Introduction to JSP © R.S.
Kang 2002
Lesson
1
Introduction to Jackson Structured
Programming (JSP)
Contents
1. Introduction
2. JSP Diagrams and Notation
2.1. Sequence
2.2. Selection
2.3. Iteration
3. Passing & returning values to and from Functions
4. JSP Recursion
5. JSP and pseudo code
6. Comparative review of JSP and traditional Flowcharting
7. Summary
1. Introduction
I remember being introduced to JSP back in 1984 whilst at College, though it was a recently
new methodology, most students found this to be fairly intuitive. I’m hoping that instructors
alike will find this to be a key tool in effectively teaching the concept of logic and design.
The Jackson Program Design Methodology, sometimes called Jackson Structured
Programming (JSP), is a method for program design and modelling. It begins with
considerations about what is known and develops a program design that becomes more
complete as the model is put through continued iterations. It is primarily based on functional
top down decomposition and can be read top-down or bottom up. It builds program design
from an incomplete model, hence termed as being a “composition type” method.
The founder of JSP, Michael Jackson developed this methodology in the early 70’s that later
emerged in industry and the academic field. By the 80’ it was found to be a de-facto in Europe
and widely taught in Colleges and Universities.
1 of 12
Introduction to JSP © R.S. Kang 2002
2. JSP Diagrams and Notations
A Jackson Structured Programming diagram is used to explain the inner workings of a
program. At a glance they seem similar to algorithm flowcharts, but the likeness is only
superficial.
To understand a JSP diagram you must read it properly. With a JSP diagram each step on
the same branch is performed top down - left to right.
Sequence
Sequence
{Sequence}
Top Down A B C
1. process A
2. process B
3. process C
Order of execution
With JSP diagrams there are three types of “box” notations used to represent the workings of
a program. All programs have three control structures Sequence, Selection and Iteration!
Action/Process Selection
Iteration
* Procedure
2 of 12
Introduction to JSP © R.S. Kang 2002
2.1. Sequence
Sequence
{Sequence}
1 process A
2 process B
A B C 3 process C
Example 1
Private Sub cmdCalculate_Click()
Dim num1 As Byte CmdCal_Click
Dim num2 As Byte
Dim answer As Byte
num1 = Val(txtNum1)
num2 = Val(txtNum2)
answer = num1 + num2 Initialize Variables
Main Process Display Answer
Num1, Num2
lblAnswer = CStr(answer)
End Sub
Input num1, num2 Answer=num1+num2
Alternative (Refined)
CmdCal_Click
Initialize Variables
Main Process Display Answer
Num1, Num2
User Input Answer=num1+num2
num1=txtNum1 num2=txtNum2
3 of 12
Introduction to JSP © R.S. Kang 2002
2.2. Selection
{Selection}
1 select
Selection ?
1.1a when c1
1.1b process A
C1 C2 NOT C1 or C2
1.2a when c2
1.2b process B
A B C
1.3a otherwise
1.3b process C
1
Example 1 : IF statement
If Age > 60 Then Age > 60
P = True
Endif
T F
P=True _
Example 2 : IF Then..Else statement
Salary > 40k
True False If Sal>40 Then
if Sal>60 Then
if Sal>100 Then
Salary > 60k Print Supervisor
Print “CEO";
Else
Print”Director"
False True
Else
Print "Manager"
Print Manager Salary > 100K
Else
Print "Supervisor"
False True
Print CEO
Print Director
4 of 12
Introduction to JSP © R.S. Kang 2002
Example 3 : CASE statement
switch(ch)
{
case 1: Add_Stud(); break;
case 2: Search_ID(); break;
case 3: Disp_All(); break;
case 0: cout<<"Exiting Student Management System...!\n\n";exit(1);
default: cout<<"Invalid choice made....try again!"; break;
}
User Choice
1 2 3 0 Default (Else)
Add_Stud() Search_ID() Disp_All() "Exit System" Err
2.3. Iteration
Iteration 1
{Iteration 2}
1 repeat
While c1 1.1 process A
A * 1 until c1
Example to model a book:
While not End of Book
*
Book
While NOT End of Book
While NOT End of Chapter
While NOT End of Chapter While NOT End of Page
While NOT End of Line
Chapter
* …..
End While
End While
While NOT End of Page
End While
Page
* End While
While NOT End of Line
Line
*
5 of 12
Introduction to JSP © R.S. Kang 2002
3. Passing & returning values to and from Functions
Sub() Sub 4
Call Sub() Call Sub4(var Z),
return Z
X W X
Sub 2 Sub 5
Call Sub2(X) Call Sub5(var W,X):
return W;
Example 1: Demonstrating calling functions with values/parameters.
Program
{Program}
Sub A B C [Link] SubA
[Link] <condition>
while <condition> [Link] SubD(X)
Z 2.2.Y = call SubE(Z)
Sub D Sub E * F 2.3.F
[Link]
X R 3.C
SubE
Function SubE(int Y)
Process B
Process C
X
Call D(X)
B C Sub D Return
return R
6 of 12
Introduction to JSP © R.S. Kang 2002
Example 2: Design below demonstrates, sequence, selection and iteration based on a I/O file
design (this example closes and opens a file for each add record task)
C1
C1 : While No More Records to ADD Add Record
*
C2 : While NOT EOF OR Record Found
PKey
Initialize Get Record Close File
C2
Open File
Record=FALSE
Input Primary Key
(Pkey)
Check Record * If Record FOUND
NOT FOUND FOUND T F
Input Data
- Record=FOUND Err
Write to file
Example below opens the file at the start of the program, and closes it the end
C1 : While No More Records to ADD
C2 : While NOT EOF OR Record Found
Main Program
C1
Open File
Add Record
* Close File
PKey
Input Primary Key
(Pkey) Get Record
Record=FALSE
C2
Check Record * If Record FOUND
NOT FOUND FOUND T F
Input Data
- Record=FOUND Err
Write to file
7 of 12
Introduction to JSP © R.S. Kang 2002
4. JSP Recursion
A recursive routine is when a procedure calls itself. Each recursive call made allows the
problem to get closer to a solution. The Tower of Hanoi is a typical example of recursion.
JSP tackles recursion simply by introducing a double line process box that represents a call to
itself.
The example below demonstrates the 10 green bottle song (or 10 beers as the Americans
would say!) with a recursive call
10
Verse(Bottle)
Bottle >0
T F
Bottle & Green
Bottles..."
_ Base Case
Bottle >1 Verse(Bottle-1) Recursive Step
T F Bottle
Bottle-1 & Green No Green
Bottles..." Bottles..."
The recursive component of a JSP object or action design is indicated by a double lined
component box. The code corresponding to this design has been included below.
main ()
{
verse(10); // Call verse
return 0;
}
int verse(int bottle)
{
if (bottle > 0 ) // Where n is number of bottles
{
cout << bottle << " green bottle";
cout << " hanging on the wall....\n\n";
cout << "and if 1 green bottle should accidentally fall? \n";
if (bottle > 1)
{
cout << "There'll be " << (bottle-1) << " green bottle";
cout << " hanging on the wall\n";
}
else
{
cout << "There'll be no green bottles hanging on the wall\n\n";
}
verse(bottle-1); //Recursive call where bottle=bottle-1 (1 less bottle)
}
return 0;
}
8 of 12
Introduction to JSP © R.S. Kang 2002
5. JSP & Pseudo code
Generally a JSP diagram is adequate to describe a program; if more detail is required you
can easily combine JSP diagrams and pseudo code. Pseudo code could reflect process
boxes in particular where complex algorithms may be difficult to represent as JSP.
The example below uses both JSP and Pseudo code to represent the design of a simple
bubble sort program.
int main()
{
for (loop = 0;loop < 6;loop++)
{
for (Scan = 0;Scan < 6;Scan++)
{
if (Entry[Scan] > Entry[Scan+1])
{
TempEntry = Entry[Scan+1];
Entry[Scan+1] = Entry[Scan];
Entry[Scan] = TempEntry;
}
}
}
return 0;
}
Sort Prog
Initialize Variables Main() End
*
Loop
*
Scan {SWAP}
TempEntry = Entry[Scan+1]
Entry[Scan+1] = Entry[Scan]
Entry[Scan] = TempEntry
SWAP
Pseudo code could also be included for each process box, dependant on the application
being developed. If this is the case, JSP should still reflect an overall break down of the
problem.
9 of 12
Introduction to JSP © R.S. Kang 2002
6. Comparative review of JSP and traditional Flowcharting
A diagrammatical comparison between traditional flowcharting and JSP methodology is
covered in this section. A snap shot picture between the designs clearly shows that JSP is
found to be simplistic, clear and avoids the fuzziness shown by the flowchart method.
The JSP methodology tends to avoid the fuzziness that can be created by too many notations
adopted by flowcharting.
Sequence: calculate the sum of two numbers
Start
Get x,y
Answer = x + y
Program
Display
Answer
Get x,y Answer=x+y Display answer
Stop
JSP Flowchart
Selection: Calculate the difference of two numbers if the first is greater than the
second, otherwise calculate the sum
Start
Program Get x,y
Yes x>y? No
Get x,y x>y Display answer
Answer = x - y Answer = x + y
Yes No
Display
Answer
answer=x-y answer=x+y
Stop
JSP Flowchart
10 of 12
Introduction to JSP © R.S. Kang 2002
Iteration: Draw a flowchart to show how to calculate the sum of the numbers 1 to 100
Start
Program Num = 1
Sum = 0
No Num <= 100?
Num=1,Sum =1 Calculate Display sum
Yes
Display
While Num<=100 * Sum
Sum=Sum + Num
Stop Num=Num+1
Sum=Sum+Num Num=Num+1
JSP Flowchart
11 of 12
Introduction to JSP © R.S. Kang 2002
7. Summary
JSP modelling does involve the emergence of input and output that eventually creates the
structure of the final design. This chapter has only focussed on how simple problems could be
broken down and, solutions designed that are later transformed into code.
This chapter concentrates on allowing students in understanding logic and how to “think
logically” prior to coding.
The application of Jackson Program Design include small-scale development of program
modules, reports, sequential file processing, and similar areas that do not involve modelling-
in-the-large. However, all types of problems can be solved using JSP, and it is a clear, robust
system for thinking about problems.
Surveys indicate JSP/JSD (Jackson Structured design) is widely recognized and used in
Europe than in the USA. JSP’s intentions have been quite clear in terms of developing
structured systems, in particular how modules are integrated to form the whole application. Its
primary implementation has close ties with SSADM (Structured Systems Analysis and Design
Methodology).
The strength of Jackson Program Design lies in its description of what a program is to do. Its
basic characteristics that give it advantages in its systematic approach include:
• Rational- based on reasonable principles, well developed and well defined: structured
programming (iteration, selection, sequence).
• Teachable- it has a well set of defined steps, it is possible to explain and show how
this system is applied and how it works.
• CONSISTENT- given a single problem for analysis, different programmers or analysts
will come up with more or less the same program design.
• SIMPLE- the method does not involve overly complex rules, cumbersome procedures
or difficult notations to remember.
• UNIVERSAL- specifications are produced that can be implemented in the language
the developer desires.
Other Resources
JAVA JSP Editor Creates C/Pascal Code from design
[Link]
OR
[Link]
12 of 12
