered the same as a cross between a three-dimensional spreadsheet and a set of query reports from a relational database. A typical four-dimensional D-Cube would contain the D-Lists: P&L, Divisions, Months, and Variance. Note: Format and Formula priority: When you create a D-Cube, it is best to choose the D-Lists in a set order to overcome any priority conflicts on formulas at a later date. In general, select D-Lists with the most calculations first, particularly when the calculations use IF...Then...ELSE logical operators. Select the aggregation D-Lists next (divisions, customers, products) then the timescale D-Lists. Finish with the versions D-List. The default formula priority is given to the later D-List in a D-Cube, whereas format priority is given to the earlier D-List.

Size Limitations
There is no software limitation on the number of cells contained in a D-Cube, but there is a hardware limitation that depends on the memory in a computer. To determine the size limitation, the number of cells is determined by multiplying the number of rows by the number of columns by the number of pages. To put it another way, the number of cells is the product of the number of D-List items contained in each D-List. Size = (no. items D-List 1)* (no. items D-List 2)* (no. items D-List 3)* . . . *(no. items D-List n). Size limitations vary greatly depending on the memory in your computer. D-Cubes containing long D-Lists of 500 items consume more memory than similarly sized D-Cubes containing D-Lists of 100 items. In general, size limitations become apparent in D-Cubes of four or more dimensions. If you have a three-dimensional D-Cube of 400 pages, adding another dimension of 20 items increases the memory usage twenty-fold. In other words, you increase the data held from 400 pages to 8000 pages. Adding a fifth dimension of 20 items increases the memory usage by twenty-fold again to 160,000 pages of data. In practice, this memory limitation is overcome by creating a series of well-populated D-Cubes of three or four dimensions rather than one sparsely populated D-Cube of five dimensions.

User Guide 147

Chapter 8: D-Cubes

Interrupt a Calculation
If a calculation is taking a long time, you can interrupt it. If you interrupt a calculation and you have Undo enabled, and the memory limits are set high enough to contain a full copy of the current D-Cube data in memory, the D-Cube editor will revert to the latest D-Cube version available in memory. If Undo is disabled when you interrupt a calculation, all current changes will be discarded and the D-Cube editor will revert to the last saved version.

Steps
1. To interrupt a calculation, click the red cross that appears in the main toolbar. This button appears only when it is safe to stop calculating. 2. To ensure the integrity of the data after interrupting a calculation when Undo is turned off, from the File menu, click Reset to return to the saved version.

Set or Clear Audit Trails

You may set an audit trail on a D-Cube that records the changes for each session. When viewing the audit trail of a cells in a D-Cube, it shows the changes for the last session in which the cell was affected. This is not necessarily the latest session, simply the last session in which the number was altered by over-typing, executing a calculation, breakback, D-Link, or other method. A session is defined as a period between saves.

Steps
1. Close all objects so that you start from a blank screen. 2. From the File menu, click Library, D-Cubes. 3. Click the D-Cubes you wish to monitor, and then click the down arrow. 4. Choose whether to set or clear the audit: To set the audit, right-click the D-Cubes, select Set Audit from the list, and then click OK. The status of the audit trail shows if the object is currently being audited. To clear the audit, right-click the D-Cube, select Clear Audit from the list and then click OK. This clears the slate entirely. It will discard all old audit information and prevent any future audit trail from being recorded until you choose Set Audit once more. 5. Close the D-Cube Library window. The program will now begin to record any changes to that object.

148 Analyst

Chapter 8: D-Cubes

View the Audit Trail of a Cell Within a D-Cube

To track changes on a more granular level, you can view the audit trail of a cell within a D-Cube.

Steps
1. Right-click a cell and select Audit. This will show the last session in which the cell was affected, not necessarily the latest session. 2. (Optional) You may browse backwards or forwards through the audit records using the Previous and Next buttons. Each record contains a list of actions between saves. 3. If you want to search for a text string within an audit record, from the File menu, click Find, and then type the text to search for. Clicking Find Next or pressing the F3 key jumps to the next occurrence of the text string within the current page. 4. If you want to print all records, from the File menu, click Print. Tip: To preview the records before printing them, from the File menu, click Print Preview.

Breakback
Breakback answers what-if problems by changing the value of any number of variables to make a formula equal to a value you specify. With breakback, you set a target for a formula, and the variables that make up that formula are changed according to the rules you specify. In the default breakback mode, totals are split pro rata according to the original values contained in the variables that make up the formula. Zeros remain at zero with one exception: if all the variables in a formula are initially zero, the total is split equally down the hierarchy.

Default Rules for Breakback

The fundamental default rule for breakback is that changes are allocated to variables in a formula in direct proportion to their initial values. In other words, the more you have initially the more you get. This is what the program does in the absence of any other rules you might set. Breakback will not work over built in functions or conditional formulas. In the event of a profile that contains mixtures of zeros and values, zeros remain at zero, the remainder being allocated pro rata across the other variables. For formulas with a mixture of additions and subtractions, the default rule increases the items that are additions and decreases the items that are subtractions by an amount proportional to the initial values. For example, suppose a D-List contains the formula: Consider a simple example containing straight addition of detail items. A simple D-List contains four detail items and a total. The formula for the total is:

User Guide 149

Chapter 8: D-Cubes Company Total = North + South + East + West Likewise, a Periods D-List is also used where Period 1 + Period 2 + . . . + Period 12 = FULL YEAR If you type data in the Company Total cell, breakback allocates the total pro rata according to an initial profile or weighting. This profile is determined by the value of the original items held in the detail items. For example, if you have a D-Cube with four divisions and a Company Total, with each division having the number 25 entered in it, the formula would allocate 25 percent of any changes in the total to each variable. If you then typed 1000 in the Company Total cell and pressed Enter, the four divisions would receive 25 percent of 1000, or 250 in each division. The profile need not be a percentage. For example, suppose each division were given an equal initial weighting of 1, then the result would split equally. If you type 1000 in the Company Total cell and press Enter, the result would still be 250 in each division. The same result could be achieved by giving an initial profile of all zeros. Before the breakback, everything reads zero; after entering a number in the Company Total cell, every item receives an equal allocation. Profit = Sales - Costs Initially Sales =60 Costs =40 So Profit = 20 The effect of increasing Profit from 20 to 30 increases Sales by six and decreases Costs by four in direct proportion to their initial values. The percentage change of each variable is determined by the relative size of the initial values. The effect of breakback on formulas containing multiplication is slightly more complex, but again, the pro rata rule applies. For example, if you have the formula: Sales = Units * Price If Sales are doubled, both Units and Price increase by the same percentage with respect to their original values. In this case, both will increase by the square root of 2. The effect of breakback on formulas containing division is similar to multiplication. The effect of an increase in the result is to increase the numerator by the same percentage as the denominator. Again, the pro rata rule applies. For example, if you have the formula: %Margin = (Margin / Sales) * 100 If % Margin is increased, Margin goes up by the same percentage as Sales go down. This keeps both variables in proportion to their original values. In fact, if % Margin doubles, then Margin goes up by a factor of the square root of two whereas Sales go down by a factor of the square root of two. In this last case, it would be better to hold or increase Sales before changing %Margin. The computer lacks the intelligence to know whether an increase in %Margin is best achieved by an increase in Sales, a reduction in Costs, or some mixture of the two.

150 Analyst

Chapter 8: D-Cubes

Breakback vs. the Solve Facilities in Spreadsheets

Breakback is similar to the Solve and Backsolve facilities in certain spreadsheets, but is far more powerful. Breakback can cope with any number of variables or any number of formula targets, it works across multiple dimensions and down multiple formula hierarchies, and the rules can be set in advance to decide which groups of variables to hold, to increase by a percentage, or to allow to float freely. If you enter a number in a total cell in a spreadsheet, it would come up with an error. By contrast, breakback begins automatically whenever you enter or copy a number into a formula cell.

Use Breakback to Set Global Targets

You can set targets across more than one dimension. In fact, you can enter a grand total in the bottom-right corner on the final page of a D-Cube, and the details are altered across all rows, columns, and pages.

Steps
1. Create or open the D-Cube to which you want to apply the breakback. 2. Set the profile. Generally, this is done by copying numbers in using a D-Link from the D-Cube where the profile is stored. 3. Enter the target in the total cell. Typically, this is done by copying numbers in using a D-Link from the D-Cube where the targets are stored. 4. Press Enter. The changes are split pro rata across the detail items using the profile as a guide.

Holds and Breakback

Applying a hold to a cell freezes its value against changes induced by breaking back a formula total. Breakback allocates any changes pro rata to the other variables making up the formula, leaving the held cell unchanged. Note: If any holds are placed on a subtotal within a D-Cube, then a breakback will automatically be performed. Only use a hold on a subtotal when necessary and when model design has been geared for holding subtotals.

Why Data Can Change Despite Being Held

Held cells are not write-protected. This means that you can type data directly into a held cell. A held cell is held only against breakback, it is not write-protected against data entry, nor is it invulnerable to having data copied into it using a D-Link or pasted into it using the clipboard. A second example in which held cells can change is when you have not given the program enough freedom to calculate. In these cases, data is changed in held cells despite the hold being applied. User Guide 151

Chapter 8: D-Cubes For example, suppose you were to apply a hold to all four divisions and then hold the COMPANY TOTAL cell as well. The program has no freedom at all. Total Company = North + South + East + West Were you to enter data in the North division without releasing any of the other divisions or the total, the Total Company cell would change despite being held. In this case, you would not be giving freedom to the program. The forward calculation would have priority over the hold on the total. If you were to enter data in the Total Company cell without releasing any holds, the total would not change. In this case, the holds on the detail items have priority over the breakback calculation. As a rule of thumb, you should release all holds after each breakback operation.

Apply a Hold
To apply a hold, use D-Cube commands (p. 173).

Remove a Hold
To remove a hold, use the D-Cube Release command (p. 173).

Example - Hold Cells

Consider a simple example in which the North cell is held by typing hold directly into the cell. The formula for the total is: COMPANY TOTAL = North + South + East + West If you enter data in the Total Company cell, breakback allocates the total pro rata according to an initial profile or weighting. However, because the North division is held, it is invulnerable to breakback and stays unchanged. Thus, the changes are allocated pro rata across the South, East, and West divisions. If each cell initially contained the value 150 with a total of 600 and you then input 750 in the total cell, the increase of 150 will be shared between South, East and West equally. Since their initial values were equal, these will all increase to 200 while North will remain unchanged. You can type or copy data using operators at the same time you apply holds.

Using Breakback with Integer Arithmetic

Breakback can be set to integer mode or decimal mode. The default mode is set to decimal mode. However, you can set breakback to integer mode to round all numbers to integers whenever a breakback is triggered. Breakback in Integer mode can be used for applications where decimals are not logical, such as for headcount allocations or for rounding data for presentation purposes where a numeric format (or scaling) is insufficient.

Rounding Errors
Frequently you hear people claim that spreadsheets do not add correctly or that they contain rounding errors. This is a false allegation against spreadsheets that calculate to decimal places. The rounding errors that occur are so minimal as to be insignificant. However, it is not a trivial issue.

152 Analyst

Chapter 8: D-Cubes There are cases where numbers displayed to too few decimal places do not seem to add correctly (see the example that follows). Many hours have been spent using spreadsheet macros and rounding facilities trying to overcome this apparent anomaly. Analyst offers a solution to rounding anomalies by using breakback in integer mode. It should, however, be used with caution. If breakback is set to integer mode, the underlying detail numbers are changed. Although it is useful for presentation purposes, if you are sending the data to be consolidated elsewhere, it is generally better to use decimal mode so that the integrity of the data is maintained. The best solution is to produce reports with numbers formatted to one more decimal place than you have been asked for. In the example below, the sum is 1.333+1.333+1.333+1.000=4.999. However, if you set the decimal places to zero, the sum reads 1+1+1+1=5. This is upsetting because it seems to contravene the rules of mathematics. Actually, it is the mathematically correct solution because each number is rounded correctly, but try telling that to the boss! Breakback in decimal mode does not round the variables at all. Breakback in integer mode rounds the variables to the nearest integer. By rounding to integers, the sum 2+1+1+1=5 seems to add correctly even when no decimal places are shown. This can be useful for presentation purposes. However, a break back must be triggered for the rounding to occur.

Set Breakback to Integer or Decimal Mode

Breakback can be set to Integer mode or Decimal mode. The default mode is set to Decimal. However, you can set breakback to Integer mode to round all numbers to integers whenever a breakback is triggered. Breakback is similar to Solve and Backsolve facilities in certain spreadsheets, but is far more powerful. Do not use Integer mode for models used to create Contributor applications.

Steps
1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Options, Break-Back tab. 3. In the Options dialog box, do the following: If you do not require rounding when performing breakback, select Decimal. If you require rounding integers when performing breakback, select Integer.

4. Click OK.

Set a Target Using Breakback

Steps
1. Open the appropriate D-Cube.

User Guide 153

Chapter 8: D-Cubes 2. Enter the profile into the detail items. Typical profiles include a seasonality pattern, last year's actual results, a headcount, or other cost driver. Alternatively, enter the absolute numbers and tweak the total using breakback. 3. Optional: Apply any holds or other commands to the detail items. 4. Enter the data into the formula cells, and press Enter. Breakback is triggered automatically, allocating the changes pro rata to the variables that make up the formula.

Create D-Cubes
Use a D-Cube to enter data, perform multidimensional analysis, and calculate and collect data. You use D-Lists to define the dimensions for a D-Cube. The D-Lists are used to perform calculations, control labels, and format data entry. Add D-Lists in order of category number to maintain the correct calculation precedence.

Order of D-List Selection

When selecting D-Lists for a D-Cube, the order of selection does not affect which D-Lists end up as rows or columns. In practice, the default setting designates the longest D-List as row labels and any timescale D-List as column labels. This can be changed by transposing rows, columns, and pages as needed. However, the order is significant when it comes to the priority of calculations. At the intersection of two totals contained in separate D-Lists, the program must distinguish between using a formula in only one D-List. For straight addition or subtraction, it is irrelevant which formula is used because both calculations produce the same result. For multiplication or division, the order should not matter provided the percentages, ratios, or per unit items have been set as weighted averages. However, for logical operators (IF THEN ELSE), the priority is significant. When there are two formulas of equal priority, precedence goes to the D-List chosen last during the creation of the D-Cube. Consequently, any D-List using a logical formula should be selected first to give it a lower priority. This ensures that the whole is equal to the sum of its parts when aggregating other dimensions. For formulas, the priority goes to the later D-List unless overridden by priority settings on individual D-List formulas. These can be set to High, Medium, or Low. The default priority is Medium. As a general rule, choose D-Lists in the following order: Calculation D-Lists such as P&L, Balance sheet D-Lists. Aggregation D-Lists such as products, customers, divisions, cost-centers, regions, or subsidiaries. Time D-Lists such as months, quarters, or years. Only one timescale D-List can be chosen in each D-Cube.

154 Analyst

Chapter 8: D-Cubes Control D-Lists such as Actual/Budget/Variance.

If the order is wrong, you can change it later from the D-Cube menu, clicking Dimensions, Reorder.

Format Priority
Format priority in a D-Cube cell is determined by the order in which you include D-Lists in the D-Cube. The first D-List in the D-Cube takes precedence.

Example - Prioritizing Formats

You have P&L and Months D-Lists. The P&L D-List contains an item, Margin %, formatted with a percent sign and zero decimal places. The Months D-List contains an item, Current Month, formatted to two decimal places. Now you build a D-Cube using these two D-Lists inserting P&L before Months. The Margin %, Current Month cell is formatted with a percent sign and zero decimal places. This is because the P&L D-List is the first D-List in the D-Cube and has format priority. Note: All formats available for individual D-List items also can be set for a D-Cube as a whole. Formats set at the D-List item level override those set at the D-Cube level. In other words, a format set in a D-Cube is active only in D-List items that have no format attribute set.

Steps to create a D-Cube

1. From the File menu, click New, D-Cube. 2. In the Create New D-Cube dialog box, select the D-Lists to include in the D-Cube. Then move each D-List to the Objects Selected box by clicking the down arrow button. You can preview any of the D-Lists you are using to build the D-Cube. Simply right-click the D-List item name. You can re-order the dimensions by using the four arrow keys located above the Cancel button. As a rule of thumb, select the D-List with the most complex calculations first, to avoid problems with priorities of calculations. Note: Holding Ctrl and clicking D-Lists lets you select multiple nonadjacent items. 3. Click OK. 4. Name the D-Cube. Note: The D-Cube name is case sensitive and must be unique within a library. 5. Click OK. 6. In the Selection dialog box, click OK to open the entire D-Cube.

User Guide 155

Chapter 8: D-Cubes Clicking OK without selecting any items in the Selection dialog box selects all items. Alternatively, you can hide rows, columns, and pages at this stage by selecting only the items you want to display.

Steps to work with a limited selection

1. In the Items available list, select the items. 2. To move the items to the Items included list, click Move>>. 3. Repeat steps 1 through 3 with the other D-Lists (if necessary). 4. Click OK. The D-Cube appears, ready for data entry.

Open D-Cubes
Open a D-Cube to populate it with real-time data and begin your analysis. You can also open a D-Cube to add some test data before the actual budget data becomes available. You can re-orient the D-Cube so that certain dimensions fall on rows, columns, and pages.

Steps
1. From the File menu, click Open, D-Cube. 2. Click the name of the D-Cube you want to open. You can choose to open a D-Cube from another library if you have access rights. 3. If you choose to open a D-Cube from another library, click the library name to select it, and then select a D-Cube name. 4. Click OK. 5. Choose the mode in which you want to open the D-Cube. Click Full to open the entire D-Cube. The full cell count of the cube appears on this screen whichever option you choose. Click Saved Selection to open a previously saved selection of rows, columns, and pages. Then choose the name of the selection. Click Edit Selection to work on a limited number of rows, columns, and pages by selecting some and hiding others.

Note: If you select Edit Selection, you must move items you want to display to the Items Included list. Select the items you want and then click Move>> to move them to the Items Included List. Repeat for each D-List by clicking the D-List tabs. Leaving the selections blank selects everything. 6. Click OK. The selected rows, columns, and pages of the D-Cube appear. 156 Analyst

Chapter 8: D-Cubes 7. To close the cube, from the File menu, click Close. If you have not saved the D-Cube, you are prompted to do so. 8. To save before closing, click Yes. If two views of the same D-Cube are open, they can be closed one at a time. However, if the selections are different, you must save the changed data when prompted.

Open Multiple D-Cubes

It is possible to have several D-Cubes open at the same time.

Steps
1. From the File menu, click Library, D-Cubes. 2. Choose the correct library from the selection box in the top left hand corner 3. Highlight the cubes you wish to open and click the down arrow to move them to the lower pane 4. Choose a different library and select more cubes, if necessary. 5. From the central toolbar, click the Open objects icon. The active D-Cube is indicated by the blue highlighted title bar. 6. To activate another D-Cube, click anywhere in the window. Tip: To change from one window to another, press Ctrl+TAB.

Expand a Subtotal
Clicking Expand in the Selection dialog box selects the variables that go into a formula. It can be used to expand selected totals in either the Items Available list or the Items Included list. By repeatedly clicking Expand, further levels of the formula hierarchy can be highlighted. It can be used to expand more than one subtotal at the same time. Note: Clicking Expand applies the selection, but does not specify whether the selected items are to be included or not. After clicking Expand the selected items must be moved to the Items included list or the Items available list by clicking Move>>.

View Multiple Slices of a D-Cube in Separate Windows

You can change the perspective of a D-Cube by slicing and dicing the cube. D-Cubes can be sliced so that any pair of dimensions can comprise the rows and columns while additional dimensions comprise the pages.

User Guide 157

Chapter 8: D-Cubes Exchanging the position of D-Lists does not affect the order of the D-List in a D-Cube in terms of calculation order. It changes only the way the D-Cube is displayed.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Selections, New Slice. A new window appears containing another view of the same D-Cube. This can be sliced independently of the first view. The active slice is indicated by the blue highlight at the top of each window. 3. If you want to change the position of the windows, from the Window menu, click Cascade, Tile Horizontally, or Tile Vertically. 4. If you want to move between windows, from the Window menu, select the relevant window. To change from one window to another, press Ctrl+Tab. 5. To close a window, make the slice you want to close into the active window, and then from the File menu, click Close.

D-Cube Data Allocations

A D-Cube data allocation uses a D-Cube as a source for allocating items from a source dimension to a target dimension within a D-Link. If you specify a selection and slice that a D-Link cannot interpret as an allocation table, you will see the "Nothing to Transfer" message when the D-Link is run. You will have to return to the D-Link and specify the selection and slice from the allocation D-Cube again. If the D-Link can interpret your selection and slice as an allocation table, but the allocation table does not match the source and/or target items, you will see a 'No items match' message. A D-Cube used in this way will contain columns of D-List formatted data. The D-Cube can have multiple columns and multiple pages but the slice used as the data allocation must consist of a single page and a single column. The column must be D-list formatted. Thus different slices of the same cube may be used in the same or many different D-Links.

Example
If you have a D-Cube that has a D-Cube data allocation that allocates cost centers to both divisions and managers as follows:

Divisions
Cost Center 1 Cost Center 2 USA Germany

Manager
Manager 2 Manager 3

158 Analyst

Chapter 8: D-Cubes

Divisions
Cost Center 3 Cost Center 4 Cost Center 5 Cost Center 6 Cost Center 7 Cost Center 8 Cost Center 9 Cost Center 10 Cost Center Total France U.K. USA Germany France U.K. USA Germany

Manager
Manager 1 Manager 1 Manager 2 Manager 4 Manager 4 Manager 1 Manager 3 Manager 3

Cost Center 1 is located in USA and is managed by Manager 2. Similarly, Cost Center 2 is located in Germany and is managed by Manager 3. In this manner, the remaining cost centers are each located in various countries (USA, Germany, England, or France) and managed by various managers (Manager 1 through Manager 4). Note that one manager can be responsible for multiple cost centers in different cities.

Enter Data
When you enter data in a D-Cube, the color of the data indicates whether the data is saved, a working copy, or not calculated. If you make a mistake while entering data, you can reset a cell to its original or saved value. By default, the Undo facility is not turned on. For large D-Cubes, turning on the Undo can affect performance, so the Reset command can be used as a viable alternative.

Enter Data into Individual Cells of a D-Cube

Data can be entered in the D-Cube by typing directly into an individual cell. After typing a number, press Enter to calculate or ESC to cancel. Column widths expand automatically to accommodate the largest number on the page you are working on. As a rule, type numbers in the blue data entry cells. Numbers that have just been entered display in green, but do not calculate until the enter key is pressed. After calculation takes place, the numbers change to red.

User Guide 159

Chapter 8: D-Cubes Entering numbers in the black formula cells triggers the breakback function. This is the equivalent to setting a target result and directing the program to break back pro rata with numbers that will meet that target. If you have changed a formula in a D-List, numbers are not recalculated in the D-Cube until you implement the changes.

Color Conventions for Data

Data that has been entered is not calculated until you press ENTER. This means you can make changes to the data without processing the changes by not pressing ENTER. Cognos Planning - Analyst uses color conventions to indicate the type of data in a cell. A summary of these conventions is given below.

Cell Type
Normal Protected Locked Held Protected and Held Protected and Locked Locked and Held

Background Color
White Yellow Light gray Light turquoise Light green Light gray Dark gray

Cell Type
Typed, but not calculated data changes Changed detail item Changed formula result Unchanged detail item Unchanged formula result

Text Color
Green Pink Red Blue Black

160 Analyst

Chapter 8: D-Cubes

View a Formula
You can view a formula to better understand complex calculations used in the D-CUbe.

Step
Move the cursor to a black or red formula cell and press F7. Alternatively, from theD-Cube menu, click Show Formulae. The formula currently being used appears first.

View the Origin of a Detail Cell

You can drill on D-Cube data that was imported by a D-Link, provided the D-Link was saved. Drilling shows the original source data regardless of its origin.

Step
To view the origin of a detail cell, click the cell, and then click the Drill Down button. Tip: Alternatively, you can press F9. If the cell is a target of a D-Link then the data which would be transferred to the cell by the D-Link opens in a separate window. If the cell is the target of multiple D-Links then multiple windows will open. The data in the cell may differ from the source data shown depending on which D-Link was the last to run and whether data has subsequently been entered into the cell. To track changes to a cell, use an audit trail (p. 148).

Edit D-Cubes
You can edit D-Cubes by copying data from within the cube or external sources outside the cube, suppressing zero rows, columns, or pages, and annotating cells,

Copy a Range on the Same Page

You can copy a range of cells from one area of the page to another.

Steps
1. Open a D-Cube. 2. Select the range you want to copy from the source D-Cube. 3. From the Edit menu, click Copy. 4. Select the range to which you want to paste the data.

User Guide 161

Chapter 8: D-Cubes You can paste single columns into selected multiple columns. If the range of the target area is smaller than the range copied to the clipboard, only the earlier rows and columns are pasted in. 5. From the Edit menu, click Paste. The data is pasted in the cells and displays in green to indicate that the numbers have not yet been entered. 6. To calculate, click the page once and press Enter.

Copy Ranges in a D-Cube from Page to Page

The recommended method for copying data from page to page is to use an internal D-Link (p. 215). This allows changes of a more global nature rather than copying to and from the clipboard. It also permits a single page to be copied to multiple pages in a single operation, rather than using multiple copy and paste commands. In fact, it allows any range, adjacent to or not, to be copied to any other range.

Steps
1. Open the D-Cube. 2. From the D-Cube menu, click D-Links, Internal. 3. Click the yellow arrow connecting the relevant D-Lists. The default color of the arrow is yellow to indicate that items are matched using a match descriptions pairing (p. 233). 4. Select Change to allocate. This changes the yellow arrow to a green and red arrow, indicating that items are matched using a local allocation table pairing. An allocation table allows you to specify any correspondence of source and target items. You can mix one-to-one, many-to-one, and one-to-many allocations in one allocation table. 5. Select the items to copy. 6. On the Source side, click the D-List item you want to copy. 7. On the Target side, click the relevant D-List item. The allocation table in the center shows how the source and target ranges correspond. Note: If you make a mistake, select the line in the allocation table and press Delete (this action deletes a single line of the table). 8. Repeat with the other items. 9. To copy the range by running the internal D-Link, click the D-Link menu, and then click Execute. Unlike the copy and paste functions, a D-Link calculates in addition to copying, so that numbers display in red to show that they have changed. 162 Analyst

Chapter 8: D-Cubes 10. From the File menu, click Save.

Copy Data Using Operators

Certain copy commands are allowed to help you copy data on the current page. These abbreviations are typed directly into the cell.

Copy the Underlying Value Contained in a Cell Using the Copy Commands
The copy commands operate on numbers on the same level in the calculation hierarchy. A copy command entered in a detail item copies the item to all subsequent details, skipping any formula items. A copy command entered on a formula copies the formula to all other formulas. For example, 4000> entered in quarter1 skips the months, but copies the number 4000 to all subsequent quarters. Then breakback decides how the quarterly value of 4000 is split by month. Copy commands are terminated by any other copy command in their path or by the colon break (:). The colon (:) acts as a break that copies up to, but does not include, the current cell contents. The following copy commands are available:

command
greater than sign (>) less than symbol (<) pipe symbol (|) 'power of' symbol (^)

description
Copies data to the right along the same row. Copies data to the left along the same row. Copies data down the same column. Copies data up the same column.

Enter a Number and Copy it Across and Down Cells at the Same Time
The copy commands can be combined to facilitate data entry. For example, using the greater than sign (>) followed by the pipe symbol (|) tells the program to copy the number across and down the cells. As with all data entry, copy commands are not processed until you press Enter.

Steps
1. In the desired cell, type the sign for the operation required. 2. Press Enter.

Copy from a Spreadsheet to Analyst

You can copy data from a spreadsheet using the Windows clipboard or using a D-Link. For more information about D-Links, see "D-Links" (p. 215). User Guide 163

Chapter 8: D-Cubes We recommend the D-Link method because it provides an audit trail back to the source data, provided the D-Link has been saved and does not rely on the relative position of cells. Instead, it effectively gives each spreadsheet cell a range name that can be directed to the correct cell in the D-Cube. For any large spreadsheet, we recommend a D-Link as the optimal choice. If the spreadsheet uses a series of sections, you can use the Follow On facility in the D-Link to direct the data to the correct page. If it is set up as an ODBC source then a 100-page spreadsheet could be copied in one step rather than 100 separate copy-and-paste steps. To be comprehensive, the copy-and-paste method is shown here, although it cannot be stressed enough that the D-Link method is better in almost every case. If you use the copy and paste method, the position of rows and columns must be identical in both the source spreadsheet and the target D-Cube. This is possible if the D-Lists have been created from the spreadsheet itself. It can be a quick method of copying data between a spreadsheet and a D-Cube; however, ensure that blank rows and columns are eliminated and that the areas to copy from and to match exactly. Note: To avoid the possibility of triggering an unwanted breakback in analyst, you should paste data into detail cells of a D-Cube only.

Steps Using the Copy and Paste Method

1. In the spreadsheet, select the range you want to copy. Exclude row and column labels; select the data only. 2. From the Edit menu, click Copy. 3. Open Analyst and select the range of cells in the D-Cube to which you will paste. 4. From the Edit menu, click Paste to paste the data. It is pasted according to cell position, not according to the range names of the cells.

Insert Lines to Separate Totals from Detail Items

It is possible to insert lines that separate the detail items in a D-List from the totals. Lines can be placed before or after totals. The color, thickness, and style also can be set.

Steps
1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Options, Lines tab. 3. Look at the Lines page and do the following: Click the list to select where to insert lines (separating columns or lines separating rows). Click Before to insert lines before totals. Click After to insert lines after totals. Note: The lines are inserted automatically to separate the formula items from detail items, regardless of the slice of the D-Cube. 164 Analyst

Chapter 8: D-Cubes Click Color and select a color from the list. Enter a line thickness in the Thickness box. Select a line style from the Style box.

4. Click OK. Note: Although the facility to insert blank lines is not really featured in Analyst, there is a slightly ambiguous method of creating them. Insert D-List items containing just the underscore character ( _ ). Then apply a numeric format (p. 112) to the underscore item and select blank if zero (p. 193). As D-List names have to be unique, the number of underscore characters must vary if more than one blank line is inserted.

Edit Undo and Redo

To recover from most errors, press ESCAPE. However, even if you have pressed Enter you can still go back one step at a time using the Undo function. The program lets you go back several steps. The Redo command takes you forward one step at a time. Note: The number of undo and redo steps available are indicated in the menu bar at the bottom of the screen. For example U2 means two steps of undo available. R2 means two steps of redo are available. The Reset button resets the settings you are working on back to the saved version.

Suppress Zero Rows, Columns, or Pages

You may hide blank rows and columns that contain all zeros.

Steps
1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Options. 3. Click the Zeros tab. 4. To hide blank rows, columns or pages, select Suppress Zero Rows, Suppress Zero Columns or Suppress Zero Pages, or any combination of these three options. This will suppress zeros based on the slice. If the Suppress Zero Rows box is selected, zero rows will be hidden no matter which D-List happens to constitute the rows at a given time. Suppress Zero Pages is used to suppress blank pages from being printed or exported to text or ASCII files. 5. Click OK. 6. Save the D-Cube.

User Guide 165

Chapter 8: D-Cubes

Reveal All Zero Suppressed Rows and Columns

Depending on the type of analysis you want to perform, you may want to see all data each time you open a D-Cube, regardless of whether the rows or columns contain zeros.

Steps
1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Options. 3. Click the Zeros tab. 4. Clear the appropriate Suppress Zero Rows, Columns, and Pages check boxes. 5. Ctrl+Click the highlighted items to clear the selection in the Suppress Zero Items for D-List check box. 6. Click OK. 7. Save the D-Cube. Note: Reset Structure resets the zero suppression settings back to the last saved version of the D-Cube.

Change the Column Width or Row Label Width

By default, Analyst sets the width of a column to nine characters. The default for the row labels is five characters.

Steps
1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Options. 3. Click the Widths tab. 4. Set the column width in number of characters. This includes the digits, comma delimiters, decimal point, currency and other suffixes and prefixes. The default is a minimum of nine characters. 5. Click Minimum to use the widths setting as the minimum column width, but allow for expansion as needed to fit the largest number. 6. Click Show Column Labels to show the column labels in full. This will automatically widen each column to fit the label. The default is to truncate the column labels according to the width setting. The Show Column Labels option is not available if the width is set to Exact. 7. Click Exact to use the width setting as an exact column width.

166 Analyst

Chapter 8: D-Cubes This sets all columns to the same width. This looks neat but will not allow for expansion of the columns to fit large numbers. If the width of the largest number exceeds the Exact width setting, then a series of ####### symbols will appear and you must increase the width setting. 8. Set the row Width in number of characters. This includes the digits, comma delimiters, decimal point, currency and other suffixes and prefixes. The default is a minimum of 5 characters for the row labels. 9. Click Minimum to use the row widths setting as the minimum number of characters, but allow for expansion as needed to fit the longest label. 10. Click Exact to use the width setting as an exact row width. 11. Click OK. The width settings may be saved with the D-Cube.

Annotate a Cell
You can attach a note to any cell regardless of its format. This is particularly useful for providing additional information regarding a particular cell.

Steps
1. Open or create a D-Cube. 2. Click the appropriate cell, then from the D-Cube menu, click Annotations, Add/Edit. 3. Type in the annotation. A red dot in the top right corner of the cell indicates that a comment or annotation is attached.

Edit or View Cell Annotations

You can view annotations for a cell or range of cells to better monitor and track data in the D-Cube. You can also edit existing annotations to ensure that they reflect the correct information for a cell or range of cells within a D-Cube.

Steps
1. To view or edit a single cell annotation, right-click the required D-Cube cell and select Edit Annotation or Show Annotation. The name of the person who last edited the annotation is displayed together with the date and time it was altered. By clicking the D-Cube menu, pointing to Annotations, and then selecting Show, you will see the annotation in a single cell. 2. To view all cell annotations, from the D-Cube menu, click Annotations, Browse All.

User Guide 167

Chapter 8: D-Cubes

Remove Cell Annotations

You can choose to remove annotations that are no longer relevant from a single cell or a range of cells.

Steps
1. To remove annotations for single cell, right-click a cell and select Delete Annotation. 2. To remove annotations for a range of cells, highlight a range of cells, then from the D-Cube menu, click Annotations, Delete.

Specify Annotate Print Options

You set the orientation, printer, column width, and font in the Print D-Cube dialog box. In addition, you can select to print cell annotations.

Steps
1. Open a D-Cube and make a selection. 2. From the File menu, click Print or Print Preview. In the Annotations area, select or clear Print at the bottom of the page.

Edit Data
You can edit the data in a D-Cube by Typing directly in a cell Applying D-Cube commands (p. 173) Running D-Links

If a cube is open when the data changes are made, they will not be saved until you save the D-Cube. Changes made to a closed D-Cube are saved automatically. You can alter the appearance of data within a D-Cube by applying formats To individual D-List items (p. 117). Globally to the whole D-Cube (p. 191).

Edit Data on the Current Page of a D-Cube

You can edit the data on the current page of a D-Cube only when the D-Cube is open to that page.

Steps to type a new number

1. Click the cell with the number you want to edit. 2. Type the new data.

168 Analyst

Chapter 8: D-Cubes 3. Press Enter.

Steps to edit a number

1. Double-click the cell with the number you want to edit. 2. Type the changes. 3. Press Enter to calculate, ESC to cancel.

Steps to edit a green number (typed in but not entered)

1. Click the cell with the number you want to edit. 2. Press F2. 3. To return a green number to the original entered number, press ESC.

Edit the Data in Individual Cells Using Operators

The following abbreviations can be entered directly into cells. They operate on the underlying number held in a cell. These can be used in conjunction with the copy commands. Reset 10k 10m add100 or +100 sub50 or -50 mul1.1 or *1.1 div1.1 or /1.1 1000> Zero per10 or %10 inc10 dec10 Resets the value to the saved version (can be shortened to res). Enters the value 10,000. The suffix k (or K) denotes thousands. Enters the value 1,000,000. The suffix m (or M) denotes millions. Adds 100 to the value (can be shortened to a100). Subtracts 50 from the value (can be shortened to s50). Multiplies by 1.1 (can be shortened to m1.1). Divides by 1.1. Sets the value to 1000 and copies it across to the right. Sets the value to zero. Takes 10 percent of the value. Increases the value by 10 percent. Decreases the value by 10 percent.

User Guide 169

Chapter 8: D-Cubes

pow2 Hold Rel Lock grow10linear

Raises the value to the power of two. Holds the value to protect it from breakback. Releases or removes the hold. Locks or write-protects the value (can be shortened to l). Grows the value by 10 percent per period linearly (works with time periods only). Grows the value by 10 percent per period compounded (works with time periods only).

grow10compound

Examples - Copy Commands

inc10> 0>| reset| Increases the entire row to the right by 10 percent. Sets everything to the right and below to zero. Resets an entire column below the current cell back to the saved numbers. Grows the value by 10 percent for each period linearly (that is, increases the value by 10 percent of the original value each period) .

gro10li

Tip: Typing the word reset over a cell resets the number back to the saved version. Also, typing reset>, reset<, reset>|, reset ^, reset| resets cells to the right, left, above, and below.

Select a Range of Cells in a D-Cube

Selecting a range is a way to highlight cells before applying commands or copying and pasting data to and from the clipboard. The selected range displays with a black background. You can select a range of cells in a D-Cube only when the D-Cube is open.

Steps to select a range of cells in a D-Cube using the mouse

1. In the D-Cube, click the cell at which you want to start the range.

170 Analyst

Chapter 8: D-Cubes 2. Drag to select the desired range.

Step to select columns or rows

In the D-Cube, click the desired column or row headings, respectively.

Step to select the entire page

Click the blank gray cell just above the first row label (the top-left corner of the window).

Steps to select using the keyboard

1. Shift+ left, right, up, down arrow keys selects a range of cells. 2. HOME, then Shift+End selects an entire row. 3. Ctrl+Home, then Shift+Ctrl+End selects an entire page.

Reset Data
You can reset an entire D-Cube if you have not saved the changes. If you reset data by typing reset in a cell, you can also combine this command with copy commands to reset an entire row or column.

Steps to reset a column or row back to the last saved version

1. Open a D-Cube. 2. Right-click the row or column label you want to reset back to the saved version. 3. Select Apply Commands. 4. In the Apply to range dialog box, select reset from the list. See "Edit Undo and Redo" (p. 165). 5. Click OK. The cells revert to the blue and black colors, indicating data that has not changed since the last save.

Step for individual cells

Type the word reset or res directly into the cell and press.

When you reset individual cells in this manner, you can use reset in conjunction with the operators used to copy data.

Steps for a range of cells on the current page

1. Select the range of cells you want to reset. 2. Right-click the highlighted range. 3. Select Reset.

User Guide 171

Chapter 8: D-Cubes The selected range of cells revert to the saved version of the data.

Steps for a range of cells globally

1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Commands. 3. In the Apply to subselection of dialog box, click reset. For information about commands, see "Edit Undo and Redo" (p. 165). 4. Click Select. 5. In the Choose subselection dialog box, click the items from the Items available list, and then click Move>> to move the items to the Items included list. 6. Click OK. You are returned to the Apply to subselection of dialog box. 7. In the Apply to subselection of dialog box, click OK. The selected range of cells revert to the saved version of the data.

Recover from Errors

Most minor mistakes such as incorrect menu selections can be overcome by pressing the Esc key or clicking Cancel. Errors in data entry range in a D-Cube from entering a single wrong number to major structural errors in the D-Cube. Analyst provides comprehensive facilities to help you retrace your steps to recover from errors, even if it has been a while since you last saved the data. In increasing levels of severity, a summary of the options for error recovery are listed below.

Command
ESC

Recovery Action
Clears a green number (numbers typed in but not yet entered). Retraces one step at a time. Goes forward one step at a time. Reverts the cell data to the last saved version. Resets a selected cell to the last saved version.

Undo Redo Reset Right-click a cell, and then click Reset Reset from Apply to subselection dialog box

Resets a D-Cube range to the last saved version.

172 Analyst

Chapter 8: D-Cubes

Command
Reset Structure

Recovery Action
Resets a D-List contained in the D-Cube without changing the data. Resets data to the last saved version; it does not reset the D-Lists. Closes the file without saving.

Reset from the File menu

Close

D-Cube Commands
Ranges of data can be operated on using commands from the Commands menu.

Commands Menu
The available commands are zero, set, add, subtract, multiply, divide, percent, increase, decrease, reset, hold, release, lock, unlock, protect, unprotect, power, random and round.

Menu Command

Syntax example

Meaning

Example (Effect on a cell containing 1000)

Zero Set Add

zero set99 add10

Set range to zero Set range to a value

0 99

Add amount to underlying 1010 values Subtract amount from underlying values 990

Subtract

subtract10

Multiply

multiply1.2

Multiply underlying values 1200 by an amount Divide underlying values by 500 an amount Take a percentage of underlying values 100

Divide

divide2

Percent

percent10

Increase

increase10

Increase underlying values 1100 by a percentage

User Guide 173

Chapter 8: D-Cubes

Menu Command

Syntax example

Meaning

Example (Effect on a cell containing 1000)

Decrease

decrease10

Decrease underlying values 900 by a percentage Reset range to last saved version Last saved version

Reset

reset

Hold

hold

Hold the range of cells Blue background against breakback (p. 151) Cells can still be changed by entering or copying data or by D-Links.

Release

release

Remove the holds on a range

Blue background removed

Lock

lock

Write protect the range. Grey background Cells can still be changed by breakback but not by entering data or by D-Links Removes the locks on a range Grey background removed

Unlock

unlock

Protect

protect

Protects the range against Yellow background data entry by manual typing. The cells can still be changed by breakback or by D-Links. Removes the protects on a Yellow background range removed Raises the underlying values 1000000 to the power specified Changes the underlying values to random integers between zero and the number specified Random integer between zero and 100

Unprotect

unprotect

Power

Power2

Random

random100

174 Analyst

Chapter 8: D-Cubes

Menu Command

Syntax example

Meaning

Example (Effect on a cell containing 1000)

Round

Round10

Rounds the underlying values to the nearest number specified If you round a total, a breakback will be triggered, altering the details so that they add up exactly to the rounded number.

1000

Apply Commands
You can apply commands to a range of cells on the same page or to a range of cells across multiple pages of a D-Cube.

Steps for a range of cells on the same page

1. Open the D-Cube 2. Highlight the range of cells to which the command is to be applied 3. Do one of the following: Right click anywhere within the highlighted area. Select Apply Commands. Choose your command, adding any numeric details needed. From the D-Cube menu, click Commands. Choose your command, adding any numeric details needed. Click Marked selection. 4. Click OK. 5. Save the D-Cube.

Steps for a range of cells over multiple D-Cube pages

1. Open the D-Cube 2. From the D-Cube menu, click Commands. 3. Choose your command, adding any numeric details needed. 4. Click Select All to apply the command to the whole D-Cube or click Select to choose the data to which the command is to be applied. 5. If necessary, make your selection from the D-Cube selection box. 6. Click OK. User Guide 175

Chapter 8: D-Cubes 7. Save the D-Cube

Edit a Range of Data on the Current Page

You can change data in a range to facilitate data entry.

Steps
1. Open a D-Cube. 2. Select the range of data you want to change. 3. Right-click the selected range. 4. Select Apply Commands. 5. In the Apply to Range dialog box, select a command from the list. This command operates on the selected range. Most of the commands require you to type a number after them. 6. To calculate, click OK. In this example, increase15 increases the values in the highlighted range by 15 percent. Increase and decrease apply a percentage change to the underlying values, not an absolute change. Most of the commands are intuitive: add100 adds 100 to the number in every cell in the range; set1000 sets the number 1000 in every cell in the highlighted range. The commands operate on the selected range only. Changed numbers display in red.

Delete the Data from an Entire D-Cube

Deleting data from an entire D-Cube is similar to setting a global range to zero. It erases all data in a D-Cube, having first made sure that there are no hidden or locked cells. Before deleting data from a D-Cube, check for the following: hidden rows, columns, or pages locked cells protected cells

Check for Hidden Rows, Columns, or Pages

Before deleting data, ensure that there are no hidden row, columns or pages that are excluded from the deletion.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Selections, Reselect. 3. To select everything in the current D-List, click Clear. 176 Analyst

Chapter 8: D-Cubes 4. Repeat steps 1 through 3 for the remaining D-Lists in the D-Cube. 5. Click OK.

Check for Locked Cells

Locked data is used to prevent data entry and data loading, but does not prevent breakback. Before proceeding with deleting data from a D-Cube, you must unlock any data that is locked. Locked data cannot be deleted.

Steps
1. From the D-Cube menu, click Commands. 2. In the Apply to subselection of dialog box, select unlock, and then click Select All. 3. Click OK.

Check for Protected Cells

Cells are protected to prevent data entry. The protection does not prevent loading or breakback. Before deleting data, check to see that all protected cells are reverted to unprotected. Protected cells will not be deleted.

Steps
1. From the D-Cube menu, click Commands. 2. In the Apply to subselection of dialog box, select unprotect, and then click Select All. 3. Click OK.

Delete Data From an Entire D-Cube

You can delete data from a D-Cube to load a new set of data for the analysis you require.

Steps
1. From the D-Cube menu, click Commands. 2. In the Apply to subselection of dialog box, select zero, and then click Select All. 3. Click OK. 4. Save the D-Cube.

Set a Column or Row to Zero

You can modify sections of a D-Cube by using D-Cube commands to zero a column or row.

Steps to set cells to zero locally

1. Open a D-Cube. User Guide 177

Chapter 8: D-Cubes 2. Right-click the row or column label you want to set to zero. 3. Select Apply Commands. For information on commands, see "Change Ranges of Data Using Menu Commands" (p. 178). 4. In the Apply to Range dialog box, select zero from the list. 5. Click OK. The entire selection is set to zero.

Steps to set cells to zero globally

1. Open a D-Cube. 2. From the D-Cube menu, click Commands. 3. In the Apply to subselection of dialog box, select the list, and then select zero. For information about commands, see "Change Ranges of Data Using Menu Commands" (p. 178). 4. To have the command operate on a selected range of the D-Cube only, click Marked selection. 5. To have the command operate on the entire D-Cube, click SelectAll. 6. Click Select, and then select the items you want the commands to operate on. 7. Click OK.

Change Ranges of Data Using Menu Commands

Use commands to speed data entry.

Steps to Access Menu Commands

1. Select a range of cells in a D-Cube. 2. Right-click the range of cells. 3. Select a command from the following list: Copy Paste Copies the highlighted range to the clipboard. Pastes the contents of the clipboard into the highlighted range.

Paste Holds, Locks, Protects, You can copy and paste Holds, Locks, and Protects patterns from or All one D-Cube to another D-Cube, or all three patterns at once. Protect You can write protect a cell against manual entry or copying from the clipboard.

178 Analyst

Chapter 8: D-Cubes

Unprotect Lock Unlock Hold Release Reset Apply Commands

Remove the write protect from a cell. Locks or write-protects a range against data entry. Removes the locks from a range of data. Holds a range of data against change through breakback. Removes the holds from a range of data. Resets the range back to the saved version. Operates on the highlighted range. The commands available are zero, set, add, subtract, multiply, divide, percent, increase, decrease, reset, hold, release, protect, unprotect, lock, unlock, power, random, and round.

Add Annotation

You can provide further explanation of data.

Step to Access the Subselection List of Commands

From the D-Cube menu, click Commands.

Menu Syntax Command Example

Zero Set Add Subtract zero set99 add10 subtract10

Meaning

Example (Effect on 1000)

Set range to zero Set range to a value Add amount to underlying values Subtract an amount from the underlying values

0 99 1010 990

Multiply Divide Percent Increase Decrease

multiply1.2 Multiply underlying values by an amount divide2 percent10 increase10 decrease10 Divide underlying values by an amount Take a percentage of underlying values Increase underlying values by a percentage Decrease underlying values by a percentage

1200 500 100 1100 900

User Guide 179

Chapter 8: D-Cubes

Menu Syntax Command Example

Reset Hold Release Lock Unlock Power reset hold release lock unlock power2

Meaning

Example (Effect on 1000)

Reset range to last saved version Hold range against breakback Remove the holds on a range of cells Write-protect the range Remove the locks on a range of cells Raise the underlying value to a specified power

Back to saved Blue background

Gray background

1000000

Locks, Protects, and Holds

Applying a hold protects a cell against breakback. Applying a lock protects a cell against data entry. Applying a protect protects a cell against manual entry or copying from the clipboard. The commands may be used in combination to ensure that a cell cannot be changed at all it must be locked and held. You can also right-click a D-Cube to copy and paste Holds, Locks, and Protects patterns from one D-Cube to another D-Cube.

Why Data Can Change Despite Being Locked or Protected

Data can change in cells despite the fact that the lock or protect has been applied. The reason for this is the two-way arithmetic function named breakback. Although locked or protected cells are write-protected against direct data entry, they are not calculation-protected against breakback. This means that if a formula result is changed, its constituent parts can also change even if the constituent items are locked. For example, suppose a D-List item, North, has been locked or protected and data is entered in the Total Company cell containing the formula, Total Company = North + South + East + West. The breakback function is activated so that the North and other divisions change to meet the target set in the Total Company cell. The North division is changed as a result of the calculation being reversed, despite being write-protected against direct data entry. In the same way, locked or protected total cells can change if one of the constituent items that make up the total has changed. To write-protect and to protect against breakback, cells must be both locked and held (dark grey background) or protected and held (green background) if amendment by D-Link is permitted. For more information about the hold command see "Holds and Breakback" (p. 151).

180 Analyst

Chapter 8: D-Cubes

Hold Data
Hold is a type of control used to prevent breakback in a cell or a range of cells, but does not prevent data entry. Any user with Write access can remove the hold control from a D-Cube.

Hold or Release Individual Cells

The Hold and Release commands can be used in conjunction with the operators used to copy data. You can place a hold on individual cells of a D-Cube without preventing breakback in other areas of the D-Cube.

Steps
1. Open a D-Cube. 2. Choose whether to hold or release a cell: To hold a cell, type the word hold or H directly into the cell and press Enter. The letter H suffices instead of hold. To release a cell from Hold, type the word rel directly into the held cell and press Enter.

Hold or Release a Range of Cells on the Current Page

You can apply a hold on a range of cells in a D-Cube without preventing breakback in other areas of the D-Cube. For example, you may be required to re-forecast for the year when the actual data for January and February becomes available. You can release the Hold for January and February and continue to retain the hold on the remaining months that contain forecasted data.

Steps
1. Select the range of cells you want to hold or release. 2. Right-click the selected range. 3. To apply a hold, select Hold. The held cells display with a light turquoise background. 4. To release the hold, click Release. The blue background disappears to indicate that the hold has been removed.

Hold a Global Range of Cells Steps

1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Commands. User Guide 181

Chapter 8: D-Cubes 3. From the list of commands, select hold. 4. Click Select. 5. In the Choose subselection dialog box, do the following: In the Items available list, select the desired items, and then click Move>> to move the selected items to the Items included list. Repeat for all D-Lists. Click OK.

6. In the Apply to subselection dialog box, click OK. The held cells display with a light turquoise background.

Release the Holds from an Entire D-Cube Steps

1. Open a D-Cube in Full mode. 2. From the D-Cube menu, click Release All Holds. The blue background disappears to indicate that the holds have been removed.

Lock or Protect the Formula Cells Only

It is possible to apply a lock or protect to only the formula cells by selecting formulas in the selection screen. This is a useful way to prevent accidental entry of data into formula cells, particularly when a large amount of manual entry is required.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Commands. 3. From the list of commands, select lock or protect. 4. Click Select. 5. In the Choose subselection dialog box, do the following: In the Show box, select Formula. In the Items available list, select the desired items, and then click Move>> to move the selected items to the Items included list. Note: As a rule, do not repeat the selection of formulas for other D-Lists or you will end up locking just a few grand totals, not the desired result at all. Instead, leave the other D-List selections blank to mean 'select all'. Click OK.

182 Analyst

Chapter 8: D-Cubes In the Apply to subselection dialog box, click OK.

Protect Data
You may write-protect a cell or range of cells to prevent data being from entered manually by applying the Protect command. Protect differs from Lock in that data can still be transferred into a protected cell via a D-Link but you may not type data into a protected cell. By contrast, locked cells prevent data entry entirely whether by typing or via a D-Link. Protected cells are still vulnerable to break back if entering data into a subtotal that has a protected cell as one of its components (The Hold command is used if you want to prevent a cell from being altered by a break back).

Three methods to protect cells:

Type the word protect over a cell (or pr for short). Highlight a cell (or range of cells), then right-click the cells, and select Protect. Click D-Cube, Commands, protect, Select. Highlight the items to protect, and then click Move. Repeat for each D-List, and then click OK twice to return to the D-Cube. The color conventions are shown below.

Command
Protect Lock Hold Protect and Hold Lock and Hold

Background Color
Yellow Light Gray Light Turquoise Dark Green/Blue Dark Gray

Lock Data
Lock is used to prevent data entry and data loading, but does not prevent breakback.

Use the Lock Command to Write-Protect

The Lock command write-protects a range of cells against data entry. This means that data can neither be typed in manually, nor can it be pasted from the clipboard or copied using a D-Link. A locked range of cells is given a gray background. A range of cells that is held and locked is given a yellow background. Three methods to apply the lock or write-protect command:

User Guide 183

Chapter 8: D-Cubes You can type the word lock directly into an individual cell. This can be used in conjunction with the operators used to copy data. You can select a range, right-click it, and then select Lock from the shortcut menu. You can apply the lock to a selection of items in each D-List by clicking Commands from the D-Cube menu, and then clicking lock. This commands method is the most common method because it applies a lock to a selection across the entire D-Cube rather than to individual cells or rows and columns. Note: Like any command, locks can be applied only to a subset of the current selection. Ensure you have reselected all the D-List items you need before applying the locks or unlocks. In particular, this applies to unlocking cells using the Select All command, which selects all items from the current selection, not necessarily the entire D-Cube.

Steps to Lock or Unlock an Entire D-Cube

1. Open a D-Cube. 2. From the D-Cube menu, click Commands. 3. Choose whether to lock or unlock a D-Cube: To lock a D-Cube, from the list of commands, select lock. To unlock a D-Cube, from the list of commands, select unlock.

4. Click Select. 5. In the Choose subselection dialog box, in the Items available list, select the desired items, and then click Move>> to move the selected items to the Items included list. 6. Repeat for all D-Lists. 7. Click OK. You are returned to the Apply to subselection dialog box. 8. In the Apply to subselection dialog box, click OK. Locked cells display with a gray background. If the D-Cube is unlocked, the gray background disappears to indicate that the lock has been removed.

Steps to Lock or Unlock a Range of Cells on the Current Page

1. Open a D-Cube. 2. Choose whether to lock or unlock a range of cells:, select the range of cells you want to lock. To lock a range of cells, right-click the selected range and select Lock. The locked cells display with a gray background to show that they are write-protected. To unlock a range of cells, highlight the range of cells, right-click the range, and then select Unlock .

184 Analyst

Chapter 8: D-Cubes The gray background disappears to indicate that the lock has been removed.

Lock Individual Cells

The Lock command can be used in conjunction with the operators available for copying.

Steps
1. Open a D-Cube. 2. Type the word lock directly into the cell and press Enter (the letter L suffices instead of lock).

Special Copy and Paste

You can copy and paste Holds, Locks, and Protects patterns from one D-Cube to another D-Cube, or all three patterns at once. You must select the exact same range of cells, or a cell in the target D-Cube so the pasted area has the same shape as the selected area.

Steps
1. Open a D-Cube. 2. Select the locked, held, or protected cell or range of cells you want to copy. 3. Right-click the selected cell or range of cells and select Copy. 4. Select the cell or range of cells where you want to copy the hold, lock or protect patterns. 5. Right-click the selected cell or range of cells and click Paste Holds, Paste Locks, Paste Protects, or Paste All.

Random Number D-Cube Command

A D-Cube command named Random generates a random number. For example, random1000 would fill all items in the selection with a random integer between 1 and 1000. The empty selection would behave in exactly the same manner as the set command.

Round Command
The Round command lets you round a selection of D-Cube cells to multiples of any number. If you type round1 into a cell, it rounds to integers. Typing round100 into a cell will round to the nearest 100. And typing round0.1 rounds to one decimal place, and so on. Any rounding factor is allowed, and will always round to the nearest multiple of the number that you entered. If you type round12, it will round to the nearest dozen. Note: Rounding should not be applied to D-List formatted or date cells as it will round the underlying ID or date value respectively.

User Guide 185

Chapter 8: D-Cubes Rounding is also available using right-click > Apply Commands, or by selecting Commands from the D-Cube menu. Rounding can be put in a macro using the @DCubeCommand macro. You can also round any selection of cells, including totals. If you round a total, a breakback will be triggered, altering the details so that they add up exactly to the rounded number. The Round command is available in Analyst, Manager and Analyst for Excel.

Export Data
You may export data from a D-Cube in a variety of formats. The data must be in a format that is readable by the program to which you are exporting.

D-Cube Export
You may export data from a D-Cube to a text file (ASCII file) or to the clipboard.

Steps to Export From a D-Cube

1. Open a D-Cube. 2. From the D-Cube menu, select Export. 3. From the Export dialog box, select the options necessary and click Export. The Export tab contains the following options: Format Separator between columns for export can be Comma, Tab, Semi-colon or Aligned Columns. Column Headings can be set to Normal, At Top, Above Each Page or None. Normal is the default behavior. In multi-column view, it puts column headings above each page, but does not show the D-List names for the row and page dimensions. In single column view, Normal means that no column headers are shown. At Top puts the D-List names and column headers at the top only. Above Each Page puts the dimension names and the column headers above each page of the export. In single-column exports, the concept of pages is meaningless, so it only puts the dimension names at the top. None does not put any column headings in at all. Mode can be set to Overwrite or Append. Overwrite will overwrite an existing file of the same name, though it will prompt you before you replace the old file. If you export to a file that already exists, Append adds the exported data onto the bottom of the existing file. Groups can be set to Single or Multiple Column. Single Column exports the data values in a single column. Multiple Column means the data values are exported as a series of columns. For multiple column exports, the items in the last dimension (marked Data) are used as the column headers. Dimension Order

186 Analyst

Chapter 8: D-Cubes If you click the Dimension Order button, the export column order is set according to the underlying order of D-Lists in a D-Cube. It is independent of the current orientation of rows, columns and pages in the view you happen to have open. You may control the dimension order manually using the arrow buttons at the bottom of the export screen. Tip: When exporting from Analyst to Cognos Planning - Contributor it is much easier if you export in Dimension Order. It means you can export directly without having to go into SQL Enterprise Manager at all. Because Dimension Order is the default, you can run the Actuals.dts directly without having to change the dimension mapping in Data Transformation Services. Plain Number Format Removes any numeric formatting for the purposes of export. It exports to as many decimal places as are needed, up to the limit stored on the computer. Negative numbers will be prefixed by a minus sign and there will be no thousand separator or percent signs currency symbols or other numeric formats that have been applied on the D-List or D-Cube. Plain Number Format uses the full-stop as the decimal separator unless Apply Regional Settings overrides this. Text, D-List or date formatting will remain as displayed in the D-Cube view. Apply Regional Settings Examines the regional options (set in Control Panel) and uses this format for the thousand and decimal separator. For example, in Germany the number one thousand two hundred and thirty four point five six gets exported as 1.234,56, whereas in the UK and the USA, the same number has 1,234.56 as the export format. Pipes As Spaces The pipe symbol is used to mark a line break for wrapping column headers in Analyst. If you check Pipes as Spaces, the pipe symbol ( | ) gets replaced by a space on export. This is useful for exporting from Analyst to Contributor. Text Qualifier You can specify the text qualifiers for exporting text strings. These can be set to none, double quote or single quote by selecting from the Text Qualifier drop-down box. The text qualifier will be put around D-List items, text, and D-List formatted data cells. The default setting is none. Plain number format relates to exporting numbers. It will not control whether or not text qualifiers appear. The format for plain number format uses the regional setting in Control panel for the decimal separator, to set the number of decimal places to the minimum number to ensure complete accuracy, to use no thousand separator, and to show negative numbers with a leading minus sign and have no prefixes or suffixes. In the DCubeExport macro, the new parameter is added: TextQualifier = single, double, or none. This is not case-sensitive.

User Guide 187

Chapter 8: D-Cubes Note: If you have old macros that relied on the plain number format putting double quotes around the text strings, you may need to change these by inserting the option TextQualifier=double in the macro, but usually this is not necessary. Special Cases for text qualifiers If the name contains the text qualifier, it will double up the offending character to avoid ambiguity. For example: If you are exporting a text field containing 1/2" Drill bits to a file that uses double quotes as the text qualifier, it gets exported as "1/2"" Drill bits". Again, if a single quote is used in the name and as a qualifier, it gets doubled up for the export. So 12' ladder exported to a file with a single quote as the text qualifier will repeat the single quote to appear as '12" ladder'. Important: The only way to export numbers with comma delimiters is to use plain number format, or select a file separator other than a comma. Leading blanks Leading blank spaces are not stripped out when you export text or D-List-formatted cells. The Header/Footer tab contains the following options: You may insert your own header/footer to appear at the very top/bottom of the export file. The Zeros tab contains the following options: Suppress Zero Pages You can suppress zero pages for the export, independently of the zero-suppression currently in force in the view you have open on the screen (set under D-Cube>Options). Zero suppression is dimension specific. To suppress zero rows, highlight the dimension labelled R. To suppress zero columns, highlight the dimension labelled C. To suppress zero pages on all dimensions, select Suppress Zero Pages. To suppress zeroes everywhere, highlight all the dimensions and select Suppress Zero Pages. Note: Zero suppression of page dimensions requires Suppress Zero Pages to be selected. It is not sufficient to just put the zero suppression on the dimension. For multiple-column exports, suppression of zero columns is disallowed. This is essential to preserve the correct number and sequence of columns exported in a multiple-column export. Note: There is a special case where a row of zeroes gets exported even though Zero Suppression is ON. This occurs if Zero Suppression is OFF for the page labels, but ON for the row dimension. In this case, a blank page containing just the first row will still get exported. Zero-suppression on export is independent of the zero suppression set on the D-Cube. The Show Det/Tot tab contains the following option:

188 Analyst

Chapter 8: D-Cubes If you have chosen an Empty selection on a dimension (empty selections meaning All Items in this context), you can hide details or totals for the purposes of export. This option is dimension specific. This allows you to cut down the data volumes for the export file.

Export from Analyst to Contributor

Two methods exist for getting data from Analyst to Contributor. In Analyst, run an Analyst to Contributor D-Link (p. 276). This creates prepared import data blocks in the Contributor application data store. You can monitor this in the Contributor Administration Console, Import, Prepared data blocks screen. When the data blocks have been written, in the Contributor Administration Console, run the Go to Production process. Export data from Analyst to Contributor by exporting to flat text (ASCII) files, then import these files using the Contributor Import facility.

Export Data to Contributor Steps

1. In Analyst, open the D-Cube you wish to export. From the D-Cube menu, choose Export. 2. In the Export dialog box, change the settings to the following: Select Separator=Tab, Column Headings=None, Mode=Overwrite. Then select Single Column, Dimension Order, Plain Number Format, and Text Qualifier=none. See "Text Qualifiers" (p. 737) for more information. 3. In the Contributor Administration Console, use the Import facility to import the data, and then run the Go to Production process. See the Contributor Administration Guide or online help for more information.

Format Prior to Export

The data must be in a form readable by the program to which you are exporting. Generally this involves removing currency symbols, removing brackets for negative numbers and percent signs (%). To preserve accuracy, the numbers may need to be formatted to several decimal places. The most readable format is a format that shows the numbers to several decimal places (for accuracy) , does not use a comma separator for thousands, and uses a minus prefix (-) for negative numbers (for example, -1234.56). Often it is useful to save such a numeric format ready for loading when exporting data. Frequently, formats have been applied to individual D-List items rather than the D-Cube. So even after the D-Cube format has been set to a standard format, there are still rows and columns that have an inappropriate local format. The format on the D-List items must be temporarily changed to a standard format for exporting, and then reset to the old format after the export has taken place.

User Guide 189

Chapter 8: D-Cubes

Export to a Spreadsheet
The procedure for exporting data to a spreadsheet via the clipboard is almost identical to exporting via an ASCII file. Then only difference is that the data recorded with the clipboard is temporary, while data recorded in an ASCII file is permanent. Before exporting data, it must first be formatted. See "Format Prior to Export" (p. 189).

Steps
1. Open a spreadsheet. 2. Export the data from your D-Cube (p. 186). If you exported to an ASCII file, click the File menu, and then click Open. The exact syntax for importing from an ASCII file will vary depending on the version and spreadsheet. Name the file as preferred. Remember to change the search to all files (*.*) or it may search for files of the type .xls, .wk4, and so on. If you exported to the Clipboard, click the Edit menu, and then click Paste.

The file will be imported into the spreadsheet. Formulas and formatting are neglected. Blank lines may need to be inserted to separate different sections.

AutoSum
If you highlight a range of cells in a D-Cube view, the sum of the numbers highlighted is automatically displayed in the status bar at the bottom of the screen. This sum stays there until a different range of numbers is highlighted. You can choose from a list of predefined functions that return a single summary value for a group of related values.

Steps
1. Open a D-Cube and from the D-Cube menu, select Options, and then click the AutoSum tab. 2. Choose the type of summary value you want to show: To show the calculated value that represents the sum of the selected data items, click Sum. To show the average value of the selected data items, click Average. To show the number of selected data items, click Count. To show the minimum value of the selected data items, click Minimum. To show the maximum value of the selected data items, click Maximum.

3. Click OK. The summary value of the highlighted cells is displayed in the status bar at the bottom of the screen.

190 Analyst

Chapter 8: D-Cubes

Find Text and Character Matches

You can search for text strings or character matches in most screens that display text. This includes free-text and D-List formatted text in D-Cubes, formulas (click a cell and press F7), and audit trails.

Steps
1. Open a D-Cube. 2. Move the cursor to the point where you want to start searching, then from the Edit menu, click Find. 3. Type in the exact text you want to find. 4. You can search down a column or across a row by clicking the check boxes in the Keep Fixed area. 5. Select the Match Case check box to match the case of the text you entered with the text displayed on screen. 6. To jump to the next occurrence of the text string, from the Edit menu, click Find Next.

Formats
A format sets the form in which data displays and can be in numeric, date/time, or D-List (text) format. Use numeric formats (p. 112) to set decimal places, insert thousand delimiters, set braces for negative numbers, and set prefixes and suffixes. Use date and time formats to display dates and times (such as 6-Jul-97) as data in a D-Cube. Use D-List formats (p. 117) to enter text, restricting the text to the codes and names displayed in a selected D-List. Free text formats (p. 118) are allowed. You can name and save formats for repeated use.

Local vs Global Formats

A local format (p. 109) involves formatting D-Lists only and applies to an individual row or column. Generally, you apply local formats in the Format Attribute screen. A global format involves formatting an entire D-Cube. Generally, you apply global formats by clicking Format on the D-Cube menu. Although the two screens are almost identical, a D-Cube format is a general format applied to all data in the entire D-Cube, not just a specific item from a D-List. However, a local format applied to an individual row or column has priority over the global D-Cube format.

Load or Remove a Global Format

Steps
1. From the D-Cube menu, click Format. 2. Choose whether to load or remove a global format: User Guide 191

Chapter 8: D-Cubes To load a global format, select a format from the Format Type box, and then click Load. Select an existing saved format, and click OK. To remove a global format, select None in the Format Type box.

3. Click OK to apply. Note: If formats still remain after removing the global D-Cube format, they are local formats applied to items in the D-List. These can be removed by opening the D-List and setting the format on each item to <None>.

Save a Global Format

Steps
1. Open an appropriate D-Cube. 2. From the D-Cube menu, click Format. 3. Click a format from the Format Type box. 4. Format the D-List as desired. 5. Click Save to save the format under the same name - or Click Save As to save the format under a different name. 6. Enter the name of the format. 7. Click OK. 8. Close the D-List.

Enter Prefixes and Suffixes

You can enter any specific prefix or suffix for both negative and positive numbers. The default option sets a negative prefix and suffix so that negative numbers are enclosed in brackets. If you specify no negative prefix - that is, delete the open bracket, negative numbers are automatically prefixed with a minus sign (-). It is recommended to limit the prefixes and suffixes to 10 characters. The prefix and suffix let you display numbers with currency and other symbols as well as displaying different formats for negative numbers. For example, to display everything in dollars with brackets for negative numbers and the k suffix to denote thousands, use the prefixes and suffixes as shown below. This displays the number -1234 as ($1,234k).

Steps
1. Open a D-List. 2. Click the Format cell of the item you want to set decimal places to. 192 Analyst

Chapter 8: D-Cubes 3. Select Numeric from the Attribute list. 4. In the Prefixes or Suffixes areas, make your changes.

Access Blank if Zero

Select the Blank if zero option if you want zero values in the item to display as blank cells in the D-Cube. The default option is to display zero values in the D-Cube as normal. Note: With the Blank if zero option set, any value rounding to zero in the D-Cube displays as a blank cell. For example, if no scaling factor is set and decimal places is set to 2, the number 0.001 displays as a blank cell (or displays as 0.00 if the Blank if zero option is not set).

Steps
1. Open a D-List. 2. Click the Format cell of the item you want to apply a numeric format. 3. Select Numeric in the Attribute list.

D-Cube Selections
Using selections from a D-Cube allows you to hide rows, columns, or pages. This enables you to work on a subset of the data contained in a D-Cube. Saved selections may be used to save a specific D-Cube orientation. When selecting D-Lists for a D-Cube, the order of selection does not affect which D-Lists end up as rows or columns. In practice, the default setting designates the longest D-List as row labels, and any timescale D-List, or if there is no timescale D-List, the second longest D-List as column labels. This can be changed by transposing rows, columns, and pages as needed. However within a saved selection you are allowed to choose the D-Lists used as rows and columns so that the cube opens with your specified orientation. When a selection of a D-Cube is open, D-Links will only update the current selection. Global commands to change the data applied across a D-Cube will only affect the D-List items in the current selection. Working with a selection provides processing advantages by limiting memory usage for very large D-Cubes. However, if you select totals, the underlying detail items also require memory. A D-Link that needs to copy data to or from a total will also require memory.

Expanded Selections
When you open a limited selection from a D-Cube, only the items you selected are displayed in the D-Cube dialog box. If you have included formula items in your selection, however, an expanded selection may be opened internally, so that the selected formula items can be recalculated and broken back properly. Selections can also be used to do the following: Hide rows, columns or pages.

User Guide 193

Chapter 8: D-Cubes Sort rows, columns, and pages. Apply commands to a selection of rows, columns, and pages. Select which items to paste into a formula. Select where to position new items. Select items to delete. Order a D-List. Select a range of rows, columns, and pages for print. Select a range of rows, columns, and pages to export to another program.

If a basic selection includes formula items, the expanded selection will also include all the items on which the selected formula items depend, either directly or indirectly. For example, consider a Months D-List that contains twelve detail items (Jan, Feb, Mar, and so on), four quarterly totals (Q1 = Jan + Feb + Mar, and so on), and one full year total (Full Year = Q1+Q2+Q3+Q4). If you open a D-Cube containing this D-List and select only Q1 from MONTHS, the expanded selection will include Q1 and Jan, Feb, and Mar. If you select only Full Year, the internal expanded selection will include all items in the MONTHS D-List. Remember that Analyst does not save calculated data in the D-Cube when it is closed; data in formula items is only recalculated when required. Selections can be saved for later use. For example, a retail chain might wish to group similar stores into a selection so that budgeting assumptions can be applied to the selection of stores without having to modify each page of data.

Creating a Selection
Various methods are available for creating a selection depending on whether the D-Cube is currently open.

Blank Selections
If the Items included list in the Selection dialog box is left completely blank, every item in the list will be selected. For saved selections, leaving the Items included list blank selects everything including future insertions of extra D-List items. There is a subtle distinction between blank selections and selecting every item: Blank selections from a D-List will select all items; so if items are added, the new items will be loaded with the saved selection. If, however, you save a selection that shows all the D-List items in the Items included list, new items will not be inserted when the saved selection is loaded.

194 Analyst

Chapter 8: D-Cubes

Reselecting after changing data

Beware of reselecting after you have changed data. If, as a result of the reselection, you hide detail items that contain altered data (colored red and pink), these changes will be lost and the cells will be reset to their saved values.

Steps if the D-Cube is Not Open

1. Click the Open D-Cube icon. 2. Click the radio button for Edit Selection and click OK. 3. Make your selection of D-List items (p. 195) to be included choosing from each D-List in the cube by clicking their tabs. 4. Click Slice to specify the D-Cube orientation. 5. Click Save to save the selection you have chosen. 6. Name your selection. 7. Click OK to the View Saved message. 8. Click OK to open your selection.

Steps if the D-Cube is Already Open

1. From the D-Cube menu choose Selections and Reselect. 2. Make your selection of D-List items (p. 195) to be included choosing from each D-List in the cube by clicking their tabs. 3. Click on Slice to specify the D-Cube orientation. 4. Click OK. 5. From the D-Cube menu choose Selections and Save current. 6. Give your selection a name. 7. Click OK.

Facilitate Selection Using the Selection Dialog Box

Within this box, items in any D-list are split into two lists: The Items included list shows items that have been selected. The Items available list shows items that are available, but have not been selected.

To facilitate selection especially from long D-lists, the following features are available:

User Guide 195

Chapter 8: D-Cubes

Show Box
The Show box allows you to reduce the number of items in the Items Available list. It does not select items in its own right, but merely determines the choice of Items Available. Select All to show the full D-List. Select Detail to show detail items only - for example, all D-List items apart from formulas. Select Formula to show only the formulas. Select Filter to show a selection based on a keyword, initial letter, or a text search criterion. It filters the text in the D-List items only, not the values contained in the cells. Wildcard characters are allowed in the filter. Click And to show items that satisfy both criteria. Click Or to show items that satisfy either criteria. If Match Case is selected, it will further refine the filter to match on capitals or small letters.

The Name box can be set using the equal symbol (=), or the not equal symbol (<>). Use the equal symbol (=) to show items that meet the criterion. Use the not equal symbol (<>) to exclude items that meet the criteria.

Order Box
The Order box sorts the Items available list to assist in selection. It does not sort the Items included list itself. The options are as follows: Normal - sorts by the normal D-List order. Alphabetic - sorts alpha-numerically. Rev Alphabetic - sorts reverse alpha-numeric. Totals - sorts by expanding the subtotals in the order they appear in the D-List.

Because the Order box only sorts the Items available list, any sorting using the Order box requires moving items from the Items available list to the Items included list by clicking Move >>. The Order by Totals feature is not available when reordering D-Lists permanently, but is available when reselecting from D-Cubes.

Steps
1. Select the appropriate items from the Items available list. Ctrl+Click to highlight non-adjacent items. Click All to select all items. Click Search to search for items (generally for long D-Lists). Click Expand to expand selected totals in either the Items Available list or the Items Included list.

196 Analyst

Chapter 8: D-Cubes Click Slice to specify the D-Lists top be used as rows and columns in the selection

2. Click Move >> to move the selected items to the Items Included list. A blank selection implies select all. 3. In the Items included list, click Sort to sort the items in the Items Included list. Use the Sort Arrows to sort the items. 4. Click Clear to clear the entire selection. 5. Click Reset to reset the selection back to its original format.

Saved Selections and D-Links

If working on a sub-selection from a D-Cube, data can not be entered or imported via D-Links into an unselected and unopened range of cells. Only an open range of cells will be updated when a D-Link is executed.

Save a Selection
Saved selections may be used to save a specific D-Cube orientation, including a selection of rows, columns and pages for later use. The selected items, sort order, and slice of the D-Cube are all saved in a named selection. When saving a selection, it is recommended that the selection name is similar to the D-Cube name.

Steps to Save the Current Selection of a D-Cube

1. From the D-Cube menu, click Selections, Save current. 2. Name the current selection.

Steps to Save a Selection in the Selection Screen

1. From the D-Cube menu, click Selections, Reselect. 2. Make your changes in the Selection dialog box (p. 195). 3. Click Save. 4. Specify the save. Click Save selection if you have just loaded and edited the selection. Click Save selection as to save the modified selection under a new name.

Note: The program only knows you are working on a saved selection when you have just loaded it, edited it, and have not left the selection screen. So if you have accessed the main screen in between loading and saving, the program will not remember the name of the original selection. Consequently it will ask you for a new name regardless of whether you choose the Save selection or Save selection as options.

User Guide 197

Chapter 8: D-Cubes 5. Name the selection. 6. Click OK when prompted. 7. To recall a saved selection, load (p. 198) the selection.

Load a Saved Selection

Steps
1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Selections, Reselect. 3. Click Load in the Selection dialog box (p. 195). 4. Select one of the previously saved selection names. 5. Click OK. 6. The saved selection is loaded. 7. Click OK in the Selection dialog box. 8. To open a different selection, repeat the process by returning to the selection screen and clicking Load once more.

Load a Saved Selection on Opening a D-Cube

Steps
1. From the File menu, click Open, D-Cube. 2. Select the D-Cube to open. 3. Select Saved Selection when prompted. Note: The D-Cube must have been saved as a selection before loading. For more information about saving a selections, see "Save a Selection" (p. 197). 4. Choose the saved selection. 5. Click OK. 6. The D-Cube will open with just the items from the saved selection.

198 Analyst

Chapter 8: D-Cubes

Edit the Selection on Opening a D-Cube

Steps
1. From the File menu, click Open, D-Cube. 2. Select the D-Cube to open, select Edit Selection when prompted, and then click OK. 3. In the Selection dialog box (p. 195) Items available list, click the appropriate D-List tab. 4. Make your selection from the Show box. Select All to show the full D-List. Select Detail to show detail items only, such as all D-List items apart from formulas. Select Filter to show a selection based on a keyword, initial letter or a text search criterion. It filters the text in the D-List items only, not the values contained in the cells. 5. Make your selection from the Order box. Normal - sorts by the normal D-List order. Alphabetic - sorts alpha-numerically. Rev Alphabetic - sorts reverse alpha-numeric. Subtotal - sorts by expanding the subtotals in the order they appear in the D-List.

Because Order sorts only the Items Available list, any sorting using Order must be followed by Move >> to move the ordered items to the Items Included list. The Order by Subtotal feature is not available when reordering D-Lists permanently, but is available when reselecting from D-Cubes. 6. Select the appropriate items from the Items Available list. Ctrl+Click to highlight non-adjacent items. Click All to select all items. Note: Clicking All applies the selection, but does not specify whether the selected items are to be included or not. After clicking All, the selected items must be moved to the Items Included list or the Items Available list using Move. Click Search to search for items (generally for long D-Lists). Wildcard characters are allowed in the filter. Use a question mark (?) to represent any single character. Use the multiplication symbol (*) to represent any series of characters. Click And to show items that satisfy both criteria. Click Or to show items that satisfy either criteria. If Match Case is selected, it will further refine the filter to match on capitals or small letters. The Name box can be set using the equal symbol (=), or the not equal symbol (<>). Use the equal symbol (=) to show items that meet the criterion. Use the not equal symbol (<>) to exclude items that meet the criteria.

User Guide 199

Chapter 8: D-Cubes Click Expand (p. 157) to expand selected totals in either the Items available list or the Items included list. Click Slice to choose which D-List you would like to appear as rows and which you would like to have as columns. The Slice command is similar to the pivot command in certain spreadsheets, but much more extensive in its scope. The orientation of a D-Cube can be changed so that pages become columns, rows become pages or columns become rows. In fact, any orientation is possible. For four and five-dimensional D-Cubes, simply choose any two of the D-Lists to make up the rows and columns, leaving the other D-Lists to become pages. The slice can be saved with a named selection. 7. Click Move >> to move the selected items to the Items included list. A blank selection (p. 194) implies select all. 8. In the Items included list, click Sort to sort the items in the Items included list. Use the Sort Arrows to sort the items. The D-List order is the default order that rows, columns, and pages will appear in the absence of any numerical or other sort. Generally, the D-List order is the order that items were typed into the D-List when it was first created. Normal is for the order held in the D-List. Alphabetical for alpha-numeric. Rev Alphabetical is for reverse alpha-numeric. 9. Click Clear to clear the entire selection. This is made permanent only when you save an object. 10. Click Reset to reset the selection back to its original format. Clicking Reset reverts to the last saved version you were working with, not necessarily the saved selection. This can be applied to data in individual cells, to a highlighted range on the current page, to a global range across a D-Cube, or to an entire D-Cube. The reset command can be applied to the data, the D-Cube structure, or both. Resetting the structure of a D-Cube allows you to cancel changes that have been implemented (not saved) in the underlying D-Lists as well as any D-Cube sorts and formats. For more information about resetting the structure of a D-Cube, see "Reset the Structure of a D-Cube" (p. 206). Note: Reset does not reset a slice. 11. Ensure any items you want to show are in the Items included list. 12. Click OK.

Edit a Saved Selection

Steps
1. In the Selection dialog box (p. 195), look at the Items available list.

200 Analyst

Chapter 8: D-Cubes 2. Click the appropriate D-List tab. 3. Make your selection from the Show box. Select All to show the full D-List. Select Detail to show detail items only, such as all D-List items apart from formulas. Select Filter to show a selection based on a keyword, initial letter or a text search criterion. It filters the text in the D-List items only, not the values contained in the cells. 4. Make your selection from the Order box. Normal - sorts by the normal D-List order. Alphabetic - sorts alpha-numerically. Rev Alphabetic - sorts reverse alpha-numeric. Subtotal - sorts by expanding the subtotals in the order they appear in the D-List.

Because Order sorts only the Items Available list, any sorting using Order must be followed by Move >> to move the ordered items to the Items Included list. The Order by Subtotal feature is not available when reordering D-Lists permanently, but is available when reselecting from D-Cubes. 5. Select the appropriate items from the Items available list. Ctrl+Click to highlight non-adjacent items. 6. Click All to select all items. Note: Clicking All applies the selection, but does not specify whether the selected items are to be included or not. After clicking All, the selected items must be moved to the Items Included list or the Items Available list using Move. 7. Click Search to search for items (generally for long D-Lists). Wildcard characters are allowed in the filter. Use a question mark (?) to represent any single character. Use the multiplication symbol (*) to represent any series of characters. Click And to show items that satisfy both criteria. Click Or to show items that satisfy either criteria. If Match Case is selected, it will further refine the filter to match on capitals or small letters. The Name box can be set using the equal symbol (=), or the not equal symbol (<>). Use the equal symbol (=) to show items that meet the criterion. Use the not equal symbol (<>) to exclude items that meet the criteria. 8. Click Expand (p. 157) to expand selected totals in either the Items available list or the Items included list. 9. Click Slice to choose which D-List you would like to appear as rows and which you would like to have as columns. The Slice command is similar to the pivot command in certain spreadsheets, but much more extensive in its scope. The orientation of a D-Cube can be changed so that pages become columns, rows become pages or columns become rows. In fact, any orientation is possible. For

User Guide 201

Chapter 8: D-Cubes four and five-dimensional D-Cubes, simply choose any two of the D-Lists to make up the rows and columns, leaving the other D-Lists to become pages. The slice can be saved with a named selection. 10. Click Move >> to move the selected items to the Items included list. A blank selection (p. 143) implies select all. 11. Look at the Items Included list: 12. Click Sort to sort the items in the Items Included list. 13. Use the Sort Arrows to sort the items in the Items Included list. Click Clear to clear the entire selection. 14. Click Reset to reset the selection back to its original format. Clicking Reset reverts to the last saved version you were working with, not necessarily the saved selection. This can be applied to data in individual cells, to a highlighted range on the current page, to a global range across a D-Cube, or to an entire D-Cube. The reset command can be applied to the data, the D-Cube structure, or both. Resetting the structure of a D-Cube allows you to cancel changes that have been implemented (not saved) in the underlying D-Lists as well as any D-Cube sorts and formats. For more information about resetting the structure of a D-Cube, see "Reset the Structure of a D-Cube" (p. 206). Note: Reset does not reset a slice. 15. Ensure any items you want to show are in the Items included list. 16. Save the selection by clicking Save on the Selection screen. 17. Click OK to return to the edited D-Cube.

Manage D-Cubes
Memory Management
For very large D-Cubes, you may see a message that states Workspace full, or Memory running low. To increase the maximum capacity, switch off the stored copy. This has the effect of switching off the colors. You have increased memory usage when the program displays changed numbers in red, because it compares two versions side by side in memory. By removing this facility, you effectively double the capacity of the largest D-Cube. The ability to reset the whole D-Cube back to a stored version remains unaffected, although reset will not be available on individual cells or selections.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Options and then select the Stored Copy tab. 202 Analyst

Chapter 8: D-Cubes 3. Clear the Create Stored copy check box. The size of the D-Cube you open increases by a factor of up to two, but it no longer displays changed numbers in different colors.

Split Column Headings onto Two Lines

Column labels may be split onto two or more lines (same as a soft return) by editing a D-List item and typing the pipe symbol | (shift+\ on most keyboards) to indicate a new line.

Steps
1. Open the D-List. 2. Double-click the item name or press f2 while in the Item name column. 3. Type a pipe symbol ( | ) where you want to indicate a new line. 4. Repeat as necessary. The D-List item splits onto a new line when it appears as a column label. The pipe symbol is not usually visible except in references to the D-List item such as in formulas. 5. Save the D-List.

Show Details or Formulas Only

You may wish to show just formula items or just details in a D-Cube. There are two methods of doing this.

Steps Using the Selection Dialog Box

1. Open a D-Cube. 2. From the D-Cube menu, click Selections, Reselect. - or Click the Reselect D-Cube button. 3. Select Formulas (or Details) from the Show box. 4. Click All to highlight the items in the Items available list. 5. Click Move to move them to the Items included list. 6. Click OK.

Steps from the D-Cube

1. Open a D-Cube. 2. Right-click the row or column labels.

User Guide 203

Chapter 8: D-Cubes 3. Choose whether to show or hide Totals or detail items: To hide totals or details, click Hide Totals or Hide Details from the menu. To show totals or details, click Show Totals or Show Details

Sort Rows, Columns, and Pages

Sort Rows or Columns Based on Numeric Values
Lists may be sorted based on the value of the data contained in the D-Cube You must decide which D-List to sort, the sort order (ascending or descending), and the criterion on which to base the sort. Multiple sort criteria are not permitted although it is possible to sort rows and columns independently.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Sort. 3. Select the D-List from the Sort Items in D-List box. 4. Click Edit Selection to select specific D-List items. Make your changes in the Select Items to Sort dialog box. 5. Click Ascending or Descending to sort the data in ascending or descending order. Note: The Select items to sort dialog box is identical to the Selection dialog box. For more information about the selection dialog box, see "Facilitate Selection Using the Selection Dialog Box" (p. 195). 6. Select the D-List from the Based on Contents of D-Lists box. 7. Click the Item Name box and select an item on which you will base the sort. 8. Click OK. 9. Click OK in the Sort Options dialog box.

Sort Rows on One Criterion, Columns on Another

It is possible to sort rows on one criterion, columns on another. For example, rows could be sorted based on the values in the Total company column, while the columns could be sorted based on the values in Q1 (the first quarter). The sorts are applied independently.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Sort. 3. Select the D-List from the Sort Items in D-List box. 204 Analyst

Chapter 8: D-Cubes 4. Click Edit Selection to select specific D-List items. Make your changes in the Select items to sort dialog box. 5. Click Ascending or Descending to sort the data in ascending or descending order. Note: The Select Items to Sort dialog box is identical (except for the title) to the Selection dialog box. 6. Select the D-List from the Based on Contents of D-Lists box. 7. Click the Item Name box and select an item on which you will base the sort. 8. Click OK. 9. Repeat this procedure for a second D-List in the same D-Cube.

Remove a Numerical Sort Steps

1. Open a D-Cube. 2. From the D-Cube menu, click Sort. 3. Locate the D-List containing the numerical sort you would like to remove. 4. Click the Item Name box. 5. Select <None> from the Item Name box. 6. Repeat for other combinations of D-Lists for which you want the sort removed. 7. Click OK. 8. Click Edit Selection to select specific D-List items.

Manipulate D-Cube Structure

You can change the perspective of a D-Cube by moving D-Lists between rows, columns, and pages, or by selecting different items on pages. Exchanging the positions of D-Lists does not affect the order of the D-List or calculation in a D-Cube. It only changes how the D-Cube is displayed.

Slice a D-Cube
You can re-orient a D-Cube to change the business perspective of your data.

Steps
1. Open the appropriate D-Cube. 2. From the D-Cube menu, click Selections, Reselect - or -

User Guide 205

Chapter 8: D-Cubes Click the Reselect D-Cube button. 3. Click Slice, and then select the D-Lists to make up the rows and the columns. 4. Click OK.

Reset the Structure of a D-Cube

The structure of a D-Cube comprises formatting, lines, sorting, zero suppressions, and, of course, the D-Lists comprising the D-Cube. In short, the structure comprises everything that makes up a D-Cube, apart from the data itself. Resetting the structure resets any changes to D-Lists contained in the D-Cube that were implemented using the Implement command from the D-List menu. This resets any calculation changes, changes to item names, averages, local D-List formats, recent insertions of rows and columns, numeric sorts, column and row widths, zero suppressions, and D-Cube formats. The data in the D-Cube is not reset although formula results might change as a result of resetting the D-Lists. Resetting the structure does not include resetting to saved selections or reverting to the saved integer or decimal mode for breakback. It does not reset D-Lists if they have been saved. It also does not reset the changes of dimensions that have been added, deleted, substituted, or reordered.

Steps to Reset the Structure of a D-Cube

1. Open a D-Cube. 2. From the D-Cube menu, click Reset structure. The underlying D-Lists are reset to their last saved versions.

Steps to Reset the Structure and Data

1. Open a D-Cube. 2. Open each D-List in the D-Cube. 3. From the File menu, click Reset for each D-List. At the prompt, click Yes. 4. With the D-Cube active, click the File menu, and then click Reset for each D-List. 5. When prompted to reset the data in the D-Cube to the saved version, click Yes.

Transpose Rows and Columns

Transposing causes the rows and columns to be swapped. Only the rows and columns are swapped, the pages remain the same.

Steps
1. Open a D-Cube. 2. Click the Transpose button. 3. To revert back to the original configuration, click the Transpose button again.

206 Analyst

Chapter 8: D-Cubes

Show a Full Selection of All Rows, Columns, and Pages

To show all rows, columns, and pages you must select all D-List items from all the D-Lists.

Steps
1. From the File menu, click Open, D-Cube. 2. Select the D-Cube to open. 3. Select Full when prompted. 4. Click OK. All D-List items are selected including any new ones that have been added since you last opened the D-Cube.

Reselect a D-Cube Steps

1. Open a D-Cube. 2. Click the Reselect button 3. Click the appropriate D-List tab. 4. Select items from the Items available box, and then click Move >>. The items are placed in the Items included list and will be displayed in your D-Cube. 5. Click OK. 6. Click Yes when prompted.

Work with Dimensions

You can choose D-Lists that will make up a D-Cube. This process includes adding D-Lists in the order of category number to maintain the correct precedence of calculation.

Add Dimensions to a D-Cube

Rows, columns, and pages (dimensions) may be inserted by adding D-Lists to a D-Cube.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Dimensions, Add. 3. Click Yes when prompted. 4. In the Select D-List to add dialog box, click the appropriate library. 5. Click the appropriate D-List.

User Guide 207

Chapter 8: D-Cubes 6. Click OK. 7. In the Select the position for the new dimension dialog box, select the order in which you would like the D-List added (row, column, or end of list). Note: When selecting the order, keep in mind that this may affect the order of D-List priority. 8. Click OK. 9. In the Select the item to which the current data belongs dialog box, select the appropriate items from the Items available box, and then click Move >>. Note: The program will incorporate the existing data from the D-Cube into the selected item. However, when adding a D-List containing subtotals, the subtotal cannot be chosen because a break back is not possible. For example, suppose the value for Monthly Sales Staff on Hand is seventy-eight. If you were to add a Months D-List and specify Full Year (which is the sum of all the months), the program would not allocate 78 pro-rata across the months. 10. Click OK. If the Items Included list in the Selection dialog box is left completely blank, all items from the Items Included list will be selected. 11. Click OK again. 12. Save the D-Cube.

Delete Dimensions (Rows, Columns, or Pages) from a D-Cube Steps

1. Open a D-Cube. 2. From the D-Cube menu, click Dimensions, Delete. 3. Select a dimension to delete In the Select dimension to delete box. 4. Click OK. 5. In the Select the item to which the current data belongs dialog box: Select the appropriate items from the Items available box, and then click Move >>. Note: The program will incorporate the existing data from the selected items to be deleted with the dimension into the D-Cube. Click OK.

If the Items included list in the Selection dialog box is left completely blank, the program will select everything. 6. Click OK. 7. Save the D-Cube.

208 Analyst

Chapter 8: D-Cubes

Substitute D-Lists Within a D-Cube

When you substitute a dimension within a D-Cube, you need to decide how to match old D-List items to the new ones that take their place. This is simple where there is a one-to-one substitution between the outgoing item and the incoming item, but there are cases where many items must be substituted by a single item. It is important to note that only detail items will preserve their data on substitution. If you replace calculated items with detail items then any data previously held against the calculated items will be lost.

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click Dimensions, Substitute. 3. Click Yes when prompted. 4. Select a D-List in the Select D-List to use as a substitute dialog box. 5. Click OK. 6. Select the dimension (a D-List from the D-Cube) to substitute. 7. Click OK. 8. Pair the items in the Old->New: Match Items dialog box. If a calculated item in the old list is being replaced by a detail item, then carry out the following procedure prior to substitution. In the old D-List, enter a new dummy detail item for every calculated item to be replaced. Use an internal D-Link to copy the D-Cube data from the calculated items to the new detail items you have inserted. On substitution match these new detail items with the detail items in the new list. Summation does not calculate weighted averages when substituting D-Lists. Assume you want to substitute a Months D-List with a Quarters D-List (as in this example): Suppose the item Unit Price is a weighted average, weighted by number of units sold. In each of the three months in a quarter the unit price is 10. When you add three months worth of Unit Price to go into one quarter, the values will be summed to display 30 ((10 + 10 + 10), which is incorrect. The correct average figure is 10. When substituting D-Lists that contain weighted averages, it is better to enter a subtotal and a corresponding detail item to which to link the data (as above) prior to substitution so that a one-to-one match is possible. 9. Click OK.

Substitute a D-List Used by Several D-Cubes

You may substitute a shared D-List across several D-Cubes in one operation. Typically this might be a new set of account codes or a new cost-center structure that must be substituted across an entire model.

User Guide 209

Chapter 8: D-Cubes Note: Data will be lost if you do not define a replacement match for each of the old D-List items. Although the substitution will make D-Links, macros and selections refer to the new D-List. The allocation table that defines how old and new D-List items correspond will not be used to update these objects. Consequently all D-Links, Selections and Macros must be checked and edited after a substitution of a D-List so that the items correspond.

Steps
1. From the File menu, click Libraries, D-Lists. 2. Click the Show objects that the selected object(s) is used by button to reveal the D-Cubes and other objects that use the selected D-List 3. Click the down arrow button to select all the D-Cubes, D-Links and other objects that use the old D-List. 4. Right-click and select Substitute D-List. 5. Select the new D-List to be used as a substitute. 6. Select the old D-List to be replaced. 7. Set up the correspondence between the old D-List and the new D-List, and then click OK. You will be prompted to confirm the substitution of the D-List in the objects selected. 8. Click Close to return to the main screen.

Effect on D-Links When Dimensions Are Altered

If you have added (p. 207), deleted (p. 208), or substituted (p. 209) a D-List, the D-Links must be edited. No attempt will be made by the program to update the source and target items in a D-Link even in cases where there is a one to one correspondence on substitution. Even in cases where you are matching descriptions in the D-Link, the substituted D-List will not be matched. If you attempt to run the D-Link you will receive the message stating "D-List xxxxx has been deleted from the D-Cube. The D-Link must be edited." If a D-List is substituted, the old D-Links will not function. For example, suppose you were to substitute a dimension in the place of an existing D-List. If you attempt to run the D-Link you will receive the message stating "D-List xxxxx has been deleted from the D-Cube. The D-Link must be edited". When you edit the D-Link a message will appear saying "The dimension xxxxx can no longer be found in the D-Cube. The D-Link will be updated accordingly". Click OK and edit the D-Link. Matching descriptions, allocation, or single item selection must be set up for the new D-List and the D-Link must be saved. When a D-List is added a similar situation occurs, although the old D-Link will still run. Most likely, however, the D-Link must still be edited. Matching descriptions, allocation, or single item selection would usually need to be set up for the new D-List. If the D-Link is not edited, a default selection for the new D-List will be left blank.

210 Analyst

Chapter 8: D-Cubes If a D-List is deleted, the old D-Links will not function. For example, suppose you were to delete an existing D-List. If you attempt to run the D-Link you will receive the message "D-List xxxxx has been deleted from the D-Cube. The D-Link must be edited." When you edit the D-Link a message will appear saying "The dimension xxxxx can no longer be found in the D-Cube. The D-Link will be updated accordingly". Click OK and edit the D-Link.

Effect on Formats when Dimensions are Altered

Formats attached to a D-List item will not be reassigned. Suppose you were to substitute a D-List with a new one. When one D-List item is substituted for another the format of the old D-List item will not be inherited, even in cases where there is a one-to-one correspondence on substitution.

Reorder the Dimensions of a D-Cube

You can set priorities, but they are not required. However, to ensure that formulas always give sensible results, follow these rules:

Priority Rule 1
Always set percentages, ratios, prices, and any per unit measure as weighted averages that are weighted by the denominator. For example, price per unit must be weighted by the number of units.

Priority Rule 2
When setting up the D-Cube, choose the D-Lists in an order that uses the calculation D-List first and the aggregation D-Lists last. For example, a typical D-Cube is set up to use four D-Lists in the order: P&L Divisions Time Actual/Budget/Variance

This ensures that the additions along the Time dimension override formulas of equal priority on the P&L dimension. Variances are calculated last. This is of particular importance when using IF THEN ELSE formulas on the P&L dimension. Generally, conditional formulas should have low priorities.

Priority Rule 3
As a last resort, change the individual priorities of formulas so that the results come out correctly. In practice, changing formula priorities is very rare, but setting weighted averages is a common occurrence.

Steps
1. Open a D-Cube.

User Guide 211

Chapter 8: D-Cubes 2. From the D-Cube menu, click Dimensions, Reorder. 3. To change or view the existing order, click Yes. Note: Although you are told that the operation cannot be undone using the undo facilities, you can always repeat the operation backward to reset the order to the original. 4. In the Reorder dimensions dialog box, select the new order of D-Lists using the arrows. 5. Click OK.

Navigate Around a D-Cube

Using the keyboard, press the following shortcut keys to move around the D-Cube.

Keys
Arrow Keys Tab Home Ctrl+Home End+Home Page Up Page Down Ctrl+Page Up Ctrl+Page Down

Action
Move one cell to the left, right, up, or down Move one cell to the right Move to the beginning of the current row Move to the top-left corner of the page Move to the bottom-right corner of the page Move up a screen, remaining on the current page Move down a screen, remaining on the current page Move to the previous page Move to the next page

Use the mouse to scroll quickly around the D-Cube by clicking the desired cell. To jump to a cell that is off the screen, drag the scroll bars. Alternatively, click the label box of a D-List and jump to a new row or column.

View a Different Slice

A D-Cube can be changed so that you can view a different slice of the D-Cube data. For example, in a D-Cube with the dimensions Centrally Allocated Items by Versions by Periods, one slice would have Centrally Allocated items as row labels, Periods as column headings, and Versions as pages. Another slice would have Versions as row labels, Periods as column headings, and Centrally Allocated Items as the pages. Yet another slice would show Centrally Allocated Items as rows, Versions as columns, and Periods as pages. There is no limit to how the data can be sliced. In four- and

212 Analyst

Chapter 8: D-Cubes five-dimensional D-Cubes you are simply selecting two dimensions to make up the rows and column labels, leaving the remaining dimensions to make up the pages. However it is important to note that within analyst it is not possible to have multiple or laminated dimensions making up the rows or columns. To do this, you must use Manager or the Analyst for Excel.

Steps
1. Click the page label box (the four-headed arrow appears). 2. Drag the page label box to the center of a column heading. In the new slice, the old page labels become the new column headings (and vice versa). To view a new slice and end up on a specified page drag the four-headed arrow from the page label box to the column label box. of the item. For example, if you have a months D-List forming the columns in your current view and Versions forming the pages, and you wish to make Versions the columns and view the October page, then drag the four headed arrow over the October columns label.

View a Different Page

Step Using the Keyboard
Choose one or more of the following keys: Ctrl+Page Up - Move to the previous page. Ctrl+Page down - Move to the next page. Ctrl+Tab - Move to a different dialog box.

Step (right-click the mouse)

Right-Click the Page Label box and select Next or Previous. Note: If you change the page, all numbers in green (typed but not entered) are lost. To prevent this, press enter to calculate before changing the page.

Steps (click the page label box)

1. Click the page label box. 2. Select the page you want to view from the list. 3. Double-click the selected page.

Save D-Cubes
Saving a current cube saves any D-Cube formatting, lines, column and row widths, breakback mode, and zero suppression settings. User Guide 213

Chapter 8: D-Cubes You can also save a copy of an open D-Cube under a different name. The old D-Cube is closed and reverts to its last saved version. Any outstanding changes are saved in the new D-Cube, not in the old one. Unlike the copy facilities in the library, D-Links are not copied when the file is saved under a different name. To save any changes to the formulas in the D-Lists, the D-Lists must be saved.

Step
Choose whether to save an existing cube or save the cube under a new name: To save the existing cube, from the File menu, click Save. Clicking Save does not save the current slice and selection. To do this, from the D-Cube menu, click Selections, Save current. You must give the selection a name. You can load the selection at a later date when you reopen the D-Cube. To confirm the save of the D-Cube, click Yes. The data color changes from pink or red to blue and black (details and formulas) to show that the data is saved. Any changes after this display in pink and red for details and formulas, respectively. To save the D-cube under a different name, from the File menu, click Save As. Enter a new name. To return to the main screen, click OK. The old D-Cube closes and you now are working on the new D-Cube with a different name.

214 Analyst

Chapter 9: D-Links
A D-Link transfers data. The target of a D-Link can be an Analyst D-Cube, a cube in a Cognos Planning - Contributor application, or a Cognos Finance dimension. Where a D-Cube is the target, the source may be another D-Cube, a cube in a Contributor application (p. 276), a dimension in Cognos Finance (p. 285), an ASCII file, an ODBC database (p. 297), or a Cognos package (p. 218). Where the target of the D-Link is a Contributor cube, certain restrictions apply and the source can only be a D-Cube or another Contributor cube. Where the target of the D-Link is a Cognos Finance dimension, the source can be an Analyst D-Cube, or a Contributor cube. Where the source of the D-Link is a Cognos Finance dimension, the target can be an Analyst D-Cube, or a Contributor cube.

Note: Contributor or Cognos Finance must be installed to use them as a target or source of data in an Analyst D-Link. To use a Cognos package as a source, you must first create a model in Framework Manager and publish the model, or elements of it as a package. When you create a D-Link, you specify how the dimensions of the source and the target correspond. Analyst is ODBC enabled, so a D-Link can import data from any source for which an ODBC driver is available, for example, Microsoft Access, Excel, and databases such as Oracle and SQL Server. You can import data in ODBC databases directly into the D-Cube because the ODBC driver presents database data in a strictly defined format. When you import data from an ASCII file, you must first create a file map (p. 731) for the ASCII file to define how the file has been structured, for example, tab separated, and how the data in the ASCII file correspond to the dimensions in the D-Cube. You do not need to open a D-Cube to update the data in it with a D-Link. However the run D-Link icon on the toolbar is only operative if a D-Cube or D-link is open. The D-Link only transfers data when the D-Link is run. There are various ways to run a single D-Link or a batch of D-Links in a specified order. You must run batches of D-Links to update your model when external data or your assumptions change. You can execute batches of D-Links using the D-Links Update list, or using macros. D-Links for which the source is a D-Cube can be run inversely to transfer the relevant data from the target to the source. You can drill down on D-Cube data that was imported by saved D-Links. This displays the original source data, regardless of its source. You can run special lookup and accumulation D-Links to reorganize your data in D-Cubes that contains D-List formatted dimensions (virtual dimensions).

User Guide 215

Chapter 9: D-Links

The D-Link Dialog Box

When you open a D-Link that has been saved, the source and target names are displayed, and their dimensions are listed in the dimension names section. Source and target dimensions may be paired together or left unpaired. Paired dimensions appear above the dividing line in the dimension names section and are connected with a pairing indicator, which indicates the pairing mode - either match descriptions (p. 233) or allocation. You can break the connection between paired dimensions, or change the pairing mode by clicking the pairing indicator. Unpaired dimensions appear below the dividing line. To view additional details of a D-Link definition, select a dimension by clicking its name. The items of the selected dimension are listed in the Dimension Items area of the D-Link dialog box. When you select an unpaired dimension, you see all items within that dimension and a separate table showing any items that have been selected from that dimension. When you select a paired dimension, the dimension it is paired with is also selected, and the dimension items section displays full details of the pairing. Initially, no dimension is selected, so the Dimension Items area is empty. The D-Link lists only unique dimension item names in the Dimension Items area. D-Lists cannot contain duplicate item names, so you will see all the D-List items in D-List order. The dimensions of an external source relates to a column in an ASCII file or a field in an ODBC database where the same item name may appear in many records (rows). The D-Link will display only the unique names, in the order in which they appear in the ASCII file or database. In either case, you may cut a subcolumn from the item names that may reduce the number of items displayed (to those which are unique within the subcolumn). In this case the D-Link only displays the first unique instance of that item. For example if you have two items A0023 Apples and A0024 Apples, and the first four characters are used in the cut, data would only be transferred to the A0023 Apples item. If these items are the source, the data in both these items would be aggregated together with the data for any other items beginning A002 and transferred to the target specified in the D-Link. When you click a dimension from an external source, Analyst will read the entire source file in order to get an up-to-date dimension item listing. With large source files you will experience a short delay while the source data is read.

Dimensions and D-Lists

The dimensions of a D-Cube are defined by the D-Lists included in the D-Cube. In a completed D-Link, the D-Lists of a D-Cube are not necessarily shown in D-Cube order. Paired D-Lists move above the dividing line in the order in which they were paired. A filemap is used to define the dimensions and data values of an ASCII file, and can have single or multiple columns. The ODBC driver effectively defines the dimensions of an ODBC database. When describing D-Links, the term dimensions will generally be used, as the D-Link is able to treat D-Lists and the dimensions of an external data source in much the same way. The term, D-Lists will only be used when a particular comment or option applies only to D-Lists. There are important differences between D-Links that use an external source and D-Links which have a D-Cube as the source.

216 Analyst

Chapter 9: D-Links In addition to normal dimensions, a D-Cube may have virtual dimensions, which are introduced when one or more of the D-Lists making up a D-Cube contains D-List formatted items. All D-Lists used as format D-Lists appear as virtual dimensions when the D-Link lists the dimensions of a D-Cube. Virtual dimensions are readily distinguished in the D-Link as the D-List name is enclosed in brackets [ ]. In normal D-Links, any virtual dimension can simply be ignored (left unpaired and no selection made). You only need to concern yourself with them if you are creating special lookup or accumulation D-Links.

Create D-Links
A D-Link is created in the following stages: Steps Select New D-Link from the menu (p. 217) Select a source and a target for the D-Link (p. 217) Pair source and target dimensions (p. 219) Select the required items from unpaired dimensions (p. 219) Change optional settings as required (p. 220) Name and save the D-Link (p. 220) These steps describe how to create a regular D-Link using a D-Cube as the source. You can also use the following as the source: mapped ASCII files (p. 731) ODBC (SQL) (p. 299) Data, Contributor data, Cognos Finance data, and a Cognos package. If you wish to create a lookup or accumulation D-Link, then you must specify the type before pairing source and target dimensions, because virtual dimensions do not appear when creating regular D-Links.

Use D-Cube as Source and Target

Steps
1. From the File menu, click New, D-Link. 2. Click Source. 3. Select D-Cube from the menu. 4. In the D-Cube Select dialog box, select a D-Cube. 5. Click OK. The D-Cube is displayed, and its dimensions are listed. 6. Click Target. 7. In the Select D-Cube dialog box, select a D-Cube.

User Guide 217

Chapter 9: D-Links 8. Click OK. If both the source and the target for the D-Link are D-Cubes, and one D-List is common to both the source and the target, these D-Lists are paired automatically using match descriptions. If required, you can change to an allocation table pairing, or break the connection by clicking the pairing indicator.

Use Cognos Package as Source in a D-Link

You may use a Cognos package as the source in a D-Link. After you pair your dimensions, you will need to specify a temporary location for the data to be stored. Note: You must first create a model in Framework Manager and publish the model, or elements of it as a package before you can use a Cognos package as a source in a D-Link.

Steps
1. Open the target D-Cube you wish to import the data into. 2. From the File menu, click New, D-Link. 3. Click Source. 4. Click Cognos Package. 5. Select a package from the drop box. 6. Select a Query Subject. 7. Select the available Query Items in the Query Subject and move them to the Selected Query Items pane. 8. Select the Display preview of selected query item check box to preview the Query Items. The Preview option only works with Query Items that have not been selected, and helps you select the correct Query Items. Click OK. The Query Items are brought into the D-Link. 9. Click Mark Data to select the columns containing the data. 10. Pair the dimensions. 11. From the D-Link menu, click Options. In the Cognos Package - Alternative temporary data storage path area, browse to a temporary location in which to store the import data. If you do not specify a location, Analyst will default to the location of the Filesys.ini file. If you specify a different default location, ensure that it is accessible and writable from all Analyst clients and Planning servers. You can also specify the temporary data storage default location before creating your D-Link. From the Tools menu in Analyst, click Options and select the General tab. Under D-Link Temporary Data Storage, you can browse to a custom default location. 12. When you have finished pairing the dimensions, from the D-Link menu, click Execute. The data is imported using the Cognos package. 218 Analyst

Chapter 9: D-Links 13. Save and close the D-Link.

Pair Source and Target Dimensions

You specify how the dimensions of the source and target correspond by pairing source and target dimensions. D-Lists that appear in both D-Cubes are already matched. If there are unmatched items, you will have to pair them.

Steps
1. Click a source dimension. The D-List items of the selected dimension are displayed in the dimension items section of the D-Link dialog box. 2. Click Ctrl + a target dimension to which you want to pair the selected source dimension. Tip: In general, dimensions belonging to the same category should be paired. For example, timescale D-Lists should be paired and products D-Lists should be paired. However, timescale D-Lists should not be paired with products D-Lists. 3. Select either Match descriptions or Allocate items: Select Match descriptions to have matched source and target items highlighted automatically, or select Allocate items to manually pair source and target items.

Select Required Items from Unpaired Dimensions

Often you will be left with unpaired dimensions in the source and/or target. If left unspecified, the unpaired items from the source will be imported into unpaired items in the target. To avoid this, you must select the required items from unpaired dimensions.

Steps
1. Click an unpaired dimension. The items of the selected dimension are displayed in the dimension items section of the D-Link dialog box, ready for you to make a selection of items. 2. Click Select if you want to make a selection of items. Instead of making a selection of items, you can leave the selection empty.The sum of the items is transferred into the designated target cells. 3. Double-click an item in the dimension items list to move it to the Selected items box. To select more than one item, highlight the required items in the dimension items list, and then click the arrow button pointing in the direction of the selected items list. 4. Repeat steps 1 through 3 for all remaining unpaired source and target dimensions.

User Guide 219

Chapter 9: D-Links

Change Optional Settings in D-Link

Optionally you can change various settings, which will modify the way a D-Link runs.

Steps
1. Select an execution mode from the Mode box: Fill Substitute Add Subtract Select a Dump Item option (this is only relevant when importing data from external sources).

2. Select the D-Link type: Regular - shows only real dimensions. Lookup - shows real dimensions and D-List formatted items for target dimensions. Accumulation - shows real dimensions and D-List formatted items for source dimensions.

3. Specify scaling and rounding options (from the D-Link menu, click Options).

Name and Save the D-Link

Steps
1. From the File menu, click Save. 2. Select a library in the Save D-Link As dialog box. 3. Type a name for the new D-Link 4. Click OK. 5. Before you close the new D-Link, you can give it a description and attach notes to it by clicking the File menu, and then clicking Summary Info. Any description entered will be displayed at the bottom of the Select D-Link box when opening a D-Link.

Open D-Links
Steps
1. Click the open D-Link button. 2. In the D-Link Select dialog box: Select a library.

220 Analyst

Chapter 9: D-Links Select the desired D-Link. Click OK.

Open More than One D-Link

Steps
1. From the File menu, click Library, D-Links. 2. In the Library Functions dialog box, select the appropriate library. 3. Select the appropriate D-Links. 4. Select the required D-Links in the Available Objects list, and then click the down arrow to select them. 5. Click the Open button. You can also run the D-Links, copy them, move them, rename them, print, or delete them by clicking the appropriate button, but only when the D-Link is not open. 6. When the D-Links have been opened, click Close.

Open a D-Link that Targets an Open D-Cube

Use this method if you have a D-Cube open and you want to open a D-Link that uses the open D-Cube as its target.

Steps
1. Make the relevant D-Cube the active D-Cube. 2. From the D-Cube menu, click D-Links, D-Links into D-Cube. 3. In the Select D-Link to Edit dialog box: Select the desired D-Link. lick Edit.

Open a D-Link that Uses an Open D-Cube as its Source

Use this method if you have a D-Cube open and you want to open a D-Link that uses the D-Cube as its source:

Steps
1. Make the D-Cube active. 2. From the D-Cube menu, click D-Links, D-Links from D-Cube, or click the Run D-Links button.

User Guide 221

Chapter 9: D-Links 3. In the Select D-Link to Edit dialog box: Select the D-Link you want to open. Click Edit.

Open D-Links Associated with Selected D-Cubes

Steps
1. From the File menu, click Library, D-Cubes. 2. In the Library Functions dialog box do the following: Select the appropriate library. Select the appropriate D-Cube. Click the Show objects that the selected object(s) is used by button, or right-click the selected D-Cube name, and then click Show Used By from the shortcut menu. 3. You will now see a list of Analyst objects that use the highlighted D-Cube. This will include the names of D-Links that use the D-Cube as their source or target. 4. Select the required D-Links in the list, and then click the down arrow. You can automatically highlight all D-Links that use the chosen D-Cube as either a source, target, or parameter by selecting Source or Target from the Highlight usage type list. You have the option to view specific precedents or dependents, depending on their function to the D-Cube. For example, after selecting Source as the usage type and clicking the down arrow, the source D-Links are highlighted. 5. To open the selected D-Links, click the Open button. Before opening the D-Links it may be useful to move the D-Link Library Functions dialog box so that you can see the D-Links as they are opened. 6. When the D-Links have been opened, click Close.

Run D-Links
Run a D-Link to transfer data. Messages are suppressed when executing macros that run ODBC D-Links from SQL7 as a source.

222 Analyst

Chapter 9: D-Links

Run an Open D-Link

Use this method if you have an open D-Link that you would like to run.

Step
Make the appropriate D-Link active and click the Run D-Link button. The open D-Link runs. If you have made changes to the D-Link but not saved the changes, only the saved version of the D-Link runs.

Run a D-Link Using a Specific D-Cube as its Source or Target

The D-Cube may or may not be open.

Steps
1. From the Tools menu, click one of the following: Run Source Link Run Target Link

2. In the D-Cube Select dialog box do the following: Select a library. Select a target D-Cube. Click OK.

3. If a D-Cube dialog box was active in step 1, this D-Cube is selected in the D-Cube Select dialog box. In the Select D-Link to Execute dialog box do the following: Select the D-Link to run. Click Execute.

Run a D-Link Using an Open D-Cube as its Target

Use this method if you have a D-Cube open and you want to run a D-Link that uses the D-Cube as its target. If a target D-Cube is open when a D-Link is run, only data assigned to cells within the open selection of the target D-Cube is transferred. Data assigned to cells outside the open selection is not transferred. Note: If the selection open does not contain the target area, the D-Link does not execute and an error message appears. If multiple views of the target D-Cube are open, each with a different selection, data is assigned to cells included in the open views. Only data assigned to cells outside all open views is not transferred. A target D-Cube is left open after a D-Link has run. To save the changed data you must save the D-Cube.

User Guide 223

Chapter 9: D-Links If a D-Link is run into a closed D-Cube, all data assigned by a D-Link is transferred to the target D-Cube. Internally, Analyst opens the target D-Cube, transfers data into it, and then saves and closes the D-Cube automatically. D-Links do not usually assign data to all cells of a target D-Cube. The selection of cells that a D-Link can affect (assign data to) is known as the target area (p. 253) for the D-Link. Opening a limited selection of the target D-Cube before running the D-Link can be used to temporarily restrict the target area for the D-Link.

Steps
1. Ensure the required D-Cube dialog box is active. 2. Click the Run D-Link button 3. In the Select D-Link to Execute dialog box, select the D-Link you want to run, and then click Execute. D-Links run in this manner will only update data within the open selection of the target D-Cube.

Run D-Links with the Source D-Cube Open

For D-Links that use an open D-Cube as their source when a D-Link is run, data from the open selection of the source D-Cube is transferred in preference to the saved data in the source D-Cube. Data outside the open selection of the source D-Cube is taken from the saved D-Cube as normal. If multiple views of the source D-Cube are open, each with a different selection, data is taken from all open views. Only data excluded in any open view is taken from the saved version of the D-Cube. If the source D-Cube is completely open, all data is taken from the open version of the source D-Cube. Of course, this only effects the data transferred by the D-Link when the open selection of the source D-Cube contains different data than the saved version.

Experiment with the source D-Cube

The behavior described above can be useful if you want to make experimental changes to data in the source D-Cube, then feed the changed data on to other D-Cubes to observe the effect, without having to make the changes permanent. Note that you will have to open these other D-Cubes before you transfer data into them if you want to be able to undo your changes. You could change data in the source D-Cube manually. You could also change the data by running experimental D-Links into the source D-Cube. For example, you could open D-Links which target this source D-Cube, modify them and then run them without saving them. To undo your changes you would simply close the D-Links without saving them, or reset them to the saved versions by clicking Reset from the File menu. You can also change data in the open source D-Cube by modifying the D-Lists of the D-Cube and implementing rather than saving the modifications. For example, you could change formula in one of the D-Lists of the source D-Cube, implement the changes (using the menu command Implement) , then transfer the changes on to other open D-Cubes to observe the effect. To undo your changes,

224 Analyst

Chapter 9: D-Links close all affected D-Cubes without saving them. You can also reset any D-Lists with implemented changes by clicking Reset from the File menu. Remember that you cannot experiment in this way with adding, deleting or substituting D-Lists in the source D-Cube. These changes to a D-Cube are immediately saved and cannot be reset.

Run D-Links with the Source D-Cube Open with a Limited Selection
Data within the open selection of the source D-Cube is taken from the open version of the source D-Cube. Data outside the open selection of the source D-Cube is taken from the saved version of the source D-Cube.

Run Update D-Links in a Single D-Cube (Manually)

Steps
1. Open a D-Cube. 2. From the D-Cube menu, click D-Links, Update. 3. Click Run All. You can also run the update D-Links using the macro command @DCubeUpdate.

Run Batches of D-Links Using Library Functions

You can create a selection of D-Links in the D-Cube Library Functions dialog box using the dependent tracing facilities, and then run the selected D-Links in order.

Steps
1. From the File menu, click Library, D-Cubes. 2. In the D-Cube Library Functions (p. 301) dialog box do the following: Select the D-Cube you wish to update. Click the Show objects that the selected object(s) is used by button. Make your selection from the Highlight Usage Type list. You have the option to view specific precedents or dependents, depending on their function to the D-Cube. Select the desired D-Links in the Objects Using D-Cube dialog box.

3. Click the Select button. When you select a new item, it is added at the bottom of the selected items list (you cannot reorder items in the selection list). When you run the D-Links, they will be run in the order they appear in the selected items list, so it is important that you select the D-Links in the correct order. Selected D-Links can be reodered using the arrows to the right hand side of the box.

User Guide 225

Chapter 9: D-Links 4. Click the Run button. If you want to run additional D-Links targeting different D-Cubes, you can run those selected so far, deselect them and repeat steps 2 to 4 for another D-Cube. 5. When all required D-Links have been run, click Close.

Memory Considerations
If you are experiencing memory problems when running a D-Link then consider use of the slice update facility which allows the D-Link to be run in smaller stages updating one slice of the target at a time. This facility is available for D-Links which target analyst or for D-Links which target contributor. For more information about the slice update facility, see @SliceUpdate (p. 700).

Open a limited selection from a D-Cube

Whenever a D-Cube is opened in Analyst, details of the base data and the calculations in the D-Cube are loaded into memory. Therefore, the memory available in your computer limits how much of the D-Cube may be opened. If you open only a limited selection of a D-Cube, less memory is required. The memory required to open a selection of a D-Cube is not directly determined by the selection you make from the D-Cube. If you include any formula items in your selection, then internally, Analyst may have to open an expanded version of your selection. The memory required to open and recalculate a D-Cube is not determined by the selection you have made from the D-Cube, but by the expansion of this selection.

What is an expanded selection?

When you open a limited selection from a D-Cube, only the items you selected are displayed in the D-Cube dialog box. If you have included formula items in your selection, however, an expanded selection may be opened internally, so that the selected formula items can be recalculated and broken back properly. If a basic selection includes formula items, the expanded selection will also include all the items on which the selected formula items depend (either directly or indirectly). For example, consider a Months D-List that contains twelve detail items (Jan, Feb, Mar, and so on), four quarterly totals (Q1 = Jan + Feb + Mar, and so on), and one full year total (Full Year = Q1+Q2+Q3+Q4). If you open a D-Cube containing this D-List and select only Q1 from MONTHS, the expanded selection will include Q1 and Jan, Feb, and Mar. If you select only Full Year, the internal expanded selection will include all items in the MONTHS D-List. Remember that Analyst does not save calculated data in the D-Cube when it is closed; data in formula items is only recalculated when required.

What is opened when a D-Link is run?

When a D-Link using a D-Cube as its source is run with the source D-Cube closed, Analyst opens the source D-Cube internally to read the data from it.

226 Analyst

Chapter 9: D-Links If the D-Link picks up only detail data from the source D-Cube, a basic selection of items, determined by the D-Link definition, is opened. The basic selection includes all items assigned by match descriptions or an allocation table, and all items selected from unpaired source D-Lists. However, if the D-Link imports calculated numbers from the source D-Cube, the selection of items opened will be the expansion of the basic selection. If the source D-Cube is fully open when the D-Link is run, Analyst does not need to open it internally as all the required data can be taken from the open version of the D-Cube. If the source D-Cube is open with a limited selection when the D-Link is run, Analyst may need to open an additional selection internally if all data transferred by the D-Link is not found in the open selection. If the target D-Cube is closed when the D-Link is run, Analyst opens the target D-Cube internally, transfers data into it, saves and closes the D-Cube automatically. The portion of the target D-Cube that Analyst needs to open internally is the expansion of the D-Link's normal target area. The target area includes target items matched by match descriptions (p. 233), target items with entries in a local allocation table (p. 242), and items selected from unpaired target D-Lists (p. 231). If the target D-Cube is open when a D-Link is run, data is only assigned within the open selection of the target D-Cube. Data assigned to cells outside the selection you have opened is not transferred, even if these cells are within the internal expanded selection. So when you run a D-Link into a selection from an open D-Cube, nothing extra is opened and no expansion takes place.

Run Batches of D-Links using Macros

You can use the following macro commands to run batches of D-Links: @DLinkExecute (p. 665) - runs one D-Link @DLinkExecuteList (p. 666) - runs a series of D-Links in order @DCubeUpdate (p. 696) - runs the update D-Links list for one D-Cube @DLinkExecSel (p. 664) - runs a D-Link into a limited selection of a target D-Cube

When you use either @DLinkExecute or @DCubeUpdate to run D-Links, the D-Links behave exactly as if they had been run manually. For example, when the @DCubeUpdate command is run, the target D-Cube may be closed, fully open, or open with a limited selection. If closed, data changed in the target D-Cube is saved automatically at the end of the update. If open with a limited selection, only the open selection of the target D-Cube will be updated.

Tun an Inverse D-Link

Any D-Link for which the source is a D-Cube can be run inversely, that is, it can transfer data from the target to the source. Note: You cannot use drill down to trace data that has been transferred by an inverse D-Link. Also remember that a D-Link run inversely does not have exactly the same the reverse effect of the D-Link run normally. To establish what a D-Link will do with data when run inversely, look at the normal D-Link definition and imagine that the source and target D-Cubes have been exchanged, while the rest of the D-Link definition is left unchanged. User Guide 227

Chapter 9: D-Links Match descriptions (p. 233) pairings are preserved with the source and target D-Lists exchanged. Any subcolumn cut (p. 259) from a D-List is preserved. The case-sensitive setting is not changed. The dump item setting is only relevant when importing data from external sources. Allocation table pairings are preserved. The source and target D-Lists and the source and target columns in the allocation table are exchanged. Any subcolumn cut from a D-List is preserved. The case sensitive setting is not changed. The dump item setting is only relevant when importing data from external sources. Unpaired source and target D-Lists (p. 231) are exchanged. The selection of items made for a D-List is preserved. The D-Link execution mode (p. 252) is unchanged. The scaling option is reversed; data transferred by the inverse D-Link is scaled by the inverse of the scaling factor set for the forward D-Link. For example, if the scaling factor (p. 257) is set to 10 (divide by ten), then the inverse D-Link will use a scaling factor of 0.1 (divide by a tenth) . The rounding option (p. 257) is unchanged; data transferred by the inverse D-Link is rounded according to the setting in D-Link options. The D-Link dump option is only relevant when importing data from external sources.

Note: Special rules apply when running accumulation D-Links inversely; the inverse accumulation D-Link will perform a break back allocation over the source data. (Lookup D-Links should not be run inversely at all). See the Lookup and accumulation D-Links section for more information.

Steps
1. Open a target D-Cube. The D-Cube which is the target for the D-Link as defined, and the source for the inverse D-Link. 2. With the target D-Cube dialog box active, click the Run D-Link button. 3. From the list, select the D-Link you want to run, and then click Inverse. Be sure you have selected the correct D-Link. When you click Inverse, the selected D-Link is run inversely - data is transferred and the source D-Cube is saved automatically. If you have the source D-Cube completely open when you run a D-Link inversely, you can save the data transferred to the source D-Cube after the D-Link has run. You can also run D-Links inversely using a macro. The @DLinkExecInv command runs one D-Link inversely. It always runs the D-Link as it is found when the macro is run, not as it was when the macro was created. The D-Link to run inversely is identified by the macro's only parameter: DLink.

Exchange Source and Target D-Lists by Running D-Links Inversely

When a D-Link is run inversely, the source and target D-Lists are exchanged. Source and target D-Lists are treated differently by the D-Link in some respects: Exchanging them can sometimes give unexpected results. The differences you must consider are listed below.

228 Analyst

Chapter 9: D-Links

Many-to-one and one-to-many allocation tables

A many-to-one local allocation table (p. 247) sums the data in multiple source items and assigns the total to a target item. However, when the source and target for a many-to-one allocation table are exchanged, the data in the target item is taken and assigned to each source item. The data in the target item is not split up and allocated over the source items, like a break back. Similarly, a one-to-many local allocation table (p. 247) takes data from one source item and assigns it to multiple target items. When the source and target for a one-to-many allocation table are exchanged, data from the multiple target items is summed, and the total assigned to the source item.

Unpaired dimensions
In an unpaired source D-List (p. 231), data is taken from all items selected and summed. In a unpaired target D-List (p. 231), data is assigned to each of the items selected.

Cut subcolumns
If multiple items in a source or a target D-List are identical within a subcolumn (the subcolumns were cut (p. 259)), only a single instance of the cut-down item name is displayed in the item names list. The cut-down name may be matched by match descriptions (p. 233) or used in an allocation table entry. In a source D-List, data from all items that match the cut-down name is summed. In a target D-List, data is assigned to the first item in the D-List that matches the cut down name.

Formula items and match descriptions

Match descriptions (p. 233) will take data from formula items in a source D-List, but will not assign data to formula items in a target D-List. Thus, when the source and target for a match descriptions pairing are exchanged, the set of matching items may be different than the original set of matching items. For example, the item, Q1, appears as a formula item in the D-List, Months, and as a detail item in the D-List, Quarters. Match descriptions will match the Q1s if Months is the source D-List, but it will not if Months becomes the target D-List.

Case sensitive option

When the case sensitive option is not selected, one allocation table entry may correspond to multiple D-List items, and match descriptions may match multiple D-List items at a time. By default the Case Sensitive option is on. Clear the Case Sensitive check box if you want match descriptions to ignore capitalization in the source item names. Typically, this is used when there are inconsistencies in the capitalization of source item names that you want the D-Link to ignore. When the Case Sensitive check box is selected, a highlighted item in the source will only have one matching item in the target. If cleared, there may be many source items highlighted that all match one target item. Data from all matching source items is summed and assigned to the matching target item.

User Guide 229

Chapter 9: D-Links The D-Link lists only unique item names. This is always case sensitive regardless of the Case Sensitive option. In a source D-List, if match descriptions matches more than one source item with one target item, or more than one item corresponds to one source allocation table entry, then data from all matching source items is summed. In a target D-List, if match descriptions matches more than one target item with one source item, or more than one item corresponds to one target allocation table entry, then data from the source item is assigned to the first matching item in the target D-List.

Dimensions
A D-Cube is made up of dimensions, which are D-Lists, or virtual dimensions.

Virtual Dimensions
In addition to normal dimensions, a D-Cube may have virtual dimensions. Virtual dimensions are introduced when one or more of the D-Lists making up a D-Cube contains D-List formatted items. When Regular is selected as the D-Link type, no virtual dimensions show in the D-Link editor. To view any virtual dimensions in the source cube, set the D-Link type as Accumulation. To view any virtual dimensions in the target cube, set the D-Link type as Lookup. In normal D-Links, any virtual dimension can simply be ignored (that is, left unpaired and no selection made). You only need to concern yourself with them if you are creating special lookup or accumulation D-Links. For information about lookup and Accumulation D-Links. Note: You cannot cut a Subcolumn when pairing virtual dimensions.

Dimensions and D-Lists

When describing D-Links the term dimensions is generally used, as a D-Link is able to treat D-Lists and the dimensions of an external data source in much the same way. The term, D-Lists is only used when a particular comment or option applies only to D-Lists.

Unvisited Dimensions
After selecting the source and target D-Cubes in a new D-Link, all dimensions are considered unvisited (not yet paired or selected). When you select an unvisited dimension in a new D-Link, the dimension items are displayed in the same way as for an empty selection. Unvisited dimensions are assumed to be empty selections; so when you are creating a D-Link, ensure you have made the appropriate selection from all unpaired dimensions. Remember that if a dimension is added or deleted from a D-Cube or external source (for example, in an ASCII file new columns may appear and existing columns may be removed or skipped), then D-Links using the D-Cube (or external source) are likely to have new unvisited dimensions. If a dimension is added, it will be unvisited in D-Links. If a dimension is deleted, any dimension to which it was paired will be unvisited.

230 Analyst

Chapter 9: D-Links Unless you edit the D-Links, unvisited dimensions are considered empty selections. If the empty selection is in the source, all detail items are aggregated. If the empty selection is in the target, all detail items are populated.

Unpaired Dimensions
Usually, you will be left with unpaired dimensions in the source and/or the target of a D-Link. It is recommended that you make a selection (p. 232) of the required items from unpaired dimensions.

Unpaired Source Dimensions

From an unmatched source dimension you should make a selection of the item (or items) that contains the data you want to be transferred in a D-Link. If you select more than one item, the data in all selected items is summed when the D-Link is run. Note: You do not have to select any items. An empty selection implies use all items. If the source dimension is a D-List, data is taken from all detail items only. If you want to take data from formula items, you will have to explicitly select them. If the source dimension relates to a column in an ASCII file, empty records may be found in this column. If the selection of source items is left empty, data in rows identified by a blank item will be imported. You cannot include a blank item in the selected items list. If the source dimension is not a D-List, the items selected are not linked to the source item names. If a source item is renamed or an item disappears, the original entry in the selected items list is not updated or removed. Note: In an accumulation D-Link there are two distinct types of unpaired source dimensions: The source Lookup D-List itself which should never be left with an empty selection, and any other remaining unpaired dimensions which will often be left unvisited.

Unpaired Target Dimensions

From an unpaired target D-List you should make a selection of the item (or items) into which the data should be transferred in a D-Link. If you select more than one item, the data is copied into each selected item (not divided and apportioned across the target items). You do not have to select any items. An empty selection implies target all detail items. If you want to target formula items, you will have to explicitly select them. When a D-Link targeting formula items is run, a break back will be performed in the target D-Cube as if the data had been entered manually in the D-Cube. One break back can change a large amount of data! You should not target formula items accidentally. Remember, if you run a D-Link into a closed D-Cube, the changed data is saved automatically. You can target a formula item and some of its dependent detail items. In this case data is first transferred to the detail items, then these items are held while a break back is performed as data is transferring to the formula. Note: In a Lookup D-Link there are two distinct types of unpaired target dimensions: The Lookup D-List itself which should never be left with an empty selection and any other remaining unpaired dimensions which will often be left unvisited.

User Guide 231

Chapter 9: D-Links

Empty Selections
In a D-Link, empty selections can be very useful, as they adapt to changes in the dimension item listing. For a D-List, all detail items are used (or targeted) even if new detail items have been added since the D-Link was defined. For other dimensions, all source items are used even if items have been renamed or added. When a source dimension relates to a column in an external source, it is important to remember that the empty selection will accept all entries found in this column.

The DATA dimension and @Count

When a Map is used as the source for a D-Link, the columns marked Use as data become the items of one dimension named DATA. The D-Link always includes an item named @Count in the DATA dimension. When a DATA dimension is left unpaired in the D-Link, the empty selection will sum all items except @Count.

Make a Selection of Items in a D-Link Steps

1. Open an appropriate D-Link. 2. Click an unpaired source or target dimension. The items of the selected dimension are displayed in the dimension items section of the D-Link dialog box. 3. Click Select to display the selected items list. 4. In the Select items box do the following: Double-click an item in the dimension items list to add to the selected items list. To add more than one item, select the required items in the dimension items list, then click the right arrow. You can highlight adjacent items by dragging or by holding shift in conjunction with end, home or arrow. You can highlight non-adjacent items by holding Ctrl and clicking.

5. To remove an item from the Selected items box, double click the item name. To remove more than one item, highlight the required items and click the left arrow. 6. Notice that selected items are not removed from the dimension items list.

Cancel a Selection of Items in a D-Link

In a D-Link, if you want to pair a dimension from which a selection of items has been made you must first cancel the selection of items. You might also cancel a selection of items in order to use an empty selection (p. 232).

Steps
1. Open an appropriate D-Link. 232 Analyst

Chapter 9: D-Links 2. Click the unpaired source or target dimension. 3. Select an item from the Selected items box. 4. Click the left arrow. 5. Repeat as necessary. When emptied, the selection list does not immediately disappear, but when you return to the dimension it will have been removed.

Match Descriptions
In a D-Link, Match Descriptions automatically matches source and target dimension items with the same name. In addition, Match Descriptions can be used to perform an allocation by date. For information about specifying a match description pairing, see "Create a Match Descriptions Pairing" (p. 233).

How Match Descriptions Pairs Data

When you first choose a Match Descriptions pairing or open a D-Link and look at an existing Match Descriptions pairing, the item names in the source and the target dimensions are compared and the matching items are highlighted. When the D-Link is run this comparison is made again and data from each highlighted source item is assigned to the matching target item. The key benefit of the Match Descriptions pairing is that it updates automatically. The D-Link will always assign data according to what matches at the time it is run, even if items in the source or target have been added, deleted, or renamed. By default, Match Descriptions will not assign data to formula items in a target D-List unless Match Calculated Target Items is selected. This exception is made to avoid unnecessary break backs. It is not possible to exclude individual matching items from the match descriptions allocation. If you want to do this you will have to use an allocation table. When you use an external source for a D-Link, the source dimensions relate to columns in an ASCII file or fields in an ODBC database, where the same item name may appear in many rows (records) . The D-Link will display only the unique names in the order in which they appear in the ASCII file or database. When the D-Link is run, Match Descriptions takes the data from all matching records, sums it, and assigns the total to the matching target item.

Create a Match Descriptions Pairing

You cannot pair a dimension that is part of an existing pairing; you must first break the existing connection. To do this, right-click the pairing indicator and select Break Connection.

User Guide 233

Chapter 9: D-Links You cannot pair an unpaired dimension from which a selection of items has been made either; you must first clear the selected items list (p. 232).

Steps
1. From the File menu, click New, D-Link. 2. Click Source and select a source. 3. Click Target and select a target D-Cube or Contributor cube. 4. Click a source dimension. The items of the selected dimension are displayed in the dimension items section of the D-Link dialog box. 5. Click Ctrl + target dimension that you want to pair with the selected source dimension. Both the source and target dimension names are now selected, and the items of both dimensions are displayed in the D-Link dialog box. 6. Click Match descriptions. The paired dimensions move above the line, joined by the match descriptions pairing indicator. Click the pairing indicator to change the pairing mode or break the connection. The matching source and target items are highlighted. If required, you can change the case sensitive option, match calculated target items, select a dump item, and cut subcolumns from the source and/or target dimensions.

Case Sensitivity
By default the Case Sensitive option is on. Clear the Case Sensitive check box if you want match descriptions to ignore capitalization in the source item names. Typically, this is used when there are inconsistencies in the capitalization of source item names that you want the D-Link to ignore. When the Case Sensitive check box is selected, a highlighted item in the source will only have one matching item in the target. If cleared, there may be many source items highlighted that all match one target item. Data from all matching source items is summed and assigned to the matching target item. The D-Link lists only unique item names. This is always case sensitive regardless of the Case Sensitive option. For example, the items, ITEM1 and item1, will both be listed whether the Case Sensitive check box is selected or cleared. Tip: While it is not recommended, you may include D-List item names that are duplicates except for capitalization (for example, ITEM1 and item1). If a D-List containing such duplicate names appears in the target and the Case Sensitive check box is cleared, then all of these items will be highlighted, but match descriptions will only assign data to the first highlighted item.

234 Analyst

Chapter 9: D-Links

Match Calculated Target Items

You can use match descriptions to match calculated totals and detail items in a D-Link. Matching formulas triggers a break back in the target D-Cube so be cautious.

Steps
1. From the File menu, click New, D-Link. 2. Click Source and select a source. 3. Click Target and select a target D-Cube. 4. Click a source dimension. The items of the selected dimension are displayed in the dimension items section of the D-Link dialog box. 5. Ctrl+Click the target dimension that you want to pair with the selected source dimension. Both the source and target dimension names are now selected, and the items of both dimensions are displayed in the D-Link dialog box. Note: If the source data contains items which match to both the detail and the calculated items in the target, and the target cube is not all zero, the calculation engine ignores the imported data for the totals and only keeps the imported data for the details. This could mean that the D-Link will not transfer data as expected. To avoid this potential problem - set the D-Link to Substitute mode, or use an allocation table to ensure that only calculated items are matched. 6. Click Match description. The paired dimensions move above the line, joined by the match descriptions pairing indicator. Click the pairing indicator to change the pairing mode or break the connection. The matching source and target items are highlighted. If required, you can change the case-sensitive option, select a dump item, and cut subcolumns from the source and/or target dimensions. 7. Select the Match Calculated Target Items check box. This matches calculated totals in the target D-Cube in addition to detail items. Note: If a contributor cube is the target of the D-Link then this option is not relevant as only detail items may be targeted in the target lists.

Allocation
When choosing to allocate items in a D-Link there are three options available Local Allocation tables (p. 242) Loaded Allocation tables (p. 248) D-Cubes used as allocation tables (p. 249)

User Guide 235

Chapter 9: D-Links The behavior of D-Links is the same regardless of the type of allocation table used, but there are some minor functional differences between the three options. Subcolumns can only be cut in a local allocation table. Wildcard characters can only be used in a local or loaded allocation tables. Local Allocation table entries may only contain items found in the source and target D-Lists (unless subcolumns are cut) in a local allocation table. Local allocation tables cannot be shared by many D-Links. Sign changing per entry is only supported in the A-Table. You cannot create mixed many-to-one and one-to-many allocations using D-Cube data allocations.

Maintain Allocation Tables

The most significant difference between the three allocation table types is how the allocation tables are maintained. Local allocation tables are maintained manually. There are also options provided for importing allocation table entries from external sources, including mapped ASCII files, two-column delimited ASCII files, ODBC databases, and Cognos packages. However, you cannot create a D-Link to refresh the allocation table entries from these external sources. In saved A-Tables, you can import the source and target items, but not the allocation table entries from external sources, including Mapped ASCII files, delimited ASCII files, ODBC databases, and Cognos packages. If you import source and/or target items from a mapped ASCII file, you can refresh the import. There are options provided to help with updating the allocation table entries. For example, the Restrict to unallocated and Show allocated items options. D-List formatted data in a D-Cube can be used as an allocation table in many D-Links. Different D-Links can use different selections of one D-Cube, so one D-Cube can effectively contain many allocation tables. You can paste the allocation table data from the clipboard, or import it from an external source using a D-Link. You can run the D-Link multiple times to refresh the allocation table data, and then drill down on the allocation table data.

Allocation Table Menu Options

By selecting Allocation Table from the D-Link menu, you can use a number of different Allocation Table options, including: Load (p. 240) Save (p. 243) Match Descriptions (p. 233) Import from a file (p. 243), or from an ODBC or Cognos package source Use D-Cube Data (p. 250)

236 Analyst

Chapter 9: D-Links Use Saved Selection (p. 250)

How Allocation Tables Assign Data

Each entry (row) in the allocation table defines the correspondence of a source and a target item. You can create one-to-one allocations, many-to-one allocations (data from many source items is summed and assigned to one target item), and one-to-many allocations (data in one source item is assigned to many target items). You can mix all three types of allocation in a single allocation table. When the D-Link is run, data is transferred according to the allocation table entries. Suppose a local allocation table is set up as follows:

Source D-List
S1 T1 S2 T2 S3 T3 S4 T4 S5 T5 S Total T Total

Local Allocation table entries

S1

Target D-List
T1

S2

T2

S3

T2

S4

T2

S5

T3

S5

T4

S5 When the D-Link is run: Data from S1 is assigned to T1.

T5

Data from S2, S3, and S4 is summed and the total allocated into T2. Data from S5 is allocated into T3, T4, and T5 (not apportioned across T3, T4, and T5).

When you use an external source for a D-Link, the source dimensions relate to columns in an ASCII file or fields in an ODBC database, where the same item name may appear in many rows (records) . The D-Link will display only the unique names, in the order in which they appear in the ASCII file or database. When the D-Link is run, the allocation table takes the data from all records for which an entry exists in the allocation table, sums it, and assigns the total to the target item. User Guide 237

Chapter 9: D-Links

Example
The source dimension in the allocation table above might relate to an ASCII file containing these two columns:

Item
S1 S1 S2 S2 S3 S3 S4 S4 S5 S5

Value
1 2 3 4 5 6 7 8 9 10

When the D-Link is run, data is assigned like this: (1+2) will be assigned to T1. (3+4 + 5+6 + 7+8) will be assigned to T2. (9+10) will be assigned to T3, T4, and T5.

Navigate Around an Allocation Table

The allocation table is simply a grid with two columns and as many rows as there are allocation table entries in a D-Link. It performs like any other grid in Analyst, similar to D-Cubes. Activate a cell in the allocation table by clicking it. Move the active cell around using standard navigation keyboard shortcuts:

Shortcut
Arrow keys Home

Description
Move the active cell in the direction indicated Makes the first cell in the current row active

238 Analyst

Chapter 9: D-Links

Shortcut
End Ctrl+Home Ctrl+End Page Down Page Up

Description
Makes the last cell in the current row active Makes the first cell in the grid active Makes the last cell in the grid active Moves the active cell down one page Moves the active cell up one page

To select a range of cells in the allocation table grid, drag the first cell to the desired destination. You can also select a range of cells by holding the shift key and pressing the arrow keys above. For example, to highlight the entire allocation table: press Home, then press Shift+Ctrl+End. Tip: To find which source dimension items match an allocation table entry, click a cell in the source side of the allocation table, and the corresponding item from the source dimension is highlighted. If more than one source item matches a source allocation table entry, if the entry contains wildcard characters or the case sensitive option is turned off, all matching items are highlighted. Wildcard characters of * and ? are allowed in a filter or an allocation table. Use a question mark (?) to represent any single character. Use an asterisk (*) to represent any series of characters. Click And to show items that satisfy both criteria. Click Or to show items that satisfy either criteria. If Match Case is selected, it will further refine the filter to match on capitals or small letters. The Name box can be set using the equal symbol (=), or the not equal symbol (<>). Use the equal symbol (=) to show items that meet the criterion. Use the not equal symbol (<>) to exclude items that meet the criteria. For example, if you have an item named P01, and you enter P01*, it will search for anything beginning with the characters P01. If you enter P?????????, it means search for any D-List item that starts with a P and is ten characters long (including spaces). By default the Case Sensitive option is on. Clear the Case Sensitive check box if you want match descriptions to ignore capitalization in the source item names. Typically, this is used when there are inconsistencies in the capitalization of source item names that you want the D-Link to ignore. When the Case Sensitive check box is selected, a highlighted item in the source will only have one matching item in the target. If cleared, there may be many source items highlighted that all match one target item. Data from all matching source items is summed and assigned to the matching target item. The D-Link lists only unique item names. This is always case sensitive regardless of the Case Sensitive option.

User Guide 239

Chapter 9: D-Links

Load an A-Table into a D-Link

You can load any A-Table into a D-Link. Its source and target dimensions do not need to be the same as the dimensions it is pairing in the D-Link. As long as some items in the paired dimensions match the allocation table entries in the A-Table the D-Link will run. If no items match, you will see a "No items match" message.

Steps
1. Open a D-Link. 2. Select an allocation table pairing. 3. From the D-Link menu, click Allocation Table, Load. 4. In the A-Table Select dialog box do the following: Select a library. Select the required A-Table. Click OK.

The D-Link dialog box will display the A-Table library and name. You can also create a new A-Table by saving a local allocation table in a D-Link as an A-Table.

Steps
1. Open the D-Link and select the required allocation table pairing. 2. From the D-Link menu, click Allocation Table, Save. 3. Select a library from the list, then type a name for the A-Table. 4. Click OK.

Change to Matched Descriptions

This feature of Analyst is useful for converting an allocation table pairing within a D-Link to a matched descriptions pairing in a few easy steps.

Steps
1. Open a D-Link. 2. Click an allocation table pairing (indicated by the allocation table pairing indicator. 3. Select Change to Matched Descriptions.

240 Analyst

Chapter 9: D-Links

Change to Allocation
This feature of Analyst is useful for converting a match descriptions pairing within a D-Link to an allocation table pairing in a few easy steps.

Steps
1. Open a D-Link. 2. Click a match descriptions pairing (indicated by the match descriptions pairing indicator. 3. Select Change to allocation. 4. Pair the source items with target items using the local allocation table.

Change Dimension Items in a D-Link

Normally entries in an allocation table corresponding to D-List item names are linked to other D-List item names. If the D-List item name changes, so does the entry in the allocation table. An exception to this is when a subcolumn is cut from a D-List; the entries in the allocation table are no longer linked to the cut down item names. If the items in an external source or Contributor target change, the local allocation table will not be updated and you will have to edit it manually. If an item is deleted from a D-List, occurrences of this item name in allocation table entries are replaced with the text "Item Deleted". The allocation table will still function (providing some complete entries are left) because deleted entries are ignored. These allocation table entries are not removed automatically so that you are aware that an entry has been removed, but when you save the D-Link, you are given the option to remove these entries. Items added to D-Lists are not automatically added to any allocation table.

Other dimensions
Entries in an allocation table corresponding to items in the dimensions of a mapped ASCII file or an ODBC database cannot be linked to the dimension item names. New text in an external source is not recognized. If an item name is deleted from the item name listing, the entries in allocation tables are left unchanged. New item names are not automatically added to any allocation table.

Target Formula Items

Unlike Match Descriptions, a local allocation table (p. 242) can target formula items in a D-List without selecting Match Calculated Target Items. When a D-Link containing an allocation to formula items is run, a break back will be performed in the target D-Cube as if the data had been entered manually. There is nothing impeding your ability to target a formula item and some of its dependent detail items. In this case, data is first transferred to the detail items, then the items are held while a break back is performed on the transferred data.

User Guide 241

Chapter 9: D-Links Do not accidentally target formula items - one break back can change a large amount of data. Remember: if you run a D-Link into a closed D-Cube, the changed data is saved automatically. After a D-Link is run, the status bar at the bottom of the Analyst screen will display calculation messages. If you see "Backward" calculation messages, the D-Link has caused a break back in the target D-Cube, either because formula items have been targeted, or because formula items are held.

Local Allocation Tables (A-Tables)

Typically local allocation tables are used to define the correspondence of a small number of items (frequently just one or two). These may be one-to-one allocations, accumulations, where a source dimension does not include the required subtotal, for example, or an allocation of one source item to a distinct subset of target items. After you have defined a local allocation table, you have the option to save it as an A-Table directly from the D-Link. You can also use D-Cube data or saved selections as a basis for allocation. Often Match Descriptions (p. 233) can replace a local allocation table, but there are many cases when match descriptions is inadequate. For example: when the items will not match, even after cutting subcolumns, when targeting formula items, when performing one-to-many allocations, when you need exclude matching items from the allocation, or when you want to ensure that new items introduced to source or target dimensions are not matched automatically.

Create a Local Allocation Table Pairing

Steps
1. From the File menu, click New, D-Link. 2. Click Source and select a source. 3. Click Target and select a target. 4. Click a source dimension. The items of the selected dimension are displayed in the dimension items section of the D-Link dialog box. 5. Ctrl+click a target dimension that you want to pair with the selected source dimension. Both the source and target dimension names are now selected, and the items contained in each dimension are displayed. 6. Click Allocate items. The paired dimensions are indicated by the allocation table pairing indicator. Click the pairing indicator to change the pairing mode, or break the connection. You can also choose an Allocation option, for example, loading or importing Allocation data, or using D-Cube Data. See Allocation Table Menu Options (p. 236).

242 Analyst

Chapter 9: D-Links 7. Click an item in the source. 8. Click an item in the target. 9. Repeat as necessary. If required, you can change the Case Sensitive (p. 234) option, select a Dump item (p. 256), and Cut Subcolumns (p. 259) from the source and/or target dimensions. To break the connection, click the pairing indicator and select Break Connection. You cannot pair a dimension that is part of an existing pairing; you must first break the existing connection. You cannot pair an unpaired dimension from which a selection of items has been made; you must first clear the selected items list.

Save a Local Allocation Table as a Saved A-Table Steps

1. Open a D-Link. 2. Select an allocation table pairing. 3. From the D-Link menu, click Allocation Table, Save. 4. In the Save A-Table As dialog box do the following: Select a library to which you will save the A-Table. Type a name for the new A-Table. Click OK.

Import a Local Allocation Table

You may import an allocation table from a mapped ASCII file, an unmapped ASCII file, providing it is a delimited file with exactly two columns or an ODBC database into a D-Link. Note: When you import an allocation table, all existing entries in the allocation table are removed. When importing an allocation table, all entries that correspond to D-List items must be valid D-List item names. For more information, see "Edit Local Allocation Table Entries" (p. 246).

Steps
1. Open the appropriate D-Link. 2. Select the existing allocation pairing or define a new allocation pairing. 3. From the D-Link menu, click Allocation Table, Import from File. 4. Select the required source type from the menu. 5. In the Specify allocation table dialog box do the following: Select the required source dimension and click Source.

User Guide 243

Chapter 9: D-Links Select the required target dimension and click Target. Click OK.

The allocation table entries are imported and are displayed in the D-Link dialog box.

Delete Entries in a Local Allocation Table

Steps
1. Open a D-Link. 2. Select the local allocation. 3. Select the entries to be deleted. Note: You must select the entire entry - both the source and the target column. 4. Press Delete to remove the selected entries from the allocation table. - or Right-click the selected range and select Delete Highlighted Rows, or Delete all Rows from the shortcut menu. The allocation table entries are deleted.

Use Wildcard Characters in Local Allocation Tables

If the source dimension for an allocation table is not a D-List, you can manually edit the source allocation table entries and introduce wildcard characters.

Steps
1. Open the appropriate D-Link. 2. Double-click a cell in the local allocation table, or press f2. 3. Type an asterisk ( * ) to represent any number of characters (including none) and a question mark ( ? ) to represent any single character. When you select a source entry containing wildcard characters, all the matching items are highlighted in the source dimension items list. As an example, look at the following allocation table. 612 Canada 613 Central America 614 Latin America Americas Americas Americas

This could be replaced by various one line entries.

244 Analyst

Chapter 9: D-Links

61*

Americas

61? *

Americas

61?*

Americas

?1?*

Americas

The asterisk ( * ) represents any character in that position and all remaining positions. Therefore, the asterisk ( * ) should only be used at the end of entries. For example, all of the following entries (beginning with characters 61) are assigned to Americas. 61* Latin America 61*Latin 61*Any text here Americas Americas Americas

Priority of source entries: As a result of including wildcard characters in source allocation table entries, one source item may be assigned to a number of target items. A matching source item will be assigned by the first matching allocation table entry. For example, the following allocation table means: Assign items beginning with characters 619 to UNALLOCATED. Then, assign items not already assigned and beginning with characters 61 to Americas. 619* 61* UNALLOCATED Americas

In contrast, the following allocation table means: Assign items beginning with characters 61 to Americas. This allocation table will never assign any records to UNALLOCATED. 61* 619* Americas UNALLOCATED

Suppose the source item 619 Test Market is found when the D-Link is run. It will be assigned by the first matching source entry in the allocation table. The first allocation table will assign this item to UNALLOCATED. The second allocation table will assign it to Americas. This can be a very useful feature, but you must remember to order the allocation table entries correctly. And always take care to avoid including ambiguous source entries accidentally.

User Guide 245

Chapter 9: D-Links

Edit Local Allocation Table Entries

When you create a local allocation table in a D-Link, you can only include entries for items present in the dimension at that time. If the source for an allocation table is not a D-list, you can build an allocation table with entries not yet present in the source dimension by editing lines within the local allocation table. Normally entries in an allocation table corresponding to D-List item names are linked to the D-List item names. If the D-List item name changes, so does the entry in the allocation table. An exception to this is when a subcolumn is cut from a D-List, the entries in the allocation table are no longer linked to the cut down item names. Entries in an allocation table corresponding to items in the dimensions of a mapped ASCII file or an ODBC database are not linked to the dimension item names. If the dimension item name changes, the entry in the allocation table does not update. If a dimension item disappears, the allocation table entry is not removed. As a result, you may edit source allocation table entries that correspond to dimensions other than D-Lists. To edit the contents of a cell in an allocation table, double-click the cell, or select the cell and press f2. Because these allocation table entries are not linked to dimension item names, they may contain virtually anything. If the source dimension relates to a column in an ASCII file, the column may be empty in some rows. You can include an empty source allocation table entry (an empty cell in the source column) if you want to assign data in these rows to a target D-List item. (Normally you would want them treated as dump items.) Because allocation table entries for a D-List are linked, you cannot edit entries in an allocation table corresponding to D-List item names.

Reorder Lines in a Local Allocation Table

You cannot reorder entries in a local allocation table. Each new pairing you create is introduced as a line at the end of the table. The order of entries can only influence the behavior of an allocation table when the source for an allocation table is not a D-List and wildcard characters have been used in source entries. The only way you can re-order is by deleting individual rows and recreating them at the end of the table.

Add Entries to a Local A-Table

Add a Single-Line Entry to a Local Allocation Table Steps
1. Open a D-Link. 2. Select the local allocation. 3. Click a source dimension item.

246 Analyst

Chapter 9: D-Links 4. Click a target dimension item. The allocation table entry will be added at the end of the existing allocation table.

Duplicate allocation table entries

You cannot add to an allocation table an entry that already exists. If you attempt to add a duplicate entry, the original entry is left in place.

Add Many-to-One Entries to a Local Allocation Table Steps

1. Open a D-Link. 2. Select the local allocation. 3. Drag to select source items. 4. Click one target item only. If you select more than one target item, you will not create many-to-one allocation table entries. The allocation table entries are added.

Add One-to-Many Entries to a Local Allocation Table Steps

1. Open a D-Link. 2. Select the local allocation. 3. Select one source item. 4. Select the target items. Non-adjacent items can be selected by Ctrl+clicking the items. Multiple adjacent items can be selected by dragging. The allocation table entries have been added.

Add Multiple-Line Entries to an Allocation Table Steps if the source items are adjacent
1. Open a D-Link. 2. Select the local allocation. 3. Drag to include items in the source range (or press Shift+Down Arrow). 4. Drag to include items in the target range.

User Guide 247

Chapter 9: D-Links Note: Highlight exactly the same number of source and target items. If the number of source and target items you select is different, then only some of the required entries will be added to the allocation table. 5. The allocation table entries are added. Note: Duplicate allocation table entries If you attempt to add a number of lines, some of which are duplicates, the new lines will be appended at the end of the existing allocation table, the duplicates lines are not added (the original entries are left in place).

Steps if the source items are not adjacent

1. Open a D-Link. 2. Select the local allocation. 3. Click an item in the source list. 4. Ctrl+click the remaining items in the source list. You can highlight adjacent items by holding Ctrl and dragging. 5. Drag to include items in the target list. Note: You cannot select multiple non-adjacent items from the target items list (the allocation table entries are added when you first release the mouse button in the target list). The allocation table entries are added.

Loaded Allocation Tables

Saved Allocation Tables
Saved A-Tables are used for larger allocation tables in a D-Link. You should use a shared A-Table if you want to use an allocation table in more than one link, or if you want different users' D-Links to use a centrally maintained allocation table. You can also use a large A-Table in a D-Link, even if only a subset of the source and/or target items are found in that D-Link. You may also use a saved A-Table simply to take advantage of the extra functions it offers - sign changing per line, highlighting unassigned items, and so on. For more details about Allocation Tables see "Local Allocation Tables (A-Tables)" (p. 242). To edit an A-Table while in the target D-Link, in the D-Link editor, double-click the name of the A-Table to open and edit it.

Copy and Paste Allocation Table Entries

You can use the Microsoft Windows clipboard to copy and paste an allocation table into a D-Link. You can copy as many allocation table entries as you want, from another D-Link or another Windows application. To paste from the clipboard, the required number of rows must already exist

248 Analyst

Chapter 9: D-Links in the allocation table. You cannot paste the contents of the clipboard if it contains more rows than the selected paste area. If both the source and target for an allocation table are D-Lists, all the entries pasted must be valid entries in both D-Lists. If they are not, they are rejected. If only the target for an allocation table is a D-List, all the pasted entries must contain valid entries in the target; the source entries may contain anything at all.

D-Cube Allocations
A D-Cube allocation uses a slice of a D-Cube consisting of the rows dimension and one D-list formatted column on a single page to act as an allocation table within a D-Link. A D-Cube allocation D-Link is created and run in the same way as a normal D-Link except that a D-Cube is used for data allocation when pairing source and target dimensions. You can run D-Cube allocations inversely and drill down on data transferred by them in the usual way.

Example
If you have a D-Cube that has a D-Cube allocation that allocates cost centers to both divisions and managers as follows:

Divisions
Cost Center 1 Cost Center 2 Cost Center 3 Cost Center 4 Cost Center 5 Cost Center 6 Cost Center 7 Cost Center 8 Cost Center 9 Cost Center 10 Cost Center Total Usa Germany France U.K. USA Germany France U.K. USA Germany

Manager
Manager 2 Manager 3 Manager 1 Manager 1 Manager 2 Manager 4 Manager 4 Manager 1 Manager 3 Manager 3

User Guide 249

Chapter 9: D-Links Cost Center 1 is located in USA and is managed by Manager 2. Similarly, Cost Center 2 is located in Germany and is managed by Manager 3. In this manner, the remaining cost centers are each located in various countries (USA, Germany, England, or France) and managed by various managers (Manager 1 through Manager 4). Note that one manager can be responsible for multiple cost centers in different countries.

Use D-Cube Data in Allocation

If you select Use D-Cube Data from the Allocation Table menu, you select an allocation D-Cube and specify the selection and slice from the allocation D-Cube.

To use D-Cube Allocations

1. Select a D-Cube in the Select D-Cube dialog box. 2. Select an item in the Select dialog box and then click Move >>. 3. Click Slice. 4. Specify the row and column in the Select Row and Column dialog box, and then click OK. 5. Click OK in Select dialog box. 6. Select DList as source or Cube-data as source in the Select Row and Column dialog box, and then click OK.

D-Cube Allocations vs. Lookup and Accumulation D-Links

You can replace any one-dimensional lookup or accumulation D-Link (that is, only one format D-List paired in the D-Link) with a normal D-Link using an allocation table. This becomes a practical option if you use D-Cube allocation tables, but remember that only an accumulation D-Link will perform a break-back allocation when run inversely.

Use Saved Selection in Allocation

As an alternative, from the D-Link menu, you can click Allocation Table, Use Saved Selection to select a predefined selection from the allocation D-Cube.

Select and Slice an Allocation D-Cube

The selection and slice taken from the allocation D-Cube must: have just one page. have just one data column. contain appropriate allocation table entries in the row headings and the data column.

250 Analyst

Chapter 9: D-Links

Enable D-Link to recognize the A-Table

When using a D-Cube as an A-Table in a D-Link, perform the following so the D-Link can recognize the A-Table.

Steps
1. Open the D-Link. 2. Click on the source dimension and click Ctrl + the target dimension. 3. Click Allocate items. 4. From the D-Link menu select Allocation table and Use D-Cube data. 5. Click an allocation D-Cube in the list. 6. Make a selection from the allocation D-Cube, and specify the slice. The selection and slice taken from the allocation D-Cube must: Have just one page Have just one data column Contain appropriate allocation table entries in the row headings and the data column

7. Specify whether the source side of the allocation table is a D-List or a data column.

Complex Allocation D-Cubes

Often allocation D-Cubes are quite simple: They are two-dimensional and contain only two data columns. The simplest possible allocation D-Cube would have just two dimensions and only one data column. In this case, making a selection is unnecessary, but the slice must still be specified correctly. You can create an allocation D-Cube containing any number of dimensions. For example, you can add a versions dimension to an allocation D-Cube if you need to use different versions of four possible allocation tables. If you create a multidimensional allocation D-Cube, you should only include one D-List containing D-List formatted items. If you use a more complex allocation D-Cube in a D-Link you must keep in mind the rules given in the Selecting and Slicing an Allocation D-Cube section when cutting down the selection and taking a slice (the D-Link must have one data column and one page). If you specify a selection and slice that a D-Link cannot interpret as an allocation table, you will see the "Nothing to Transfer" message when the D-Link is run. You will have to return to the D-Link and specify the selection and slice from the allocation D-Cube again. If the D-Link can interpret your selection and slice as an allocation table, but the allocation table does not match the source and/or target items, you will see a 'No items match' message.

A Default Slice in a D-Cube Allocation D-Link

When you use a D-Cube selection as an allocation table in a D-Link the default slice is defined like this: If there is a timescale D-List in the D-Cube, the timescale D-List will form the columns and User Guide 251

Chapter 9: D-Links the first D-List included in the D-Cube will form the rows. If there is no timescale D-List in the D-Cube, the first D-List included in the D-Cube will form the rows and the second D-List included in the D-Cube will form the columns. If you are not sure of the order of D-Lists in a D-Cube, they can be reordered by opening the D-Cube, clicking the D-Cube menu, pointing to Dimensions, and then clicking Reorder. You can eliminate the need to re-slice an allocation D-Cube by including the D-Lists in the following order when you create the D-Cube.

Steps
1. The D-List that you will use most frequently as the source (or the target) in allocation tables. Usually, this is the D-List at the lowest level of the hierarchy. 2. The D-List containing the D-List formatted items. 3. Any other D-Lists.

Editing a D-Cube Allocation D-Link

When you use a D-Cube selection as an allocation table in a D-Link, you have the option when editing the D-Cube allocation either to view the slice which has already been defined and then edit it, or to choose a new cube completely and define a new slice.

Step
With the D-Link open, click on the pairing you wish to edit.

From the D-Link menu click Allocation Table and Use D-Cube Data. The question Do you wish to edit the existing selection? appears. To view and edit the existing selection click Yes. To choose a new cube click No and select a new cube.

Execution Modes
The execution mode determines how data transferred by a D-Link is combined with existing data within a target area (p. 253) of a target D-Cube. There are four D-Link execution modes: Fill (p. 253), Substitute (p. 253), Add (p. 253), and Subtract (p. 253). To set the execution mode, select a setting from the Mode list. The Fill and Substitute modes have a special meaning when applied to Lookup (p. 266) and Accumulation D-Links (p. 272).

Run D-Links inversely

If you run a D-Link inversely (p. 227), the execution mode still applies. The definition of the target area, and the behavior of each of the modes is the same as when a D-Link is run normally, except that the source for the normal D-Link becomes the target for the inverse D-Link (and vice-versa).

252 Analyst

Chapter 9: D-Links Note: You cannot inversely run a D-Link that has an external source because a D-Link cannot target an external source (which also means that the source for an inverse D-Link will always be a D-Cube).

Fill Mode
Fill is the default execution mode. All numbers within the target area (p. 253) of the D-Cube are replaced with the transferred numbers. For a D-Link that uses a D-Cube as its source, Fill and Substitute are identical in behavior for regular D-Links because the target area of the target cube consists only of cells for which data can be found in the source cube. If a D-Link is from an external source or is a look-up or accumulation D-Link then fill mode will set untargeted cells to zero whereas substitute will leave them untouched. In Analyst, for look up and accumulation D-Links (p. 262) any cell within the target area of the cube is set to zero if no data exists in the source cube for that cell. The fill is applied first to zero the data, and then the data is written as if in substitute mode. In Contributor, for a look up and accumulation links that targets both detail and calculated items at the same time, the zeros to fill and the data to set are merged into a single update of the target cube. The different methods for Fill for Analyst and Contributor causes different breakback behavior to occur. If the Analyst method is required for Contributor, you can use one link to zero the data, then an accumulation link running in substitution mode. For a D-Link using an external source, any cell within the target area of the cube is set to zero if no data exists in the source file for that cell.

Substitute Mode
Numbers within the target area of the D-Cube are replaced by the transferred data, but if no data is found in the source for a particular cell, the data in that cell is left unchanged. For look up and accumulation D-Links any cell within the target area of the cube is left untouched if no data exists in the source cube for that cell. For a D-Link using an external source, any cell within the target area of the cube is left untouched if no data exists in the source file for that cell.

Add Mode
Transferred data is added to existing data within the target area of the D-Cube.

Subtract Mode
Transferred data is subtracted from existing data within the target area of the D-Cube.

The Target Area

The target area is the maximum portion of a D-Cube that can be affected by a particular D-Link.

User Guide 253

Chapter 9: D-Links The target area can be described as a selection of items from each of the D-Lists making up the D-Cube, like any other D-Cube selection. The treatment of each target D-List in a D-Link determines which items are within the target area selection.

Type of Target D-List

Unpaired

Target Area Selection

The items selected In an empty selection: all detail items

Allocation table pairing Match descriptions pairing

Items with an entry in the target side of the allocation table With a D-Cube as source: Matched detail items. With an external source: All detail items, except on the DATA dimension where only items matched in the source are selected.

Notice that the type of source completely changes the target selection for a match descriptions pairing. This means that when importing data from an external source, all unmatched target items in a match descriptions pairing are set to zero if the execution mode is set to Fill (p. 253). If a D-Cube is open, a D-Link targeting the D-Cube can only change data within the open selection. Thus, you can run D-Links into limited selections of D-Cubes if you reduce the target area of the D-Link.

Dump Options
The dump option determines the treatment of all unassigned records found in an external source. It is not applicable to D-Links that use a D-Cube as their source. There are four D-Link dump options: Ignore, Edit, Print, and File. The default setting is Ignore, where unassigned records are simply overlooked (any dump item set in an allocation table or match descriptions pairing is still operative). The other three settings will produce a report listing all unassigned records.

To set or change the dump option

1. Create or open a D-Link. 2. Match source and target items. 3. Select a setting from the Dump drop-down box. For allocation table and match descriptions pairings, you can set a dump item in the target D-List. The dump item and the D-Link dump options are not independent - data assigned to a dump item is excluded from the unassigned records report.

254 Analyst

Chapter 9: D-Links

When will records be unassigned?

Items in an external source may be unassigned by the D-Link for the three reasons that follow: For an unpaired source dimension: The items are not to be included in the selected items list. For a match descriptions pairing: The items are not matched (not selected in the source items list). For an allocation table pairing: The items do not have an entry in the source side of the allocation table. Note 1: In the case of the DATA dimension, the source items in the map are cut down so that only those matched with target items by the link are considered to be part of the source, so the unmatched items from the DATA dimension are not considered unassigned and will not be recorded by the Dump Option. Note 2: Records containing bad data will be imported. Any entry in a data column that cannot be interpreted as a number is assumed to be zero and is imported as such. Bad data will not be included in any unassigned records report (unless the record is bad for some other reason). If you drill down on a cell to which bad (zero) data has been assigned, you will see the bad data in the drill down report.

Edit
Unassigned records are presented in a dialog box within Analyst. The field (or fields) within each record that could not be assigned are highlighted in red, enabling you to identify the errors in the source file. You can copy a selection from this report, or you can copy the entire report to the Windows clipboard if required.

Print
Unassigned records are sent directly to the default printer.

File
Unassigned records are written to a delimited ASCII file of your choice.

Steps
1. Open or create a D-Link. 2. Match source items with target items. 3. Select File from the Dump drop-down box. 4. In the Select Dump file dialog box, select a destination directory and a name for the file. 5. Click OK.

User Guide 255

Chapter 9: D-Links The chosen directory and filename are then shown at the top of the D-Link dialog box, along with a button marked with an ellipsis (. . .). Click this button to change the dump file location and/or name.

Dump Item
This option allows you to assign data from all source items in a D-Link that do not have an entry in an allocation table to a single dump item in a target D-List. When the D-Link has run, you can drill down on any data assigned to the dump item to see which records were not matched. If necessary, you can edit the source data file or the local allocation table. You may also need to add an item to the target D-List. The dump item is only applicable when data is being imported from an external source, either a mapped ASCII file, an ODBC database, or Contributor data. If the source for a D-Link is a D-Cube, the dump item is inactive; source items excluded from the allocation table will not be transferred to the target.

To set a dump item within a D-Link

Select an item from the target D-List from the Dump Item box. Do not select a formula item from the target D-List as the dump item. Unmatched records will not be assigned to any formula item.

To discontinue using a dump item

Select None from the Dump Item box. The dump item is one of the tools provided in the D-Link for dealing with bad records. The other is the Dump option, which allows you to control how bad records are treated by the D-Link as a whole. They may be ignored, edited in a screen that highlights the problem area, printed, or output to an ASCII file of your choice. Dump items and Dump options are not independent: Any records assigned to a dump item are excluded from the bad records report. Data in an external source will often contain total lines, which can be very useful for reconciliations. For example, you could add an extra CHECK TOTAL item to a target D-List and allocate a grand total found in the source file to it. When the D-Link has run, you can compare the values in CHECK TOTAL to the values in TOTAL (an existing D-List item that calculates a total for the imported data). Frequently an extra D-List item is added to perform this comparison (containing the formula CHECK TOTAL - TOTAL, for example).

256 Analyst

Chapter 9: D-Links

Drill Down on Data Assigned to Dump Items

If you want to know what data has been assigned to various dump items, drill down on the appropriate numbers transferred into the D-Cube.

Step
Select a cell or a range of cells, then click the Drill Down button.

Dump Items Used with Dump Options

Data assigned to a dump item is excluded from the unassigned records report. An unmatched record cannot be caused by a D-List item that is not included in the selection of items from the source dimension. Because it is not included in the selection, it is not imported into the target D-Cube at all and cannot get assigned to a dump item.

Scale and Round Data within a D-Link

A D-Link can scale and/or round numbers when it is run. All numbers transferred by a D-Link will be scaled and/or rounded by the same factor. It is not possible to set scaling (p. 257) or rounding factors (p. 257) that affect only a subset of data transferred by a D-Link. In a saved A-Table, it is possible to change the sign for selected lines (multiply data assigned by selected allocation table entries by -1). Click the button to use the edit signs feature. If you want to scale or round only some of the numbers a D-Link transfers, you will normally have to create a D-Link, or edit the existing D-Link and save a copy. An alternative is to open only the required limited selection of the target D-Cube, temporarily set a scaling or rounding factor for the D-Link, and then run it. Finally, remove the scaling or rounding factor.

Scaling Factors
You can enter as a scaling factor any number you like, positive or negative. When a D-Link is run, all data transferred is divided by this number. If a rounding factor is also set, data is scaled before it is rounded. Note: Scaling factors can also be set in a D-List. For more information about scaling factors within D-lists, see "Set Scaling Factor within a D-List" (p. 113).

Rounding Factors
The available rounding factors range from six decimal places to millions. When the D-Link is run, all data transferred is rounded to the chosen accuracy. For example:

User Guide 257

Chapter 9: D-Links

Rounding Factor
6 decimals 4 decimals 1 decimal Units Tens Millions

Normal Number
1.123456789 1.123456789 1.123456789 123456789.1 123456789.1 123456789.1

Rounded Number
1.123457 1.1235 1.1 123456789 123456790 123000000

Data midway between two rounded numbers is rounded up. For example:

Rounding Factor
Units Units Tens Tens Millions Millions

Normal Number
0.5 0.499 -5 -5.01 1500000 1499999

Rounded Number
1 0 0 -10 2000000 1000000

If a scaling factor is also set, data is scaled before it is rounded.

Set Scaling and Rounding

Steps
1. Open the appropriate D-Link. 2. From the D-Link menu, click Options. 3. In the D-Link Options dialog box: Select the Use Scaling check box. Type a number (positive or negative) in the Scaling Factor box. Select the Use Rounding check box.

258 Analyst

Chapter 9: D-Links Select a setting from the Rounding Factor list.

4. Click OK. When you next run the D-Link the scaling and rounding options you have set will be active. Remember to save the D-Link if you want to make the change permanent.

Subcolumns
By default, Match Descriptions (p. 233) analyzes the entire text of the dimension item names. However, it is possible to match descriptions on a limited portion of the item names by cutting a subcolumn (p. 259). This is particularly useful when you wish to match items or exclude invalid descriptions. When a subcolumn is cut, the D-Link dialog box displays the cut-down item names. Only entries that are unique within the subcolumn are listed. Often subcolumns are cut when dimension item names contain a unique code and a description (for example, a product code and description). When importing D-List items into a D-List on a regular basis, it is often useful to define which part of a D-List item contains a unique code. This lets you test whether it is a genuinely new item or just a slightly different spelling of the description. The unique portion of the D-List item is case sensitive and takes into account leading and trailing spaces. After you have set the unique part of a D-List item you can not type, paste, or import any duplicate items using a D-Link beginning with that code. For example, a typical product code consists of a code followed by a description such as P03 Camping Gear (P03 is the code, Camping Gear is the description). If, the following month, you are updating the list of product codes and an item named P03 Campin' Gear displays, the program recognizes the code (P03) and ignores the different spelling of the description (Campin' Gear). This prevents duplication of the same product. Remember, if you need to cut subcolumns from the dimensions of a Mapped ASCII file, you can do it in the D-Link as described in the Cut a Subcolumn examples, or you can do it in an ASCII file Map. Note: When working with virtual dimensions, it is impossible to cut a subcolumn.

Cut a Subcolumn
You can cut a subcolumn from any dimension in a D-Link except an unpaired dimension (p. 229) where a selection (p. 232) of items has already been made. However, it is recommended that you pair the dimensions before cutting subcolumns. For a match descriptions pairing (p. 233), this enables you to see which items match as you cut the subcolumns.

User Guide 259

Chapter 9: D-Links If you change the pairing mode (Match Descriptions or Allocate items), or break the connection between paired dimensions, any cut subcolumns are preserved. Even if you change the target for a D-Link, subcolumns cut from source dimensions will be preserved (and vice versa).

Steps
1. From the File menu, click New, D-Link. 2. Click Source and select a source. 3. Click Target and select a target D-Cube. 4. Click a source dimension. The items of the selected dimension are displayed in the dimension items section of the D-Link dialog box. 5. Ctrl+click the target dimension that you want to pair with the selected source dimension. Both the source and target dimension names are now selected, and the items of both dimensions are displayed in the D-Link dialog box. 6. Click Match Descriptions. 7. Click the Cut Sub-Column icon. In the Sub-Column dialog box, click at the start of the subcolumn (to the left of the first character of the subcolumn). Note: If the subcolumn starts at the beginning of the item name (at character 1), you do not need to define the start of the subcolumn; you can skip to step 9. The dimension items are displayed in a non-proportional font (each character is the same width) . A ruler at the top of the screen indicates the character number at the end of a column of characters. Left-click - to create a column break Click+drag - to move a column break Right-click - to delete a column break

8. Click at the end of the subcolumn (to the right of the last character of the subcolumn). 9. Click OK. You will return to the D-Link dialog box. The character range in square brackets after the dimension name indicates that a subcolumn has been cut, and displays which characters are included in the subcolumn.

Change the Position of an Existing Subcolumn

Steps
1. Open the relevant D-Link.

260 Analyst

Chapter 9: D-Links 2. Select the required pairing. 3. Double-click the name of the dimension that has the subcolumn you want to change. 4. In the Subcolumn dialog box, drag the column break you want to move. If the existing subcolumn starts at the beginning of the item name and you want the new subcolumn to start at another character, click the point at which you want the subcolumn to start. 5. Click OK. You will return to the D-Link dialog box. The character range in square brackets after the dimension name indicates which characters are included in the new subcolumn.

Clear a Subcolumn
Steps
1. Open the relevant D-Link. 2. Select the required pairing. 3. Double-click the name of the dimension that has the subcolumn you want to change. 4. In the Subcolumn dialog box, right-click the column breaks that you want to remove. If the existing subcolumn starts at the beginning of the item name, you will only need to right-click the end of subcolumn break. 5. Click OK. You will return to the D-Link dialog box. The absence of a character range in square brackets after the dimension name indicates that the subcolumn has been removed.

Duplicate Target Items

When you cut a subcolumn from a target D-List, it is important that the D-List contains unique codes within the subcolumn. Items with duplicated codes are removed from the item listing, and Match Descriptions (p. 233) will only assign data to the first instance of the matching item. For example, consider a target D-List containing the following three items: A001 Description 1 A001 Description 2 A002 Description 3 If the first four characters from this D-List are cut as a subcolumn, the D-Link will display only two items: A001 A002

User Guide 261

Chapter 9: D-Links If A001 is highlighted by match descriptions and the D-Link is run, data will only be assigned to the item A001 Description 1. You may find that duplicate formula items are removed from the list when a subcolumn is cut. This does not matter, as match descriptions does not target formula items unless otherwise specified by clicking Match Calculated Target Items. For example, many D-Lists have a structure like the following: A001 Xx Xxxx A002 Xxxx Xxx Total Group A B001 Xxxx Xx B002 Xxx Xxxxx Total Group B If the first four characters from this D-List are cut as a subcolumn, the D-Link will display only one instance of "Total."

Lookup and Accumulation D-Links

Database D-Cube
A database D-Cube differs from an ordinary D-Cube in that it contains numeric data and text data, which classifies the numeric data. To enable cells to contain text data, The D-List items must be D-List formatted. A D-List format lets you enter text from another D-List in a row or a column.

Terminology
An item in a D-List that has been given a D-List format is called a D-List formatted item. A D-List containing a D-List formatted item is referred to as a lookup D-List. The D-List chosen for the format is called the format D-List.

Sparse D-Cube
A sparse D-Cube contains limited dimensions and sparse data. For example, employee cost data can be held in an ordinary three-dimensional D-Cube, containing the D-Lists Employees, Levels, and Divisions. The following table shows one page of this cube for Employee A.

Division 1
Level 1 0

Division 2
0

Division 3
0

Total
0

262 Analyst

Chapter 9: D-Links

Division 1
Level 2 Level 3 Level 4 0 0 0

Division 2
0 80,000 0

Division 3
0

Total
0 80,000

A sparse D-Cube stores the salary data inefficiently. Each page of this D-Cube contains just one number; thus, this D-Cube is a sparse D-Cube. However, you can transfer data from this D-Cube into a D-Cube containing a Divisions D-List using a normal D-Link. Alternatively, the data in the D-Cube above can be held in a simpler two-dimensional D-Cube.

Employees
A B C D E

Salary
80,000 50,000 45,000 60,000 20,000

This is also a sparse D-Cube, and there are two problems with it: First, you can no longer see which level or division an employee belongs to. Secondly, it is awkward to create a normal D-Link to transfer this data into another D-Cube that contains a Divisions D-List. You have to create an allocation table to assign each employee separately. You can store this allocation of Employees to Divisions in a saved A-Table and use this in D-Links. Alternatively, you can create a special allocation D-Cube, and use the Allocate Using D-Cube Data option when you create the D-Link. For more information see "Use D-Cube Data in Allocation " (p. 250). The latter approach is preferable because an allocation D-Cube is easier to maintain and update than a saved A-Table. However it would be more efficient to store the data in a two dimensional cube as follows. Here the D-List forming the columns would have two D-List formatted items, Division, formatted on a Divisions D-List and Level formatted on a Levels D-List. Only the Salary item would be numeric.

Employees
A B

Division
Division 3 Division 2

Level
Level 2 Level 4

Salary
80,000 50,000

User Guide 263

Chapter 9: D-Links

Employees
C D E

Division
Division 1 Division 4 Division 2

Level
Level 5 Level 3 Level 6

Salary
45,000 60,000 20,000

Lookup and Accumulation D-Link Restrictions

There are certain restrictions imposed when creating lookup or accumulation D-Links. You cannot create a simultaneous lookup and accumulation D-Link. If both the source and target D-Cubes contain D-Lists with D-List formatted items, the D-Link will display format D-Lists in both the source and target D-Cubes. However, after you have paired a format D-List in the source, you cannot pair a format D-List in the target.

You cannot use multiple lookup D-Lists. If a D-Cube contains multiple D-Lists each with D-List formatted items, the D-Link will display the format D-Lists belonging to more than one lookup D-List. After you have paired a format D-List belonging to one lookup D-List, you cannot pair other format D-Lists that belong to a different lookup D-List. You may, however pair multiple format D-Lists if they all belong to the same lookup D-List. You can only create lookup and accumulation D-Links when the source for the D-Link is a D-Cube or Contributor cube. You cannot use the scaling or rounding options (p. 257) in a lookup or accumulation D-Link. You cannot use a saved A-Table that uses the sign changing option in a lookup or accumulation D-Link. You cannot cut subcolumns (p. 259) from a format D-List. You cannot make many to one allocations on the real dimensions of a lookup D-Link

Lookup D-Links
Lookup D-Links are D-Links that literally "look up" data from a source D-Cube based on text data using a database D-Cube as a target. You define a D-Link as a lookup D-Link by selecting the Link Type, pairing a normal D-List from a source D-Cube with a format D-List from a target D-Cube. Normally source and target D-Lists are formatted to use the same D-List. For example create a two dimensional cube containing Salaries based on levels.

264 Analyst

Chapter 9: D-Links

Level
Level 1 Level 2 Level 3 Level 4 Level 5

Salary
60,000 70,000 80,000 85,000 100,000

Use this as a source D-Cube and set up a D-Link to transfer the salary to a Costs D-Cube based on Level. When this D-Link is run, data is transferred to the employees in the Employee Costs D-Cube based on level:

Employees
A B C D E

Division
Division 1 Division 1 Division 2 Division 3 Division 2

Level
Level 3 Level 2 Level 5 Level 1 Level 1

Salary
80,000 70,000 100,000 60,000 60,000

In this example, the Employee Costs D-Cube is the database D-Cube, and the Annual Salaries D-Cube is the source D-Cube. The paired format D-List (Levels) drives the lookup D-Link when it runs.

One-dimensional Lookup D-Links

A one-dimensional lookup D-Link is based on two criteria: It must contain at least one D-List formatted item that can be used as a target. Only one format D-List is paired in the D-Link.

Two-dimensional Lookup D-Links

A two-dimensional lookup D-Link is based on two criteria. It must contain at least two D-List formatted items that can be used as targets.

User Guide 265

Chapter 9: D-Links Two format D-Lists are paired in the D-Link.

Create a Two-dimensional Lookup D-Link

Creating a two-dimensional lookup D-Link is the same as creating a one-dimensional lookup D-Link, except that instead of pairing only one virtual dimension, you pair two virtual dimensions.

Lookup D-Link Execution Modes

The Fill and Substitute execution modes (p. 253), set using the Mode drop-down box in a lookup D-Link, have a special function when applied to lookup and accumulation D-Links. If a dimension formatted on a D-List contains an entry that is not matched by a D-Link, the corresponding data will be set to zero if the mode is set to Fill, and left untouched if the mode is set to Substitute.

Unpaired Dimensions in Lookup D-Links

If you leave a lookup D-List unpaired, you should never leave it with an empty selection. Nor should you select the D-List formatted item that is used to drive the D-Link from the lookup D-List.

Steps
1. Click an unpaired dimension. The items of the selected dimension are displayed in the dimension items section of the D-Link dialog box, ready for you to make a selection of items. 2. Click Select to make a selection of items. 3. Double-click an item in the dimension items list to move it to the Selected items box. To select more than one item, highlight the required items in the dimension items list, then click the arrow button pointing in the direction of the selected items list. Repeat steps 1 through 3 for all remaining unpaired source and target dimensions.

Run Lookup D-Links Inversely

You can run lookup D-Links inversely the same as normal D-Links.

Steps
1. Open the target D-Cube - the target database for the normal D-Link, which will be the source when the D-Link is run inversely. 2. Click the Run D-Link button. 3. Select a D-Link from the list of those targeting the D-Cube and click Inverse. A lookup D-Link run inversely does not become an accumulation D-Link. For example, the source D-Cube (Salary by Level) for a lookup D-Link contains the D-Lists, Levels and Salaries, and the following data.

266 Analyst

Chapter 9: D-Links

Salary by Level D-Cube Level

Level 1 Level 2 Level 3 Level 4 Level 5

Salary
60,000 70,000 80,000 85,000 100,000

The target D-Cube (Employee Costs) for the lookup D-Link contains the D-Lists, Employees and Employee Attributes. Employee Attributes contains two D-List formatted items Division and Levels. It contains the following data.

Employee Attributes D-Cube Employees

A B C D E

Division
Division 1 Division 1 Division 2 Division 3 Division 2

Level
Level 3 Level 2 Level 5 Level 1 Level 1

Salary
80,000 70,000 100,000 60,000 60,000

In the lookup D-Link to transfer data from Salary by Level to Employee Costs, the item Salary within the target D-List Employee Attributes, is paired with the item Salary in the source. The target D-List Employees is left with an empty selection (p. 232) (auto-allocated D-Lists). The source D-List Levels, is paired with the target format D-List Levels using match descriptions. In a lookup D-Link, unpaired D-Lists in the target D-Cube that do not contain a selection are referred to as auto-allocated D-Lists (with the exception of the lookup D-List). Each (detail) item in an auto-allocated D-List contains one entry from the driving format D-List (that is, in the driving D-List formatted item in the lookup D-List). This entry determines which data is taken from the driving source D-List. If this lookup D-Link is run inversely with the execution mode set to Substitute, the data in Salary by Level is unchanged.

User Guide 267

Chapter 9: D-Links However, if it is run with the mode set to Fill, the data in Salary by Level will be changed as follows because there is no data in the Employee Attributes cube for level 4 - so this is set to zero.

Level
Level 1 Level 2 Level 3 Level 4 Level 5

Salary
60,000 70,000 80,000 0 100,000

If Employee Costs contains inconsistent data, the last matching entry (based on D-list order) from this database D-Cube is transferred. For example, if Employee Costs contains the following data:

Employees
A B C D E

Division
Division 1 Division 1 Division 2 Division 3 Division 2

Level
Level 3 Level 2 Level 5 Level 1 Level 1

Salary
80,000 70,000 100,000 60,000 45,000

and the lookup D-Link is run inversely, then the following data is transferred to the Salary by Level D-Cube:

Execution Mode Set to Substitute Level

Level 1 Level 2 Level 3 Level 4

Salary
45,000 70,000 80,000 85,000

268 Analyst

Chapter 9: D-Links

Level
Level 5

Salary
100,000

Execution Mode Set to Fill Level

Level 1 Level 2 Level 3 Level 4 Level 5

Salary
45,000 70,000 80,000 0 100,000

These same results would be obtained even if the D-Cube were re-sorted in any way based on the cube data, as the last matching item is based on the order of the Employees D-List.

D-Lists Used in Lookup D-Links

In a normal D-Link, all D-Lists are equivalent: Format D-Lists (p. 109). Virtual dimensions may appear, but as long as they are not paired, they will play no part in the D-Link. In lookup D-Links, however, the treatment of D-Lists depends on which format D-Lists have been paired. D-Lists are treated like this:

Normal D-Lists paired with Format D-Lists

These pairings drive the lookup D-Link. The normal D-List in the source that is paired with a format D-List in the target is referred to as the driving source D-List; the format D-List in the target is referred to as the driving format D-List. Data is taken from the driving source D-List and allocated to target auto-allocated D-Lists (see below for a description of auto-allocated D-Lists) according to the contents of the driving D-List formatted item in the target lookup D-List.

Unpaired Format D-Lists

Unpaired format D-Lists have no effect in a lookup D-Link. If you do not pair any target format D-Lists, you will create a normal D-Link.

Normal D-Lists Paired with Normal D-Lists

These pairings are treated the same way as a normal D-Link.

User Guide 269

Chapter 9: D-Links

The Target Lookup D-List

The target lookup D-List contains the D-List formatted items used to drive the D-Link. This D-List is treated the same way as a normal D-Link, except that you should not leave it unpaired with an empty selection or target the driving D-List formatted items. If you do target the driving D-List formatted items, the data transferred by the D-Link will destroy the data used to drive the D-Link. You must identify the item or items in a target lookup D-List, which use the data transferred by the D-Link. Data is not allocated automatically into this D-List. You can use only one lookup D-List in a lookup D-Link. For information about restrictions, see "Lookup and Accumulation D-Link Restrictions" (p. 264).

Unpaired D-Lists in the Target

These are referred to as the auto-allocated D-Lists, with the exception of a lookup D-List. Each detail item in an auto-allocated D-List will contain one entry from the driving format D-List (the driving D-List formatted item in the lookup D-List). This entry determines which data is taken from the driving source D-List.

Unpaired D-Lists in the Source

These D-Lists are treated the same way as a normal D-Link.

Accumulation D-Links
Accumulation D-Links are D-Links that consolidate data from a source database D-Cube to a target sparse D-Cube (p. 262) based on D-List formatted text. You can run an accumulation D-Link inversely to perform a breakback (p. 149) allocation in the database D-Cube. You define a D-Link as an accumulation D-Link by pairing a format D-List in a source D-Cube with a normal D-List in a target sparse D-Cube and selecting Accumulation as the D-Link type. Normally source and target D-Lists are formatted to use the same D-List. You need to specify the D-Link as an accumulation D-Link based on how the dimensions of the source and target correspond by pairing a formatted D-List with a normal target D-List dimension. You can create an accumulation D-Link in more than one dimension by pairing more than one source format D-List. Be cautious because certain restrictions (p. 264) apply.

D-Lists Used in Accumulation D-Links

In a normal D-Link, all D-Lists are equivalent. Format D-Lists (virtual dimensions) may appear, but as long as they are not paired, they will play no part in the D-Link. In accumulation D-Links, however, the treatment of D-Lists depends on which format D-Lists have been paired. In an accumulation D-Link (that is, a format D-List in the source paired with a normal D-List in the target), D-Lists are treated like this:

270 Analyst

Chapter 9: D-Links

Format D-Lists paired with normal D-Lists

These pairings drive the accumulation D-Link. The format D-List in the source is referred to as the driving format D-List, and the normal D-List in the target is the driving target D-List. Data in the source auto-consolidated D-Lists is consolidated into a driving target D-List according to the contents of the driving D-List formatted item in the source lookup D-List. In an accumulation D-Link, unpaired D-Lists in the source D-Cube that do not contain a selection are referred to as auto-consolidated D-Lists (with the exception of the lookup D-List). Each (detail) item in an auto-consolidated D-Lists will contain one entry from the driving format D-List (that is, in the driving D-List formatted item in the lookup D-List). This entry determines which item in the driving target D-List will take data.

Unpaired Format D-Lists

Unpaired format D-Lists play no part in the D-Link. If you do not pair source format D-Lists, you will create a normal D-Link rather than an accumulation D-Link.

Normal D-Lists paired with normal D-Lists

These pairings are treated in the same way as for a normal D-Link.

The source Lookup D-List

The lookup D-List contains the D-List formatted items used to drive the D-Link, and is treated in the same way as a normal D-Link, except that you should not leave it unpaired with an empty selection; otherwise it will take data from the driving D-List formatted items. If you do, you will include the D-List formatted data in the data transferred by the D-Link. You must identify the item, or items in this D-List containing the data you want the D-Link to transfer. Data is not consolidated automatically from this D-List. There can only be one lookup D-List in an accumulation D-Link. For information about restrictions, see "Lookup and Accumulation D-Link Restrictions" (p. 264).

Unpaired D-Lists in the source

These are referred to as the auto-accumulated D-Lists with the exception of the lookup D-List. Each detail item in an auto-accumulated D-Lists will contain one entry from the driving format D-List (the driving D-List formatted item in the lookup D-List). This entry determines which item in the driving target D-List will take data.

Unpaired D-Lists in the target

These D-Lists are treated the same way as for a normal D-Link.

One-dimensional Accumulation D-Links

A one-dimensional accumulation D-Link is based on two criteria.

User Guide 271

Chapter 9: D-Links It must contain at least one D-List formatted item (p. 117) that can be used as a source to consolidate data from a database D-Cube to a target D-Cube based on text data. Only one format D-List is paired in the D-Link.

Two-dimensional Accumulation D-Links

A two-dimensional accumulation D-Link is based on two criteria. It must contain at least two D-List formatted items (p. 117) that can be used as sources to consolidate data from a database D-Cube to a target D-Cube based on text data. Two format D-Lists are paired in the D-Link.

You define a D-Link as a two-dimensional accumulation D-Link by pairing two format D-Lists in a source D-Cube with two normal D-Lists in a target D-Cube. Normally source and target D-Lists are formatted to use the same D-Lists.

Create a Two-dimensional Accumulation D-Link

Creating a two-dimensional accumulation D-Link is the same as creating a one-dimensional accumulation D-Link, except that instead of pairing only one virtual dimension, you pair two virtual dimensions.

Accumulation D-Link Execution Modes

The Fill and Substitute execution modes (p. 253) (set using the Mode drop-down box in an accumulation D-Link) have a special function when applied to lookup and accumulation D-Links. If the mode is set to Fill, all data in the target D-Cube is replaced and data that is not represented in the database D-Cube is set to zero. When the mode is set to Substitute, data that is not represented in the database D-Cube is left untouched. For example, the source D-Cube (Employee Costs) used for an accumulation D-Link contains the D-Lists, Employees and Employee Attributes. The D-List Employee Attributes contains two D-List formatted items Division and Levels. It contains the following data:

Employees
Employee A Employee B Employee C Employee D Employee E

Division
Division 1 Division 1 Division 2 Division 3 Division 2

Level
Level 3 Level 2 Level 5 Level 1 Level 1

Salary
80,000 70,000 100,000 60,000 45,000

The target D-Cube (Levels by Divisions) for the accumulation D-Link contains the D-Lists, Divisions and Levels. Initially, it contains the following data: 272 Analyst

Chapter 9: D-Links

Division 1
Level 1 Level 2 Level 3 Level 4 Level 5 1,000 1,000 1,000 1,000 1,000

Division 2
1,000 1,000 1,000 1,000 1,000

Division 3
1,000 1,000 1,000 1,000 1,000

Division 4
1,000 1,000 1,000 1,000 1,000

Division 5
1,000 1,000 1,000 1,000 1,000

Total
5,000 5,000 5,000 5,000 5,000

In the accumulation D-Link to transfer data from Employee Costs to Levels by Divisions, the source D-List, Employee Attributes, is left unpaired with the item Salary, selected. The source D-List, Employees, has been left an empty selection (auto-allocated D-Lists). The source format D-List, Divisions, is paired with the target D-List, Divisions, using match descriptions (p. 233). The source format D-List, Levels, is paired with the target D-List, Levels, using match descriptions. If the execution mode for the D-Link is set to Fill, the following data is transferred to the Levels by Divisions cube:

Division 1
Level 1 Level 2 Level 3 Level 4 Level 5 0 70,000 80,000 0 0

Division 2
45,000 0 0 0 100,000

Division 3
60,000 0 0 0 0

Division 4
0 0 0 0 0

Division 5
0 0 0 0 0

Total
105,000 70,000 80,000 0 100,000

Some of the cells are set to zero because no record was found in the database D-Cube. If the execution mode for the D-Link is set to Substitute, the following data is transferred to the Levels by Divisions cube:

Division 1
Level 1 Level 2 Level 3 1,000 70,000 80,000

Division 2
45,000 1,000 1,000

Division 3
60,000 1,000 1,000

Division 4
1,000 1,000 1,000

Division 5
1,000 1,000 1,000

Total
108,000 74,000 84,000

User Guide 273

Chapter 9: D-Links

Division 1
Level 4 Level 5 1,000 1,000

Division 2
1,000 100,000

Division 3
1,000 1,000

Division 4
1,000 1,000

Division 5
1,000 1,000

Total
5,000 104,000

The cells without records stay the same (the data is still 1,000).

Run Accumulation D-Links Inversely

You can run accumulation D-Links inversely the same as normal D-Links. For more information, see "Tun an Inverse D-Link" (p. 227). An accumulation D-Link run inversely does not become a Lookup D-Link. Instead, it performs a break back allocation over the source data. For example, the source D-Cube (Employee Costs 2) for the accumulation D-Link contains the D-Lists, Employees and Employee Attributes. The Employee Details D-List contains the D-List formatted items, Divisions and Levels, formatted on the D-Lists, Divisions and Levels, respectively. Initially, it contains the following data.

Division
Employee A Employee B Employee C Employee D Employee E Division 1 Division 1 Division 2 Division 3 Division 5

Levels
Level 3 Level 2 Level 5 Level 1 Level 1

Basic Salary Benefits

80,000 70,000 100,000 60,000 60,000 8,000 8,000 4,000 4,000 4,000

Total Salary
88,000 78,000 104,000 64,000 64,000

The target D-Cube (Overheads) for the accumulation D-Link contains the D-Lists, Levels and Divisions. It contains the following data.

Division 1

Division 2

Division 3

Division 4

Division 5

Divisional Total
658,000 24,000 0

Basic Salary Benefits Other Overheads

223,000 8,000 0

112,000 4,000 0

193,000 4,000 0

55,000 4,000 0

75,000 4,000 0

274 Analyst

Chapter 9: D-Links

Division 1

Division 2

Division 3

Division 4

Division 5

Divisional Total
682,000

Total Overheads

231,000

116,000

197,000

59,000

79,000

In the accumulation D-Link to transfer data from Employee Costs 2 to Overheads, the source D-List, Employee Attributes 2, is paired with the target D-List Overheads and the items, Basic Salary and Benefits, are matched. The source D-Lists, Employees and Levels, has been left with an empty selection (p. 232) (an auto-accumulated D-List). The source format D-List Divisions is paired with the target D-List Divisions using match descriptions (p. 233). In a lookup D-Link, unpaired D-Lists in the target D-Cube that do not contain a selection are referred to as auto-allocated D-Lists (with the exception of the lookup D-List). Each (detail) item in an auto-allocated D-List contains one entry from the driving format D-List (that is, in the driving D-List formatted item in the lookup D-List). This entry determines which data is taken from the driving source D-List. The data in Overheads has been transferred from Employee Costs 2 using the accumulation D-Link described above.

Division 1

Division 2

Division 3

Division 4

Division 5

Divisional Total
370,000 28,000 0

Basic Salary 150,000 Benefits Other Overheads Total Overheads 16,000 0

100,000 4,000 0

60,000 4,000 0

0 0 0

60,000 4,000 0

166,000

104,000

64,000

64,000

398,000

But, suppose we change the number in the Division 2 column, Benefits row from 4,000 to 8,000. And then run the accumulation D-Link inversely. The following data is transferred to Employee Costs 2 D-Cube.

Division
Employee A Employee B Employee C Division 1 Division 1 Division 2

Levels
Level 3 Level 2 Level 5

Basic Salary Benefits

80,000 70,000 100,000 8,000 8,000 8,000

Total Salary
88,000 78,000 108,000

User Guide 275

Chapter 9: D-Links

Division
Employee D Employee E Division 3 Division 5

Levels
Level 1 Level 1

Basic Salary Benefits

60,000 60,000 4,000 4,000

Total Salary
64,000 64,000

Note that Employee C's benefits have increased from 4,000 to 8,000. When an accumulation D-Link is run inversely, any holds set in the source D-Cube are respected. Also, where there are a number of detail items, the total is broken back in the same ratio as the detail items.

Analyst<>Contributor Links
You can transfer data between Contributor and Analyst using Analyst's D-Link function. All the standard features of a normal D-Link are available. For example, the use of A-tables, D-Cube allocations, local allocation tables, match descriptions, and matching on codes. There are three types of links available: Analyst>Contributor Contributor>Analyst Contributor>Contributor

When to use Analyst<>Contributor Links

For small amounts of data, an Analyst<>Contributor link can be a quick and effective method of transferring data. Analyst <> Contributor links use the Analyst link engine, which was designed to efficiently move small amounts of data. If you attempt to move large amounts of data, it can result in performance issues. For large amounts of data, it can be more efficient to import data in and out of Contributor using Contributor's import and publish facilities. This is because Contributor can be scaled out to make use of multiple computers and processors. Analyst<>Contributor links work in the same way as a standard Analyst D-Link. They treat Contributor as a single large cube which means that with large models, you can quickly run into memory problems. Because of these limitations, we recommend that Analyst<>Contributor links should only be used for doing ad-hoc transfers of small amounts of data of no more than 10 - 20 e.List items at a time.

Using the @SliceUpdate Macro

You can avoid memory problems for links that target an entire e.List in Contributor by using the @SliceUpdate macro. This macro processes the link in slices of the e.List, making it a much more scalable solution.

276 Analyst

Chapter 9: D-Links

How Analyst<>Contributor links differ from Standard D-Links

Most D-Links that have Contributor as a source or target behave the same as normal Analyst D-Links. The few exceptions are listed below: Lookup D-Links are not allowed where Contributor is the target. This is because Lookup D-Links depend on the data in the target cube, but in a Web environment, this data can be changing all the time. You are not allowed to target calculated items in Contributor. This includes totals on the e.List dimension as well as any total in other D-Lists. Match descriptions in Analyst D-Links to/from Contributor will treat the pipe symbol as a blank. The pipe symbol is used in Analyst as a line-feed for column headers. It is stripped out when you create a Contributor application from an Analyst model. You cannot target an Assumptions Cube in Contributor, although you can have one as a source. For Analyst>Contributor links, each e.List item of a Contributor cube can only be targeted by a single D-Link for each time you Go to Production. If you inadvertently run two D-Links that target the same e.List item of a Contributor cube, the D-Link you run last will be the one to run, and any earlier D-Links will be ignored. However two different D-Links may target the same cube provided that they target different e.List items. You cannot target No Data cells as defined by access tables in Contributor. However, as the Administrator, you are not subject to the restrictions imposed on Contributor users entering data via the Web client, so hidden and read-only cells are not applicable. You can write to these cells just as you can using normal import routines. Analyst<>Contributor links cannot be run inversely.

Otherwise, most D-Link types will be permitted. You can use Match Descriptions, local allocation tables, A-tables, and D-Cube allocations. You can cut subcolumns, letting you match on codes. You can run accumulation links both ways, but lookup links run from Contributor to Analyst only. Note: If you use a saved allocation table and rename when using Contributor as a source or target in a D-Link, the allocation table must be manually updated in order for the D-Link to work like it did before the item was renamed.

Installation
Analyst users who do not have the Contributor Administration Console installed are not able to run Analyst<>Contributor D-Links. Note: When you have installed Client tools onto a workstation, it is installed only for the user doing the installation.

Security
To run a Contributor<>Analyst link, users must: Have Analyst and the Contributor Administration Console installed.

User Guide 277

Chapter 9: D-Links Belong to a user, group or role which gives them rights to Analyst and the appropriate Contributor applications.

In addition, organizations may lock down access to the database or the Web server using the IP address, limiting who can run these links. Note: D-Links from ASCII and ODBC directly into Contributor are not allowed. You must use Contributor Import to do this.

How Analyst<>Contributor and Contributor<>Contributor Links Work

You set up a Analyst<>Contributor link in the same way as you would a standard D-Link, choosing Contributor as the source or target of a D-Link. All the data is prepared in Analyst. The steps for setting up any Analyst<>Contributor or Contributor<>Contributor link are: In the Analyst D-Link editor, choose Contributor Data as the source or target. Select a Datastore server (if more than one is available). Select a Contributor application. You may need to click the Refresh button to the right of the Name list to get the list of available Contributor applications. Click Test Connection, then click OK. Select whether to target the production system or the development system of the Contributor application. Select a Contributor cube. Pair dimensions against the target (or source) cube as you would for a normal D-Link.

Analyst>Contributor Links
These links can target either the production or development version of Contributor. If targeting the development version, they only appear on the Contributor screens after the Go to Production process is completed. If targeting the production version, you do not need to run a full Go to Production process. Instead, an activate process is run which moves the data into the import production queue and creates a snapshot of the data at the time the link was executed. Then a reconcile job is triggered, which updates the e.List items with the new data. The following steps occur when you target the development application. When you run an Analyst>Contributor Link, the data is read out of Analyst when you run the D-Link. The prepared data is written directly to the import queue in the Contributor application datastore as a prepared data block, e.List item by e.List item when you run the Go to Production process in the Contributor Administration Console, or through Automation.

278 Analyst

Chapter 9: D-Links You can have multiple links target the same e.List in the same Contributor cube. This is done by having each link create its own bucket in the import queue, so multiple links in the same import queue can target the same e.List. If the same link is run twice, the last run will overwrite the first run in the bucket.

There may be a delay between the Go to Production process and the data being reconciled in the Web client. If, in the meantime, a planner has edited one of the cells targeted by the link, that cell will be overwritten when the reconciliation takes place. This behavior is very similar to the reconciliation that takes place when you import data into Contributor from text files or via DTS. The following occurs when targeting the production application: The data is read out of Analyst when you run the D-Link. An activate process is run that applies the data to a cube. If running the link through the Analyst interface, the activate process happens automatically. If running the link using macros, you must run the @DLinkActivateQueue macro to activate the link. The import queue targeting the production system can be activated from Analyst.

Recommendation - Use Macros for Multiple Analyst >Contributor Links

If you have multiple Analyst > Contributor links that target the same Contributor application, and that target production, we recommend that you use Analyst macros. This means that you can run multiple links one after the other in the macro that all target the same Contributor application, and then run a single activation step. This is much quicker than running multiple Analyst >Contributor links from the Analyst user interface because after each link, the activate process runs automatically. It is also quicker than running a link that targets development and then running Go to Production.

Contributor>Analyst Links
The following steps occur when a Contributor>Analyst Link is run: A snapshot is taken of the production version of the Contributor Application. To ensure a consistent read if you are using the @SliceUpdate macro, take the Contributor application offline, or use the @DLinkExecuteList macro. The link process causes a prepare publish job to be run. You can monitor this in the Contributor Job Management screen. Once the Prepare publish job has run, a publish job is created and immediately set to ready. Note that the job is not run, this is because the Contributor job executor is not used. Analyst executes the transfer of data. A Contributor session is loaded and the entire data block is loaded for each e.List item. If the link is set up for more than one e.List item, it is the equivalent of loading a multi-e.List item view; a very memory intensive process. The data is written directly to the Analyst cube data file (H2D file).

User Guide 279

Chapter 9: D-Links

Contributor>Contributor Links
These links go from the Production version of a Contributor source into the development version of the Contributor target. They are typically used between separate applications. If the applications are small, Contributor>Contributor links can be fast. However, if you transfer data between larger applications this way, you may run into problems due to memory use and scalability problems. You can avoid these issues by using the @SliceUpdate macro. You can also use Administration Links in the Contributor Administration Console. These copy data between Contributor cubes and applications. This process is scalable and so can move large volumes of data into either the development or production version of the Contributor application.

Copying Analyst<>Contributor Links

There are three methods for making copies of links. Save As, Library Objects, and Library Copy Wizard. These three methods of copying Analyst<>Contributor links affect whether the link refers to the original application or a new application.

Save As Method
This method results in a copy which refers to the original Contributor application(s) and/or Analyst D-Cubes.

Steps
1. Open the link. 2. From the File menu, click Save As. 3. Choose a Library in which to copy the link. 4. Enter a name for the link copy. 5. Click OK.

Library Method
This method lets you select the link with or without other objects and choose either to copy or move the link. This results in a link which refers to the original Contributor application(s) although the source or target Analyst D-Cube (if it is not a Contributor > Contributor link) could be changed by this method if certain reference options are chosen when copying.

Steps
1. From the File menu, click Library, Objects. 2. Select the link with or without other objects and move it down. 3. Click the Copy selected objects button.

280 Analyst

Chapter 9: D-Links 4. In the Copy dialog box, enter a new name for the link, select a target library in which to copy the link, and select how to remap references. 5. Click OK.

Library Copy Wizard Method

This method lets you make a copy of an Analyst<>Contributor link which refers to a new application based on a copied library.

Things to Watch Out For

Using the Library Copy wizard to create duplicate objects within the same library will not work for making copies of links because if you copy template D-Cubes containing the e.List, and include the e.List itself in the selection of objects, the e.List will be copied. Thus when you synchronize Contributor, the new cube you have created will be an assumption cube as it does not contain the original e.List. A link can only be pointed to an application where it will refer to template cubes which were copied at the same time. You cannot copy a link into a library which already contains suitable template cubes and then refer the link to an application based on that library. If you make a copy of a link using this method and copy the link and its associated objects at the same time, then you will not be able to refer the link back to the original application. You will have to make a new application based on the copied library and refer the link to this new application. If you copy macros which refer to Contributor applications using the Library Copy wizard, then the macros will continue to refer to the original application. You must open the copied macros and manually edit them to refer to any new applications based on copied libraries.

Steps
1. Use the Library Copy wizard to copy the link and any related Analyst template cubes at the same time. 2. Create a Contributor application based on the copied Analyst template D-Cubes. 3. Point the link to your new application by using one of two ways. Open the link and then select your new Contributor application when prompted. From the File menu, click Library, Object. Double-click the link to move it down and then right click the link and select Change Contributor source on D-Links.

Factors That Can Affect Memory Usage

Memory limits can be hit, depending on the following: Density of data being transferred by links.

User Guide 281

Chapter 9: D-Links Available RAM. The more available, the better. Multi-e.List item views with access tables, are not cut down as much as a single e.List item view with access tables. Maximum workspace setting (MAXWS) in Analyst.

This is the amount of space reserved for use by Analyst. As a general rule, this should not be more than half the available RAM. Setting this option too high can grab so much memory for the Analyst process that it does not leave enough for the Contributor process.

Opening a Link From a Computer that Does not Have Access to the Original Datastore
If a Contributor cube is used as a source or target, and the link is opened from a computer different from the one the link was created on (different datastore), you are prompted to reselect the connection and application to point to the data store and application name that holds the cube the link was built on. All matching is then preserved. You save the link, and the new connection will be saved, and the link will run in the future. This allows multiple data sources to be used. If the two applications are built from the same Analyst library, the GUIDs will match when pointing the link to the original datastore. Running a link from a workstation that does not have access to the original datastore will require manually opening the link and reselecting the connection. You can also update the connection for several links at once. Note: Analyst <>Contributor and Contributor <>Contributor links are associated with Contributor applications using unique identifiers (GUIDs). If you change the source or target Contributor application, you must remap the D-Link connections because the GUIDs have changed.

Steps
1. Click File, Library, Objects and select the links that you want to update and move then to the bottom pane. 2. In the bottom pane, right click and select Change Contributor Source on D-Links. 3. Enter the connection details for the new datastore. 4. Select the appropriate substitution option. This will update all the selected links with the new connection details.

Running Batches of D-Links using the @DLinkExecuteList macro

@DLinkExecuteList is a macro designed to run a series of D-Links in order. The @DLinkExecuteList macro behaves similarly to a series of @DLinkExecute steps, with a subtle difference when D-Links have Contributor as a source. When the macro runs the first D-Link that has Contributor as a data source, it logs the time and takes a consistent read of the database. All subsequent D-Links that have the same Contributor source use the data from that point in time

282 Analyst

Chapter 9: D-Links rather than the actual time when the D-Link is run. This ensures consistency across D-Links coming from the same Contributor data source. If a subsequent D-Link in the macro has a different Contributor data source, the old source is closed and the new one opened. Similarly, if you have two @DLinkExecuteList steps in the same macro or in a sub-macro called via a @MacroExecute step, each time the old data source is closed and the new one is opened.

Steps to set up @DLinkExecuteList

1. Click Edit Object List in the Macro wizard. 2. Select the D-Links you want to run. You must select only D-Links, because anything other than D-Links are ignored. You may include any set of D-Links, whether they be normal Analyst > Contributor, or Contributor > Analyst D-Links.

Running D-Links While Making Model Changes

If you want to make changes to the Contributor model and import data into Contributor using Analyst>Contributor D-Links, it is important that you synchronize first, then run the D-Link. For example, if you have inserted a new product as part of a model change, you cannot import data into the new product until the Contributor model has been synchronized. You must then run Go to Production so that the model and data changes become live. So, if you make any changes to an Analyst model, then want to populate the Contributor cube, the order is as follows: Synchronize Run the D-Link Run Go to Production.

You can use Contributor>Contributor links to preserve data during cube dimensional restructuring, for example when adding a dimension.

Steps
1. Take the Contributor application offline. 2. Run a link from the production version of the Contributor application to the import queue of the development application. 3. Run Go to Production.

Fill and Substitute Mode

In Contributor<>Analyst D-Links, Fill and Substitute Modes behave in the same way as normal D-Cube to D-Cube D-Links in Analyst. Fill and Substitute modes generally only apply to lookup and accumulation D-Links. In Substitute mode, the data in the untargeted area of the D-Link remains unchanged. In Fill mode, the untargeted cells get set to zero when the D-Link is run. User Guide 283

Chapter 9: D-Links However, this only applies to lookup and accumulation D-Links. In normal D-Links, if an item is not included in the right-hand side of an allocation table, it is assumed to be untargeted, so the original target data is kept unchanged, irrespective of whether you use Fill or Substitute mode. Similarly, on normal real dimension pairings that use match descriptions, unmatched items are kept untouched when the D-Link is run. This applies to both Fill and Substitute modes.

Action Applied to Untargeted Cell D-Cube or Contributor Cube asSource

Fill Allocation Tables lookup Accumulation Keep Zero Zero Substitute Keep Keep Keep Keep

File Map (ODBC or ASCII) as Source

Fill Keep Not Applicable Not Applicable Zero Keep Substitute Keep

Match Descriptions Keep

Effect of Access Tables in Contributor

If Contributor is the source, cells marked as No Data are treated as zero when running a D-Link into an Analyst or Contributor target D-Cube. If Contributor is the target, you cannot target No Data cells as defined by access tables in Contributor. However, as the Administrator, you are not subject to the restrictions imposed on Contributor users entering data via the Web Client, so hidden and read-only cells are not applicable. You can write to these cells just as you can using normal