On 14 Apr 2008 08:14:24 -0700, [EMAIL PROTECTED] (Clark Morris) wrote: >Very definitely, you should at least be checking for 00 and 97. >Depending on the files and any recent compiler changes, other >conditionally successful opens should be checked for.
We had a purchased system that included a called program to handle KSDS input. I have no idea why they did that, maybe because the concepts of OO were being developed and they wanted on the bandwagon. Well, it accepted parms to open, read, or close - and it passed a return of "7" if the VSAM return code wasn't "00" on the open command. This program was called by every program that read a KSDS file (which is now down to zero programs). Every once in a while, it would abort and our procedure would be to restart. That was because it didn't check for "97". In a CoBOL shop, we had standards about having everything that used that program being extensively tested by users before our fixes went into production - so it just wasn't cost effective to fix this called program. This program had all of the problems of OO, but none of the advantages. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
