Re: CL Parameters sharing memory?

LOL - now that IS weird. I agree - but it is even cleaner to put a simple command front end on it and submit that instead of calling the program directly.

Many of our customers have never run into this issue, because their character parameters are not usually that long - not over 32. But when you start messing with path names and URLs, well, we get lots of calls saying - it worked fine when I call it, but when I submit it, things go wrong - first clue to this symptom

Later
Vern

Dennis Lovelady wrote:

Not sure which is more evil - RQSDTA does not strip trailing blanks.
CMD
gives you promptability but is weird when using CALLs - as Tom
suggests,
maybe CMD parameter should not process CALLs.

So you want to compare weirdness, huh? :)

Which is weirder: RQSDTA('CALL MYPGM PARM(''A'' 123 ''= ''''*NONE'''''')')
Or
CMD(CALL MYPGM PARM('A' 123 '= ''*NONE'''))

(Note, I believe the above RQSDTA above to be a correct match for the CMD
below it. I haven't actually ported to the i to compare.) This is not an
exaggeration (at all) when you are passing commands like OPNQRYF command as
parameters themselves. In fact, this is mundane compared to some I have
encountered. Yes, recently.

Dennis Lovelady
http://www.linkedin.com/in/dennislovelady
--
The first meeting of the Apathy Society has been postponed due to lack of
interest.