Perl 6 - the future is here, just unevenly distributed

IRC log for #pogl, 2017-02-01

| Channels | #pogl index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 CandyAngel Oh, I see.
00:03 CandyAngel I have to call glewInit() after glutCreateWindow() to use it!
00:59 CandyAngel chm: If OpenGL::Modern is missing any function I'm using, should I just import them from OpenGL and add them to the github issue?
01:44 chm joined #pogl
01:46 chm CandyAngel: Well that is good feedback.  At least calling glewInit() should be documented.
01:47 chm Eventually, it should not be something users worry about as it should be called transparently.
01:47 chm I'm not sure the best way to do that.  Opening an issue to track.
01:48 CandyAngel Do you read the channel logs or something?
01:49 chm CandyAngel: OpenGL::Modern should have bindings to all OpenGL APIs so missing functions are bad.
01:50 chm That means you have a function glXxxx or glXxxx_c which work.
01:51 CandyAngel I don't think Mithaldu uses it, so glGenRenderBuffers_c is missing
01:51 CandyAngel Or at least.. seems to be missing..
01:52 CandyAngel It's not listed in any of those required-api*.txt files you posted
01:54 chm It is glGenRenderbuffers with a lowercase b and there is glGenRenderbuffers_c
01:55 chm If possible, it would be nice if you could confirm or disprove that the available bindings work.
01:55 chm If there is a better bindings (more perlish, packed strings, whatever) in OpenGL, please report that.
01:56 chm I'm prioritizing development so the ones being used can be done first.
01:56 CandyAngel Oh yeah, I typed it wrong
01:56 * CandyAngel facepalms
01:56 CandyAngel I'm sorry!
01:56 chm I read the channel logs since I'm not always able to get to #pogl.
01:57 CandyAngel Ah, I'm just using ZNC so I'm "always" connected
01:58 chm You can also send email to the mailing lists at the sf.net site: https://sourceforge.net/p/pogl/mailman/?source=navbar
02:02 CandyAngel Should I still be using OpenGL::Shader for shader things or is that in OpenGL::Modern now?
02:02 chm The plan is to spin off the original/legacy OpenGL components so they can be used with OpenGL::Modern
02:03 chm rather that re-inventing the wheel.  Regarding OpenGL::Shader, I don't have maintainer or co-maintainer
02:04 chm status and the current owner has not been able to be reached.
02:04 chm Ideally, we could update that package to go with OpenGL::Modern but I think it might be cleaner
02:05 chm to do an update as a new module.  If you're using legacy functions, definitely let us know so
02:05 chm that they stay on the radar.
02:06 chm I anticipate that we might want a more general OO framework for OpenGL::Modern since,
02:06 chm if you look at it, most of the new functionality is organized like objects with properties and methods.
02:07 chm FYI: This issue https://github.com/devel-chm/OpenGL-Modern/issues/7 has a table of implemented funs near the end.
02:08 CandyAngel Yeah, that's why I thought glGenRenderbuffers was missing, it isn't listed there
02:09 chm The table only lists the routines reported as being used.  Not every implemented routine.
02:09 CandyAngel Ah okies, that makes sense
02:09 chm For example, all glXxx() routines taking a pointer/array type argument and/or returning a pointer/array value
02:10 chm has an glXxx_c binding.  It isn't listed in the table unless it was "of interest".
02:10 chm That is why verification of routines you use is important.
02:10 * CandyAngel nods
02:11 chm For example, there are 1329 OpenGL routines which have a single pointer/array argument.
02:11 chm Needlesstosay, I haven't done much checking.  :-)
02:12 chm The good news is that there are 1263 that don't have any pointer/array args so should be ok.
02:13 CandyAngel Coolies
02:13 CandyAngel Thank you so much for doing this :)
02:33 CandyAngel Hm.. will the constants be provided as they are now (e.g. GLUT_KEY_F1) or will there be variable versions of them ($GLUT_KEY_F1) too?
02:33 chm At the moment, it is the standard "constant sub" approach.
02:34 chm I was thinking about implementing $GL_CONSTANTS to match GL_CONSTANTS.
02:35 chm Rather than cluttering up the base module, it would be easy to add as a suppliment
02:35 CandyAngel That would be good, because state $name_for = { GLUT_KEY_F1 => 'whatever' }; doesn't work >.<
02:36 chm I mean supplement.
02:36 chm Doesn't {GLUT_KEY_F1() => 'work'} ?
02:41 chm Maybe the way to go is to use the standard AUTOLOAD to generate the constant sub
02:42 chm and have the AUTOLOAD routine have the side effect of creating the same value as readonly scalar.
02:46 chm That way, we don't need massive amounts of code/text duplication.
02:46 CandyAngel Oh, that sounds neat
02:46 CandyAngel Okies, everything is working except the shader using OpenGL::Shader (doesn't seem to do anything..)
02:48 ilbot2 joined #pogl
02:48 Topic for #pogl is now Perl OpenGL discussion and development | https://p3rl.org/OpenGL | Mailing lists: https://sourceforge.net/p/pogl/mailman/ | Channel is logged by ilbot2: https://irclog.perlgeek.de/pogl/today |
02:48 CandyAngel Oh, yeah, I did something stupid
02:52 CandyAngel Okie, simple OpenHMD with OpenGL::Modern test script is working :)
02:52 CandyAngel And on that note, I should head off to bed (is 3am, have work in a few hours!)
02:53 CandyAngel See you tomorrow :)
02:59 chm o/
12:37 chm joined #pogl
21:26 CandyAngel chm: Is there anything you would like me to do next?
22:42 chm joined #pogl
22:49 chm CandyAngel: Were you able to finish your checking?  Which routines were covered and what was the result?
22:49 chm CandyAngel: On the same note, I've attached a list of the priority function signatures to issue 7
22:50 chm at https://github.com/devel-chm/OpenGL-Modern/files/745902/priority.txt and I would like
22:50 chm to have a test/demo program[s] that show each function working.
22:51 chm For multiple routines with the same base name, we only need to verify 1
22:51 chm as the other routine should be basically cut-and-paste.
22:52 chm I would like to verify that all the priority routines (with and without _c as implemented) work.
22:52 chm The ones needed checks are the _c variants.
22:53 chm The test programs can be then modified to verify that any _p or nicer routine flavors work as well.
22:54 CandyAngel The test script just implements the basics of HMD rendering, which is all working fine
22:55 CandyAngel https://gist.github.com/CandyAngel/e589e9c7d8de2a50e11aad566ec27455
23:46 chm Great, that gives a list of the _c routines uses.

| Channels | #pogl index | Today | | Search | Google Search | Plain-Text | summary