Re: When ctl-opt main() is missing...

Yes - one would think, that the compile can "sense" this and emit an error message.

In my memory, I have a compiler error in mind, that would read somehow like this: "The compiler can not detect how the program is ended." And I thought this error would come, if no return, no *INLR and no primary file was coded in the program.

But my memory might trick me on this - I really don't know. And even if this kind of error message exists (or existed) this might be a constellation that the compiler simply cannot detect.

Kind regards,
Daniel

Am 31.10.2024 um 16:40 schrieb Martijn van Breden <m.vanbreden@xxxxxxxxxxxxxxxxxxxxxxxxxx>:

Hi Daniel,

Your explanaition seems to be quite to the point, thanks! I didn't think of it this way. The program is just very busy doing nothing, as there were no "C-specs" outside of any procedure and there were only usropn file specifications.
Still... the compiler might notice this IMHO


Kind regards,



Martijn van Breden

lead software architect







________________________________
Van: RPG400-L <rpg400-l-bounces@xxxxxxxxxxxxxxxxxx> namens Daniel Gross <daniel@xxxxxxxx>
Verzonden: donderdag 31 oktober 2024 16:17
Aan: rpg400-l@xxxxxxxxxxxxxxxxxx <rpg400-l@xxxxxxxxxxxxxxxxxx>
Onderwerp: Re: When ctl-opt main() is missing...

CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe.



Hi Martjin,

this program works indeed as intended.

If you do not code the "main(..)" keyword, you end up with a "traditional" ILE-RPG program which uses the cycle main "procedure".

This cycle main "procedure" is not enclosed in "dcl-proc/end-proc" and it is normally repeated for every record in a primary or secondary file. The cycle main "procedure" ends, when the last record was processed and the *INLR indicator is *ON when it reached the end of the cycle main "procedure".

But you might have no primary or secondary file - so the LR indicator doesn't switch *ON automatically, so you have to code *INLR = *ON somewhere in the cycle main "procedure".

So if all your code is enclosed inside of "real" procedures, you have effectively an empty cycle main "procedure" - which is repeated until *INLR is *ON ... which never happens - resulting in an endless loop.

And even as the new Power10 processors are really fast - they really just can't finish an endless loop.

HTH and kind regards,
Daniel

Am 31.10.2024 um 15:40 schrieb Martijn van Breden <m.vanbreden@xxxxxxxxxxxxxxxxxxxxxxxxxx>:

Hi all,

I've just helped out a very experienced colleague with a program that just wouldn't run. In debug it halted at **free. It showed status RUN in WrkActJob eaiting quite some CPU. In the call stack it stopped just past program entry point _QRNP_PEP.... We were not even able to force it into an error with the wrong number of parameters.
After a couple of hours testing we found that the ctl-opt main() keyword was missing.

I would say that the compiler should have been able to tell us about this situation or that it could have just ungracefully failed runtime.

Just out of curiosity: is there a reason why a program with this condition does indeed compile and can be called without failing?

Thank you for any shared thoughts


Kind regards,



Martijn van Breden

lead software architect






--
This is the RPG programming on IBM i (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/rpg400-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related questions.

--
This is the RPG programming on IBM i (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/rpg400-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related questions.

--
This is the RPG programming on IBM i (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/rpg400-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related questions.