|
View previous topic :: View next topic
|
| Author |
Message |
William Thompson
Global Moderator
Joined: 18 Nov 2006
Posts: 3156
Location: Tucson AZ
|
|
|
|
| ami777us wrote: |
| CICS GUY---IN UR WHOLE MSG U HAVENT TOLD ME IF MY SYNTAX WAS RIGHT.IF U DONOT UNDERSTAND MY PROBLEM THEN U BETTER DONT RESPOND.LET THE MORE INTELLIGENT GUYS LIKE DICK AND OTHERS ANSWER.OK!! |
You don't have to yell....
| CICS Guy wrote: |
| ami777us wrote: |
| Please let me know what am i doing wrong.I am sure I am missig something. |
Yes, at least the EIB in the PRGA. |
|
|
| Back to top |
|
 |
ami777us
New User
Joined: 06 Apr 2007
Posts: 33
Location: USA
|
|
|
|
Dick-I have used Strobe to determine that the parser is consuming a high CPU.I hae learnt that by changing subscript to index within the parser the performance can be improved.I have a doubt on the index now.
There are MOVE statements for the subscript value to another working storage variable for some length calculation.
My Quesion is "Can I just move the Index to the working storage or there is any other way to get around it"?? I think that Index values cannot be just moved.Can you please confirm. |
|
| Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006
Posts: 6966
Location: porcelain throne
|
|
|
|
use a SET statement to populate a numeric variable with the 'occur #' of an index.
| Code: |
05 SAVE-OCCURANCE-NUMBER PIC S9(9) COMP.
SET SAVE-OCCURANCE-NUMBER TO INDEX-1
|
you can use any numeric PIC clause, but COMP converts faster. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
| Quote: |
| I hae learnt that by changing subscript to index within the parser the performance can be improved |
While this may be true in some cases, i expect that the parsing approach the code uses currently will not significantly change cpu utilization by switching from subscript to index.
If you work with an index, you would use SET rather than MOVE.
As i mentioned previously
| Quote: |
If you post your fields that are being parsed and the output field(s) expected from some representative sample inputs, we may be able to offer alternate schemes for implementing.
|
|
|
| Back to top |
|
|
|
|
