Camelia, the Perl 6 bug

IRC log for #parrot, 2011-09-23

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:45 plobsing ~~
01:05 whiteknight hello plobsing
01:06 plobsing kinda dead in here tonight
01:06 plobsing what are you working on ATM whiteknight?
01:07 whiteknight nothing really. I wanted to help nine with that gsoc_threads build failure tonight but I've been busy
01:07 plobsing i replied to his email on parrot-dev. hopefully that helps.
01:08 whiteknight yeah, I saw that. I had the same basic thought
01:08 whiteknight make bootstrap-ops is messy and always makes me angry. I'm happy to see a workaround suggested
01:20 cotto_work Somehow his branch has opsc generating invalid C code.
01:23 whiteknight awesome
01:25 whiteknight what is it generating wrong?
01:25 cotto_work return0;
01:25 whiteknight ...fail
01:26 cotto_work not sure how or why
01:26 benabik I'd bet he merged the checked in bootstrap code.
01:27 cotto_work that's after copying known-good code into the branch and trying to rebootstrap
01:28 benabik Hm.
01:28 benabik Fascinating.
01:28 cotto_work yeah
01:28 cotto_work eyes welcome
01:29 whiteknight I'll have to stare at it tomorrow. Goodnight
01:30 woosley joined #parrot
01:33 cotto_work It's quite odd.  If it were on parrot.git I'd push something to get the non-bootstrap part of the build working.
02:30 dalek parrot/kid51/release_update_version: d0e4c28 | jkeenan++ | / (4 files):
02:30 dalek parrot/kid51/release_update_version: Encapsulate all functionality in tools/release/auto_release.pl.  Move subs into new package Parrot::Release::Functions.  Import those functions into tools/release/auto_release.pl.
02:30 dalek parrot/kid51/release_update_version: review: https://github.com/parrot/parrot/commit/d0e4c28860
02:30 dalek parrot/kid51/release_update_version: 8e43c3a | jkeenan++ | / (2 files):
02:30 dalek parrot/kid51/release_update_version: tools/release/update_version.pl is now subsumed within auto_release.pl.
02:30 dalek parrot/kid51/release_update_version: review: https://github.com/parrot/parrot/commit/8e43c3a032
02:40 dalek parrot: ecdb453 | jkeenan++ | MANIFEST:
02:40 dalek parrot: Rerun mk_manifest_and_skip.pl.
02:40 dalek parrot: review: https://github.com/parrot/parrot/commit/ecdb4537ff
02:40 cotto_work seen kid51
02:40 aloha kid51 was last seen in #parrot 15 days 14 hours ago saying "afk".
02:40 cotto_work anyone know what the benefit of moving that code into a module is?
02:52 soh_cah_toa i can't really think of a significant benefit other than increasing encapsulation
02:52 soh_cah_toa which in a script like auto_release.pl, isn't *that* necessary
02:52 soh_cah_toa but it is nice and clean so i can't complain too much ;)
03:01 Coke \o/ # dayjob cram month over.
03:02 soh_cah_toa nice :)
03:10 soh_cah_toa anybody know if Pod::Usage is a core module? how do you find out that sorta thing?
03:19 benabik soh_cah_toa: corelist
03:20 benabik soh_cah_toa: corelist Pod::Usage gives me Pod::Usage was first released with perl v5.6.0
03:20 soh_cah_toa oh neat, i think i remember that tool now
03:21 soh_cah_toa we might wanna think about using that in our developer scripts
03:29 soh_cah_toa benabik: how much code did you send as your code sample? an entire file? just a few lines? what?
03:35 dalek parrot: 169f00b | soh_cah_toa++ | tools/release/auto_release.pl:
03:35 dalek parrot: Added --test_jobs switch for specifying number of parallel test harnesses as suggested by benabik++. Also added support for $TEST_JOBS environment variable as suggested by dukeleto++.
03:35 dalek parrot: review: https://github.com/parrot/parrot/commit/169f00b213
03:35 dalek parrot: bfa7f50 | soh_cah_toa++ | tools/release/auto_release.pl:
03:35 dalek parrot: Changed --test_jobs to --test-jobs because that's what I originally meant to name it.
03:35 dalek parrot: review: https://github.com/parrot/parrot/commit/bfa7f506c5
03:36 benabik soh_cah_toa: I think I sent in a copy of the `git log --author="Brian Gernhardt" --since="some date" --before="some date"
03:37 benabik There was some voice on the GSoC student list about how to do it
03:37 benabik *noise
03:37 soh_cah_toa ok
03:43 cotto ~~
03:50 soh_cah_toa benabik: i can't seem to find the date format the --from and --until switches accept. i thought i could use iso 8601
03:51 soh_cah_toa fatal: unrecognized argument: --from=2011-05-19
03:55 soh_cah_toa uh duh, --from isn't a switch. i meant --since :(
03:55 soh_cah_toa nevermind
03:55 benabik Yeah.
04:00 nopaste "soh_cah_toa" at 192.168.1.3 pasted "git log --author="soh_cah_toa' --since="2011-05-18" --until=2011-08-21"" (1264 lines) at http://nopaste.snit.ch/81976
04:00 soh_cah_toa benabik: so you're saying something like that ^ is ok?
04:02 benabik Oh.  Add -p
04:02 soh_cah_toa ok
04:10 soh_cah_toa alright. done
04:11 soh_cah_toa that could've been pretty bad if i hadn't cleaned up my email and noticed that old email
04:11 JimmyZ joined #parrot
04:14 benabik I think you just wouldn't have gotten the t-shirt
04:14 soh_cah_toa or the certificate it said
04:14 soh_cah_toa plus, i totally want that t-shirt
04:14 soh_cah_toa i've been waiting and waiting ;)
04:21 benabik You want to be certifiable?
04:22 soh_cah_toa well, of course :\
04:29 dalek Heuristic branch merge: pushed 21 commits to parrot/kill-useless-defaults by plobsing
04:39 alvis joined #parrot
05:09 perlite joined #parrot
05:22 alvis joined #parrot
05:22 SHODAN joined #parrot
06:05 jsut_ joined #parrot
07:13 awwaiid joined #parrot
07:18 rfw joined #parrot
07:26 mj41 joined #parrot
07:41 autark joined #parrot
08:06 dalek winxed: b6fb889 | NotFound++ | winxedst1.winxed:
08:06 dalek winxed: don't use a bare identifier as hash key if it's found in scope
08:06 dalek winxed: review: https://github.com/NotFoun​d/winxed/commit/b6fb889a22
08:06 NotFound msg whiteknght Fixed in b6fb889a22
08:06 aloha OK. I'll deliver the message.
08:06 NotFound msg whiteknight Fixed in b6fb889a22
08:06 aloha OK. I'll deliver the message.
08:12 lucian joined #parrot
08:17 Patterner "get_pointer() not implemented in class 'GLOBALish'" is bad, right?
08:20 moritz Patterner: are you trying to compile nqp?
08:20 Patterner guilty...
08:20 moritz it means you should stick to parrot 3.8.0 for now
08:21 moritz there is a fix for that problem in an nqp branch (kill-useless-defaults), but it will only be merged after the rakudo and nqp releases
08:42 cotto moritz, I'm glad Rakudo's users won't see any impact from this, but it's substantially lta on the parrot side.
08:43 cotto gotta iterate the process though
09:34 cotto joined #parrot
09:38 woosley left #parrot
09:55 dalek rakudo/nom: 7a51746 | moritz++ | src/core/Buf.pm:
09:55 dalek rakudo/nom: [Buf] remove cruft noticed by JimmyZ++
09:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/7a51746e53
11:32 woosley joined #parrot
11:40 Psyche^ joined #parrot
11:53 dalek rakudo/nom: 09a598f | moritz++ | / (2 files):
11:53 dalek rakudo/nom: implement bitwise operators for Buf
11:53 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/09a598fed2
12:02 whiteknight joined #parrot
12:03 whiteknight good morning, #parrot
12:03 tadzik good morning whiteknight
12:05 moritz good am, wk
12:17 whiteknight hello tadzik, moritz. how are you two doing today?
12:33 bluescreen joined #parrot
12:41 * Coke yawns.
12:48 * moritz is also half asleep
12:48 moritz it's been a hard week with far too little sleep
12:50 whiteknight I would be sleeping just fine if it weren't for the kid
12:50 whiteknight he is not sleeping well at all
12:53 moritz :(
12:53 nine Does a return 0; actually make any sense in an op? Haven't seen a return in any op in master, only the one in gsoc_threads: http://paste.scsys.co.uk/144348
12:53 whiteknight nine: which op?
12:54 whiteknight ops return the address of the next opcode to execute. return 0 would be used to signify the end of the runloop
12:56 nine whiteknight: pasted the op's code
12:57 whiteknight no, I don't think that really makes sense
12:58 whiteknight branch op is stupid
13:16 Coke in git-svn, is there a way to force it to do an svn rename when you dcommit?
13:16 Coke I want to rename a file, but in the past, I've had git-svn not catch that I did a "git mv", and just delete and readd the file.
13:17 Coke meh. I'll just do the rename in svn. bother.
13:19 PerlJam Coke: Later git-svn supposedly handle renaming files correctly  (I haven't tested it though)
13:22 Coke PerlJam: I think my problem is a recent git-svn but an old (1.4) svn server.
13:22 PerlJam could be
13:22 Coke and I have very little control over the server.
13:23 Coke easier to drop to raw svn if I can't force the issue, it's an infrequent thing.
13:30 tadzik whiteknight: pretty nice. I have no kids yet, so I have a decent amount of sleep :)
13:42 preflex_ joined #parrot
13:43 whiteknight tadzik: I always tell people this: Having kids is a really really great thing, if you're ready for it
13:44 whiteknight in no way is it easy
13:45 * moritz couldn't agree more
13:46 PerlJam "if you're ready for it"?  Most people don't know if they're ready until it's too late :)
13:46 atrodo He didn't say that you'd know if you were ready
13:46 whiteknight PerlJam: Let me put together a counter-example. A friend of mine just had a kid with a guy who is uneducated, unemployed, and lists "hate to be hassled about being unemployed" on facebook
13:47 whiteknight those two, not ready
13:47 atrodo whiteknight++ # Perfect example
13:47 PerlJam whiteknight: yes, but are they self-aware of this fact?
13:48 whiteknight PerlJam: They are rapidly becoming aware of it
13:48 whiteknight hindsight is 20/20
13:48 atrodo Another example: everyone on 16 and pregnant.
13:48 whiteknight atrodo: yes, they are the postal children
13:48 whiteknight poster
14:06 nbrown joined #parrot
14:18 not_gerd joined #parrot
14:18 dalek nqp: a6899ad | moritz++ | src/QAST/Compiler.nqp:
14:18 dalek nqp: [QRegex] do not quote method names in nqp
14:18 dalek nqp:
14:18 dalek nqp: no functional changes
14:18 dalek nqp: review: https://github.com/perl6/nqp/commit/a6899ad70d
14:18 not_gerd hello, #parrot
14:19 whiteknight hello not_gerd
14:19 tadzik hello
14:19 alvis joined #parrot
14:21 not_gerd cotto: ping
14:26 woosley left #parrot
14:41 woosley joined #parrot
14:44 jsut joined #parrot
14:50 schmooster joined #parrot
14:55 not_gerd msg cotto I'm working on a m0+/- implementation ( https://github.com/gerdr/m0-pm )
14:55 aloha OK. I'll deliver the message.
14:55 not_gerd msg cotto it already compiles NOOP ;)
14:55 aloha OK. I'll deliver the message.
16:09 jsut_ joined #parrot
16:53 plobsing_ joined #parrot
17:06 dmalcolm joined #parrot
17:12 alvis joined #parrot
17:13 Khisanth joined #parrot
17:39 Tene joined #parrot
17:56 cotto_work ~~
18:06 zby_home joined #parrot
18:12 cotto_work pear--
18:14 whiteknight banana--
18:15 cotto_work If there were a package manager called banana, I'm sure it'd be better than pear.
18:18 whiteknight it leaves slippery peels around your computer, and causes other processors to slip
18:21 not_gerd cotto_work: in case you don't get messages addressed to 'cotto': http://irclog.perlgeek.de/p​arrot/2011-09-23#i_4469063
18:22 cotto_work hi not_gerd.  nice to see you around again
18:23 cotto_work shiny
18:26 not_gerd results on x86 are mediocre because of register spilling, but the generated code for x86-64 looks nice...
18:27 cotto_work not_gerd: I'm excited to dig into that.
18:40 Coke aloha, ask kid51 if we can close https://rt.perl.org/rt3/Ti​cket/Display.html?id=76828
19:03 dmalcolm joined #parrot
19:22 mj41 joined #parrot
20:20 bluescreen joined #parrot
21:00 zby_home_ joined #parrot
21:31 perlite_ joined #parrot
21:39 nbrown joined #parrot
22:56 Drossel joined #parrot
23:33 rfw joined #parrot

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

Parrot | source cross referenced