Camelia, the Perl 6 bug

IRC log for #parrot, 2010-06-20

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:01 dalek rakudo: 683a745 | jonathan++ | t/spectest.data:
00:01 dalek rakudo: Turn on S14-roles/parameterized-mixin.t.
00:01 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/6​83a745c069a0be1514fa3da58d618f8ff2e2417
00:05 theory joined #parrot
00:29 zarchne joined #parrot
00:44 mikehh_ joined #parrot
00:47 mikehh opbots, names
00:51 * mikehh need a reboot
00:53 pmichaud msg bacek +1 to the foo:bar<baz>() patch.
00:53 purl Message for bacek stored.
00:54 bacek pmichaud, ok, committing.
00:56 bacek pmichaud, should I regenerate bootstrap files?
00:57 pmichaud if you want to go ahead and port it into parrot's nqp, then yes.
00:57 bacek ok.
00:58 dalek nqp-rx: 6c707c6 | bacek++ |  (3 files):
00:58 dalek nqp-rx: Implement foo:bar<baz>() sub calls.
00:58 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/6​c707c60f0ad8798b7cb1a3bac090e514ba0acad
01:04 dalek nqp-rx: 5142577 | bacek++ | src/stage0/ (4 files):
01:04 dalek nqp-rx: Bootstrap
01:04 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/5​1425776876a5b1f2688c8a83d78d1eded8e5cfe
01:04 dalek rakudo: f9943a2 | (Martin Berends)++ | src/Perl6/Module/Loader.pm:
01:04 dalek rakudo: [Perl6/Module/Loader.pm] correctly derive pre-compiled .pir filename from both
01:04 dalek rakudo: .pm and .pm6 source names
01:05 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/f​9943a285ad7ff416c8eef0fe98414fe649ee96a
01:06 atrodo joined #parrot
01:08 dalek parrot: r47718 | bacek++ | trunk/ext/nqp-rx/src/stage0 (3 files):
01:08 dalek parrot: Update nqp to latest version
01:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47718/
01:18 bacek pmichaud, any suggestions how to implement pir's .include with nqp?
01:18 sorear bacek: to include .nqp files, .pir, or .pasm?
01:18 bacek sorear, pir.
01:19 bacek pirate?
01:19 purl pirate is a career that is washed up these days...and raping and pillaging just doesn't have a good reputation these days.... or http://www.eff.org/IP/DRM/piratead/CEA_ad.png or http://www.wired.com/politics/securi​ty/magazine/17-07/ff_somali_pirates or PIR parser (and compiler) implemented in NQP http://github.com/bacek/pir
01:19 bacek sorear, last one :)
01:21 dalek pir: 77cfd33 | bacek++ | src/PIR/Actions.pm:
01:21 dalek pir: Reduce noise by collapsing few subs. Requires bleading edge parrot (r47719).
01:21 dalek pir: review: http://github.com/bacek/pir/commit/77c​fd3313532c0cc83a6a52d3115a8781a700eea
01:21 dalek nqp-rx: 78a6c1d | bacek++ |  (2 files):
01:21 dalek nqp-rx: Implement dotty version of colonpair named methods
01:21 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/7​8a6c1dc530b1c5ee59bef461cfe8860f4b4cada
01:21 dalek nqp-rx: b6defec | bacek++ | src/stage0/ (3 files):
01:21 dalek nqp-rx: Bootstrap files
01:21 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/b​6defec922257e2072a1132160e641532880a78d
01:24 dalek parrot: r47719 | bacek++ | trunk/ext/nqp-rx/src/stage0 (3 files):
01:24 dalek parrot: Update nqp to latest versioni again
01:24 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47719/
01:30 mikehh joined #parrot
01:54 snarkyboojum joined #parrot
01:54 cotto bacek, ping
02:03 nopaste "cotto" at 192.168.1.3 pasted "running PIRATE manually broken?" (11 lines) at http://nopaste.snit.ch/21406
02:06 somebody_ joined #parrot
02:13 somebody__ joined #parrot
02:15 bacek_mobile joined #parrot
02:16 bacek_mobile cotto, it should work. Did make test passed?
02:40 theory joined #parrot
02:51 cotto msg bacek PIRATE's test pass.  If you don't get there first, I'll see what's breaking manual PIR compilation.
02:51 purl Message for bacek stored.
03:00 janus joined #parrot
03:18 Andy joined #parrot
04:01 preflex joined #parrot
05:44 confound joined #parrot
07:15 snarkyboojum joined #parrot
07:51 bacek cotto, You shouldn't run PBC files with parrot-nqp :)
08:02 sorear hah.
08:03 fperrad joined #parrot
08:12 integral joined #parrot
08:25 cotto *headdesk*
08:25 fperrad joined #parrot
08:26 cotto I was hoping it wasn't something silly like that.
08:30 cotto bacek, should a slurpy param show up as such in POST?
08:31 bacek cotto, it's nyi.
08:31 bacek Actions.param_decl.
08:34 dalek pir: 2d21ed4 | bacek++ |  (3 files):
08:34 dalek pir: Store Sub.params separately
08:34 dalek pir: review: http://github.com/bacek/pir/commit/2d2​1ed4e9c27cdce391df78e7cefda9fed400993
08:42 cotto Why do multis need to be our-scoped?
08:43 bacek afaik, perl6's subs are "my" scoped by default.
08:45 cotto more specifically, why do most methods not need it when multis do?
08:46 bacek cotto, ah. They probably not required our scoping in nqp.
08:47 bacek better to check with perl6 gurus.
08:48 cotto I'll file that question away for the next time pmichaud is around.
08:50 clinton joined #parrot
08:51 cotto Do you have a plan for integrating PIRATE into Parrot once it's reasonably complete?
08:53 bacek cotto, yes, of course.
08:54 cotto Ooh.  What does it look like?
08:55 plobsing joined #parrot
08:55 bacek Tall guy, one eyed with black patch on second eye. Squaking parrot on left shoulder.
08:55 cotto It's hard to say that they're not integrated.
08:56 cotto unless it's a mutant pirate/parrot hybrid
08:56 bacek I didn't think about integrating yet :)
08:57 cotto Getting from POST to PBC is the tricky part, but you already knew that.
08:58 bacek cotto, not actually. It's pretty straight forward.
08:58 sorear I would have thought the tricky part was in emulating all of imcc's bugxsd
08:58 bacek Getting right POST is much harder.
08:59 sorear getting right POST is quite easy, POST has an "inline PIR" code ;)
08:59 cotto If that code is called Poster, we could have PIRATE/Poster be our offical PIR compiler.
09:00 cotto not hugely creative, but it conjures up a nice image
09:01 cotto bacek, is there a nice way to figure out what the POST for a given chunk of PIR should look like, or is the lack of such tool what makes it hard?
09:02 bacek cotto, I tried to put TODO into Actions where I cut corners.
09:02 dalek parrot: r47720 | khairul++ | branches/gsoc_instrument (3 files):
09:02 cotto I saw a few of those
09:02 dalek parrot: added tests for instrumentop.
09:02 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47720/
09:02 bacek "./installable_pir --target=postshortcut hello.pir" is easiest way to check POST
09:02 dalek parrot: r47721 | NotFound++ | trunk/t/pmc/arrayiterator.t:
09:02 dalek parrot: fix and enhance ArrayIterator test for backwards out of bounds
09:02 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47721/
09:02 dalek parrot: r47722 | bacek++ | trunk/src/pmc/packfilefixupentry.pmc:
09:02 dalek parrot: Add PackfileFixupEntry.init_pmc
09:02 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47722/
09:02 dalek parrot: r47723 | bacek++ | trunk/src/pmc/packfilerawsegment.pmc:
09:02 dalek parrot: Small helper functions for Packfile PMCs
09:02 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47723/
09:03 cotto sure, but is there an easy way to check what the *correct* POST is, i.e. what's equivalent to what imcc generates?
09:03 bacek cotto, just implement something missing. Check post with command-line. If it looks reasonable - add to test
09:03 cotto I realize that I've made a logical contradiction there.
09:04 bacek imcc doesn't generate "POST"
09:04 bacek We are on clean ground here
09:04 cotto That's why I say equivalent.
09:04 cotto ok
09:04 cotto bacek is also blazing new trails
09:04 purl okay, cotto.
09:05 cotto sounds like fun
09:07 cotto that gives me a better framework for understanding PIRATE
09:08 dalek pir: 00a6ddc | bacek++ |  (4 files):
09:08 dalek pir: Add Vanilla register allocator. Not used yet.
09:09 dalek pir: review: http://github.com/bacek/pir/commit/00a​6ddcb27fb1dd71e7218ae78e40024d25e6ba6
09:09 dalek pir: a083fa8 | bacek++ | pir.pir:
09:09 dalek pir: Include VanillaAllocator before Compiler.
09:09 dalek pir: review: http://github.com/bacek/pir/commit/a08​3fa8d938f7345ae0f5a9026b391ee88aa0ce8
09:09 dalek pir: f20f5cc | bacek++ | src/POST/Compiler.pm:
09:09 dalek pir: Initialize VanillaAllocator properly.
09:09 dalek pir: review: http://github.com/bacek/pir/commit/f20​f5cc0eeed4ddf835b79fbde0874759ef4d7be
09:09 sorear yum, vanilla
09:13 cotto a new alligator
09:14 cotto bacek, I don't think your allocator is as added as you think it is
09:14 Coke cotto: what the hell are you doing up at this hour?
09:14 Coke aren't you on the left coast?
09:14 cotto sure
09:15 * Coke sits at the airport consuming some lovely free wifi.
09:15 bacek My sister going to visit me in next couple of weeks. She is probably on left coast too :)
09:15 cotto I'm sitting here poking at PIRATE, contemplating if iced tea late in the evening was a good idea.
09:16 cotto verdict: ok as long as I don't have to be anywhere tomorrow morning
09:16 sorear left coast of which country and in which map projection?
09:17 cotto probably this projection: http://camp4.files.wordpress.c​om/2007/11/aussieworldmap.jpg
09:18 dalek parrot: r47724 | NotFound++ | trunk (2 files):
09:18 dalek parrot: fix a bug in ByteBuffer resizing and add more test for it
09:18 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47724/
09:20 dalek pir: 96d4284 | bacek++ | src/POST/VanillaAllocator.pm:
09:20 dalek pir: Bah... Add VanillaAllocator. cotto++, bacek--
09:20 dalek pir: review: http://github.com/bacek/pir/commit/96d​42848ac3a332205132c68bdce266f6032e225
09:20 cotto also, the build is sad
09:21 cotto nm.  I'm stupid
09:22 bacek Yay! It's proper map! :)
09:23 cotto Are such maps popular in Australia?
09:25 bacek nope :)
09:26 Coke cotto: yah, i have similar caffiene issues of late. :P
09:26 cotto That allocator lives up to its name.
09:26 dalek rakudo: d6309ea | Whiteknight++ |  (8 files):
09:26 dalek rakudo: updates for the ns_func_cleanup branch that just merged in Parrot
09:26 dalek rakudo: Also bump PARROT_REVISION.
09:26 dalek rakudo: Signed-off-by: Moritz Lenz <moritz@faui2k3.org>
09:26 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/d​6309ea028690f66ffcbcf8f4aac1e8423054c36
09:27 * Coke should double check he has a working copy of parrot before boarding, he supposes.
09:27 cotto It'll most likely change while you're in the air anyway.
09:29 Coke yes, but if it's borked now, with svn I can't revert easily. =-)
09:31 mao http://i.imgur.com/XkKHh.png wtf
09:31 cotto speaking of which, do you know if tractest.parrot.org is supposed live up to its name?  It's currently pointing at the Drupal site (or a copy)
09:32 Coke cotto: I don't know what tractest is.
09:32 cotto ok
09:33 sorear I like the way PIR::Compiler uses a temp file
09:35 * NotFound likes the smell of 100% coverage in the morning
09:35 dalek parrot: r47725 | NotFound++ | trunk/t/pmc/schedulermessage.t:
09:35 dalek parrot: test SchedulerMessage invalid initialization
09:35 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47725/
09:37 nopaste "bacek" at 192.168.1.3 pasted "Yarr!!! Register allocator works! :)" (19 lines) at http://nopaste.snit.ch/21408
09:37 dalek pir: 1dd0ff0 | bacek++ | src/PIR/ (2 files):
09:37 dalek pir: Change grammar slightly for simplify handling of op_params
09:37 purl dalek: that doesn't look right
09:37 dalek pir: review: http://github.com/bacek/pir/commit/1dd​0ff0a21a54e5ab6c4aaec598d1edd9513c110
09:37 dalek pir: 220aed4 | bacek++ | src/PIR/Actions.pm:
09:37 dalek pir: Start desugarizing pir
09:37 dalek pir: review: http://github.com/bacek/pir/commit/220​aed46e3eca237ede54b72891c027f9246ad1b
09:38 bacek sorear, ?
09:38 purl i heard sorear, was there something about it that doesn't work?
09:38 bacek ah. There is TODO item to extend OS PMC...
09:39 cotto bacek, nice
09:39 NotFound BTW os.pmc has 30.2% coverage
09:40 cotto bacek, does that mean you know how to fix make_hello_world_pbc?
09:40 bacek cotto, nope...
09:48 szabgabx joined #parrot
09:52 cotto bacek, should I be able to do that with what you've pushed to github?
09:52 bacek cotto, yes
09:54 nopaste "cotto" at 192.168.1.3 pasted "no PIRATE love for me" (16 lines) at http://nopaste.snit.ch/21409
09:55 cotto updating parrot...
09:55 cotto I just remembered that you made some possibly relevant changes
09:56 bacek They are :)
09:56 cotto Works now.
09:56 cotto That's really shiny.  bacek++
09:57 jnthn bacek: Is this doing...post straight to PBC?
09:57 jnthn *POST?
09:57 bacek jnthn, yes.
09:57 jnthn yay
09:58 jnthn bacek++
09:58 bacek jnthn, it's kinda special version of POST. Updating PAST->newPOST in todo list.
10:00 dalek rakudo: 964d497 | jonathan++ | src/metamodel/ClassHOW.pir:
10:00 dalek rakudo: Give ClassHOW an attribute for storing handles fallback info, and a method for
10:00 dalek rakudo: adding new ones, to encapsulate such functionality neatly in the metamodel
10:00 dalek rakudo: rather than it being magic-y like in alpha.
10:00 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/9​64d497a52fdb2e3d7115accb1521a01f3e44443
10:00 dalek rakudo: 13fabba | jonathan++ | src/glue/handles.pm:
10:00 dalek rakudo: When we have a handles that doesn't match any of the criteria where we can
10:00 dalek rakudo: generate methods, add it as a fallback instead.
10:01 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/1​3fabba8520713151fb19663ebe2a6be33db9487
10:01 dalek rakudo: 5624fe3 | jonathan++ | src/ (3 files):
10:01 dalek rakudo: Bring dispatcher up to date with latest handles dispatch fallback handling
10:01 dalek rakudo: approach. Also, can't just go on truth for if we have a $!handles value because
10:01 dalek rakudo: type objects are undefined and untrue. With this, we handle all the cases the
10:01 dalek rakudo: alpha did again.
10:01 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​624fe3b98182f2ae1e225701471564f380f1326
10:01 cotto bacek, are src/actions.pm and src/pir.pg there for a reason?
10:01 bacek cotto, leftovers from "alpha"
10:01 bacek can be removed now.
10:02 cotto same for src/pasm?
10:02 bacek yes
10:02 cotto I figured as much.
10:03 cotto The caffeine is wearing out.
10:03 cotto night
10:03 bacek night, cotto
10:06 sorear bacek, src/PIR/Compiler.pm lines 19, 24, and 30
10:06 dalek pir: 219a5cb | bacek++ | src/PIR/Actions.pm:
10:06 dalek pir: Fix desugarizing of sym<assign>.
10:06 dalek pir: review: http://github.com/bacek/pir/commit/219​a5cb5ae28043902f254d8c9a2a26bee93435a
10:06 dalek pir: c59fc0a | bacek++ | t/ (2 files):
10:06 dalek pir: Add first test for desugarizing pir
10:06 dalek pir: review: http://github.com/bacek/pir/commit/c59​fc0af1034acc7f13be2af672c5dcb9f2feee5
10:06 dalek pir: 7bc9791 | bacek++ | TODO:
10:06 dalek pir: Add more TODO items.
10:06 dalek pir: review: http://github.com/bacek/pir/commit/7bc​979134e39154657d3b19c50031f8536132b27
10:07 bacek sorear, there is todo item for OS PMC on line 18.
10:07 dalek rakudo: deef2c9 | moritz++ | t/spectest.data:
10:07 dalek rakudo: we pass another smart match test
10:07 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/d​eef2c9361bf3d6234e5aa33338c49686f5de7b8
10:08 sorear bacek: mkstemp makes it less insecure but I would rather eval("2 + 2") not hit the disk at all ...
10:08 dalek parrot: r47726 | NotFound++ | trunk/t/pmc/float.t:
10:08 dalek parrot: fix Float basic assignment tests and add one for get_repr
10:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47726/
10:09 bacek sorear, it will require more changes in parrot's guts than I want to do atm.
10:09 bacek but yes
10:09 sorear oh of course
10:09 bacek We have to have way of unpack PBC from memory.
10:10 sorear "I like the way XXX" is usually my way of saying "Oh man, I feel your pain for all those horrible workarounds"
10:10 bacek yeah...
10:11 bacek Best thing is switch ParrotInterp to use Packfile* PMCs internally. But it's about 2 weeks of boring coding for me.
10:13 dalek rakudo: cdc9433 | jonathan++ | t/spectest.data:
10:13 dalek rakudo: Remove test file from regressed list that was renamed and is now run.
10:13 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/c​dc94335115331530503b46be3232142c6e17205
10:18 dalek rakudo: 8e95548 | jonathan++ | t/spectest.data:
10:18 dalek rakudo: Turn S12-attributes/recursive.t on again.
10:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/8​e955486442bcf402025077eb0391e82f4238752
10:20 bacek sorear, http://github.com/bacek/pir/commit/ef2​5afcf83bc8afb630a69d0f640a1ff8e5a0ef9
10:21 bacek feel free to share my pain :)
10:23 dalek pir: ef25afc | bacek++ | src/PIR/Compiler.pm:
10:23 dalek pir: Add "rationale" for POST::Compiler.pbc. sorear++ for motivating.
10:23 dalek pir: review: http://github.com/bacek/pir/commit/ef2​5afcf83bc8afb630a69d0f640a1ff8e5a0ef9
10:31 whiteknight joined #parrot
10:48 whiteknight good morning, #parrot
10:53 bacek whiteknight, aloha
10:53 whiteknight hello bacek, how are you tonight?
10:54 bacek whiteknight, pretty sober :)
10:55 jnthn oh noes...the vodka ran out?
10:55 bacek jnthn, yes... And scotch too
10:55 jnthn Aww!
10:57 dalek parrot: r47727 | NotFound++ | trunk/lib/Parrot/Pmc2c/PMCEmitter.pm:
10:57 dalek parrot: subtle change in Parrot_${classname}_get_isa generation to shorten generated code
10:58 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47727/
11:02 lucian joined #parrot
11:07 kjeldahl joined #parrot
11:17 dalek pir: f2d8c62 | bacek++ |  (2 files):
11:17 dalek pir: Desugarized 'foo = bar' should use POST::Op.pirop instead of .name
11:17 dalek pir: review: http://github.com/bacek/pir/commit/f2d​8c62e3f66c303aefa449e289600645393ff68
11:47 dalek TT #1686 created by cosimo++: Null PMC access in find_method('signature') using Digest/MD5 from rakudo
11:47 dalek TT #1686: http://trac.parrot.org/parrot/ticket/1686
12:09 dalek pir: 930cd72 | bacek++ | src/PIR/Actions.pm:
12:09 dalek pir: Initial brain-dump for handling Labels
12:09 dalek pir: review: http://github.com/bacek/pir/commit/930​cd722df79dde0ac98921c56f49393a6e4e04d
12:11 dalek rakudo: ee7620d | (Solomon Foster)++ | src/core/Numeric.pm:
12:11 dalek rakudo: Modify the Numeric version of infix:�<=>� to extend values with zeros, so that
12:11 dalek rakudo: (for Real $x) $x == $x + 0i.
12:11 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​e7620db73446b05a7f5852eeeed88c1398dfcaa
12:34 kjeldahl joined #parrot
12:45 dalek rakudo: 7a05ae5 | jonathan++ | t/spectest.data:
12:45 dalek rakudo: Turn on S12-introspection/walk.t.
12:45 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/7​a05ae5916be3647deb397692c1b46c7f5c86672
12:45 dalek rakudo: 97f2caa | jonathan++ | src/core/Mu.pm:
12:45 dalek rakudo: Various fixes to Mu.WALK.
12:45 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/9​7f2caadc26a6d20b76ac2c01f7a5843f7c4f420
12:46 kjeldahl joined #parrot
12:57 dalek rakudo: 0adb7fc | jonathan++ | t/spectest.data:
12:57 dalek rakudo: Turn on the now re-fudged indirect-notation.t; will hopefully be able to unfudge
12:57 dalek rakudo: a bunch more of it soon, but good to test the things that do already work.
12:57 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/0​adb7fc7e285742aa4d45b6fa6916bb7dd199a55
13:02 khairul joined #parrot
13:04 kjeldahl joined #parrot
13:06 ambs joined #parrot
13:09 kjeldahl joined #parrot
13:12 dalek rakudo: 63b92f6 | jonathan++ | src/pmc/perl6multisub.pmc:
13:12 dalek rakudo: Remove comments about todos that are todone.
13:12 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/6​3b92f6599d5fae34ed15739b98c1885e7f7fb46
13:28 dalek parrot: r47728 | mikehh++ | trunk/t/pmc/float.t:
13:28 dalek parrot: fix copyright line
13:28 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47728/
13:28 dalek parrot: r47729 | mikehh++ | trunk/src/pmc/packfilerawsegment.pmc:
13:28 dalek parrot: fix codetest failure - parentheses should not have space immediately after the opening parenthesis nor immediately before the closing parenthesis
13:28 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47729/
14:00 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34495), fulltest) at r47715 - Ubuntu 10.04 amd64 (g++)
14:00 mikehh arrgh wrong one
14:01 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34504), fulltest) at r47729 - Ubuntu 10.04 amd64 (g++ with --optimize)
14:32 whiteknight joined #parrot
14:35 mikehh rakudo (63b92f6) builds on parrot r47729 - make test PASS, spectest_smolder -> #34506 (pugs r31393) FAIL - Ubuntu 10.04 amd64 (g++ with --optimize)
14:35 mikehh 18 TODO PASSes in 6 files
14:35 mikehh damnit that should be PASS not FAIL
14:35 mikehh rakudo (63b92f6) builds on parrot r47729 - make test PASS, spectest_smolder -> #34506 (pugs r31393) PASS - Ubuntu 10.04 amd64 (g++ with --optimize)
14:35 mikehh 18 TODO PASSes in 6 files
14:38 dalek rakudo: e298e50 | (Solomon Foster)++ | src/core/ (5 files):
14:38 dalek rakudo: Remove all existing numeric ACCEPTS methods in place of one new Numeric.ACCEPTS
14:38 dalek rakudo: method.  (Will replace a few of the older ones in a future patch, for
14:38 dalek rakudo: efficiency's sake.)
14:38 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​298e50d6768315e3177d5d1f2929f9b5fc12e0b
14:42 tetragon joined #parrot
14:43 kid51 joined #parrot
14:47 mikehh partcl - get a build failure / partcl-nqp PASS
14:47 mikehh PIRATE - PASS
14:48 * mikehh ok need a breeak
14:48 mikehh can't even get that right :-{
14:49 dalek parrot: r47730 | NotFound++ | trunk (5 files):
14:49 dalek parrot: add basic URI escape functions to runtime library
14:49 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47730/
15:10 mikehh joined #parrot
15:19 atrodo joined #parrot
15:22 dalek parrot: r47731 | NotFound++ | trunk/t/pmc/stringhandle.t:
15:22 dalek parrot: test StringHanlde is_tty method
15:22 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47731/
15:23 atrodo joined #parrot
16:05 arnsholt Is there a way to get a list of .subs stored in a namespace?
16:16 dalek winxed: r505 | julian.notfound++ | trunk/winxedst1.winxed:
16:16 dalek winxed: a little optimization in assignments of same type
16:16 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=505
16:18 mdiep joined #parrot
16:30 arnsholt Nm. namespace.'get_class' and then methods on that class
16:44 dalek parrot: r47732 | khairul++ | branches/gsoc_instrument (8 files):
16:44 dalek parrot: Update probes on dynop loading so that hooks for dynops can be set + tab removal
16:44 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47732/
16:52 mdiep joined #parrot
16:57 whiteknight arnsholt: that's not a complete list
16:57 whiteknight methods are not stored in the namespace. neither are vtables
16:58 whiteknight functions in the namespace are likewise not stored in the class
17:00 dalek parrot: r47733 | khairul++ | branches/gsoc_instrument/t/dynpmc (2 files):
17:00 dalek parrot: Updated tests to reflect previous changes.
17:00 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47733/
17:23 PacoLinux joined #parrot
17:24 GodFather joined #parrot
17:25 arnsholt whiteknight: Yeah, I noticed. Although it's pretty good for what I want to do
17:26 arnsholt (I'm trying to write a utility that checks crosschecks a grammar class and an actions class)
17:26 arnsholt Only thing really missing is getting methods from superclasses
17:27 arnsholt Which shouldn't be that hard
17:36 tetragon joined #parrot
17:37 dalek winxed: r506 | julian.notfound++ | trunk/winxedst1.winxed:
17:37 dalek winxed: allow const statements with multiple declarations in stage 1
17:37 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=506
17:42 dalek winxed: r507 | julian.notfound++ | trunk/winxedst1.winxed:
17:42 dalek winxed: allow const in multiple declarations to use previous values in the same
17:42 dalek winxed: statement in stage 1
17:42 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=507
18:03 * Coke is @yapc.
18:06 dalek winxed: r508 | julian.notfound++ | trunk/examples/xlibtest.winxed:
18:06 dalek winxed: some cleanup and use new features in Xlib example
18:06 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=508
18:15 mdiep Coke: Ohio welcomes you.
18:20 theory joined #parrot
19:08 kthakore oh hai
19:08 Coke \o/
19:09 Coke I and mdiep will be hacking at morrison.
19:11 Coke hey, I wonder if the OSU OSL folks are about.
19:20 kid51 joined #parrot
19:25 dalek partcl: 5f0d1a6 | Whiteknight++ | src/ (2 files):
19:25 dalek partcl: fixes after ns_func_cleanup
19:25 dalek partcl: review: http://github.com/partcl/partcl/commit/5​f0d1a6b4f7466cd03c9395a7b252a0aef963be5
19:25 Coke mikehh: there!
19:26 Coke mdiep++ found the magic button we needed to press to apply whiteknight++'s patch.
19:27 cotto Coke, I thought Parrot was hosted at *Oregon* State University.
19:27 dalek parrot: r47734 | NotFound++ | trunk (4 files):
19:27 dalek parrot: always fill isa_hash to avoid making special cases for null in several places
19:27 dalek parrot: review: http://trac.parrot.org/parrot/changeset/47734/
19:27 cotto s/Parrot/parrot.org/
19:28 Coke cotto: oh. =-)
19:28 LoganLK joined #parrot
19:39 kid51 Too man Os have SUs.
19:40 kid51 s/man/many/
19:46 kthakore kid51: SUs?
19:46 purl rumour has it SUs is Software Update Services or at http://www.susserver.com/
19:46 kthakore ah
19:49 dalek partcl-nqp: c38a414 | Coke++ | Configure.pl:
19:49 dalek partcl-nqp: Warn builder that ICU is really really really really nice to have.
19:49 dalek partcl-nqp: review: http://github.com/partcl/partcl-nqp/commit​/c38a414eff31bff84ba04cb1ba613553ce8c0724
19:53 kid51 kthakore: State University(ies)
19:54 whiteknight really really really really
19:54 spinclad joined #parrot
19:55 kthakore whiteknight: ??
19:55 whiteknight kthakore: in response to Coke's last commit message
19:57 kthakore kid51: lots of OS have state universities?
20:03 whiteknight Coke: has anything else happened with the board nominations? Last I heard we were working on taking a membership census, but I haven't seen any other information after that
20:04 kjeldahl joined #parrot
20:05 Psyche^ joined #parrot
20:12 ambs joined #parrot
20:15 * ambs sends a subliminally message to Coke
20:22 Coke ambs: hush.
20:22 Coke whiteknight: Think about whether you want to run.
20:22 Coke we need to give 10 days notice before holding the election, so it's not happenign this week.
20:22 whiteknight ok. Thought about it.
20:22 Coke whiteknight: you running?
20:23 whiteknight Yeah, I think I will
20:23 whiteknight The more, the merrier
20:25 whiteknight who else is? Anybody else put their name in yet?
20:26 eternaleye joined #parrot
20:32 NotFound A tailcall should increase the recursion level?
20:32 whiteknight NotFound: No, I don't think it should
20:32 NotFound Looks like it does.
20:32 whiteknight oh
20:34 whiteknight I mean, it is a recursion, but it doesn't create a new context and doesn't increase the cost of retuning back up the call chain, so I don't see any practical reason to increase the recursion count
20:35 nopaste "NotFound" at 192.168.1.3 pasted "tailcall optimization increases recursion level?" (51 lines) at http://nopaste.snit.ch/21421
20:36 NotFound This program dies with recursion level exceded
20:36 whiteknight NotFound: I call it a bug. Create a ticket or just fix it
20:40 moritz uhm
20:40 moritz why has getattribute signature pps and getprop psp?
20:41 pmichaud moritz: don't ask for consistency in the opcode set... it doesn't exist.
20:42 * moritz asks anyway
20:42 NotFound Please don't suggest an incompatible change for aestetic reasons.
20:42 * moritz just asks
20:43 ambs aestetic++
20:44 ambs at least, girls aestetic
20:44 ambs (A)
20:44 ambs O;_)
20:44 cotto Warning: cat-like typing detected.
20:45 ambs :D
20:45 NotFound Looks like a problem is that the recursion count for tailcall is decreased at get_args. No arg, no dec.
20:46 bacek Good morning, humans
20:46 NotFound But using args it dies after creating a lot of contexts.
20:46 cotto hi bacek
20:46 bacek aloha, cotto
20:48 cotto bacek, what'd be involved in making ParrotInterp use PackFiles internally and what would be the win from that?
20:48 cotto s/PackFiles/PackFile PMCs/
20:48 bacek cotto, basically, rewriting PackFile_* functions from scratch.
20:49 bacek (Well, not from scratch, but almost)
20:49 bacek Then switch ParrotInterp to use Packfile PMCs instead of current PF_* structs.
20:50 bacek And 2 wins: 1) no more memory leaks from PackFiles. 2) Ability to attach freshly generated Packfile PMC to Interp.
20:50 cotto and the win from that would be that we'd know that PackFile PMCs could be used for PBC generation?
20:51 bacek We already know it!
20:51 cotto ok
20:51 NotFound bacek: Have you found the sub problem?
20:52 bacek NotFound, unfortunately...
20:52 cotto If you could write up a brief tasklist, that sounds like something I might be able to take on.
20:52 cotto (wiki, here, parrot-dev, wherever)
20:52 NotFound Well, the we already know that we can generate PBC that doesn't work ;)
20:53 bacek NotFound, they does :)
20:53 NotFound bacek: yeah, for some value of 'work'
20:53 bacek cotto, Packfile tasklist?
20:54 bacek NotFound, I can load generated PBC into "fresh" parrot and it works...
20:54 patspam joined #parrot
20:54 cotto sure.  does it already exist?
20:54 dalek winxed: r509 | julian.notfound++ | trunk/winxedst1.winxed:
20:54 NotFound bacek: And it can call generated subs?
20:54 dalek winxed: experimental tailcall optimization in stage 1
20:54 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=509
20:55 bacek NotFound, at least one of it. I didn't implement PCC in PIRATE yet.
20:55 GeJ Bonjour everyone.
20:55 bacek cotto, include/parrot/interpeter.h, line 215, 216
20:56 bacek cotto, start with replacing them with "PMC *" :)
20:56 bacek GeJ, G'Day
20:57 NotFound bacek: that is orthogonal with locating the sub in the namespace, wich is the problem.
20:58 bacek NotFound, I'll probably hit this in next couple of days. I don't clearly understand why it's broken...
20:58 NotFound bacek: that is the reason I resist to say 'it works' for anow.
20:59 bacek NotFound, fair enough.
21:00 NotFound The majority of people that looked of the problem don't even clearly understand it.
21:00 NotFound (and I think that group of people is you and me) ;)
21:00 bacek :)
21:01 cotto I tried looking at it, but it was more like show a dog a card trick. ;)
21:01 whiteknight wha problem?
21:01 cotto why examples/pir/make_hello_world_pbc.pir is broken
21:01 whiteknight when did that break?
21:02 patspam joined #parrot
21:03 mikehh whiteknight: TT #1685
21:04 mmcleric joined #parrot
21:16 dalek rakudo: 88210a9 | tene++ | src/glue/run.pir:
21:16 dalek rakudo: Put a Perl 6 Hash in %*VM
21:16 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/8​8210a9e8455f7578aaa0f9cd9a7e38af1ae57b1
21:26 patspam joined #parrot
21:37 dalek pir: 858294c | bacek++ | src/POST/Sub.pm:
21:37 dalek pir: Add storage for Sub's Labels.
21:37 dalek pir: review: http://github.com/bacek/pir/commit/858​294cd1b68430af5149432866e85ffe4fe2310
21:37 dalek pir: 622c8b3 | bacek++ | t/ (2 files):
21:37 dalek pir: Rename 03-sugar to 04-sugar. 03 will be used for labels.
21:37 dalek pir: review: http://github.com/bacek/pir/commit/622​c8b31d8c44ed0b73c6df0d785599a06145ef6
21:37 dalek pir: ce1e3ba | bacek++ |  (3 files):
21:37 dalek pir: Add stub for POST::Label
21:37 dalek pir: review: http://github.com/bacek/pir/commit/ce1​e3bad30c45573840719b7a7dca8f429518a73
21:37 dalek pir: 4d64280 | bacek++ | src/POST/Label.pm:
21:37 dalek pir: Add Label.declared
21:37 dalek pir: review: http://github.com/bacek/pir/commit/4d6​428059ff97911520a3cc4d798eff99facea5e
21:37 dalek pir: 5265a72 | bacek++ |  (3 files):
21:38 dalek pir: Store labels in Sub.labels
21:38 dalek pir: review: http://github.com/bacek/pir/commit/526​5a72d582e43bd9eb7f390c584478526c8ff96
21:38 dalek pir: 589b599 | bacek++ |  (2 files):
21:38 dalek pir: Add test for label redeclaration and fix typo
21:38 dalek pir: review: http://github.com/bacek/pir/commit/589​b599842097d5ee0f987dcb671b82032798ef5
21:38 dalek pir: 30ed00b | bacek++ |  (2 files):
21:38 dalek pir: Handle undeclared labels
21:38 dalek pir: review: http://github.com/bacek/pir/commit/30e​d00b037fc3b73d550931adf70d0a518afaf20
21:39 patspam joined #parrot
21:42 patspam joined #parrot
21:44 patspam joined #parrot
21:49 lucian_ joined #parrot
21:54 dalek pir: 524ff56 | bacek++ | TODO:
21:54 dalek pir: Update TODO.
21:54 dalek pir: review: http://github.com/bacek/pir/commit/524​ff562309aa36d36c2f2f981da87e89e204286
22:04 bacek cotto, ping
22:10 lucian joined #parrot
22:11 jrtayloriv joined #parrot
22:13 mikehh PIRATE - builds/tests ok
22:14 bacek msg cotto some kind of "pbc testing framework" for pirate will be helpful. With workaround for TT#1685. E.g. test_pbc($pir_text) { compile; write tmp.pbc; capture stdout; spawn parrot tmp.pbc }. Capturing output is optional. We can write tests to output TAP.
22:14 purl Message for cotto stored.
22:14 bacek mikehh, thanks!
22:16 cotto ohai
22:18 bacek cotto, aloha :)
22:19 cotto sounds doable
22:19 bacek cotto, indeed. You can grab t/post.nqp as sample of "testdata" grammar to implement something similar.
22:20 cotto sure
22:20 bacek afk # time for prepare to $dayjob
22:22 casan joined #parrot
22:23 mikehh partcl builds/test PASS, partcl-nqp builds/test PASS
22:24 mikehh however partcl - Files=74, Tests=1382, partcl-nqp - Files=46, Tests=672
22:30 casan joined #parrot
23:17 davidfetter joined #parrot
23:52 eternaleye_ joined #parrot
23:55 bacek_at_work joined #parrot

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

Parrot | source cross referenced