Camelia, the Perl 6 bug

IRC log for #parrot, 2010-05-17

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:06 bacek_at_work ~~
00:06 bacek_at_work Coke, which code?
00:30 Coke bacek_at_work: the JAPHs in his most recent blog post.
00:31 bacek_at_work Coke, url?
00:32 mikehh bacek_at_work: http://www.effectiveperlprogramming.com/
00:38 bacek_at_work mikehh, thanks
00:49 rurban_ joined #parrot
00:54 abqar joined #parrot
01:04 TiMBuS joined #parrot
01:05 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#33869), fulltest) at r46720 - Ubuntu 10.04 amd64 (g++)
01:19 Coke yay, my son is hooked on doctor who!
01:29 mikehh Coke: I remember the original series which I must have seen in the early '60s
01:32 * mikehh needs sleep
01:33 Coke I watched the 5th doctor in the 70s. Starting him on the 9th.
01:35 mikehh in those days it was in B&W, although I was living in Africa at the time (Rhodesia as it was then)
01:38 tcurtis I want to get into Doctor Who, but I worry that if I do, I would want to watch every single episode ever created. And I'md not sure I have the time for that.
01:41 mikehh tcurtis - it could take quite a while
01:49 tcurtis For this reason, I refrain from Doctor Who and other well-reputed lengthy series.
01:49 mikehh damnit nearly 3am here - realy must sleep
02:02 JimmyZ joined #parrot
02:04 agricola joined #parrot
02:05 pjcj joined #parrot
02:21 Andy joined #parrot
02:34 dalek parrot: r46721 | petdance++ | trunk/src/gc/alloc_resources.c:
02:34 dalek parrot: removed unused variables and commented out code
02:34 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46721/
02:34 dalek parrot: r46722 | petdance++ | trunk/src/dynpmc/gziphandle.pmc:
02:34 dalek parrot: removed dead vars and fixed some POD
02:34 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46722/
02:34 dalek parrot: r46723 | petdance++ | trunk/src/gc/alloc_resources.c:
02:34 dalek parrot: removed dead var
02:34 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46723/
02:34 Andy That Andy, he is just so industrious!
02:42 janus joined #parrot
03:02 sorear Andy: Doesn't he know revision numbers are a finite resource?  He's being quite wasteful with them
03:02 Andy Something should be done.
03:03 Andy Perhaps a stern word.
03:13 cotto Sheesh.  I suppose he's also wasting gravity.
03:27 tcurtis Do GSoC students get commit bits?
03:28 cotto if they send in a signed CLA
03:35 Andy cotto: If the gravity is being affected by Andy's reckless commits, by all means he should be disciplined.
03:38 agricola1 joined #parrot
04:49 snarkyboojum joined #parrot
05:15 cotto Is there any reason not to make opsrenumber part of ops2c?
05:17 cotto It's a small bit of extra work and a little extra redundant header modification.
05:18 cotto A nice thing is that it'd mean one less thing to think about when messing with ops.
05:18 cotto just update the .ops files, update ops.num and rebuild
05:26 bacek_at_work cotto, wfm
05:31 cotto ok
05:32 cotto clock?
05:32 purl cotto: LAX: Sun 10:32pm PDT / CHI: Mon 12:32am CDT / NYC: Mon 1:32am EDT / LON: Mon 6:32am BST / BER: Mon 7:32am CEST / IND: Mon 11:02am IST / TOK: Mon 2:32pm JST / SYD: Mon 3:32pm EST /
05:33 cotto slowbot is slow
05:33 purl i already had it that way, cotto.
05:36 plobsing slowbot?
05:36 purl slowbot is slow
05:43 cotto slowbot is also purl
05:43 purl okay, cotto.
05:44 nopaste "plobsing" at 192.168.1.3 pasted "Op Audit" (13 lines) at http://nopaste.snit.ch/20575
05:44 plobsing I went through the ops and those were the ones that jumped out at me
05:44 plobsing do we really need a factorial op in core?
05:46 sorear need? no
05:47 cotto How else are you going to optimize factorialization?
05:47 plobsing Oh I don't know? Make op dispatch faster? JIT?
05:48 plobsing there were a few other ops I'm on the fence about. like noop.
05:51 bacek_at_work plobsing, noop is quite useful and inexpensive!
05:51 cotto how else do you want to do nothing?  add 0?
05:51 cotto that's way too inefficient
05:52 cotto ;)
05:52 plobsing I understand how it is useful in some instruction sets, but I don't see how in PBC
05:53 plobsing for example, there is no need to wait for changes to propagate
05:53 sorear istr some of our code monkeypatches nops into packfile segments
05:53 sorear so run fulltest after axing...
05:55 plobsing do we have any computed goto? noop might be useful for playing games with alignment there.
05:55 cotto cgoto is history
05:56 sorear our goto ops all seem to take integers
05:56 plobsing not the op dispatcher, like goto $I0
05:56 sorear and ExceptionHandler is set up with an integer
05:57 plobsing as I said, I'm on the fence. I hadn't seen any good uses (monkey-patching seems like a decent use case)
05:58 sorear yep, jump $I0 is valid PASM...
05:58 sorear not sure if it's expressable in pir
05:58 plobsing this presents many opportunities for evil. good.
06:00 sorear local_branch and local_return are also quite nice
06:00 sorear push and pop IP off a RIA
06:01 plobsing they seem questionable to me.
06:01 plobsing but I'm not hard set on killing them
06:02 plobsing enternative on the other hand...
06:04 cotto not much purpose left for that one
06:05 plobsing do I have to deprecate ops that are marked "not for external use"?
06:06 sorear well, you're breaking pbc compat
06:08 dalek parrot: r46724 | plobsing++ | branches/libjit_framebuilder2:
06:08 dalek parrot: functionality now available in a library at �http://github.com/plobsing/parrot-libjit-fb
06:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46724/
06:09 plobsing breaking PBC compat doesn't require deprecation
06:09 plobsing IIUC
06:14 cotto best to wait until after the release
06:15 uniejo joined #parrot
06:15 dalek TT #1105 closed by plobsing++: change to a libjit based frame builder
06:15 dalek TT #1105: http://trac.parrot.org/parrot/ticket/1105
06:15 plobsing sure. I'm already holding back on a couple of things.
06:31 dalek TT #1644 created by plobsing++: [RFC] Text format serializer for PASM/PIR constants
06:31 dalek TT #1644: http://trac.parrot.org/parrot/ticket/1644
06:32 * sorear wonders if 1644 would allow Rakudo to store meta objects in PBCs and kill the startup cost of constructing them
06:33 Coke joined #parrot
06:34 plobsing sorear: you already can do that
06:34 plobsing I talked to the rakudo devs a couple weeks back to get the requirements for killing that startup cost.
06:35 plobsing You need PBC to be able to store: lexpads, dynpmcs, dynops
06:36 plobsing which are all likely hard to do.
06:36 plobsing but storing arbitrary constants in PBC is already doable
06:36 plobsing with :anon :immediate magic subs
06:37 iblechbot joined #parrot
06:39 plobsing what TT #1644 buys us is round-trip PIR->PBC->PASM or PIR->PBC->PIR
06:39 plobsing or gets us closer at least
06:42 fperrad joined #parrot
06:47 aukjan joined #parrot
06:49 fperrad_ joined #parrot
06:56 plobsing hmmmm... the more I look at my new deserialization, the more I realize I hate our 2-pass approach to freeze/thaw.
06:57 cotto Freeze it with fire.
06:57 cotto wait.  that doesn't really work
06:57 cotto nm
06:59 plobsing I never said I thought I could do it better. I just think it can be done better ... somehow.
07:26 fperrad ping bacek
07:47 uniejo joined #parrot
07:49 integral joined #parrot
08:05 bacek fperrad, pong
08:06 fperrad bacek, in LWP.pir, I use a StringBuilder as receive buffer, I compute a progress rate,
08:06 fperrad (I known the Content-Length), get_integer gives a wrong value (not the same as the length of get_string)
08:07 bacek SB.get_integer returns capacity of storage, not string length.
08:07 bacek Just fetch string and take length of it
08:10 fperrad bacek, if I fetch string after each fragment, I lose the optimization of SB. I want the length without compute the string
08:10 bacek fperrad, hmm. Why do you need it?
08:10 bacek (in the middle of computation)
08:12 fperrad I compute/update a progress rate after the reception  of each fragment
08:12 fperrad try :
08:12 fperrad $ parrot examples/io/get.pir http://www.parrot.org/
08:14 bacek rebuilding parrot
08:14 bacek looks cool :)
08:17 bacek I can add something like StringBuilder.get_string_length method
08:20 moritz wouldn't it be sensible for get_integer to return the length, and have a longer name report the memory usage?
08:21 bacek moritz, I slightly disagree.
08:22 bacek SB is not String.
08:22 moritz ok
08:22 bacek And have large subset of String's VTABLE can lead to wrong usage of it.
08:22 moritz ah right, String's get_integer tries to parse as a number, right?
08:23 bacek moritz, yes.
08:25 bacek fperrad, r46725
08:27 fperrad bacek, thanks
08:29 aukjan joined #parrot
08:37 dalek parrot: r46725 | bacek++ | trunk (2 files):
08:37 dalek parrot: Add StringBuilder.get_string_length method to fetch string length without cloning buffer
08:37 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46725/
08:37 dalek parrot: r46726 | fperrad++ | trunk/runtime/parrot/library/LWP.pir:
08:37 dalek parrot: [LWP] refactor parse http response
08:37 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46726/
08:49 rurban_ joined #parrot
08:52 moritz http://geekandpoke.typepad.com/.a/6a0​0d8341d3df553ef0133ed964b97970b-800wi we should get an insurance for Coke++ :-)
08:53 dalek parrot: r46727 | fperrad++ | trunk/runtime/parrot/library/LWP.pir:
08:53 dalek parrot: [LWP] fix content without Content-Length
08:53 dalek parrot: "GET �http://fperrad.googlepages.com/home HTTP/1.1" reworks
08:53 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46727/
09:10 dalek parrot: r46728 | fperrad++ | trunk/runtime/parrot/library/LWP.pir:
09:10 dalek parrot: [LWP] refactor with StringBuilder.get_string_length(), bacek++
09:10 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46728/
09:12 viklund joined #parrot
09:12 moritz what's the reason for having LWP in parrot? plumage?
09:14 bacek moritz, smolder?
09:16 moritz ah, right
09:16 moritz reinventing the wheels (bad), eating own dogfood (good)
09:18 bacek it's not quite "reinventing". It's kind of rebuilding old wooden one with alloys :)
09:25 snarkyboojum joined #parrot
09:32 bacek msg pmichaud I've pushed new branch "multis" into nqp-rx. http://gist.github.com/403584 is example of work :)
09:32 purl Message for pmichaud stored.
09:33 * bacek dancing around support of multi in nqp
09:35 bacek msg pmichaud http://gist.github.com/403587 is full diff. I have to remove <typename> cause it's eat "multi" as type...
09:35 purl Message for pmichaud stored.
09:37 gaz joined #parrot
09:37 * bacek look forward to reimplement PCT in new shiny NQP :)
09:46 * JimmyZ wonders what benefit to parrot?
09:57 bacek JimmyZ, -Ofun
10:15 dalek parrot: r46729 | gerd++ | trunk (2 files):
10:15 dalek parrot: Add a little program that can be used by the release manager to increment the version number
10:15 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46729/
10:31 bacek purl, parrot roadmap?
10:31 purl well, parrot roadmap is http://www.perlfoundation.org/pa​rrot/index.cgi?release_planning
10:32 JimmyZ parrotroadmap?
10:32 JimmyZ roadmap?
10:32 purl roadmap is http://icanhaz.com/parrotroadmap  or  http://trac.parrot.org/parrot/roadmap
10:33 JimmyZ no, parrot roadmap is see roadmap
10:33 purl okay, JimmyZ.
10:51 mj41 joined #parrot
10:54 bacek mj41, aloha
10:55 mj41 bacek, cau
10:55 bacek mj41, how hard it will be to add few HLLs to TapTinder?
10:57 mj41 bacek, here ( http://dev.taptinder.org/wiki​/ToDo/Another_project_to_test ) is related summary
11:00 bacek mj41, no... Little bit more complex than that. Test HLLs against Parrot's trunk.
11:01 mj41 afk, meeting
11:08 mj41 back, it was a short one
11:21 dalek parrot: r46730 | NotFound++ | trunk/src/ops/experimental.ops:
11:21 dalek parrot: avoid longjmp if destination runloop is already current in unroll
11:21 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46730/
11:25 GodFather joined #parrot
11:27 khairul joined #parrot
11:37 ruoso joined #parrot
11:38 JimmyZ joined #parrot
11:50 whiteknight joined #parrot
12:01 whiteknight good morning, #parrot
12:04 TiMBuS joined #parrot
12:11 Coke (breaking PBC compat) doesn't require /another/ deprecation.
12:12 bacek whiteknight, aloha
12:14 bacek msg pmichaud I finished :multi generating in nqp. Finally ready for review (and hopefully merge). About 130 lines diff + tests.
12:14 purl Message for pmichaud stored.
12:38 * Coke spies bacek's picture on github. it's like the Terminator.
12:40 bacek Coke, sigh... No I have to destroy you to keep this secret...
12:40 * Coke misses the sarah conner chronicles.
12:41 dalek winxed: r468 | julian.notfound++ | trunk/winxedst1.winxed:
12:41 dalek winxed: use PIR .const directive for const string in stage 1
12:41 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=468
12:41 bacek Coke, http://www.facebook.com/Bace​k?v=photos&amp;sb=4#!/photo.php?pid=30172557&id=1408388236 :)
12:43 Coke ZOMG STEERING WHEEL ON WRONG SIDE!
12:46 bacek It's RIGHT side!
12:47 pmichaud bacek: In general, I'd prefer the Foo::Bar conversion to take place inside of PCT, not NQP
12:47 pmichaud (I think.)
12:47 pmichaud maybe not
12:47 pmichaud I'm pretty sure PCT does that conversion other places....
12:47 * pmichaud looks
12:47 bacek pmichaud, yes, it does.
12:47 pmichaud I'm certain that NQP doesn't want to be the thing constructing ['Foo';'Bar'] keys.
12:47 bacek but I want to ensure that overall things looks good before touching PCT.
12:48 pmichaud okay.
12:49 pmichaud I'll take a more detailed look shortly -- I may have to run a few errands this morning first.
12:49 bacek (and touching PCT before after release isn't quite reasonable :)
12:49 bacek pmichaud, ok.
12:49 pmichaud yes, I think we'll put multi in after release
12:49 pmichaud that's saner/safer
12:51 bacek OTOH, I can create branch for PCT now.
13:09 plobsing joined #parrot
13:10 atrodo joined #parrot
13:17 dalek parrot: r46731 | bacek++ | branches/pct_multi_support:
13:17 dalek parrot: Branch for implementing support of :multi generating in PCT
13:17 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46731/
13:32 Coke gerd - what is the purpose of the update revision # script?
13:33 Coke seen gerd?
13:33 purl gerd was last seen on #parrot 25 days, 5 hours, 8 minutes and 38 seconds ago, saying: but I will send an email with this question, bye  [Apr 22 08:24:23 2010]
13:33 Coke msg gerd the new version update script should probably 1) be referenced in the release guide, 2) actually update /all/ the version numbers the release manager has to update.
13:33 purl Message for gerd stored.
13:33 dalek parrot: r46732 | coke++ | trunk/tools/util/crow.pir:
13:33 dalek parrot: Rip out broken method, replace with less PIR.
13:33 dalek parrot: (probably broken due to recent string changes)
13:33 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46732/
14:07 dalek parrot: r46733 | bacek++ | branches/pct_multi_support/compilers/pct/src (2 files):
14:07 dalek parrot: Add PAST::Block.multi and POST::Sub.multi.
14:07 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46733/
14:07 dalek parrot: r46734 | bacek++ | branches/pct_multi_support/compi​lers/pct/src/PAST/Compiler.pir:
14:08 dalek parrot: Pass PAST::Block.multi to POST::Sub.new
14:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46734/
14:08 dalek parrot: r46735 | bacek++ | branches/pct_multi_support/compi​lers/pct/src/POST/Compiler.pir:
14:08 dalek parrot: Generate :multi in POST::Compiler
14:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46735/
14:08 dalek parrot: r46736 | bacek++ | branches/pct_multi_support/t/compilers/pct/post.t:
14:08 dalek parrot: Add test for :multi generating from POST::Sub.
14:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46736/
14:08 dalek parrot: r46737 | bacek++ | branches/pct_multi_support/compi​lers/pct/src/POST/Compiler.pir:
14:08 dalek parrot: Fix generating :multi
14:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46737/
14:08 dalek parrot: r46738 | fperrad++ | trunk/runtime/parrot/library/HTTP/Message.pir:
14:08 dalek parrot: [LWP] request POST (step 1)
14:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46738/
14:08 dalek parrot: r46739 | bacek++ | branches/pct_multi_support/com​pilers/pct/src/PAST/Node.pir:
14:08 dalek parrot: Add PAST::Var.multitype
14:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46739/
14:09 PacoLinux joined #parrot
14:19 bacek msg pmichaud It's kind of done. Branches in parrot and nqp.
14:19 purl Message for pmichaud stored.
14:20 patspam joined #parrot
14:22 jsut joined #parrot
14:36 aukjan joined #parrot
14:38 bubaflub joined #parrot
14:39 whiteknight t/distro/file_metadata.t fails on test 4
14:39 whiteknight #  svn ps svn:eol-style native tools/util/inc_ver.pir
14:39 whiteknight Anybody looking for cheap karma to fix that?
14:40 bubaflub whiteknight: i gotta bring my svn repo up to date
14:40 bubaflub but
14:40 bubaflub i can fix that
14:40 whiteknight awesome
14:40 whiteknight bubaflub++
14:42 Coke whiteknight: that's the thing I just messaged gerd about.
14:42 Coke seems LTA in its current state.
14:43 whiteknight Coke: oh, ok. I'm starting my fulltest runs now and that's the only error I'm seeing
14:43 whiteknight so it doesn't matter to me who fixes it, so long as it's done today
14:43 whiteknight (and I can't really do it now from work)
14:47 moritz jnthn just reported some new Rakudo troubles on parrot HEAD
14:47 moritz I'll see if I can duplicate them here
14:49 whiteknight great
14:49 * whiteknight mumbles angrily
14:50 whiteknight moritz: if you can duplicate them, can you email info to the parrot-dev list ASAP?
14:51 moritz whiteknight: I'll just tell jnthn to do it :-)
14:51 whiteknight that's fine too. So long as the list sees the report soon
14:51 ash_ joined #parrot
14:51 whiteknight want to get as many eyes on any problems as possible
14:52 whiteknight Anybody here on win32?
14:53 ash_ there is a great post about the llvm being integrated with ghc, http://blog.llvm.org/2010/05/glasg​ow-haskell-compiler-and-llvm.html i can't wait to start working on the llvm backend for parrot, i am planning on starting this week on the libffi work
14:53 whiteknight ash_: nice
14:57 aukjan joined #parrot
15:08 bubaflub joined #parrot
15:08 ash_ has anyone tried adding AutoZone to the list of supported gc's?
15:09 bubaflub whiteknight: i svn ps'd that file you asked for, dalek should report it anytime now
15:09 whiteknight bubaflub: thanks!
15:09 bubaflub whiteknight: np
15:09 whiteknight bubaflub: are you on win32?
15:09 bubaflub whiteknight: lemme know if you need anything else
15:09 bubaflub whiteknight: nay, OS X
15:09 * Coke guesses that if a TT is owned by someone, but only by default, then there is no harm in stealing that ticket.
15:09 whiteknight bubaflub: okay. t/profiling/profiling.t has a failure on win32, and you're listed as the last editor of that file
15:10 whiteknight I'll dig into it later
15:10 bubaflub whiteknight: i'm checking the logs, lemme see what i did to it
15:11 whiteknight okay, thanks. It looks like an error in the test logic itself, not a problem with the VM, so it's not a show-stopper for the release
15:11 whiteknight I opened a ticket for it too, in case you need more info. It's giving me the error that "C:/Documents" is not a valid path
15:11 bubaflub whiteknight: i fixed a codingstd test, cotto was before me
15:12 bubaflub at r45320
15:12 moritz path escape issue?
15:12 moritz should probably be "C:/Documents and Settings/" or so
15:12 bubaflub moritz: yeah, that looks right
15:13 whiteknight moritz: yeah, I'm sure that's the case
15:13 whiteknight probably doesn't help that my parrot checkout is in My Documents, and not in a space-free path off the root
15:14 dalek parrot: r46740 | bubaflub++ | trunk/tools/util/inc_ver.pir:
15:14 dalek parrot: set svn ps svn:eol-style native for tools/util/inc_ver.pir
15:14 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46740/
15:14 dalek parrot: r46741 | fperrad++ | trunk/runtime/parrot/library/HTTP/Message.pir:
15:14 dalek parrot: [LWP] request POST (step 2)
15:14 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46741/
15:18 moritz whiteknight: neither me nor colomon could reproduce the rakudo bugs on amd64 linux... jonathan will try again with a cleaned/nuked build
15:19 whiteknight moritz: okay, I will try it myself as well. Does it have to be on amd64, or can it be x86?
15:19 moritz whiteknight: jnthn observed it on x86
15:19 dalek rakudo: 96fbd6f | (Solomon Foster)++ | t/spectest.data:
15:19 dalek rakudo: Turn on the new S05-modifier/global.t.
15:19 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/9​6fbd6fb500d7c333496bf9fbfcb57fb5d176838
15:19 dalek rakudo: ec37acc | (Solomon Foster)++ | src/core/Cool-str.pm:
15:20 dalek rakudo: In .match's global path, copy the match variable, then call "take" with the
15:20 dalek rakudo: copy.
15:20 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​c37accfda6533d2d926abfdec8e038a7e41fcf6
15:20 whiteknight ah, okay. I can run it locally, but it's a VM so I'm sure it will take for-damn-ever
15:21 shockwave joined #parrot
15:22 shockwave Hi
15:22 purl hello, shockwave.
15:22 shockwave This page: http://docs.parrot.org/parrot/l​atest/html/src/ops/cmp.ops.html
15:22 whiteknight hello shockwave
15:22 shockwave Mentions of these ops:
15:22 shockwave not(out INT, in INT)
15:22 shockwave not(out PMC, invar PMC)
15:22 shockwave It would be greate to also have: not(out INT, invar PMC)
15:23 shockwave That makes the generated code more natural.
15:23 * moritz points shockwave to the 'box' opcode
15:23 shockwave If someone doesn't have a specific objection on why that would be a good idea, I'll create a request ticket.
15:23 moritz or is that a dynop somewhere?
15:24 shockwave moritz, It makes me sad to have to box so many things.
15:24 shockwave It would be nice _not_ to have to box some things that feel natural.
15:24 moritz having not(PMC, in) would also box
15:25 shockwave what?
15:25 shockwave I don't get it. I'm requesting this: not(out INT, invar PMC)
15:25 moritz oh
15:25 moritz wrong way round
15:25 moritz sorry
15:26 Coke shockwave: let them complain on an RFC ticket.
15:26 shockwave It looks like the code should be mostly there, because the current two overloads should take care of most functionality.
15:27 shockwave Coke, alrighty.
15:34 Coke whiteknight: snuck in an update to skeleton.pir, only developer-facing.
15:35 whiteknight Coke: which one is that?
15:35 Coke the skeleton pir file you get when you edit a new .pir file in vi
15:36 Coke http://trac.parrot.org/par​rot/ticket/1526#comment:2
15:36 Coke simplifies creation of new .pir when hacking on parrot.
15:36 Coke (and was borked before that commit.)
15:40 dalek TT #1526 closed by coke++: editor/skeleton.pir cleanup.
15:40 dalek TT #1526: http://trac.parrot.org/parrot/ticket/1526
15:40 dalek TT #1645 created by shockwave++: Request for another overload to 'not' opcode.
15:40 dalek TT #1645: http://trac.parrot.org/parrot/ticket/1645
15:40 Coke (/me manages to keep the # of tickets constant. :|
15:46 shockwave purl paste
15:46 purl paste is (see: nopaste) or like glue but a little safer to sniff. or http://nopaste.snit.ch:8001/ or http://scsys.co.uk:8001/ anywhere shadowpaste is or mmm, sticky paste or You there! Eating the paste. or <see> 2 girls, 1 paste or App::Nopaste or toxic for bots and humans
15:47 dalek parrot: r46742 | coke++ | trunk (5 files):
15:47 dalek parrot: generate skeleton.pir to avoid hardcoding wrong values (fix TT#1526)
15:47 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46742/
15:48 Andy joined #parrot
15:49 ash__ joined #parrot
15:50 gbacon_ joined #parrot
15:53 Topic for #parrotis now Parrot 2.3.0 Released | parrot.org | Log: irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | Priority : apply deprecations, merge branches, finish exceptions_refactor branch | GSoC students, please read http://trac.parrot.org/par​rot/wiki/GSoCersStartHere
15:54 agricola joined #parrot
15:56 theory joined #parrot
16:12 nopaste "shockwave" at 192.168.1.3 pasted "Some comment that's too long for IRC" (19 lines) at http://nopaste.snit.ch/20576
16:12 shockwave I was writing a comment, but it was too long for the window. So I pasted it somewhere.
16:13 shockwave It's just that, a comment. Maybe a justification, I not sure anymore.
16:13 shockwave Anyhow, it's not vital to read, but if you do. Don't hesitate to comment back.
16:13 shockwave Thanks
16:14 Coke argh, scrollfail.
16:17 shockwave Coke: Oops, apparently that pasting service has no linewrap. They're probably just <pre>....</pre>'ing everything. Sorry.
16:19 Coke shockwave: there's a word wrap button. all good. =-)
16:35 whiteknight we give our baby milk early, to try and prevent him from crying
16:36 shockwave :-P
16:36 dalek parrot: r46743 | fperrad++ | trunk/runtime/parrot/library (2 files):
16:36 dalek parrot: [LWP] request POST (step 3)
16:36 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46743/
16:39 whiteknight urg, I don't have ICU on the VM I am using for testing
16:39 iblechbot joined #parrot
16:41 shockwave Time for lunch. Laterz
16:41 shockwave left #parrot
16:43 whiteknight shockwave: put together a patch to add the new opcode you want to see added. we can look at it and apply it if it's good
16:43 whiteknight oh, damnit
16:51 rurban_ joined #parrot
16:52 bubaflub whiteknight: don't forget tests
16:52 bubaflub (with that patch for a new opcode)
17:03 cotto_work good morning, parrots
17:38 rurban joined #parrot
17:46 tcurtis joined #parrot
17:52 dalek TT #1646 created by doughera++: Failure due to missing external definition of inlined function ...
17:52 dalek TT #1646: http://trac.parrot.org/parrot/ticket/1646
17:57 Mokurai joined #parrot
18:12 spinclad joined #parrot
18:13 gaz joined #parrot
18:15 dolmen joined #parrot
18:26 joeri joined #parrot
18:30 rockon joined #parrot
18:30 rockon left #parrot
18:35 dalek rakudo: a797463 | (Solomon Foster)++ | src/core/Cool-str.pm:
18:35 dalek rakudo: Allow :global as well as :g for match method.
18:35 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/a​797463edb84597afa451f3980accebbdfd67a28
18:35 dalek rakudo: 816a4fc | (Solomon Foster)++ |  (2 files):
18:35 dalek rakudo: Better support for :c / :continue in match, and turn on S05-modifier/continue.t.
18:35 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/8​16a4fc01fb86df24bca53c04d5ee429da192835
18:40 cotto_work I'm not seeing any of the failures jnthn mentioned when using Parrot r46743 and the latest rakudo.
18:42 dalek parrot: r46744 | fperrad++ | trunk (2 files):
18:42 dalek parrot: [LWP] add an example: post.pir (don't work)
18:42 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46744/
19:02 tcurtis_ joined #parrot
19:02 dalek joined #parrot
19:08 whiteknight joined #parrot
19:11 whiteknight 2 hours later, I aborted the rakudo spectest run
19:12 whiteknight I don't know if the last test hung, or if it was just extremely slow like all the other ones
19:24 Myhrlin joined #parrot
19:33 tcurtis All I'm getting on r46744 of parrot is 9 failures in t/spec/S05-mass/properties-derived.rakudo and 3 TODOs passed in t/spec/S32-str/uc.rakudo.
19:35 moritz the latter is random in time
19:36 moritz tcurtis: do you have icu installed, and detected by parrot?
19:36 moritz $ ./parrot_config has_icu
19:37 tcurtis moritz: yes.
19:37 moritz that's a weird one; it fails for some people, but works for most
19:37 moritz and I haven't figured out why yet
19:38 moritz tcurtis: which version of icu?
19:39 cotto_work left #parrot
19:40 tcurtis 4.4
19:41 cotto_work joined #parrot
19:41 moritz I have 3.6 here
19:41 moritz I'm going to collect some version numbers, and try to corellate
19:43 bluescreen joined #parrot
19:44 aukjan joined #parrot
19:56 nopaste "coke" at 192.168.1.3 pasted "rakudo spec test failures with recent parrot/rakudo" (8 lines) at http://nopaste.snit.ch/20577
19:57 * Coke writes "Hello World" in C#
19:58 Coke slick IDE.
20:00 tcurtis Coke: I think moritz will be wanting your icu version number.
20:00 moritz aye
20:10 dalek rakudo: 2d0fc2b | moritz++ |  (2 files):
20:10 dalek rakudo: implement :p/:pos in Regex.match
20:10 dalek rakudo: With help from colomon++
20:10 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/2​d0fc2b2729773a9987636cae65327480f54fd47
20:11 szabgab joined #parrot
20:13 gbacon_ joined #parrot
20:22 bluescreen joined #parrot
20:26 elmex joined #parrot
20:40 GeJ Good morning everyone.
20:41 cotto_work hio GeJ
20:53 GeJ heya cotto.
20:54 cotto_work Are all tests passing for you?
20:57 Coke $ icu-config --version
20:57 Coke 4.3.4
20:59 Whiteknight joined #parrot
21:01 dalek parrot: r46745 | Util++ | branches/pbc_to_exe_direct_output:
21:01 dalek parrot: Creating a branch for reducing the high memory usage of pbc_to_exe.
21:01 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46745/
21:21 hercynium joined #parrot
21:34 dalek parrot: r46746 | Util++ | branches/pbc_to_exe_direct_out​put/tools/dev/pbc_to_exe.pir:
21:34 dalek parrot: Move the open() of outfh to before the generation of the large PBC string.
21:34 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46746/
21:52 ruoso joined #parrot
22:06 dalek parrot: r46747 | Util++ | branches/pbc_to_exe_direct_out​put/tools/dev/pbc_to_exe.pir:
22:06 dalek parrot: Add outfh to the parameters of all the generate_* subs.
22:06 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46747/
22:14 bacek Good morning, humans.
22:15 GeJ Good Day bacek.
22:15 cotto_work On noes.  It's the bacek from the FUTURE.
22:16 bacek G'Day GeJ
22:17 bacek cotto_work, GeJ is from future too. But he using FreeBSD which is slightly suspicious about how bright it is...
22:19 darbelo clock?
22:19 purl darbelo: LAX: Mon 3:19pm PDT / CHI: Mon 5:19pm CDT / NYC: Mon 6:19pm EDT / LON: Mon 11:19pm BST / BER: Tue 12:19am CEST / IND: Tue 3:49am IST / TOK: Tue 7:19am JST / SYD: Tue 8:19am EST /
22:20 GeJ Being more future-ish than you, I can tell you that the future is cloudy with chances of showers. I doubt however that FreeBSD has anything to do with it.
22:20 darbelo bacek: are you sure GeJ's future is after your future?
22:20 * GeJ is on NCT
22:20 bacek GeJ, it's showers here already...
22:21 * darbelo definitely has a shower in his future.
22:21 GeJ bacek: damn, did I drive over 88 mph again?
22:22 darbelo Maybe the train pushed your car too fast.
22:22 bacek GeJ, yeah... You have to fix centuries selector on your car.
22:23 cotto_work bacek: did you mean to commit that to trunk?
22:23 dalek parrot: r46748 | bacek++ | trunk/compilers/pct/src/PAST/Node.pir:
22:23 dalek parrot: Add PAST::Block.multi accessor to store multi information.
22:23 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46748/
22:23 dalek parrot: r46749 | bacek++ | trunk (2 files):
22:23 dalek parrot: Fix decorators on Parrot_str_rep_compatible. Closes #1646.
22:23 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46749/
22:23 bacek cotto_work, bah... No...
22:25 GeJ From the code-tests-are-failing dept. : Could someone add an Id keyword line to examples/io/post.pir ?
22:26 bacek GeJ, done.
22:28 GeJ bacek: thank you.
22:28 dalek TT #1646 closed by bacek++: Failure due to missing external definition of inlined function ...
22:28 dalek TT #1646: http://trac.parrot.org/parrot/ticket/1646
22:31 pjcj joined #parrot
22:33 dalek parrot: r46750 | bacek++ | trunk/compilers/pct/src/PAST/Node.pir:
22:33 dalek parrot: Revert "Add PAST::Block.multi accessor to store multi information." This
22:33 dalek parrot: one doesn't belong to trunk.
22:33 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46750/
22:33 dalek parrot: r46751 | Util++ | branches/pbc_to_exe_direct_out​put/tools/dev/pbc_to_exe.pir:
22:33 dalek parrot: Change all "push codestring" to "print outfh".
22:33 purl dalek: that doesn't look right
22:33 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46751/
22:33 dalek parrot: r46752 | bacek++ | trunk/examples/io/post.pir:
22:33 dalek parrot: Add svn $Id$. GeJ++
22:33 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46752/
22:33 dalek parrot: r46753 | Util++ | branches/pbc_to_exe_direct_out​put/tools/dev/pbc_to_exe.pir:
22:33 dalek parrot: Remove codestring StringBuilder.
22:33 dalek parrot: review: http://trac.parrot.org/parrot/changeset/46753/
22:38 Whiteknight t/codingstd/svn_id.t ........... 1/1
22:38 Whiteknight #   Failed test 'Id keyword line exists'
22:38 Whiteknight #   at t/codingstd/svn_id.t line 82.
22:38 Whiteknight # No Id keyword line found in 1 files:
22:38 Whiteknight # /home/andrew/projects/parrot/examples/io/post.pir
22:38 Whiteknight who is working on post.pir?
22:39 davidfetter joined #parrot
22:39 bacek Whiteknight, r46752
22:41 cotto_work That's fperrad
22:44 Mokurai joined #parrot
22:48 Whiteknight anybody able to fix it? I have a few more tests to run
22:50 darbelo Whiteknight: It's already fixed. r46752
22:50 Whiteknight oh, thanks. I misunderstood
22:50 darbelo The bacek did it.
22:51 Whiteknight The bacek!
22:51 purl the bacek is THE MANIAC or some sort of magical coding robot or probably not mailto:pmichaud@pobox.com
22:57 Myhrlin joined #parrot
22:58 Psyche^ joined #parrot
23:15 bacek purl, bacek is also thebacek@gmail.com
23:15 purl okay, bacek.
23:15 bacek the bacek?
23:15 purl i guess the bacek is THE MANIAC or some sort of magical coding robot or probably not mailto:pmichaud@pobox.com or mailto:thebacek@gmail.com
23:15 bacek :)
23:23 GeJ as of r46752, fulltest pass on FreeBSD 8
23:26 tetragon joined #parrot
23:41 Whiteknight fulltest passes for me too on linux-amd64, and linux-x86
23:41 Whiteknight both good signs
23:46 Whiteknight everytime I'm about to do a release I spend a lot of time looking at pictures of Parrots
23:47 darbelo Name picking?
23:48 Whiteknight yeah
23:49 darbelo Finding a good quote is harder for me.
23:49 Whiteknight go to wikipedia, search "parrot", click on links for about 30 minutes until I'm looking at a completely unrelated page
23:49 Whiteknight then I panic because I wasted 30 minutes and haven't found a suitable name, and I make up something stupid
23:49 darbelo http://xkcd.com/214/
23:50 Whiteknight I've never done a quote with a release before. I'm particularly uncreative with that
23:50 darbelo I've found that most of the quoutes that I can think of translate poorly into enghlish.
23:51 Whiteknight most of the quotes I know contain some amount of vulgarity or profanity
23:52 darbelo That's the part that trasnlates poorly, yeah.
23:52 darbelo ;)
23:52 bacek_at_work Whiteknight, http://www.parrot-bird.com​/eastern-rosella-info.html
23:53 Whiteknight I might have found a decent quote, but it's border-line
23:56 Coke quotes are completely optional.
23:56 Coke as are release names.
23:57 Whiteknight Coke: balderdash!
23:57 Whiteknight it's a time-honored tradition

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

Parrot | source cross referenced