Perl 6 - the future is here, just unevenly distributed

IRC log for #pdl, 2016-10-01

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

All times shown according to UTC.

Time Nick Message
09:56 [1]mohawk joined #pdl
09:58 chansen__ joined #pdl
09:58 sivoais_ joined #pdl
10:02 preaction_ joined #pdl
10:06 drrho joined #pdl
10:08 plicease joined #pdl
10:11 lungching joined #pdl
10:47 shadowpaste joined #pdl
10:58 shadowpaste joined #pdl
12:13 shadowpaste joined #pdl
19:08 chm joined #pdl
19:21 Mithaldu joined #pdl
19:22 Mithaldu chm: 8245c5211facbb7f991fffc1b473148b3a17203c still needs to be added, and the releasenotes to need a second pair of human eyes to verify them against the change log and possibly add items not apparent in the notes
19:24 Mithaldu chm: as for bleeding: i don't know. it's things i am implementing for the express purpose of making them available to the things i am doing. due to both time constraints and lack of knowledge on my side i am not currently doing more validation on them than is required to create functionality on my system
19:24 Mithaldu this might often be the correct thing, but i cannot say if it is so in the general case
19:25 chm Mithaldu: hi!
19:25 Mithaldu if possible i'd rather have someone experienced read these commits to see if they need changes for inclusion in master and let me know if so, or do it themselves
19:25 chm Mithaldu: I think you are right in that bleeding may be too bleeding
19:25 Mithaldu also hi :)
19:26 Mithaldu there's also the whole thing in that glew, if i understand it right, will obviate much of the code we have
19:26 chm Would you be ok with a wrap up based on _91 with 8245c5211facbb7f991fffc1b473148b3a17203c and cleaned up release notes?
19:26 Mithaldu 8245 is release notes, so yes, absolutely
19:26 opkick [pdl] wchristian pushed 1 new commit to master: https://git.io/vPqUK
19:26 opkick pdl/master c4eb23a Chris Marshall: qw{} has problems with dmake, use qw[] instead
19:26 chm ok, couldn't tell from the #
19:27 Mithaldu that's why i explained just now :)
19:27 chm My plan was to have a better POGL 1 release (that you could work with for your bleeding, etc...) and then
19:27 chm get to the GLEW POGL2 stuff.
19:27 Mithaldu aye
19:28 chm And yes, the GLEW bindings should make a lot of this go away.
19:28 Mithaldu btw, regarding the mac issues, just so the knowledge is spread a little
19:28 Mithaldu the main problem in mac is that by default it comes only with GLUT, which of course lacks all the bits of freeglut
19:28 Mithaldu but worse
19:29 chm Alright, I'll take a look at the release notes vs the change logs and prep a final rc2 developers release
19:29 Mithaldu it comes with its own set of header files that deviate strongly from the khronos standard ones, which are currently used to generate our headers
19:29 Mithaldu that's why i changed the mac stuff to simply use our own headers and hope for the best
19:29 chm I've pinged bfree a couple of times but it looks good to me.
19:30 Mithaldu didn't expect much response from him, though i hoped at least for a grunt
19:30 Mithaldu my main problem is i looked at this code for so long, i'm a little blind to it ;)
19:30 chm In my experience, folks who are too busy to check things out ahead of a release squeak loudly if something breaks!
19:30 chm Not that I prefer that but 0.70 is really nice...  you've made a shine appear on the old POGL1 codebase.  :-)
19:31 Mithaldu shine is good, yea, much of it was just polishing :)
19:32 chm I was thinking to try some preliminary GLEW bindings experiments this weekend.  Will keep you posted.
19:32 Mithaldu awesome, i'll stick around to see what i hear
19:32 Mithaldu i can also run tests on windows and a mac vm
19:32 Mithaldu i assume you're linux?
19:32 chm Regarding the code for POGL2, the difficult part is that I need to have better GLUT bindings and, ideally GLFW...
19:33 Mithaldu what does better glut bindings mean?
19:33 chm and the existing callback structure is *really* difficult to follow so I was hoping to clean that up.
19:33 Mithaldu callback structure?
19:34 chm All the user interfaces use some type of callback which need to have perl bindings so you can have a perl callback routine that works as if it were C.
19:34 Mithaldu oh you mean for the event loop
19:34 chm Basically some tricky wrapper XS code.  Regular patterns to it but needs to be right.
19:34 Mithaldu i have not looked at that
19:35 Mithaldu though i did notice apple complaining about it since SV or AV or something like that was being passed to a function as a callback param and converted to an int and then later on back again
19:35 chm We're also wrapping up for a PDL-2.017 release which has a number of bugs fixed and some cleanup and polish as well.
19:35 Mithaldu \o/
19:36 chm Yes, that is broken code from when IVs were 32bits.
19:36 Mithaldu oh good
19:36 Mithaldu OH YEAH
19:36 Mithaldu that reminds me
19:36 chm BTW, I've been looking into the COS (C Object System) and it is very nice.
19:36 Mithaldu why test.pl was fucked, important to know
19:37 Mithaldu first off, the event loop uses threads
19:37 chm My hopes of having a better OO C api for PDL next gen look pretty good.
19:37 Mithaldu this is important
19:37 Mithaldu in glut using exit to leave the event loop is fine, it doesn't mind that
19:37 Mithaldu but under freeglut things break if the event loop thread just gets shot in the face like that
19:38 Mithaldu one must signal freeglut to end things, then leave the callback to hand back control and let freeglut do its thing
19:38 chm The good news for FreeGLUT is that we've added error and warning handlers so the perl bindings won't have exit() called in them.  Can't do anything about the GLUT on the Mac.
19:38 Mithaldu uh
19:38 Mithaldu are you talking about my commit regarding that or something else?
19:39 Mithaldu and well, GLUT doesn't mind exit in the event loop, as i said
19:39 Mithaldu gotta admit, i'm aware i'm being a chatty dude right now, but i feel you're skipping some of the things i say :P
19:39 chm The recent FreeGLUT library now has user installable handler functions
19:39 Mithaldu freeglut 3?
19:40 Mithaldu you work on freeglut?
19:40 chm I think I've caught up but the interleaved chat is confusing.
19:40 Mithaldu just handle it one line by one line, resist the urge to scan ahead ;)
19:40 chm I was the one who contributed the functionality so we have GLUT back compatibility and no exit in language bindings.
19:40 Mithaldu to freeglut?
19:40 Mithaldu err
19:41 Mithaldu contributed to freeglut?
19:42 chm Yes, glutInitErrorFunc and glutInitWarningFunc
19:42 Mithaldu nice, i wasn't aware you had done work on freeglut
19:42 chm in since freeglut 2.8.1, I just haven't done new GLUT/FreeGLUT bindings since the plan was to spin off from POGL
19:43 chm Just trying to get POGL to work.... :-)
19:43 chm Had to push the fix upstream and all.  I don't know how many folks have used it.
19:43 Mithaldu ah, gotcha :)
19:44 Mithaldu well, as of now, pogl does
19:44 chm Did you use it in your fix?
19:44 Mithaldu oh wait, sorry, misunderstanding
19:45 Mithaldu i was conflating it with glutLeaveMainLoop
19:45 chm Nice chatting but I'd like to get to the final POGL release candidate.
19:45 Mithaldu chm: still curious to know what systems you have available
19:46 chm There was actually code in Mac GLUT that was just enough to handle it.
19:46 chm I had the glutCloseFunc mapped out since there was a crash.
19:46 chm I should ensure that the close capability is used again if available.
19:46 Mithaldu someone had put it back in, but i was still getting crashes, so i removed it from this release
19:46 Mithaldu might merit revisiting
19:47 Mithaldu oh yeah, bob did
19:47 Mithaldu look at the questionable_ branch
19:47 chm I have Strawberry Perl Portable and cygwin64 both on win7
19:47 Mithaldu oh, windows user
19:47 Mithaldu well in that case i'll only bother you for cygwin testing :)
19:47 chm I'm in the process of building a linux DVR so I might actually have linux soon.
19:48 Mithaldu with a real graphics card would be nice
19:48 chm I'll push the rc2 before I send it to pause.perl.org for you to confirm.
19:48 Mithaldu i have a debian vm, but that card it has is two sticks at an angle
19:48 Mithaldu aight
19:48 chm Catch up with you later. o/
19:50 Mithaldu chm: oh wait
19:50 Mithaldu one thing
19:50 chm still here...
19:51 Mithaldu please consider using https://www.irccloud.com/ for irc in the future, it has many benefits for someone with your use profile, and might even be worth the 4€/mo for always-on connectivity
19:51 Mithaldu has browser/mobile functionality as well
19:51 chm not really a power irc user but thanks for the suggestion, will take a look
19:52 Mithaldu it's exactly that why i suggest it
19:52 Mithaldu and thanks for considering :)
19:52 Mithaldu good luck with the rc :D
19:53 travis-ci PDLPorters/pdl#506 (master - c4eb23a : Chris Marshall): The build passed.
19:53 travis-ci Change view : https://github.com/PDLPorters/pdl/c​ompare/90660cc55e94...c4eb23a6ddfd
19:53 travis-ci Build details : https://travis-ci.org/PDLPo​rters/pdl/builds/164314958
20:26 opkick [pdl] wchristian pushed 1 new commit to master: https://git.io/vPqIE
20:26 opkick pdl/master fccf70e Chris Marshall: Update Changes and Known_problems
20:55 travis-ci PDLPorters/pdl#507 (master - fccf70e : Chris Marshall): The build passed.
20:55 travis-ci Change view : https://github.com/PDLPorters/pdl/c​ompare/c4eb23a6ddfd...fccf70ea7640
20:55 travis-ci Build details : https://travis-ci.org/PDLPo​rters/pdl/builds/164326317
21:01 opkick [pdl] wchristian pushed 2 new commits to master: https://git.io/vPqtY
21:01 opkick pdl/master c3c2225 Derek Lamb: Add better rle output size handling and documentation (SF feat req 80)....
21:01 opkick pdl/master 73d5355 Derek Lamb: Make the signature of rle accurate for the new usage....
21:26 travis-ci PDLPorters/pdl#508 (master - 73d5355 : Derek Lamb): The build passed.
21:26 travis-ci Change view : https://github.com/PDLPorters/pdl/c​ompare/fccf70ea7640...73d5355a491f
21:26 travis-ci Build details : https://travis-ci.org/PDLPo​rters/pdl/builds/164332129
22:22 sivoais_ just saw PDL 2.016_03 go up! chm++
22:26 opkick [pdl] wchristian pushed 1 new commit to master: https://git.io/vPqOv
22:26 opkick pdl/master 3efdac2 Chris Marshall: Update Changes for PDL-2.016_03 developers release
22:26 opkick [pdl] wchristian deleted sf080_rle at 2151584: https://git.io/vPqOf
22:31 opkick [pdl] wchristian tagged v2.016_03 at 4742954: https://git.io/vPqOO
22:41 mohawk w00t
22:51 travis-ci PDLPorters/pdl#509 (master - 3efdac2 : Chris Marshall): The build passed.
22:51 travis-ci Change view : https://github.com/PDLPorters/pdl/c​ompare/73d5355a491f...3efdac24e80c
22:51 travis-ci Build details : https://travis-ci.org/PDLPo​rters/pdl/builds/164344051

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