Camelia, the Perl 6 bug

IRC log for #parrot, 2012-07-21

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:19 whiteknight good evening, #parrot
00:24 whiteknight blah
00:25 moritz \o whiteknight
00:25 whiteknight hello moritz. How are you doing
00:26 moritz you don't want to know
00:28 moritz https://en.wikipedia.org/wiki‚Äč/Hand,_foot_and_mouth_disease (child, wife and $self)
00:28 benabik Um, ew?
00:28 whiteknight heh. We were at a family reunion a few weeks ago and found out, after the fact, that my nephew has hand foot and mouth and nobody told us
00:29 whiteknight we were.....slightly angry
00:29 dalek parrot: 11f9d7c | Whiteknight++ | src/pmc/packfile (2 files):
00:29 dalek parrot: The init_pmc vtable I added should have been in Packfile PMC, not PackfileView. benabik++ for catching the mistake
00:30 dalek parrot: review: https://github.com/parrot/parrot/commit/11f9d7cabf
00:32 moritz "slightly"
00:32 moritz and the small one won't let me sleep, which is why I IRC at 2:30 AM
00:34 whiteknight ouch
00:37 moritz having another go at sleep&
00:41 whiteknight good luck!
00:45 whiteknight okay, great. In my branch now, FileHandle.readline() is reading the entire contents of the file
00:52 whiteknight that is...suboptimal
01:07 whiteknight actually, it looks like most of this is plumbing.
01:14 whiteknight plumbing, and clear evidence of my IQ having dropped by several dozen points
01:42 dalek parrot/whiteknight/io_cleanup1: 6c44a3a | Whiteknight++ | / (5 files):
01:42 dalek parrot/whiteknight/io_cleanup1: Change to the readline logic, in pursuit of a permanent fix. A few other small changes
01:42 dalek parrot/whiteknight/io_cleanup1: review: https://github.com/parrot/parrot/commit/6c44a3a9a8
01:42 dalek parrot/whiteknight/io_cleanup1: 2a953ca | Whiteknight++ | src/ (3 files):
01:42 dalek parrot/whiteknight/io_cleanup1: several misc build fixes. Several fixes to readline, though it's still not 100%. Now, in some tests, the last readline of a file without a trailing newline cuts off the last character. Also, some failing tests look like they're complaining about bad string encodings (which might be the same issue)
01:42 dalek parrot/whiteknight/io_cleanup1: review: https://github.com/parrot/parrot/commit/2a953cac16
02:36 Psyche^ joined #parrot
04:00 fperrad joined #parrot
08:35 lucian joined #parrot
08:57 tuxit joined #parrot
10:50 woosley joined #parrot
11:10 whiteknight joined #parrot
11:35 whiteknight good morning, #parrot
12:03 whiteknight aha!
12:03 whiteknight I think I've got this thing figured out, finally
12:07 JimmyZ joined #parrot
12:32 dalek nqp/toqast: 6ad982a | jnthn++ | src/QAST/Operations.nqp:
12:32 dalek nqp/toqast: Get while/until loops able to handle next/last/redo control exceptions again. Also a fix to support the 3-operand version that loop wants.
12:32 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/6ad982aeb1
12:32 dalek rakudo/toqast: cddd89c | jnthn++ | src/QPerl6/Actions.pm:
12:32 dalek rakudo/toqast: Fix loop.
12:32 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/cddd89c373
12:53 whiteknight joined #parrot
13:19 PacoAir joined #parrot
13:49 fperrad_ joined #parrot
13:54 lucian joined #parrot
15:20 lucian joined #parrot
15:53 dalek rakudo/nom: 5b56cfc | moritz++ | src/ (6 files):
15:53 dalek rakudo/nom: remove abs as a prefix op, make it a normal sub instead
15:53 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5b56cfcf92
15:58 whiteknight multibyte encodings are the most obnoxious things ever
15:59 whiteknight on the bright side, getting semantics right at this level means they will be right for every other user
15:59 whiteknight ...and I think I just about have got them right
16:00 dalek nqp/toqast: 0cf1ec5 | jnthn++ | src/QAST/Operations.nqp:
16:00 dalek nqp/toqast: Handling for 2 and 3 arg version of substr.
16:00 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/0cf1ec5e5d
16:00 dalek nqp/toqast: 839ab97 | jnthn++ | src/QAST/Operations.nqp:
16:00 dalek nqp/toqast: Re-instate nqp::die.
16:00 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/839ab97cb4
16:00 dalek nqp/toqast: a3fbc99 | jnthn++ | src/QAST/Operations.nqp:
16:00 dalek nqp/toqast: Add nqp::findmethod.
16:00 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/a3fbc99ffc
16:00 dalek nqp/toqast: 1668e70 | jnthn++ | src/QAST/Compiler.nqp:
16:00 dalek nqp/toqast: Make use of QAST::Want v when processing statement lists.
16:00 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/1668e70283
16:00 dalek nqp/toqast: 74c1e10 | jnthn++ | src/QAST/ (2 files):
16:00 dalek nqp/toqast: Contextual QAST::Want nodes when compiling ops.
16:00 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/74c1e10a63
16:01 whiteknight And more importantly than just getting the semantics right, I'm documenting my fail as I go, so the next guy can get them even more right
16:01 whiteknight or get them right in less code
16:04 dalek parrot/whiteknight/io_cleanup1: 655a880 | Whiteknight++ | src/io/buffer.c:
16:04 dalek parrot/whiteknight/io_cleanup1: Fix a major mistake I added in readline.
16:04 dalek parrot/whiteknight/io_cleanup1:
16:04 dalek parrot/whiteknight/io_cleanup1: When reading from the buffer, we have to return the number of bytes up to the delimiter, not the number of characters. For multi-byte characters, readline was omitting some bytes. This fixes most remaining test failures.
16:04 dalek parrot/whiteknight/io_cleanup1: review: https://github.com/parrot/parrot/commit/655a880575
16:04 dalek parrot/whiteknight/io_cleanup1: 8a10dbe | Whiteknight++ | src/ (3 files):
16:04 dalek parrot/whiteknight/io_cleanup1: Final fixes for readline. coretest now passes 100% again.
16:04 dalek parrot/whiteknight/io_cleanup1:
16:04 dalek parrot/whiteknight/io_cleanup1: Fix Handle.encoding() to update the encoding of the record_separator STRING ATTR too. This prevents, for example, cases where we are trying to readline '\n' on a UTF-16 string and getting an odd number of bytes back from the buffer.
16:04 dalek parrot/whiteknight/io_cleanup1: review: https://github.com/parrot/parrot/commit/8a10dbed83
16:06 dalek rakudo/toqast: 7253071 | jnthn++ | src/QPerl6/Actions.pm:
16:06 dalek rakudo/toqast: Toss a needless check, which could cause problems.
16:06 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/72530716f1
16:06 dalek rakudo/toqast: edf1a73 | jnthn++ | src/core/ (2 files):
16:06 dalek rakudo/toqast: Pair/Whatever/WhateverCode need to be stubbed really early now.
16:06 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/edf1a73cbe
16:06 dalek rakudo/toqast: 35d8ceb | jnthn++ | src/QPerl6/Actions.pm:
16:06 dalek rakudo/toqast: Couple of some AST building tweaks to enable better code-gen.
16:06 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/35d8ceb6a2
16:06 dalek rakudo/toqast: aff12ef | jnthn++ | src/QPerl6/ (2 files):
16:06 dalek rakudo/toqast: Fix up constant folder, thus fixing compilation of Order enum.
16:06 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/aff12ef765
16:42 jashwanth joined #parrot
17:04 brrt joined #parrot
17:11 dalek nqp/toqast: 212b72c | jnthn++ | src/QAST/Operations.nqp:
17:11 dalek nqp/toqast: Add a couple of missing ops.
17:11 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/212b72c873
17:11 dalek nqp/toqast: 876c9d5 | jnthn++ | src/QAST/Operations.nqp:
17:11 dalek nqp/toqast: Fix some op signatures.
17:11 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/876c9d53e4
17:11 dalek nqp/toqast: 818b756 | jnthn++ | src/QAST/Compiler.nqp:
17:11 dalek nqp/toqast: Add to QAST::VM so we can handle pir::const::foo.
17:11 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/818b756de5
17:11 dalek rakudo/toqast: e6609f3 | jnthn++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
17:11 dalek rakudo/toqast: Add a missing bit of circularity sawing for Str. This may well have been at the heart of many of the not-fun-to-track-down infinite recursions related to stringification when compiling CORE.setting.
17:11 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/e6609f3ff6
17:11 dalek rakudo/toqast: 6f79391 | jnthn++ | src/core/control.pm:
17:11 dalek rakudo/toqast: Fix a bunch of incorrect pirop signatures that the slightly stricter processing in QAST catches.
17:11 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/6f79391463
17:11 dalek rakudo/toqast: b2aa14b | jnthn++ | src/QPerl6/Actions.pm:
17:11 dalek rakudo/toqast: Update pir::const compilation.
17:11 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/b2aa14bdea
17:20 dalek parrot/whiteknight/io_cleanup1: 2f37147 | Whiteknight++ | src/io/api.c:
17:20 dalek parrot/whiteknight/io_cleanup1: Small codestd line-length fix
17:20 dalek parrot/whiteknight/io_cleanup1: review: https://github.com/parrot/parrot/commit/2f3714761f
17:26 dalek parrot/whiteknight/io_cleanup1: 5105eec | Whiteknight++ | src/io/userhandle.c:
17:26 dalek parrot/whiteknight/io_cleanup1: small codestd fix to the pod in src/io/userhandle.c. All codestd tests pass
17:26 dalek parrot/whiteknight/io_cleanup1: review: https://github.com/parrot/parrot/commit/5105eecb38
17:40 wagle joined #parrot
17:42 lucian joined #parrot
19:08 moritz whiteknight: is the io_cleanup1 branch ready for rakudo testing?
19:08 whiteknight moritz: my answer is a cautious "yes"
19:08 whiteknight I haven't added tests to Parrot's suite yet for Socket.readline("\r\n"), or for FileHandle.readline("\r\n") yet
19:09 whiteknight if you can wait about ~15 minutes I might have that issue addressed
19:09 brrt1 joined #parrot
19:10 moritz i can :-)
19:10 zby_home joined #parrot
19:10 moritz I can't even build the parrot branch right now
19:11 moritz http://perlpunks.de/paste/show/500afed8.528.163
19:11 whiteknight yeah, I think I missed a commit somewhere
19:11 whiteknight I'll have things sorted out soon
19:12 moritz no rush, I have other stuff to test too
19:13 whiteknight okay, good. I think I have already found a bug
19:14 brrt1 \o #parrot
19:15 moritz brrt1 o/
19:16 brrt1 my phone works against me today, logging in in my name
19:19 whiteknight hello brrt
19:20 brrt1 hi whiteknight
19:20 whiteknight how did the move go?
19:20 brrt1 still halfway
19:20 brrt1 most stuff is over
19:20 brrt1 floor is half laying
19:20 brrt1 (that came out bad... i mean theres half a floor on my floor)
19:21 brrt1 but the house is absolutely beatiful, so i'm happy as can be
19:23 whiteknight good!
19:23 brrt1 hows life on the other side of the atlantic?
19:23 brrt1 as far as mod_parrot is concerned, i had all sorts of fun new ideas
19:24 whiteknight fun new ideas make the world go 'round
19:24 whiteknight things here are good. We've finally had a break in the heat, so I've been outside working
19:24 whiteknight last week was miserably hot
19:25 brrt1 hasn't been hot here for a long time, fortunately
19:29 whiteknight it was briefly hotter in than it was for my brother in kuwait
19:30 brrt1 in pennsylvania? well, i guess that happens
19:35 mtk joined #parrot
19:47 dalek nqp/toqast: e688001 | jnthn++ | src/QAST/Operations.nqp:
19:47 dalek nqp/toqast: A few more missing nqp:: ops.
19:47 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/e688001516
19:48 dalek rakudo/toqast: f9b1ed7 | jnthn++ | src/core/ (4 files):
19:48 dalek rakudo/toqast: We need to provide explicit signatures for pir::ops now; this was mostly happening, and this tweaks the rest, hopefully.
19:48 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/f9b1ed7291
19:48 dalek rakudo/toqast: 8fb3afe | jnthn++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
19:48 dalek rakudo/toqast: Fix copy-pasto; benabik++.
19:48 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/8fb3afe057
19:48 dalek rakudo/toqast: ed9292f | jnthn++ | src/QPerl6/World.pm:
19:48 dalek rakudo/toqast: Fix a leftover PAST::Op usage.
19:48 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/ed9292f8f7
19:48 contingencyplan joined #parrot
19:58 whiteknight hmmm... for some reason, STRING_index is NOT returning the index of a substring that I know exists
20:14 brrt1 joined #parrot
20:17 lucian joined #parrot
20:24 whiteknight msg moritz it's going to take longer than 15 minutes. I'm running into one last little problem that's a bit of a head-scratcher
20:24 aloha OK. I'll deliver the message.
22:27 dalek nqp/toqast: 653152e | jnthn++ | src/QAST/Operations.nqp:
22:27 dalek nqp/toqast: Fix a couple of signatures; v is bad.
22:27 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/653152ec98
22:27 dalek nqp/toqast: aa9f9e7 | jnthn++ | src/QAST/Compiler.nqp:
22:27 dalek nqp/toqast: Add some missing coercion cases.
22:27 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/aa9f9e7f50
22:27 dalek nqp/toqast: 366f84f | jnthn++ | src/QAST/Operations.nqp:
22:27 dalek nqp/toqast: A bunch of op additions and signature fixes.
22:27 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/366f84f350
22:27 dalek nqp/toqast: 939c9ff | jnthn++ | src/QAST/Operations.nqp:
22:27 dalek nqp/toqast: index and ord also work with varying numbers of arguments.
22:27 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/939c9ffa03
22:27 dalek nqp/toqast: 60afa48 | jnthn++ | src/QAST/ (2 files):
22:27 dalek nqp/toqast: Fix yet more incorrect op signatures.
22:27 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/60afa48c7f
22:27 dalek nqp/toqast: e179076 | jnthn++ | src/QAST/Operations.nqp:
22:27 dalek nqp/toqast: Add can and rindex.
22:27 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/e17907694c
22:27 dalek rakudo/toqast: 605574c | jnthn++ | src/QPerl6/Actions.pm:
22:27 dalek rakudo/toqast: Detect and complain about missing pir::op signature.
22:27 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/605574cde5
22:27 dalek rakudo/toqast: e3f8021 | jnthn++ | src/core/terms.pm:
22:27 dalek rakudo/toqast: Fix a missing pirop sig.
22:27 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/e3f80211f6
22:27 dalek rakudo/toqast: c02c86d | jnthn++ | src/QPerl6/Ops.pm:
22:27 dalek rakudo/toqast: Toss an op that's now in the default set.
22:27 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/c02c86daad
22:27 dalek rakudo/toqast: c0ca5a3 | jnthn++ | src/core/Str.pm:
22:27 dalek rakudo/toqast: Eliminate a double boxing spotted while bug hunting.
22:27 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/c0ca5a3c0a
22:43 dalek nqp/toqast: b6049ef | jnthn++ | src/QAST/Compiler.nqp:
22:43 dalek nqp/toqast: Better diagnostics.
22:43 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/b6049ef54f
22:43 dalek nqp/toqast: de65b03 | jnthn++ | src/QAST/Operations.nqp:
22:43 dalek nqp/toqast: CAPITALS FOR RESULT TYPES. *sigh*
22:43 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/de65b03b14
22:47 lucian joined #parrot
22:48 drift joined #parrot
23:00 dalek rakudo/toqast: bb7a618 | jnthn++ | src/ (2 files):
23:00 dalek rakudo/toqast: Fix some pir::op sigs.
23:00 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/bb7a618e15
23:04 dalek nqp/toqast: 6c19396 | jnthn++ | src/QAST/Operations.nqp:
23:04 dalek nqp/toqast: Add qlist, findnotcclass.
23:04 dalek nqp/toqast: review: https://github.com/perl6/nqp/commit/6c19396802
23:17 dalek rakudo/toqast: 34ad03b | jnthn++ | src/core/ (2 files):
23:17 dalek rakudo/toqast: Last couple of tweaks needed to get CORE.setting compilation to PIR to complete. The generated PIR doesn't quite work yet.
23:17 dalek rakudo/toqast: review: https://github.com/rakudo/rakudo/commit/34ad03b3e1

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

Parrot | source cross referenced