Adam Taylor-Williams Profile
Adam Taylor-Williams

@adam_on_i

Followers
275
Following
786
Statuses
799

@KatoIntegration IBM i dev. #ibmi enthusiast, PASE aficionado, ILE fan. @GatewayTech grad. He/Him. 🏳️‍🌈

Wisconsin, USA
Joined February 2017
Don't wanna be here? Send us removal request.
@adam_on_i
Adam Taylor-Williams
6 months
It'd be nice if RPG had a figurative constant that was equivalent to *Allx'00' which worked on all (non-pointer) data types.
1
0
1
@adam_on_i
Adam Taylor-Williams
7 months
@notesofbarry I think this has to be right - I tried using _NPMPARMLISTADDR and was able to get the full parameter list regardless of the prototype, and access the unprototyped parm values.
0
0
3
@adam_on_i
Adam Taylor-Williams
7 months
@ImRickyT I'd forgotten they added these BIFs! I don't think I can do anything with them inside of the "OldProc" though b/c they'll hit a compile error due to no second parm to reference, but this does give me an idea of something different to test!
0
0
0
@adam_on_i
Adam Taylor-Williams
7 months
@kadlerio So far it doesn't look like ARGOPT(*YES) makes a difference, at least not for the test cases I've tried so far. Also tried with OPTIMIZE(*FULL) for good measure in case that would have an impact.
0
0
0
@adam_on_i
Adam Taylor-Williams
7 months
@k0ku3n @notesofbarry That's kind of what I've been wondering!
0
0
0
@adam_on_i
Adam Taylor-Williams
7 months
@AlexanderStix Just threw a writeup and a snippet up on Gitlab
0
0
0
@adam_on_i
Adam Taylor-Williams
7 months
@AlexanderStix Pretty much yeah, the compiler injects another subproc at compile-time in place of the on-exit.
0
0
0
@adam_on_i
Adam Taylor-Williams
7 months
@OSSJesseG That's impressive, thanks! I'd been contemplating building out a Python script to do something like this, but this is way easier.
0
0
0
@adam_on_i
Adam Taylor-Williams
7 months
It's wild that CL hasn't added support for <> or != as ways to handle not-equal logical comparisons. Using ¬︎= is wonky, and using *NE just makes code harder to read. The compiler clearly supports multiple methods for not-equal already, how hard could adding <> and/or != be?
0
0
5
@adam_on_i
Adam Taylor-Williams
7 months
I guess that also extends to the control spec CCSID keyword as well, if I'm being honest.
0
0
2
@adam_on_i
Adam Taylor-Williams
8 months
@richardschoen Honestly fascinating! Do you remember how long ago that was? Kind of curious if it's something that might have gotten fixed via a PTF or if it's still buggy and something I could open a case on.
1
0
0
@adam_on_i
Adam Taylor-Williams
8 months
@richardschoen Thanks for pointing me at those! I'd been looking at an older fork of QSHONI which didn't have the SRVPGM. Just curious, is there a reason why QSHONI programs are all OPM?
1
0
0
@adam_on_i
Adam Taylor-Williams
8 months
Pillow does not seem to want to build for me inside a virtualenv on the IBM i and I don't know enough to know what I'm missing. Feels irresponsible to just rely on the system package?
0
0
1