Parrot 6.3.0 "Black-cheeked Lovebird" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 15 April 2014.
00:05 rurban joined 00:08 rurban1 joined 01:10 rurban joined 01:12 rurban1 joined 01:52 FROGGS_ joined 02:12 rurban joined 02:35 rurban joined 02:37 rurban1 joined 04:44 rurban joined 04:46 rurban1 joined 05:40 rurban joined 05:46 rurban1 joined 06:41 FROGGS joined 06:58 rurban joined
diginet I swear this channel used to be much more active 07:22
FROGGS diginet: well, you could help there 07:29
diginet hahah
FROGGS :o)
diginet I am so impressed by the documentation/tools available for writing compilers for Parrot
by far the best I've ever seen, because even I could understand it :P 07:30
I've been fascinated by the possibility of a sort of semantic analogue of EBNF, but I'm not even sure if that's pragamtically possible
FROGGS hmmm, I must admit that I usually do not read documentation... I prefer to read the implementation and the test :/ 07:38
tests*
08:00 rurban joined 08:03 basiliscos joined 09:04 rurban joined 09:51 TonyC_ joined 10:07 rurban joined 11:08 rurban joined 12:02 basiliscos joined 12:11 rurban joined 12:46 rurban joined
dalek rrot: 382c74d | util++ | t/dynoplibs/debug.t:
[test] t/dynoplibs/debug.t - Skip debug_break on darwin also. See GH#1071
13:55
rrot: d0e2d9b | util++ | / (6 files):
Prepare for the 6.4.0 release.
14:05 rurban joined 14:19 travis-ci joined
travis-ci [travis-ci] parrot/parrot#1027 (master - d0e2d9b : Bruce Gray): The build passed. 14:19
[travis-ci] Change view : github.com/parrot/parrot/compare/6...e2d9be76d4
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/25696625
14:19 travis-ci left
pmichaud if working on the 6.4.0 release, note that I just bumped github.com/parrot/parrot/issues/1014 14:22
Util pmichaud: Ack! I just pushed the tarballs out 2 minutes ago. Sorry! 14:24
pmichaud I just found the problem about 30 mins ago 14:25
Util pmichaud: Would I need Kubuntu specifically to create a fix, or will any Ubuntu do?
pmichaud I suspect it's any Ubuntu
I'll add my error output to the ticket 14:26
added. 14:27
Util pmichaud: It certainly looks fixable by the next release next month. Will that be sufficient, or would you like an 6.4.1 release earlier?
All: Release is complete. OK to commit to master branch.
Release was delayed due to debug.t failing on Darwin.
I will send the release announcement shortly.
pmichaud Util: I'll leave it up to you to decide. Just know that it means rakudo/parrot isn't building on debian-based systems at the moment. 14:28
Util OK, thanks! pkg-config is listed as deprecated for Parrot since 3.3, but is in use for libFFI. I will dig in and see how long the fix will take. 14:31
Expect an answer at the next #parrotsketch meeting, with thr release the same day if all looks good.
rurban_ pmichaud: Just the get the build out I symlinked the missing amd64 headers to the global incdir. 14:41
pmichaud rurban_: I don't follow that, sorry. 14:42
rurban_ multiarch moved the headers to /usr/include/x86_64-linux-gnu/unicode
cd /usr/include; ln -s /usr/include/x86_64-linux-gnu/unicode 14:43
FROGGS that is the workaround, yes
but not the solution :o)
pmichaud I'll just wait for parrot to fix the problem. 14:44
rurban_ ok
it's indeed a wrong parrot probe 14:49
dalek rrot/rurban/icuheaders-gh1014: fc280ef | rurban++ | / (2 files):
[config] fix failing ICU headers probes on multi-arch

We must not die when perl does not find the icuheaders. cc needs to find it, and it does it in the probe if they are in the proper cc searchdir.
icu-config does not store the full headerpath, only pkg-config. Fixes [GH #1014] on debian/ubuntu
15:10
rurban_ I removed my symlink now
FROGGS rurban++ 15:13
will test that after rakudo release this week 15:14
pmichaud testing locally 15:23
got past the configure stage
Parrot builds, "make test" passes. 15:27
rurban_ I'll test it on more weird system with a special icu --prefix path 15:32
my solaris is weird enough for this test
15:35 FROGGS joined
rurban_ I'll also merge the BSD __APPLE__ fix for freeglut today 16:20
dalek rrot: e3462ec | rurban++ | / (2 files):
[config] fix failing ICU headers probes on multi-arch

We must not die when perl does not find the icuheaders. cc needs to find it, and it does it in the probe if they are in the proper cc searchdir.
icu-config does not store the full headerpath, only pkg-config. Fixes [GH #1014] on debian/ubuntu
16:26
rrot: 475ed1a | rurban++ | / (2 files):
Merge branch 'rurban/icuheaders-gh1014'

Confirmed ok by pmichaud on Kubuntu 14.04, and also tested ok on solaris
16:32 Chirag joined
dalek rrot: c646e43 | rurban++ | / (2 files):
[opengl] Fix BSD problem which does not define __APPLE__

Fix opengl on bsd which does not have __APPLE__ defined as 0. freeglut from 2.0 to at least 2.8 relies on that define. Fixes [GH #1070].
16:45
17:00 travis-ci joined
travis-ci [travis-ci] parrot/parrot#1028 (master - 475ed1a : Reini Urban): The build passed. 17:00
[travis-ci] Change view : github.com/parrot/parrot/compare/d...5ed1a59115
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/25710333
17:00 travis-ci left 17:21 travis-ci joined
travis-ci [travis-ci] parrot/parrot#1029 (master - c646e43 : Reini Urban): The build passed. 17:21
[travis-ci] Change view : github.com/parrot/parrot/compare/4...46e43181c4
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/25711896
17:21 travis-ci left
Chirag Hey! 18:17
FROGGS hi Chirag 18:18
Chirag Hi!
I needed a little help
I am trying to make install after adding WB to my branch.. but it fails.. rurban & I know that its the pmc2c thats misbehaving.. But i cant figure out the point of failure 18:20
to put the question in context, this is what I get - 18:21
./parrot-nqp --target=pir --output=compilers/opsc/gen/Ops/Compiler/Actions.pir compilers/opsc/src/Ops/Compiler/Actions.pm
PAST::Compiler can't compile node of type ResizableIntegerArray 18:22
18:22 basiliscos joined
FROGGS I guess PAST::Compiler wants PAST:: types 18:22
but yeah, questions is: where does that array sneak in... 18:23
Chirag probably when a testcase creates a new ResizableIntegerArray pmc?
FROGGS but that ResizableIntegerArray should never land in code-gen 18:25
so, something calls $compiler.as_post(...) with an ResizableIntegerArray 18:26
moderator Parrot 6.4.0 "Double-eyed Fig Parrot" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC 18:26
Chirag rurban suggested that I step through pmc2c but I am not sure which executable to use for gdb (I am not very good at that) 18:28
maybe parrot with a testcase that uses ResizableIntegerArray 18:29
rurban perl -d … stepping, not gdb 18:31
Chirag oops 18:32
18:33 kid51 joined
kid51 parrot builds and tests fine for me at c646e43181c4b2e64ec5ab979c2fa0ede4dde3a7 on Linux/i386 (4 variants) and Linux/x86_64 18:34
rurban_ kid51: thanks 18:45
we really need a freebsd smoker now
revived my freebsd vm now... 18:55
kid51 And Parrot continues to buid on Mac OS X 10.4 (Darwin/PPC)! 19:09
rurban_ that's one of my stable smokers: perl514.cpanel.net/build/one_line_per_build 19:10
just testO1__t/pmc/socket.t failed on my centos5 smoker: perl514.cpanel.net/build/builders/p...logs/stdio 19:11
19:17 TonyC joined 19:40 TonyC joined 19:52 TonyC joined 21:18 davidfetter joined 22:39 rurban joined