| View previous topic :: View next topic |
| Author |
Message |
deepa12
Beginner
Joined: 05 Apr 2005
Posts: 131
Topics: 64
Location: chennai
|
 Posted: Mon Aug 25, 2014 2:03 pm Post subject: Numeric deedit |
 |
|
Consider the following
WS-A PIC S9(V4)V99 COMP.
WS-A1 PIC S9(V4)V99.
WS-B PIC ZZ99.99
Consider the following code
move ws-b to ws-a
what i understand is using numeric dedit in this fashion is not correct coding practice. Is this true
move ws-b to ws-a1 - is this safer than the above?
i.e using DISPLAY field rather than COMP Field
_________________
deepa |
|
| Back to top |
|
 |
kolusu
Site Admin
Joined: 26 Nov 2002
Posts: 12375
Topics: 75
Location: San Jose
|
| Posted: Mon Aug 25, 2014 2:27 pm Post subject: Re: numeric deedit |
|
|
| deepa12 wrote: | Consider the following
WS-A PIC S9(V4)V99 COMP.
WS-A1 PIC S9(V4)V99.
WS-B PIC ZZ99.99 |
Deepa,
Can you please enlighten me as to which COBOL version supports the PIC S9(V4)V99 COMP definitions?
| deepa12 wrote: |
Consider the following code
move ws-b to ws-a
what i understand is using numeric dedit in this fashion is not correct coding practice. Is this true
move ws-b to ws-a1 - is this safer than the above?
i.e using DISPLAY field rather than COMP Field |
Safer in what sense? No matter which Move Instruction you use you will end with a 0C7 abend if either WS-B or WS-A1 contain uninitialized/Garbage values.
You need to remember that every line you code is internally translated to a machine language by your compiler. So your main goal should be to reduce the number of instructions that your code generates. In order to see the machine instructions generated you just need to use the compiler option LIST
Run this sample JCL and look at the sysprint output from this step and determine which move is appropriate.
| Code: |
//STEP0100 EXEC PGM=IGYCRCTL
//SYSPRINT DD SYSOUT=*
//SYSTERM DD SYSOUT=*
//SYSUT1 DD UNIT=SYSDA,SPACE=(CYL,(10,2),RLSE)
//SYSUT2 DD UNIT=SYSDA,SPACE=(CYL,(10,2),RLSE)
//SYSUT3 DD UNIT=SYSDA,SPACE=(CYL,(10,2),RLSE)
//SYSUT4 DD UNIT=SYSDA,SPACE=(CYL,(10,2),RLSE)
//SYSUT5 DD UNIT=SYSDA,SPACE=(CYL,(10,2),RLSE)
//SYSUT6 DD UNIT=SYSDA,SPACE=(CYL,(10,2),RLSE)
//SYSUT7 DD UNIT=SYSDA,SPACE=(CYL,(10,2),RLSE)
//SYSLIN DD DSN=&&LOADSET,
// DISP=(MOD,PASS),
// UNIT=SYSDA,
// SPACE=(CYL,(1,1),RLSE)
//SYSIN DD *
CBL LIST
IDENTIFICATION DIVISION.
PROGRAM-ID. TEMP1.
ENVIRONMENT DIVISION.
DATA DIVISION.
WORKING-STORAGE SECTION.
01 WS-A PIC S9(4)V99 COMP.
01 WS-A1 PIC S9(4)V99.
01 WS-B PIC ZZ99.99.
PROCEDURE DIVISION.
MOVE 12.34 TO WS-A1
WS-B
PERFORM 1000-WS-B-MOVE
PERFORM 2000-WS-A1-MOVE
GOBACK
.
1000-WS-B-MOVE.
MOVE WS-B TO WS-A
.
2000-WS-A1-MOVE.
MOVE WS-A1 TO WS-A
.
/* |
_________________
Kolusu
www.linkedin.com/in/kolusu |
|
| Back to top |
|
|
deepa12
Beginner
Joined: 05 Apr 2005
Posts: 131
Topics: 64
Location: chennai
|
| Posted: Mon Aug 25, 2014 7:34 pm Post subject: |
|
|
Sorry that was aa typo
The folln should have been WS-A PIC S9(4)V99 COMP
_________________
deepa |
|
| Back to top |
|
|
William Collins
Supermod
Joined: 03 Jun 2012
Posts: 437
Topics: 0
|
| Posted: Tue Aug 26, 2014 9:35 am Post subject: |
|
|
What is the content of your data? If you have an actual decimal-point in your data, the de-editing is one way to MOVE the data to a numeric field - using a numeric field whilst there is still an actual decimal-point lurking will get you a S0C7.
There is a more-code-but-validate-the-data solution if you have an atcual decimal-point as well. |
|
| Back to top |
|
|
deepa12
Beginner
Joined: 05 Apr 2005
Posts: 131
Topics: 64
Location: chennai
|
| Posted: Tue Aug 26, 2014 11:39 pm Post subject: |
|
|
WS-B PIC ZZ99.99 will have data like <space><space>12.34 i.e it will have actual decimal point
Now we have actually changed the code to the folln:
MOVE WS-B TO WS-A1
instead of
MOVE WS-B TO WS-A
Pls suggest
_________________
deepa |
|
| Back to top |
|
|
William Collins
Supermod
Joined: 03 Jun 2012
Posts: 437
Topics: 0
|
| Posted: Wed Aug 27, 2014 2:31 am Post subject: |
|
|
An appropriate data-type depends on what you are going to do with the thing.
If I have edited numbers arriving, I de-edit them (a different way) to a USAGE DISPLAY field (so a PIC 9 or PIC X of the correct length). If I want to do something else with that (like use it in a calculation) they I'd MOVE it to a packed-decimal.
Long ago I was told not to use binary fields with decimal places. So I don't. Which now means I don't know if that is good advice or not.
As to which of your examples is "safer", if the source field contains data which will cause a S0C7 then the MOVE to the binary field will cause the S0C7 (on a CVB (convert to binary) instruction) and the MOVE to the USAGE DISPLAY item will not cause a S0C7 - until you do something with the field which causes a decimal instruction (or CVB) to be used. |
|
| Back to top |
|
|
deepa12
Beginner
Joined: 05 Apr 2005
Posts: 131
Topics: 64
Location: chennai
|
| Posted: Wed Aug 27, 2014 3:35 am Post subject: |
|
|
Many thanks
_________________
deepa |
|
| Back to top |
|
|
|
|
|
FAQ
Search
Quick Manuals
Register
Profile
Log in to check your private messages
Log in
