Camelia, the Perl 6 bug

IRC log for #parrot, 2011-09-09

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:09 nbrown joined #parrot
00:35 benabik joined #parrot
01:10 whiteknight chromatic: We're trying to replace most PIR flags with the new :tag() syntax. So :load becomes :tag("load"), etc. You can use any string you want, and you can get a reference to subs by name name from the PackfileView PMC
01:10 whiteknight then, you can execute all your :load and :init subs from PIR, without nested runloops. That's what the parrot frontend does now
01:12 chromatic Good, that's what I thought you were doing.
01:14 whiteknight I guess that without the dep policy we can start to fast-track some of that
01:15 JimmyZ joined #parrot
01:17 chromatic In theory IMCC could rewrite the existing syntax to those flags.
01:22 whiteknight I have that code already, but commented out
01:22 whiteknight it's the lookup mechanisms aren't compatible right now, so we can't make the switch yet
01:26 whiteknight the :tag stuff is designed to be looked up from the PackfileView PMC, not from do_sub_pragmas. I could implement that, but I haven't yet
01:34 dalek Rosella: fcadbb5 | Whiteknight++ | s (5 files):
01:34 dalek Rosella: Add in a new FileSystem.FileIterator tto iterate over a steam line by line. Add in Take and Skip Stages for Query.Queryable.Stream
01:34 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/fcadbb5716
01:53 woosley joined #parrot
02:32 contingencyplan joined #parrot
04:04 fperrad joined #parrot
04:17 cotto ~~
04:26 SHODAN joined #parrot
04:37 Shozan joined #parrot
04:37 dalek rakudo/nom: 433a731 | Coke++ | t/spectest.data:
04:37 dalek rakudo/nom: track failure mode
04:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/433a7311cc
04:45 Coke arrrrgh, I've just had a horrible idea.
04:46 tadzik go on..
04:46 Coke if I don't tell you, maybe it'll go away.
04:49 tadzik oh, one of those
04:50 tadzik seems that S02-whitespace_and_comments/comments.t suffers the the same GC bug as wherefore-gc-bug branch
04:50 tadzik also S03-smartmatch/capture-signature.t
04:51 dalek rakudo/nom: 3053e41 | pmichaud++ | docs/announce/2011.0 (3 files):
04:51 dalek rakudo/nom: Add announcements from master into nom branch.
04:51 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3053e410e9
04:53 tadzik Coke: do you have a script automating tracking failure reasons?
04:53 Coke tadzik: mayyyybe.
04:53 tadzik Coke: it treats warnings as errors, see "Useless declaration of a has-scoped method in mainline"
04:53 tadzik I'm fixing those now
04:53 Coke I figured it would be unneeded once nom caught up.
04:54 Coke tadzik: it's handrolled. I should share it. momentito.
04:55 Coke tadzik: https://gist.github.com/1205523
04:55 Coke Feel free to unhackify it.
04:56 Coke ... also, wrong channel?
04:56 tadzik oh, sure
04:56 * Coke installs Mojo as step one of evil plan.
04:57 tadzik those dalek notifications confuse me all the time
05:00 dalek rakudo/nom: 15813b2 | tadzik++ | t/spectest.data:
05:00 dalek rakudo/nom: Fix a couple of failure reasons in spectest.data
05:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/15813b242a
06:05 JimmyZ_ joined #parrot
06:06 he_ joined #parrot
06:08 JimmyZ_ joined #parrot
06:15 Coke mojo++
06:36 benabik_ joined #parrot
06:41 JimmyZ joined #parrot
06:55 JimmyZ_ joined #parrot
06:56 JimmyZ__ joined #parrot
06:58 JimmyZ___ joined #parrot
06:59 mj41 joined #parrot
07:06 JimmyZ joined #parrot
07:10 benabik mojo?
07:10 benabik oh, that was an hour ago.  Maybe he's gone.  I should be.  Blarg.
07:10 nbrown_ joined #parrot
07:11 tadzik mojolicious and friends, I suppose
07:19 JimmyZ_ joined #parrot
07:29 woosley1 joined #parrot
07:31 woosley joined #parrot
07:45 JimmyZ_ joined #parrot
07:48 JimmyZ joined #parrot
07:57 JimmyZ_ joined #parrot
08:02 woosley joined #parrot
08:12 JimmyZ_ joined #parrot
08:17 SHODAN joined #parrot
08:20 JimmyZ_ joined #parrot
09:06 JimmyZ_ joined #parrot
09:20 JimmyZ_ joined #parrot
09:40 JimmyZ joined #parrot
10:07 JimmyZ joined #parrot
10:10 dalek nqp: c6f6882 | jonathan++ | / (7 files):
10:10 dalek nqp: Add archetypes to the various NQP meta-objects.
10:10 dalek nqp: review: https://github.com/perl6/nqp/commit/c6f68825f9
10:11 ambs joined #parrot
10:24 dalek rakudo/nom: ae8a6dd | moritz++ | src/SAFE.setting:
10:24 dalek rakudo/nom: forbid mkdir in SAFE.setting
10:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ae8a6dd767
10:33 preflex_ joined #parrot
10:38 dalek rakudo/nom: 1ff410a | jonathan++ | tools/build/NQP_REVISION:
10:38 dalek rakudo/nom: Bump NQP_REVISION to get archetypes patch.
10:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1ff410a06d
10:38 dalek rakudo/nom: 9a0650a | jonathan++ | src/Perl6/Metamodel/ (3 files):
10:38 dalek rakudo/nom: Make roles inheritalizable; implement inheritalize (which just returns the pun), so class Foo is SomeRole { } works again.
10:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9a0650ace9
10:38 dalek rakudo/nom: 8c9dd1a | jonathan++ | src/core/traits.pm:
10:38 dalek rakudo/nom: Catch invalid inheritance/composition operations, and fill out support for types that know how to produce inheritable/composable types.
10:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8c9dd1af6d
10:44 dalek rakudo/nom: 21089d9 | jonathan++ | src/Perl6/Metamodel/RolePunning.pm:
10:44 dalek rakudo/nom: Only make a pun once.
10:44 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/21089d981d
11:25 nopaste joined #parrot
11:34 Psyche^ joined #parrot
12:07 whiteknight joined #parrot
12:07 whiteknight good morning, #parrot
12:08 tadzik good morning whiteknight
12:12 ambs joined #parrot
12:17 whiteknight hello tadzik
12:21 dalek rakudo/nom: c8ebd81 | pmichaud++ | src/old/ (142 files):
12:21 dalek rakudo/nom: Remove the we-have-not-needed-it-for-a-long-time old/ subdirectory.
12:21 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/c8ebd81c2b
12:36 jsut joined #parrot
12:36 darbelo joined #parrot
12:41 mls good afternoon
12:42 mls I tried that "put continuation into the callcontext" hack and it made the setting compilation 6% faster
12:42 mls So I don't know if it's worth to do or not
12:44 mls (It doesn't work for tailcalls yet, so maybe I can get the number to 7% or 8%)
12:44 mls still not the 10% I had hoped for
12:46 tadzik worth it
12:46 tadzik well, are there any cons?
12:47 whiteknight internally I'm sure the code is more messy
12:47 whiteknight or, more complicated I should say
12:48 dalek rakudo/nom: ac4d97c | masak++ | / (2 files):
12:48 dalek rakudo/nom: [src/core/control.pm] add &sleep
12:48 dalek rakudo/nom:
12:48 dalek rakudo/nom: It's the same as in Beijing, modulo $Inf hack.
12:48 dalek rakudo/nom:
12:48 dalek rakudo/nom: Also, uncommented sleep.t in spectest.data.
12:48 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ac4d97cb71
13:00 atrodo aloha: msg dukeleto it's not that I don't receive the post-commit, I do.  The server saves it to disk, but doesn't insert it into the database.
13:00 aloha atrodo: OK. I'll deliver the message.
13:02 bluescreen joined #parrot
13:03 mls whiteknight: it's not super-messy, but it makes things more confusing
13:28 TonyC joined #parrot
13:34 nopaste joined #parrot
13:47 mls does imcc reuse registers if they are no longer needed?
13:49 mls the documentation says it does
13:50 mls but I wonder how continuations can work correctly if the registers get reused, as the continuation does not snapshot the registers
13:50 benabik_ joined #parrot
13:50 dalek Rosella: 8f265cc | Whiteknight++ | src/harness/View.winxed:
13:50 dalek Rosella: Small fix to Harness.View when displaying a list of files with premature exits
13:50 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/8f265cc0a0
13:51 mls hi benabik!
13:57 dalek rakudo/nom: 948ca74 | masak++ | tools/build/Makefile.in:
13:57 dalek rakudo/nom: [Makefile.in] s/run/shell/
13:57 dalek rakudo/nom:
13:57 dalek rakudo/nom: flussence++ for fossil spotting.
13:57 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/948ca744f5
14:04 SHODAN joined #parrot
14:06 Coke so, aside from "not crashing", "using less disk space", "using fewer inodes", what do we need from a smolder-like service?
14:07 SHODAN joined #parrot
14:10 whiteknight Coke: good question
14:10 Coke I'll work on those 3 for now. ;)
14:10 whiteknight Coke++
14:10 Coke do we have access to a DB via OSUOSL if needs be?
14:11 Coke (i'll still w/ sqllite for now otherwise.)
14:12 tadzik so the crazy idea worked out? :)
14:12 dalek rakudo/nom: 6c2607a | masak++ | src/core/Real.pm:
14:12 dalek rakudo/nom: [src/core/Real.pm] implemented .isNaN
14:12 dalek rakudo/nom:
14:12 dalek rakudo/nom: Same as in Beijing.
14:12 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6c2607ae3a
14:12 dalek rakudo/nom: 88e14a0 | masak++ | src/core/Numeric.pm:
14:12 dalek rakudo/nom: [src/core/Numeric.pm] made NaN ~~ NaN true
14:12 dalek rakudo/nom:
14:12 dalek rakudo/nom: Borrowed the implementation from Beijing.
14:12 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/88e14a0bdd
14:13 Coke tadzik: it hasn't gotten far, but mojo is pretty much exactly what I want in a framework to write a perl MVC app, so wth.
14:14 benabik mls: Good morning!
14:14 Coke Once I have something that isn't just "hello world", I'll push to github.
14:14 Coke I figure for now I'll consume the same zipfiles smolder does.
14:15 dalek jaesop: 160ecd4 | Whiteknight++ | stage0/js/ (2 files):
14:15 dalek jaesop: Implement basic try/catch block. Some other cleanups
14:15 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/160ecd4817
14:15 dalek jaesop: 3d93f0a | Whiteknight++ | / (2 files):
14:15 dalek jaesop: mostly whitespace cleanups and small fixes
14:15 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/3d93f0afe5
14:15 dalek jaesop: 70ea936 | Whiteknight++ | stage0/js/ (2 files):
14:15 dalek jaesop: Implement basic throw statement. Fix top-level error handler to deal with JS-alike exceptions
14:15 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/70ea9366d3
14:15 dalek jaesop: 074ab40 | Whiteknight++ | t/stage0/try_catch.t:
14:15 dalek jaesop: add in a test for throw/try/catch
14:15 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/074ab4038e
14:23 contingencyplan joined #parrot
14:26 PacoLinux joined #parrot
14:40 dalek rakudo/nom: 46e7cc8 | jonathan++ | src/Perl6/Metamodel/ (7 files):
14:40 dalek rakudo/nom: Big refactor of role type checking. Includes an initial implementation of role introspection (.^roles), supporting :local in classes (don't show me my parent's roles) and :transitive (show me the roles that my roles do too).
14:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/46e7cc8343
14:40 dalek rakudo/nom: ca870df | moritz++ | t/00-parrot/09-pir.t:
14:40 dalek rakudo/nom: update 09-pir.t to current Q:PIR policy
14:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ca870df127
14:47 dalek nqp: 7d5eec2 | jonathan++ | src/how/NQPConcreteRoleHOW.pm:
14:47 dalek nqp: Update NQPConcreteRoleHOW to match new role type checking API.
14:47 dalek nqp: review: https://github.com/perl6/nqp/commit/7d5eec2966
14:49 dalek rakudo/nom: 133508e | jonathan++ | tools/build/NQP_REVISION:
14:49 dalek rakudo/nom: Bump NQP_REVISION.
14:49 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/133508ed61
14:52 mls argh, gcc hates me
14:53 mls there seems to be a bug in the __nonnull__ attribute handling
14:53 Coke and smolder is down.
14:59 whiteknight mls: what kind of bug? What version of GCC are you on?
14:59 dod joined #parrot
15:15 ambs joined #parrot
15:21 dalek rakudo/nom: ce13be6 | moritz++ | t/spectest.data:
15:21 dalek rakudo/nom: run role introspection tests
15:21 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ce13be61c8
15:31 Coke and smolder is back.
15:32 whiteknight Coke++
15:32 Coke osuosl did that one. trying to find out what they did so I can just do it next time.
15:33 whiteknight they're magicians
15:41 Coke they had restarted it earlier to free RAM and something wedged, even a restart wouldn't fix it.
15:41 Coke we'll just limp along for a little while longer. I am excited about the mojo thing, so expect some progress there.
15:44 dalek jaesop: 484a8a2 | Whiteknight++ | stage0/README.md:
15:44 dalek jaesop: Add in a quick README for stage0 showing the status of some features that do and do not work
15:44 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/484a8a2abf
15:47 mj41 joined #parrot
16:06 dmalcolm joined #parrot
16:08 cotto ~~
16:12 pyrimidine left #parrot
16:23 mls whiteknight: 4.5.1 20101208
16:24 mls I'll open a bug with our gcc guys
16:27 mls (I've spent the last hour creating a test case)
16:35 lateau joined #parrot
16:36 atrodo Coke++
16:36 lateau joined #parrot
16:44 cotto_work ~~
17:00 mls whiteknight: http://rudin.suse.de:8888/show_bug.cgi?id=717014 in case you're interested
17:01 mls oops, make that https://bugzilla.novell.com/show_bug.cgi?id=717014
17:01 mls (rudin is my internal bugzilla proxy)
17:02 mls anyway, weekend! afk..
17:03 cotto_work have fun
17:07 dalek rakudo/nom: 36d06e1 | jonathan++ | src/Perl6/Metamodel/CurriedRoleHOW.pm:
17:07 dalek rakudo/nom: Add first cut of curried role type check handling. Works for some cases, but not all yet.
17:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/36d06e1142
17:07 dalek rakudo/nom: 3e24ff0 | jonathan++ | src/Perl6/Metamodel/C (2 files):
17:07 dalek rakudo/nom: Make sure we don't lose imformation about how we arrived at a given concrete role; we need it in the type check list.
17:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3e24ff0823
17:07 dalek rakudo/nom: d89d8c9 | jonathan++ | t/spectest.data:
17:07 dalek rakudo/nom: We now pass S14-roles/instantiation.t.
17:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d89d8c9b45
17:07 dalek rakudo/nom: 5d26134 | jonathan++ | NOMMAP.markdown:
17:07 dalek rakudo/nom: Update nommap.
17:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5d261341a9
17:09 whiteknight wow, that's a creepy bug
17:10 whiteknight and very subtle
17:23 Coke atrodo: ?
17:30 lateau joined #parrot
18:00 dalek rakudo/nom: f9d94fe | pmichaud++ | src/core/operators.pm:
18:00 dalek rakudo/nom: Fix failure when unable to determine a sequence (RT #98678).
18:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f9d94fe1b5
18:16 preflex_ joined #parrot
18:16 davidfetter joined #parrot
18:28 darbelo joined #parrot
18:36 atrodo Coke> For smolder work
18:37 alester joined #parrot
18:39 Coke oh, that's nothing. thank me if the new thing goes anywhere. ;)
18:56 mj41 joined #parrot
19:03 Tene joined #parrot
19:08 * Coke has a request - can folks save him some .tar.gz files that get sent to smolder and leave them on feather for me somewhere?
19:14 * sorear wonders what smolder is and if ey should care.
19:23 cotto_work sorear: TAP aggregation, except that it's delicate enough to come close to offsetting its usefulness
19:23 sorear cotto_work: what do you mean by 'TAP aggregation'?
19:24 cotto_work aloha: smolder?
19:24 aloha cotto_work: smolder is not automatic, but smolder clients can be
19:24 cotto_work sorear: http://smolder.parrot.org/a​pp/projects/smoke_reports/1
19:24 lateau joined #parrot
19:46 ambs joined #parrot
20:20 ambs joined #parrot
20:21 bluescreen joined #parrot
20:48 Coke joined #parrot
20:48 Coke .
21:15 Tene !
21:23 SHODAN joined #parrot
21:27 perlite joined #parrot
21:57 nbrown joined #parrot
22:00 jsut_ joined #parrot
22:04 dalek parrot/kid51/resubmit_smolder: c274632 | jkeenan++ | / (2 files):
22:04 dalek parrot/kid51/resubmit_smolder: Create tools/dev/resubmit_smolder.pl, then create new 'make' target
22:04 dalek parrot/kid51/resubmit_smolder: 'resubmit_smolder".
22:04 dalek parrot/kid51/resubmit_smolder: review: https://github.com/parrot/parrot/commit/c2746329da
22:08 bluescreen joined #parrot
22:23 darbelo joined #parrot
22:27 plobsing joined #parrot
22:30 darbelo joined #parrot
22:51 dalek rakudo/nom: 569deeb | pmichaud++ | src/core/operators.pm:
22:51 dalek rakudo/nom: Fix problem with two sequence operators being active at the same time (RT #98790).
22:51 dalek rakudo/nom:
22:51 dalek rakudo/nom: This commit doesn't fix the actual underlying problem, which is
22:51 dalek rakudo/nom: that lexically nested subs aren't being cloned properly.  jnthn++
22:51 dalek rakudo/nom: is working on that issue.  Instead, this patch simply inlines the
22:51 dalek rakudo/nom: nested sub instead of calling it separately.
22:51 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/569deeb7f7
23:13 dalek rakudo/nom: efe1f21 | jonathan++ | src/Perl6/Metamodel/ParametricRoleHOW.pm:
23:13 dalek rakudo/nom: Tiny refactor in preparation for implementing multiple roles with the same shortname.
23:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/efe1f21253
23:13 dalek rakudo/nom: 00a1159 | jonathan++ | src/Perl6/ (2 files):
23:13 dalek rakudo/nom: Fix closure handling bug with nested subs; should fix it for my-scoped methods/regexes too.
23:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/00a11597b2
23:19 rfw joined #parrot
23:38 dukeleto kernel.org is "down for maintenance"

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

Parrot | source cross referenced