Camelia, the Perl 6 bug

IRC log for #parrot, 2010-06-12

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:21 Khisanth joined #parrot
00:27 whiteknight joined #parrot
00:32 snarkyboojum joined #parrot
00:35 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34321), fulltest) at r47559 - Ubuntu 10.04 amd64 (g++)
00:39 dalek TT #1135 closed by mikehh++: t/op/annotate-old.t fails using -f core
00:39 dalek TT #1135: http://trac.parrot.org/parrot/ticket/1135
00:48 dalek rakudo: e22e80f | (Solomon Foster)++ | src/core/ (2 files):
00:48 dalek rakudo: Tidy up Numeric and Real.
00:48 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​22e80f018f1ed4bd034497047c4c902eb51c5af
00:50 whiteknight I wonder if Chandon ever figured out that segfault he was seeing
00:53 whiteknight purl msg Chandon the "$P0 = new ['Scheduler']" line is unnecessary. The scheduler is an interpreter-global singleton that is created at program startup.
00:53 purl Message for chandon stored.
01:03 tetragon_ joined #parrot
01:09 mikehh kid51: BTW I was looking at ticket TT #292 in rewlation to t/codingstd/pod_description.t - has anything ever been done about that
01:09 mikehh s/rewlation/relation/
01:10 TiMBuS joined #parrot
01:13 mikehh kid51: # Found 234 files without DESCRIPTION sections.
01:14 whiteknight purl msg Chandon I fixed your segfault issue in r47560. Tricky GC-related order-of-destruction bug. Your example should work fine now
01:14 purl Message for chandon stored.
01:20 whiteknight the tests in t/pir/*.t are not run as part of coretest
01:22 dalek parrot: r47560 | whiteknight++ | trunk/src (2 files):
01:22 dalek parrot: [gc] Gix order-of-destruction bug discovered by Chandon++. The check for aliveness of the scheduler in Parrot_cx_delete_taskwas failing, so in Scheduler.destroy I null out the ->scheduler field of the interpreter to ensure it gets around the segfault here. I don't think gc finalization adds PMCs back onto the free list, so the test there was failing
01:22 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47560/
01:25 whiteknight ...don't appear to run for make test either
01:25 whiteknight anybody around who knows enough about the test targets to fix that omission?
01:25 kid51 mikehh:  It's effectively stalled.
01:26 kid51 Reason:  In order to transform the POD in the way Allison wanted, you'd have to be able to measure your progress ...
01:26 kid51 ... which, IIRC, would require modifications to CPAN module Pod::Simple, ...
01:26 kid51 ... whose maintainer, at least then, was:  Allison!
01:30 whiteknight kid51: Do you have any insight into my problem? I can't quite figure out how the test targets work, or how they decide which tests get run
01:33 kid51 whiteknight:  sorry, I've been looking at tickets; didn't see your posts here
01:34 whiteknight it's okay
01:34 kid51 t/pir/*.t:  Added by fperrad Apr 25 of this year.
01:35 kid51 He probably forgot to add a target.
01:35 whiteknight kid51: how do targets even get added? I can't figure it out
01:35 kid51 But, let me ask:  Looking at them, do you think they need to be in a directory of their own?
01:35 whiteknight idiomatic perl, like in t/harness, is incredibly difficult for me to grok
01:35 kid51 First, let's see what they do.
01:35 whiteknight kid51: those two tests by themselves are not a big deal, no. I'm planning to add new tests
01:36 kid51 Tonight is my first look at them.
01:36 whiteknight I have tests to add that don't fit neatly into t/pmc or t/ops
01:36 kid51 What do your tests do?
01:36 whiteknight the one tests an order-of-destruction bug in the GC
01:37 whiteknight but it involves a few specific ops, and a few specific pmcs
01:37 kid51 regular or 'dyn'?
01:37 whiteknight regular
01:37 kid51 Should they be runnable after 'make corevm'?
01:38 whiteknight yes. They should be in the coretest target
01:38 kid51 r u working in trunk or in a specific branch?
01:39 whiteknight trunk
01:39 whiteknight I just fixed a bug in r47560, and want to add a test for it
01:42 mikehh whiteknight: they prove ok after make corevm - could be added to the make coretest target
01:42 whiteknight yeah, I've been running them here to make sure
01:42 kid51 vi +61 ./lib/Parrot/Harness/DefaultTests.pm
01:43 kid51 They should probably go in one of the directories included in @runcore_tests.
01:43 kid51 They => your 47560 correction tests.
01:44 kid51 Probably t/op/ or t/pmc/.
01:44 kid51 Take your pick.
01:44 whiteknight I'm trying it now
01:44 kid51 There's no cop on that beat to give you a ticket if you select the wrong one.
01:45 dalek TT #227 closed by jkeenan++: t/op/pushaction fails with runcore PBC
01:45 dalek TT #227: http://trac.parrot.org/parrot/ticket/227
01:46 whiteknight works. Kid51++
01:47 kid51 Now, can you take a look at those two tests in t/pir/*.t?  Do we need a directory just for them?
01:48 whiteknight yes, I think we do
01:48 kid51 k
01:48 whiteknight t/pmc are tests for individual pmcs (one file per pmc, etc), t/op is tests for specific ops, and t/pir is for emergent or compound behavior
01:48 whiteknight at least, that's my "vision" for it
01:48 kid51 BTW, you'll need a reconfigure or make realclean for your change to DefaultTests to take effect.
01:49 kid51 And should t/pir/*.t tests require 'make' or just 'make corevm'?
01:50 whiteknight kid51: I didn't need to realclean. It just worked
01:50 whiteknight kid51: make corevm, I think
01:53 * kid51 has to start thinking about where he's going to watch World Cup matches tomorrow.
01:53 whiteknight you a world cup fan?
01:54 kid51 Yes
01:54 kid51 It's about the extent of my contact with the sporting world.
01:54 kid51 Two draws in Group A today.
01:55 dalek parrot: r47561 | whiteknight++ | trunk/lib/Parrot/Harness/DefaultTests.pm:
01:55 dalek parrot: the files from t/pir/ should be run as part of coretest. There aren't many. kid51++ and mikehh++ for assistance
01:55 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47561/
01:55 dalek parrot: r47562 | whiteknight++ | trunk (2 files):
01:55 dalek parrot: [t] Add a test for my fix in r47560. Test adapted from a test case provided by Chandon++
01:55 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47562/
01:55 whiteknight kid51: I played a lot of organized soccer as a kid, so naturally I hate it now
01:56 whiteknight of course, I'm not much of a fan of any sport
01:57 kid51 I didn't play any except in graduate school (the first time).
01:57 kid51 All the foreign grad students would put the Americans into the game for comic relief.
02:00 kid51 But that was a long time ago ... before we even had a whiteknight!
02:01 particle joined #parrot
02:02 mikehh whiteknight: fixed codetest failures with your new test and coretest passes with it
02:03 whiteknight damnit
02:03 whiteknight stupid codetest stupid failures
02:05 particle oh, yeah, don't blame your stupid fingers.
02:06 eternaleye joined #parrot
02:06 whiteknight holycrap, I broke all the codetests
02:06 whiteknight or, a good portion of them
02:06 mikehh whiteknight: fixed and make test passes
02:06 szabgabx__ joined #parrot
02:07 whiteknight mikehh: you fixed?
02:07 mikehh whiteknight: yes
02:07 whiteknight nice, mikehh++
02:11 dalek parrot: r47563 | mikehh++ | trunk/t/pir/timer_exit.t:
02:11 dalek parrot: add svn properties, copyright, Id lines and coda
02:11 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47563/
02:18 mikehh whiteknight: fulltest passes as well
02:21 whiteknight what does this error mean, in perl 5.10: Can't call method "say" without a package or object reference
02:22 whiteknight I thought say was a built-in in 5.10
02:23 mikehh It needs use 5.10;
02:24 mikehh or use Modern::Perl;
02:24 whiteknight ah, use 5.10. That's what I'm missing
02:24 mikehh if you get chromatic's module
02:32 dalek rakudo: 20a512b | (Solomon Foster)++ | src/core/ (2 files):
02:32 dalek rakudo: Tidy Int and Rat.
02:32 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/2​0a512bc25e59fb83d164c10abe0d432febd9c59
02:32 dalek rakudo: 006bdf0 | (Solomon Foster)++ | src/core/ (2 files):
02:32 dalek rakudo: Move Num operators from Real.pm to Num.pm.
02:32 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/0​06bdf0adc3e4e26951e481c498f9444c64789ce
02:34 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34323), fulltest) at r47563 - Ubuntu 10.04 amd64 (g++)
02:39 bacek aloha, humans
02:39 bacek seen cotto
02:39 purl cotto was last seen on #parrot 20 hours, 12 minutes and 11 seconds ago, saying: braids?
02:39 bacek seen cotto_work
02:39 purl cotto_work was last seen on #parrot 6 hours, 34 minutes and 27 seconds ago, saying: We've got layers.
02:40 bacek msg cotto Do you have github id?
02:40 purl Message for cotto stored.
02:41 mikehh hey bacek
02:41 dalek pir: c7ca31b | bacek++ | src/PIR/Compiler/Grammar.pm:
02:41 dalek pir: Don't capture C<panic> during parsing.
02:41 dalek pir: review: http://github.com/bacek/pir/commit/c7c​a31b0416ce572ac219e9f2b25c80c0493eedb
02:41 dalek pir: 273817b | bacek++ | src/PIR/Compiler/Grammar.pm:
02:41 dalek pir: Shorten binary_* rule names to "LTM" over them to <call>
02:41 dalek pir: review: http://github.com/bacek/pir/commit/273​817bf8d1412298d818f89d60cf68f70d5d0d8
02:41 dalek pir: 4a8b2e2 | bacek++ | src/PIR/Compiler/Grammar.pm:
02:41 dalek pir: Change names of various rule for manual LTMing...
02:41 purl dalek: that doesn't look right
02:41 dalek pir: review: http://github.com/bacek/pir/commit/4a8​b2e2db94325420f7b2f01c4e13fe45720a1e7
02:41 dalek pir: 5f7852c | bacek++ | src/PIR/Compiler/Grammar.pm:
02:41 dalek pir: Add keyword rule
02:41 dalek pir: review: http://github.com/bacek/pir/commit/5f7​852c84c180854ad1b84941404b8be06137ed5
02:42 dalek pir: a1e7ac5 | bacek++ | t/0 (22 files):
02:42 dalek pir: Move old tests into old_t
02:42 dalek pir: review: http://github.com/bacek/pir/commit/a1e​7ac5bbc9dc7139e8f8401a8e6329aa76e5d59
02:43 bacek mikehh, aloha
02:46 mikehh bacek: how are you supposed to use/test pirate
02:47 bacek mikehh, "parrot setup.pir test"
02:47 bacek "use" is far-far away...
02:47 whiteknight good morning bacek, throw another shrimp on the barbie?
02:47 mikehh so you need to work from an installed parrot
02:47 bacek mikehh, yes.
02:48 bacek whiteknight, it's a good plan. Weather is quite good today :)
02:48 janus joined #parrot
02:48 whiteknight bacek: now that you're an Aussie, I have to speak your language?
02:48 whiteknight ...that wasn't supposed to be a question
02:49 bacek whiteknight, mate, not everyone speaks "ozzy" here :)
02:50 whiteknight :)
02:50 mikehh bacek: but you got the paspers to show you do now
02:50 mikehh papers
02:51 whiteknight bacek: what was your citizenship before Australia?
02:51 bacek mikehh, http://twitpic.com/1vhszc
02:51 bacek whiteknight, I still have Russian citizenship.
02:51 whiteknight bacek: ah, okay
02:52 mikehh bacek: there you go
02:52 bacek whiteknight, Russia allow dual citizenship.
02:54 tetragon_ joined #parrot
02:54 * bacek want "pir coverage tool"...
02:54 bacek and "nqp coverage"
02:54 bacek and world domination
02:54 purl i guess world domination is the eventual goal
02:54 bacek and pony...
02:56 bacek Time to make some lunch
02:56 whiteknight bacek: what do you mean "pir coverage tool"?
03:00 bacek whiteknight, similar to any other coverage tools. Run test, show how many time each line was executed.
03:00 whiteknight ah, okay.
03:00 bacek Similar to Devel::Cover
03:00 whiteknight that shouldn't be too hard, especially not with an instrumentation tool like what khairul is developing
03:01 bacek Yes. But I want it NOW :)
03:02 whiteknight :)
03:02 whiteknight okay, it's late and I'm going to bed. Enjoy the lunch, bacek
03:02 bacek whiteknight, g'night
03:24 ash_ joined #parrot
04:09 rurban_ joined #parrot
04:32 dalek TT #1596 closed by coke++: PIR methods fail to be added to PMCs unless marked :nsentry
04:32 dalek TT #1596: http://trac.parrot.org/parrot/ticket/1596
04:38 Coke msg notfound - can we close TT #300 ?
04:38 purl Message for notfound stored.
04:48 dalek TT #1458 closed by coke++: Don't add probed warnings to config information.
04:48 dalek TT #1458: http://trac.parrot.org/parrot/ticket/1458
05:05 dalek TT #1558 closed by coke++: Request wiki edit-access
05:05 dalek TT #1558: http://trac.parrot.org/parrot/ticket/1558
05:21 dalek TT #900 closed by coke++: NQP parses functions beginning with 'make' as keyword
05:21 dalek TT #900: http://trac.parrot.org/parrot/ticket/900
05:21 dalek TT #933 closed by coke++: building NQP on ppc stops with 'Segmentation fault'
05:21 dalek TT #933: http://trac.parrot.org/parrot/ticket/933
05:21 dalek TT #842 closed by coke++: [DEPRECATED]  NQP  PIR q:to:...
05:21 dalek TT #842: http://trac.parrot.org/parrot/ticket/842
06:11 dalek TT #837 closed by jimmy++: [DEPRECATED] NQP quoting rules going to be close to Perl6.
06:11 dalek TT #837: http://trac.parrot.org/parrot/ticket/837
06:15 JimmyZ joined #parrot
06:27 dalek TT #1079 closed by jimmy++: Chaining operator ':=' is ignored in NQP
06:27 dalek TT #1079: http://trac.parrot.org/parrot/ticket/1079
06:27 dalek TT #1051 closed by jimmy++: [TODO]  NQP should report use of undeclared variable
06:27 dalek TT #1051: http://trac.parrot.org/parrot/ticket/1051
06:27 dalek TT #459 closed by jimmy++: $(...) syntax in NQP
06:27 dalek TT #459: http://trac.parrot.org/parrot/ticket/459
06:27 dalek TT #1066 closed by jimmy++: Add %(...) expression to NQP
06:27 dalek TT #1066: http://trac.parrot.org/parrot/ticket/1066
07:09 cotto msg bacek I'm cotto on github
07:09 purl Message for bacek stored.
07:09 cotto helllo
07:09 dalek parrot: r47564 | khairul++ | branches/gsoc_instrument (49 files):
07:09 dalek parrot: Sync with trunk
07:09 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47564/
07:12 tcurtis_ joined #parrot
07:16 bacek joined #parrot
07:16 aloha joined #parrot
07:19 cotto Wow.  jimmy++ for going nuts on old nqp tickets
07:23 bacek aloha, cotto
07:25 bacek cotto, I've added you to pirate contributors. Feel free to hack it :)
07:29 cotto He left too early.  A coverage tool is essentially trivial with the existing profiling runcore.
07:32 cotto as far as the info from imcc is accurate, that is
07:34 cotto msg bacek PIR coverage should be pretty trivial with the profiling runcore.  You'll have to do some post-processing though.
07:34 purl Message for bacek stored.
07:41 dalek rakudo: 8fc1ef5 | dukeleto++ |  (3 files):
07:41 dalek rakudo: Add test for loading perl6.pbc bytecode
07:41 dalek rakudo: Signed-off-by: Moritz Lenz <moritz@faui2k3.org>
07:41 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/8​fc1ef57fae03889c38a566dcc8daa4c0dda3250
08:04 dalek rakudo: a63c7e0 | moritz++ |  (4 files):
08:04 dalek rakudo: add stresstest target, and move day 17 of the advent tests to stresstest
08:04 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/a​63c7e050fef6a648af234ac551e6b0d04db9fe4
08:12 szabgabx joined #parrot
08:22 dalek rakudo: 5ad963b | moritz++ | src/ (5 files):
08:22 dalek rakudo: s/CREATE_HASH_LOW_LEVEL/CR​EATE_HASH_FROM_LOW_LEVEL/
08:22 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​ad963bdc8d4116e86793c7e7c4db83830f13455
08:22 dalek rakudo: edf0072 | moritz++ | t/harness:
08:22 dalek rakudo: default to one test job at a time, due to complaints from avar++ and sorear++
08:22 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​df00722595864310f52ab1f4eb757aff323e09f
08:24 szabgabx_ joined #parrot
08:26 szabgabx joined #parrot
08:46 dalek website: tcurtis++ | It's Finally Time to Write Some Optimizations (Almost)
08:46 dalek website: http://www.parrot.org/content/its-final​ly-time-write-some-optimizations-almost
09:06 mmcleric joined #parrot
09:09 bacek mmcleric, привет :)
09:09 bacek cotto, do we have anyone who can write coverage report?...
09:10 radu joined #parrot
09:24 JimmyZ_ joined #parrot
09:26 aukjan joined #parrot
09:43 dalek rakudo: aecd470 | moritz++ |  (2 files):
09:43 dalek rakudo: early, hacky implementation of callframe($level) and class CallFrame
09:43 dalek rakudo: Currently it only supports things like these:
09:43 dalek rakudo: callframe().line  # the current line
09:43 dalek rakudo: callframe(1).line # caller's line
09:43 dalek rakudo: callframe(1).file # caller's file
09:43 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/a​ecd47061298d5967201fe9bcc008db4026354dc
09:46 mmcleric bacek, привет :) а у нас тут Джонатан - рядом сидит
09:48 dalek rakudo: 6fe0817 | (Vyacheslav Matjukhin)++ | src/core/Any-list.pm:
09:48 dalek rakudo: #RT75626 - [TODO] Implement Any.all (and .any, .one, .none) in Rakudo
09:48 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/6​fe08170cf519da533f3434f1ec0121ec68f3336
09:50 bacek joined #parrot
09:51 aloha joined #parrot
09:54 dalek rakudo: 46e9ac2 | jonathan++ | src/ (3 files):
09:54 dalek rakudo: Start to store attribute type constraints so we can introspect them.
09:54 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/4​6e9ac2d8b8fb54c47cb668388918ba1f92f7b89
10:29 mj41 joined #parrot
10:33 bacek mj41, heya! Where is my dependency projects in TapTinder?
10:36 clinton joined #parrot
10:37 whiteknight joined #parrot
10:39 mj41 bacek, ahoj! Sorry. No time to code anything, but I'm thinking about it a lot. Preparing wedding day during weekends :-).
10:40 bacek mj41, Congratulations :)
10:42 mj41 bacek Thanks. Is "dependency projects" support useful without Git support for you? Which project are your favourite? All from http://proto.perl6.org/ :-) ?
10:45 mj41 bacek I need to automate testing TapTinder a bit first. Should we move this discussion to irc://irc.freenode.net/#taptinder ?
10:47 bacek mj41, I would like to see git support. Almost all HLLs use git.
10:48 bacek mj41, (I'm little busy now for more discussions... :/ )
10:52 whiteknight good morning, #parrot
10:59 mj41 bacek, ok np
11:00 whiteknight ACCEPTSEXACTLY() ? I don't know what ACTLY is, but does an optimizer really need to be ACCEPTing SEX?
11:00 whiteknight seems suboptimal to me
11:01 moritz an _ won't hurt
11:03 bacek whiteknight, why you reject ACTLY sex???
11:07 mikehh opbotsa, names
11:07 mikehh opbots, names
11:09 whiteknight tcurtis++ # blog
11:11 radu joined #parrot
11:26 dalek rakudo: 92b1f42 | (Vyacheslav Matjukhin)++ | src/core/RangeIter.pm:
11:26 dalek rakudo: RangeIter works with values which are idempotent to .succ (closes RT#75526)
11:26 dalek rakudo: Signed-off-by: Moritz Lenz <moritz@faui2k3.org>
11:26 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/9​2b1f42333fa022b58159c9b2ac71648dff76b02
11:28 lucian joined #parrot
11:32 aukjan1 joined #parrot
12:06 khairul joined #parrot
12:09 rurban_ joined #parrot
12:09 radu_ joined #parrot
13:03 * whiteknight is starting the process of installing icc again
13:19 snarkyboojum joined #parrot
13:23 dalek rakudo: 0d0672b | (Solomon Foster)++ | src/core/Numeric.pm:
13:23 dalek rakudo: Add stubs with fail for the Real-only methods, so that using them incorrectly
13:23 dalek rakudo: gives you a decent error message rather than an infinite recursion.
13:23 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/0​d0672bd0a1dbb439148be2ac3de819b4d14ace6
13:23 dalek rakudo: b502e87 | (Solomon Foster)++ | src/core/Complex.pm:
13:23 dalek rakudo: Tweak Complex a bit, including requiring Real values be passed to new, moving
13:23 dalek rakudo: sqrt, roots, and prefix:<->, and deleting sub log(Complex), sub sign(Complex),
13:23 dalek rakudo: and Complex.sign.
13:23 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/b​502e87d32f8985faf1acf44087f10a7d539fa47
13:26 JimmyZ joined #parrot
13:28 dalek parrot: r47565 | gerd++ | trunk/NEWS:
13:28 dalek parrot: add two news for the next release
13:28 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47565/
13:48 * moritz finds the documentation of most PMCs very C centric
13:49 moritz take the LexPad PMC
13:49 moritz PMC *get_pmc_keyed_str(STRING *name)
13:49 moritz the user is assumed to know that this is what you can access as $P0[$S1]
13:50 moritz which is not very friendly for the casual user, who wants to use PIR most of the time
13:51 whiteknight moritz: good observation
13:54 moritz do we want to do anything about it? if yes, what?
13:55 whiteknight Well, we do have two audiences that really do need documentation
13:55 whiteknight the C coders do need proper docs for their work
13:55 whiteknight but the PIR coders do too
13:56 moritz can we automatically translate it?
13:58 whiteknight I don't know. How do you propose that?
13:58 moritz we know that get_pmc_keyed_str translates to $P0[$S1]
13:59 moritz so from PMC *get_pmc_keyed_str(STRING *name) we could generate
13:59 moritz C access: PMC *get_pmc_keyed_str(STRING *name)
13:59 moritz PIR access: .local str name \n $P0[name]
14:00 whiteknight we could try that, I assume. But the automatic translation would have to play nicely with the existing documentation, where available
14:01 whiteknight if you and your perl6 magic can make it happen, power to you
14:02 moritz well, I guess perl 5 magic would be more appropriate
14:02 moritz might be worth a try, but other ideas would be welcome too
14:07 * mikehh needs to reboot
14:13 mikehh joined #parrot
14:19 szabgabx joined #parrot
14:21 davidfetter joined #parrot
14:27 tetragon joined #parrot
14:35 whiteknight damnit. I cannot get icc installed today
14:36 whiteknight or, to be more precise, I've barely gotten it installed, but I can't get it to compile anything
14:41 dalek rakudo: 5c78c71 | jonathan++ | src/Perl6/Compiler/Package.pm:
14:41 dalek rakudo: Set an attribute type to Any if there's no supplied type constraint.
14:41 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​c78c71ec36130ed25c4cc6068904105517cb963
14:41 dalek rakudo: 52873b9 | jonathan++ |  (3 files):
14:41 dalek rakudo: A little tweak so we can augment P6metaclass classes in the setting as well as
14:41 dalek rakudo: ClassHOW ones. Use this to augment Attribute to add has-accessor.
14:41 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​2873b9b20a549549290a4bc3ee930d5c9bb03c6
14:41 dalek rakudo: 8731a52 | jonathan++ | src/metamodel/Attribute.nqp:
14:41 dalek rakudo: Attribute.build should not return a Parrot Undef; it explodes leakily in Perl 6.
14:41 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/8​731a52c5adc7c6efa41b90e97d0a06b7c26c4a2
14:58 dalek rakudo: 02d164d | (Vyacheslav Matjukhin)++ | src/Perl6/Actions.pm:
14:58 dalek rakudo: colonpair shortcuts with twigils fixed (closes RT#73908)
14:59 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/0​2d164d9fb828701edb0badc0121fdaab8c1f9be
15:00 kid51 joined #parrot
15:08 ambs joined #parrot
15:22 dalek rakudo: c19f148 | moritz++ |  (3 files):
15:22 dalek rakudo: callframe().my returns a LexPad.
15:22 dalek rakudo: You can index a lexpad with .{ }, or obtain its .keys.
15:22 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/c​19f148791b9e06b08f3f41bb8a5477548be4e35
15:27 dalek rakudo: 57ec550 | jonathan++ | src/Perl6/Compiler/Package.pm:
15:27 dalek rakudo: Default to unconstrained, not Any, for attribvutes in order to be consistent
15:27 dalek rakudo: with variables.
15:27 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​7ec55031ffdb7d3462c4fdc4b957fff79f7cf60
15:27 dalek rakudo: ae1300f | jonathan++ | src/core/operators.pm:
15:27 dalek rakudo: Fix infix:<...> with a list to the right so it doesn't flow away data. Patch
15:27 dalek rakudo: courtesty of (Alexey Grebenschikov)++.
15:27 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/a​e1300f47ebafedd6c4255227e32cff19d405128
15:56 patspam joined #parrot
16:04 hercynium joined #parrot
16:16 theory joined #parrot
16:24 dalek parrot: r47566 | chromatic++ | trunk/config/gen/makefiles/root.in:
16:24 dalek parrot: Added IO dynop dependency to Message.pbc and distutils (Andy Dougherty, TT
16:24 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47566/
16:48 darbelo joined #parrot
16:57 ambs joined #parrot
16:57 dalek parrot: r47567 | jkeenan++ | branches/cfunctionsdocs (9 files):
16:57 dalek parrot: [codingstd] Provide documentation for C functions.
16:57 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47567/
17:00 ambs joined #parrot
17:03 ambs_ joined #parrot
17:04 szabgabx joined #parrot
17:28 tcurtis joined #parrot
17:42 LoganLK joined #parrot
17:46 dalek parrot: r47568 | darbelo++ | branches/gsoc_nfg (2 files):
17:46 dalek parrot: Add NFG tests.
17:46 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47568/
17:46 dalek parrot: r47569 | darbelo++ | branches/gsoc_nfg/t/op/string_nfg.t:
17:46 dalek parrot: Transcoding tests.
17:46 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47569/
18:01 Italianguy85 joined #parrot
18:15 Italianguy85 left #parrot
18:16 lucian_ joined #parrot
18:19 dalek parrot: r47570 | tcurtis++ | branches/gsoc_past_optimization (4 files):
18:19 dalek parrot: Update MANIFEST.generated, fix a dependency loading problem, and add support for bare string and integer children with PAST::Walker.
18:19 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47570/
18:20 LoganLK joined #parrot
18:22 cotto bacek, I'm afk for most of the weekend but I don't think it'd be difficult to hack up a quick PIR coverage tool.  The output of the profiling runcore is a simple text-based format.
18:22 cotto docs/dev/profiling.pod
18:23 cotto If you don't beat me to it, I'll take a shot at writing something.
19:09 dalek parrot: r47571 | tcurtis++ | branches/gsoc_past_optimization/runtime​/parrot/library/PAST/Pattern/Node.nqp:
19:09 dalek parrot: Factor out ACCEPTSGLOBALLY.
19:09 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47571/
19:42 dalek parrot: r47572 | tcurtis++ | branches/gsoc_past_optimizatio​n/runtime/parrot/library/PAST (4 files):
19:42 dalek parrot: Move option handling and :g logic of .ACCEPTS into PAST::Pattern.
19:42 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47572/
19:47 aukjan joined #parrot
20:04 Psyche^ joined #parrot
20:43 ambs_ joined #parrot
20:45 bacek aloha, humans.
20:46 eternaleye joined #parrot
20:56 darbelo bacek, aloha.
20:57 bacek darbelo, aloha
20:57 darbelo No wait. Wrong order.
20:57 darbelo aloha, bacek :)
20:57 * darbelo get confised by multiple robots.
21:13 bacek seen chromatic
21:13 purl chromatic was last seen on #parrot 3 days, 19 hours, 25 minutes and 29 seconds ago, saying: I thought we had one, courtesy of duke.  [Jun  9 01:48:29 2010]
21:36 dalek parrot: r47573 | bacek++ | branches/gc_massacre/src/gc/gc_ms2.c:
21:36 dalek parrot: Fix typo in threshold variable name. Lock liveset size.
21:36 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47573/
21:36 dalek parrot: r47574 | bacek++ | branches/gc_massacre/src/string (3 files):
21:36 dalek parrot: Avoid useless Parrot_str_copy call. Just create proper string.
21:36 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47574/
22:09 dalek parrot: r47575 | bacek++ | branches/gc_massacre/src/pmc/imageio.pmc:
22:09 dalek parrot: Clean allocated memory in ImageIO. GC doesn't do it implicitely anymore.
22:09 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47575/
22:16 ruoso joined #parrot
22:26 dalek parrot: r47576 | bacek++ | branches/gc_massacre/t/op/gc-leaky.t:
22:26 dalek parrot: Increase amount of allocated object in GC test.
22:26 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47576/
22:26 dalek parrot: r47577 | bacek++ | branches/gc_massacre/src/gc/gc_ms2.c:
22:26 dalek parrot: Cleanup allocated string storage as workaround for PackFile garbage packing...
22:26 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47577/
22:26 dalek parrot: r47578 | darbelo++ | branches/gsoc_nfg (50 files):
22:26 dalek parrot: Sync with trunk.
22:26 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47578/
22:50 tcurtis PAST::Pattern is, I think, usable for optimizations now. About to write some documentation.
22:50 bacek tcurtis++ # Hooray!
22:51 bacek tcurtis, you can probably claim http://trac.parrot.org/parrot/ticket/620
22:59 dalek parrot: r47579 | tcurtis++ | branches/gsoc_past_optimization (3 files):
22:59 dalek parrot: Add :p, fix the deep-matching children bug, and no-op 'walk'ing non-nodes(NQP-rx appears to explode if I box them into PAST::Vals).
22:59 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47579/
23:00 tcurtis bacek, I don't think PAST::Pattern is quite lightweight enough to qualify for that(still need to manually add the stage), nor is it really a language. However, I think it's a convenient and high-level enough interface to justify writing the docs and writing optimizations in it now. Designing a syntax for it seems like more of a "it would be nice" thing, and it would probably be better to wait until some optimizations have been written to start worrying abou
23:00 tcurtis and we better know what is needed for it.
23:01 tcurtis That was a somewhat excessively lengthy reply: summary: maybe, but it's probably better that I wait to worry about that.
23:03 bacek fair enough
23:07 darbelo tcurtis++ # Work now, worry later :)
23:16 Coke msg bacek and plobsing, can you check out TT #1628?
23:16 purl Message for bacek stored.
23:17 Coke msg plobsing and bacek, can you check out TT #1628?
23:17 purl Message for plobsing stored.
23:17 Coke msg hachi - is there any way to purl-msg two people with the same message?
23:17 purl Message for hachi stored.
23:20 Coke seen austin?
23:20 purl austin was last seen on #parrot 39 days, 17 hours, 20 minutes and 26 seconds ago, saying: http://gitorious.net/kakapo/kak​apo/blobs/master/src/Syntax.nqp  [May  4 05:59:49 2010]
23:20 darbelo Coke: That's an 'Track experimetal status' ticket.
23:21 * tcurtis thought he had a bug because --target=past with his modified NQP-rx was putting out the unoptimized PAST.
23:21 darbelo Parrot_str_unescape sucks, so Parrot_str_unescape_string was added (experimentally)
23:22 darbelo Look near line 130 of DEPRECATED.pod
23:23 tcurtis Of course, given that my optimization stage runs after "past", it would be quite strange if --target=past produced the optimized PAST.
23:26 bacek ~~
23:32 Coke ask++
23:32 dalek TT #1431 closed by coke++: Register assigned to .local string variable is mis-remembered
23:32 dalek TT #1431: http://trac.parrot.org/parrot/ticket/1431
23:32 Coke darbelo: it doesn't say what was added.
23:33 Coke at least not in a way that I can understand. it reads like we need to add "a function that does..."
23:33 Coke Can you clarify the verbiage?
23:33 Coke ah.
23:33 Coke you mention Parrot_str_unescape_string here, but not in the ticket. =-)
23:35 darbelo I can recite most of src/string/api.c from memory by now.
23:37 darbelo I don't have time to properly pretty up the ticket *now*. But I'll add it to my TODO, for tomorrow morning.
23:40 darbelo purl: msg darbelo can you check out TT #1628?
23:40 purl Message for darbelo stored.
23:48 dalek parrot: r47580 | darbelo++ | branches/gsoc_nfg/src/string (3 files):
23:48 dalek parrot: Make replace nfg-aware. Adjust the API to make that a little more convenient.
23:48 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47580/

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

Parrot | source cross referenced