Parrot 4.1.0 "Black-headed Parrot" | parrot.org | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 28 February 2012.
00:29 kid51 joined
dalek rrot/remove_sub_flags: 95c870b | Whiteknight++ | runtime/parrot/library/ (21 files):
Fix remaining load_bytecode_s and load_language_s references in runtime. Build completes.
 00:34
00:42 travis-ci joined
travis-ci [travis-ci] parrot/parrot#142 (remove_sub_flags - 95c870b : Whiteknight): The build is still failing. 00:42
[travis-ci] Change view : github.com/parrot/parrot/compare/2......95c870b
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/843242
00:42 travis-ci left 00:49 Patterner joined 01:06 alester joined
dalek rrot/remove_sub_flags: 9b9e1d5 | Whiteknight++ | / (7 files):
re-add the load_bytecode_s opcode with explicitly automagic triggering of :tag('init') Subs
 01:20
rrot/remove_sub_flags: ae6ba06 | Whiteknight++ | src/ops/core (2 files):
fix throwback load_bytecode_s op. I was hoping it would fix some tests, it didn't
01:32 travis-ci joined
travis-ci [travis-ci] parrot/parrot#143 (remove_sub_flags - ae6ba06 : Whiteknight): The build is still failing. 01:32
[travis-ci] Change view : github.com/parrot/parrot/compare/9......ae6ba06
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/843460
01:32 travis-ci left
whiteknight I'm starting to really dislike travis 01:35
benabik It seems to be ignoring the "branches: only: - master" declaration 01:36
whiteknight let me try something 01:37
dalek rrot/remove_sub_flags: c962702 | Whiteknight++ | .travis.yml:
Use the same branches/only/master declaration as is on master, to try and get travis to ignore this branch
whiteknight anyway, I'm out for the night. Later 01:38
dalek rrot: 7baf549 | petdance++ | frontend/parrot_debugger/main.c:
consted some local vars
 01:47
rrot: 1ea49e0 | petdance++ | frontend/parrot_debugger/main.c:
Don't pass potentially-NULL pfraw into Parrot_pf_get_packfile_pmc()
02:00 lateau joined 02:08 travis-ci joined
travis-ci [travis-ci] parrot/parrot#144 (master - 1ea49e0 : Andy Lester): The build was broken. 02:08
[travis-ci] Change view : github.com/parrot/parrot/compare/a......1ea49e0
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/843595
02:08 travis-ci left
ttbot Parrot 1ea49e0e MSWin32-x86-multi-thread make error tt.taptinder.org/cmdinfo/72488 02:17
bacek_at_work alester, looks like you broke win32 build. ^^^ 02:42
alester How now
Argh
bacek_at_work "declaration before code" 02:44
alester Same as the C90 thing earlier in the week.
Why is GCC happy with that? Hmm
bacek_at_work my gcc isn't happy either 02:45
4.6.2
alester Are you GCC or G++?
Trying again with gcc 02:46
bacek_at_work gcc
alester So there we go. It failed for me under GCC as well. But not G++
ok
bacek_at_work c++ is much more relaxed. And it's perfectly valid C++
frontend/parrot_debugger/main.c:198:13: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement]
cc1: some warnings being treated as errors
alester Right, I know. 02:47
Wondering if I can get G++ to complain about that. 02:48
g++ is pickier about other things, too. So I do g++ for cage-y things.
OK, I'm working on a fix to make g++ complain more. 02:55
benabik ... We're trying to get compilers to complain _more_? I didn't think we had that problem. ;-) 03:09
alester Yes, because the more we can complain, the more we can quiet the little complaints and turn on the watches for that failing code. 03:14
Strict = good.
dalek rrot: 6c09f3b | petdance++ | config/auto/warnings.pm:
adding another flag to see if we can get more pickiness
 03:26
rrot: fea6177 | petdance++ | frontend/parrot_debugger/main.c:
fix the code-before-declarations error
 03:30
rrot: 52a9527 | petdance++ | / (4 files):
add all the frontend programs to be headerized. A headerized program is a happy program.
 03:43
03:54 travis-ci joined
travis-ci [travis-ci] parrot/parrot#145 (master - 6c09f3b : Andy Lester): The build is still failing. 03:54
[travis-ci] Change view : github.com/parrot/parrot/compare/1......6c09f3b
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/844087
03:54 travis-ci left
dalek rrot: 1c3c908 | petdance++ | frontend/pbc_dump/main.c:
adding headerizer macros
 04:02
04:02 travis-ci joined
travis-ci [travis-ci] parrot/parrot#146 (master - fea6177 : Andy Lester): The build was fixed. 04:02
[travis-ci] Change view : github.com/parrot/parrot/compare/6......fea6177
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/844104
04:02 travis-ci left
dalek rrot: fd023f8 | petdance++ | frontend/pbc_dump/main.c:
Adding headerizer macros
 04:08
04:09 travis-ci joined
travis-ci [travis-ci] parrot/parrot#147 (master - 52a9527 : Andy Lester): The build was broken. 04:09
[travis-ci] Change view : github.com/parrot/parrot/compare/f......52a9527
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/844132
04:09 travis-ci left 04:14 preflex_ joined
dalek rrot: 59b35dc | petdance++ | frontend/parrot_debugger/main.c:
added headerizer macros
 04:16
04:48 travis-ci joined
travis-ci [travis-ci] parrot/parrot#150 (master - 59b35dc : Andy Lester): The build is still failing. 04:48
[travis-ci] Change view : github.com/parrot/parrot/compare/f......59b35dc
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/844275
04:48 travis-ci left
ttbot Parrot 6c09f3bc MSWin32-x86-multi-thread make error tt.taptinder.org/cmdinfo/72652 05:20
06:48 fperrad joined 07:58 preflex_ joined 08:46 mj41 joined 09:22 lucian joined 10:34 schm00ster joined 12:05 alin joined 12:13 lateau joined 12:23 bluescreen joined 12:45 JimmyZ joined 12:48 colomon joined
colomon I'm terribly afraid this is somehow my fault, but the latest parrot fails at the configuration stage on OS X for me. 12:54
moritz tries parrot master on linux 12:55
colomon gist.github.com/2021664
moritz: it also happens for me in rakudo's version of parrot 12:56
moritz wow, that's incredibly noisy 12:57
colomon: can you compile a very simple C file with gcc on your machine? 13:08
something like int main(int argc, char** argv) { return 0 }
colomon int main(int argc, char** argv) { printf ("see, moritz\\n"); return 0; } 13:09
Wynne:parrot colomon$ gcc frip.c
Wynne:parrot colomon$ ./a.out
see, moritz
no prob
moritz hm, so it's not a very simple gcc problem 13:10
Coke colomon: what arguments did you invoke the build with?
colomon Coke: I did perl Configure.pl 13:11
Coke what gcc are you using? what perl?
colomon: did you do that in parrot or rakudo's build dir? 13:12
colomon Coke: parrot
once I realized rakudo's issue was parrot, I created a fresh clone of parrot
13:40 alin_ joined 13:46 alin joined 14:41 lateau joined, fperrad joined 14:43 alester joined 14:50 contingencyplan joined 15:19 bluescreen joined 15:45 Psyche^ joined
cotto ~~ 15:45
Coke colomon: so, what gcc are you using? What perl? 15:49
colomon This is perl 5, version 14, subversion 0 (v5.14.0) built for darwin-thread-multi-2level 15:50
i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
15:56 cogno joined 16:22 cogno joined 16:29 dngor joined
dukeleto ~~ 16:40
16:46 cogno joined 16:50 wagle joined 17:06 alin joined
Coke colomon: sorry, those seem normalish. 17:07
17:10 johbar_work joined
colomon oooo... if I turn off the perlbrew perl, configure works fine 17:10
Coke O_o 17:18
colomon yup, full parrot build, full rakudo build on top of it. 17:24
Coke O/\\o 17:27
... that's either a high five or admiral ackbar.
colomon "It's a trap!"
any idea why a perlbrew-built perl won't work for the configuration? surely I cannot be the only person out there using perlbrew... 17:29
17:30 kthakore left
Coke well, Configure.pl relies heavily on perl5 for config probes, both code and data. 17:32
did you get any sort of config hash the first time that you could diff with the working one? 17:33
tadzik good evening #parrot
colomon Coke: "config hash"? 17:38
Coke dobry wieczór, tadzik.
colomon: look for files named *config* in your working root dir. 17:39
colomon Coke: there's a config/ directory... and a config_lib.pir file in parrot root 17:40
Coke e.g. "config_lib.pir"
do you get any of that with the broken run?
also, you could run Configure.pl --with-some-option-that-enables-verbose-output but I forget what it's called.
benabik I would hope it's --verbose
colomon let me try again with the perlbrew perl 17:43
Coke: here's the diff of the broken one with the working one: gist.github.com/2023598 17:45
Coke broken first? 17:46
what's the -m32 option to gcc?
benabik Hm. 17:47
moritz align stack frames to 32 bits 17:48
benabik No.
Generate 32b code.
colomon what benabik says
benabik It seems to compile with -m32 but link with -arch x86_64
colomon yeah, that does sound like trouble, doesn't it? 17:49
benabik Which is why it can't find _main at link time.
18:04 lucian joined 18:05 dmalcolm joined
dukeleto How have I never heard of this before? packagist.org 18:08
benabik dukeleto: By not using PHP? 18:10
dalek kudo/nom: bd67428 | moritz++ | src/core/ (2 files):
fix list assignment to object hashes

This moves the key type coercion to STORE_AT_KEY, which is replaced by a non-coercing method in the case of object hashes		 18:56
19:04 mj41 joined 19:18 johbar_work joined
cotto so much promise, yet they recommend blindly executing a script retrieved over http 19:30
tadzik well, so does cpanminus 19:31
dukeleto and so does rvm... 20:01
if you are using SSL, it is a bit better. But still...
cotto if everyone else was installing jump_off_a_cliff.exe...
20:07 brambles joined
dukeleto Completely OT: I spent the weekend setting up a webcam for my turtle: unstablemable.github.com 20:09
alester not "mabel"? 20:15
dukeleto alester: seems like Mable and Mabel are used. I like "Mable" :) 20:17
alester Mabel isn't the name of the turbtle/
?
dukeleto alester: yes, "Unstable Mable" is the name of the turtle :) 20:18
20:37 perlite_ joined
cotto reminder: #PS is an hour later tomorrow for US folks due to DST 20:58
21:20 dngor joined 21:30 PacoAir joined 21:37 trey joined 21:49 dngor joined 22:04 preflex_ joined 22:12 PacoAir joined 22:13 kid51 joined, dngor joined 22:40 alin joined 22:42 whiteknight joined
whiteknight good evening, #parrot 22:59
tadzik good evening whiteknight 23:00
whiteknight hello tadzik
23:05 lucian joined 23:20 jsut joined 23:29 davidfetter joined 23:40 dngor_ joined 23:50 dngor joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!