Camelia, the Perl 6 bug

IRC log for #parrot, 2011-01-16

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 dalek TT #1289 closed by jkeenan++: Review/Consolidate test infrastructure in t/codingstd and t/distro
00:09 dalek TT #1289: http://trac.parrot.org/parrot/ticket/1289
00:09 dalek TT #97 closed by jkeenan++: test installable config
00:09 dalek TT #97: http://trac.parrot.org/parrot/ticket/97
00:12 whiteknight I didn't realize until a few days ago that IMCC really had multiple entry points, and that they are duplicating a hell of a lot of code between them
00:13 whiteknight compilers/imcc/main.c:imcc_run is called by the frontend, but compilers/imcc/parser_util.c:imcc_compile_file is called by the PIR compreg
00:13 whiteknight the two do nearly identical things
00:14 whiteknight the later sets up a new context and updates some crap for the namespace and all
00:15 whiteknight very soon I think we are going to need to unify
00:19 mtk left #parrot
00:21 dukeleto whiteknight: sounds good
00:23 plobsing +1
00:24 plobsing why does 'plumage projects' run so slowly? it takes ~1 second per line of output.
00:25 plobsing that's not going to scale well
00:26 mtk joined #parrot
00:27 dukeleto plobsing: not sure, nqp-rx is still pretty slow
00:27 dukeleto plobsing: perhaps something can be cached
00:29 plobsing I suspect the json parsing, but I have no evidence to support that assumption yet.
00:29 plobsing that speed is unacceptable if we're going to try pushing stuff out of core and grow our ecosystem
00:30 dukeleto plobsing: yes, we will have to fix that soon
00:51 kid51 left #parrot
01:04 dalek nqp-rx/nom: 7852255 | jonathan++ | src/metamodel/multi_dispatch.c:
01:04 dalek nqp-rx/nom: Finish up getting type-based multiple dispatch to work.
01:04 dalek nqp-rx/nom: review: https://github.com/perl6/nqp-rx/commit/7852255ed5
01:19 mtk left #parrot
01:36 dalek nqp-rx/nom: b6b076d | jonathan++ | src/metamodel/multi_dispatch.c:
01:36 dalek nqp-rx/nom: Get multi-dispatch on definedness to work.
01:36 dalek nqp-rx/nom: review: https://github.com/perl6/nqp-rx/commit/b6b076d6af
01:37 dalek nqp-rx/nom: f4fe15c | jonathan++ | t/nqp/55-multi-method.t:
01:37 dalek nqp-rx/nom: Add multi-method tests from nqpclr, which now pass in nqp-rx/nom.
01:37 dalek nqp-rx/nom: review: https://github.com/perl6/nqp-rx/commit/f4fe15c2d7
01:53 dalek nqp-rx/nom: 78031bf | jonathan++ | src/stage0/ (3 files):
01:53 dalek nqp-rx/nom: Update bootstrap with multiple dispatch and signature parsing improvements.
01:53 dalek nqp-rx/nom: review: https://github.com/perl6/nqp-rx/commit/78031bfdc9
01:53 dalek nqp-rx/nom: b0fa8db | jonathan++ | src/metamodel/how/NQPMu.pm:
01:53 dalek nqp-rx/nom: Make ACCEPTS a multi and correct the signature of it.
01:53 dalek nqp-rx/nom: review: https://github.com/perl6/nqp-rx/commit/b0fa8db53f
01:54 lucian joined #parrot
02:04 lucian left #parrot
02:15 rurban left #parrot
02:23 chromatic left #parrot
02:26 whiteknight tests in t/src/embed/* are not being run in coretest
02:27 rurban joined #parrot
02:27 Coke joined #parrot
03:03 dalek parrot: 3532e07 | Whiteknight++ | lib/Parrot/Harness/DefaultTests.pm:
03:03 dalek parrot: add t/src/embed/*.t tests to the coretest run target. They weren't being run before
03:03 dalek parrot: review: https://github.com/parrot/parrot/commit/3532e07a39
03:03 dalek parrot: 9d084b0 | Whiteknight++ | / (2 files):
03:03 dalek parrot: Merge branch 'master' of github.com:parrot/parrot
03:03 dalek parrot: review: https://github.com/parrot/parrot/commit/9d084b029e
03:30 whiteknight blah. The imcc_cleanups branch removes a file from the build. Every time I switch branches I need to configure again tor regen the makefile
03:31 whiteknight let me tell you how old this has gotten
03:33 cotto whiteknight, I've resorted to using a separate clone and screen sessions for branches that do that.
03:33 whiteknight yeah, I'm about 90% of the way to making that same decision
03:35 cotto If you know what decision you'll make, you can save some time by making it now instead of later.
03:36 cotto I'm chatting with the osuosl folks about fixing parrot.org.  The drupal site is back.  trac seems to be trickier.
03:36 whiteknight what's broken about it?
03:37 cotto other than it being unresponsive?
03:38 whiteknight no, I mean what's the root cause of the brokenness?
03:38 whiteknight bad widget? Broken doodad?
03:38 cotto They say they're going to grow the disk.  That seems to imply something about space.
03:40 cotto it's back
03:41 whiteknight I wonder if all those smolder reports filled us up?
03:44 jsut left #parrot
03:44 dalek TT #1970 created by cotto++: test
03:45 dalek TT #1970: http://trac.parrot.org/parrot/ticket/1970
03:45 dalek TT #1970 closed by cotto++: test
03:45 dalek TT #1970: http://trac.parrot.org/parrot/ticket/1970
03:45 cotto whiteknight, when did you notice trac being broken?
03:46 whiteknight kid51 noticed this morning. I haven't really been looking at it today
03:49 cotto It looks like it wasn't completely broken for very long.  The github hook caught all but the most recent batch of commits.
03:49 cotto anyway, now's a good time to file that Eval deprecation
03:51 whiteknight and the PIR compreg?
03:53 cotto Is that #1967 Deprecate IMCC compreg ?
03:56 sorear don't we need to provide an alternate way to compile PIR?
03:56 sorear we could deprecate "all uses of the PIR compreg other than VTABLE_invoke", then ensure that the replacement class has a vtable override for at least one dep cycle
03:56 whiteknight cotto: yes
03:57 cotto whiteknight, ok.  What are you thoughts on pmichaud's suggestion?
03:57 whiteknight sorear: yes, we are going to have a replacement
03:57 whiteknight cotto: I haven't see his suggestion. Is it on the ticket?
03:57 cotto yes
03:57 cotto basically, go with pdd31, which defines a standard interface for compilers
03:58 whiteknight cotto: yes, that was basically my idea. I didn't list all the specifics in the ticket
03:58 whiteknight but yes, we are going to make a PDD31-alike PMC type for the PIR compiler object
03:58 whiteknight and use that to replace the current NCI PMC that it uses
03:59 cotto whiteknight: by all means, go for it
03:59 whiteknight awesome. I'm running some tests right now, I'll put in the deprecation notices shortly thereafter
03:59 cotto those pdds aren't going to become true of nobody follows them
04:01 whiteknight I do want to review that PDD in particular and make sure it is up to snuff
04:03 cotto atrodo, ping
04:07 cotto msg atrodo Have you given any thought to what a struct would look like in your Lorito prototype?
04:07 aloha OK. I'll deliver the message.
04:13 whiteknight is there an option for the c_output_is test functions where it won't delete the binaries after the test is over?
04:14 whiteknight I have a problem with a test, but I can't debug it because the binaries are being cleaned up by prove
04:15 cotto whiteknight, that doesn't sound like a bad idea.  +1 to add such an option if you document it.
04:15 whiteknight blah
04:16 whiteknight and for the life of me I can't figure out how to compile these files myself
04:16 whiteknight When I compile with "clang $(./parrot_config libparrot_linkflags) $(./parrot_config libparrot_soname) $(./parrot_config linkflags)" it says it can't find libparrot
04:16 whiteknight so that frustrates me
04:17 whiteknight fuggedaboutit. I'm going to bed
04:17 cotto 'night
04:17 whiteknight goodnight
04:17 whiteknight left #parrot
04:17 Coke (easiest way to deprecate compreg "PIR" and replace it with a new compreg object is to call the new one something other than PIR.
04:18 Coke then you avoid having to dance.
04:27 cotto atrodo++
04:34 KatrinaTheLamia left #parrot
04:38 Kristaba left #parrot
04:54 KatrinaTheLamia joined #parrot
05:45 fbrito joined #parrot
05:47 rurban_ joined #parrot
05:50 rurban left #parrot
05:50 rurban_ is now known as rurban
06:20 dukeleto ~~
06:21 dukeleto Coke: perhaps we should call it PIR++ ...
06:23 plobsing call it what it is, IMCC
06:23 cotto +1 to imcc
06:28 dalek parrot: 9dad052 | dukeleto++ | NEWS:
06:28 dalek parrot: [doc] Add some meat to NEWS
06:28 dalek parrot: review: https://github.com/parrot/parrot/commit/9dad0521fa
06:29 * cotto remembers that updating NEWS was on his todo list for today.
06:29 dukeleto cotto: yeah, a lot has happened during GCI that is not in NEWS
06:31 contingencyplan left #parrot
06:40 * dukeleto starts up the debian-sparc64 smoker, now that smolder seems to be back up
06:51 cotto dukeleto, how would you summarize the testing you're doing with the gcc compile farm?
06:51 cotto that should go in NEWS
07:02 dukeleto cotto: i will add something
07:02 dukeleto cotto: http://smolder.parrot.org/app​/projects/report_details/2988 <-- first smolder report from the sparc smoker in the gcc compile farm
07:04 cotto dukeleto, ok
07:07 cotto dukeleto, is there an easy way to get a list of everyone who's contributed to Parrot since the last release, including people who had branches merged in from external repos?
07:09 dukeleto cotto: yeah
07:09 dukeleto cotto: https://github.com/visionmedia/git-extras
07:09 dukeleto cotto: and use the "git summary" tool
07:09 cotto dukeleto++
07:18 ilbot2 joined #parrot
07:18 Topic for #parrot is now Parrot 2.11.0 Released | http://parrot.org | Log: irclog.perlgeek.de/parrot/today | Goals: write GettingStartedWithPlumage, review html documentation, test HLLs, review deprecations | no major changes after Tuesday, only critical fixes and documentation updates after Friday
07:18 d4l3k_ is now known as dalek
07:19 theory joined #parrot
07:20 Infinoid joined #parrot
07:24 preflex joined #parrot
07:24 slavorg left #parrot
07:24 slavorg joined #parrot
07:36 cotto bluescreen is Mariano Wahlmann, right?
07:38 cotto verified
07:39 dukeleto cotto: CREDITS should have most of the info you want, and if it doesn't, please add it there
07:47 cotto dukeleto, right.  In this case it didn't.  The #ffmpeg folks were helpful though.
07:51 dukeleto Rubinius was ported to HaikuOS and NetBSD in the last few weeks.
07:53 fbrito left #parrot
07:58 cotto dukeleto, are you adding anything more to NEWS?
07:59 cotto git seems lta at merging NEWS changes
08:02 cotto seen nwellnhof
08:02 aloha nwellnhof was last seen in #parrot 14 hours 9 mins ago joining the channel.
08:08 fperrad joined #parrot
08:33 cotto msg kid51 Could you set up a poll for the next PDS time?  It's coming up pretty quickly.
08:33 aloha OK. I'll deliver the message.
08:52 Patterner left #parrot
08:52 Psyche^ joined #parrot
08:52 Psyche^ is now known as Patterner
09:07 fperrad left #parrot
09:09 fperrad joined #parrot
09:13 dalek parrot: 86dce60 | cotto++ | NEWS:
09:13 dalek parrot: add a couple misc news items
09:13 dalek parrot: review: https://github.com/parrot/parrot/commit/86dce602aa
09:13 dalek parrot: 778fb90 | cotto++ | CREDITS:
09:13 dalek parrot: add gci students to CREDITS
09:13 dalek parrot: review: https://github.com/parrot/parrot/commit/778fb901a7
09:13 dalek parrot: e8dca43 | cotto++ | NEWS:
09:13 dalek parrot: add more NEWS items, tweak formatting some more
09:13 dalek parrot: review: https://github.com/parrot/parrot/commit/e8dca4384c
10:01 tadzik joined #parrot
10:09 theory left #parrot
12:01 tadzik https://github.com/parrot/par​rot/blob/e8dca4384c/NEWS#L15 -- shouldn't that be "now lives"?
12:11 mberends left #parrot
12:19 lucian joined #parrot
12:21 lucian left #parrot
12:41 simcop2387 left #parrot
13:00 dalek parrot: ea7ad98 | bacek++ | NEWS:
13:00 dalek parrot: Fix typo. tadzik++
13:00 dalek parrot: review: https://github.com/parrot/parrot/commit/ea7ad981d1
13:01 simcop2387 joined #parrot
13:05 whiteknight joined #parrot
13:06 dalek nqp-rx: e657b28 | bacek++ | / (2 files):
13:06 dalek nqp-rx: Add default value for  into RPA.join to follow Perl6 setting closely
13:06 dalek nqp-rx: review: https://github.com/perl6/nqp-rx/commit/e657b280d6
13:11 dalek nqp-rx: 2613f07 | bacek++ | t/setting/01-resizablepmcarray.t:
13:11 dalek nqp-rx: Add test for RPA.map
13:11 dalek nqp-rx: review: https://github.com/perl6/nqp-rx/commit/2613f07cb2
13:20 whiteknight good morning, #parrot
13:24 bacek whiteknight, aloha
13:24 whiteknight good morning, bacek. How are you doing today?
13:25 bacek whiteknight, very sleepy of first 25 minutes of today. It's Monday here already :)
13:26 whiteknight yeah, I've been up since 3:30 AM, so I'm very sleepy
13:26 whiteknight but at least it isn't monday
13:27 bacek whiteknight, welcome to the future :)
13:43 Kristaba joined #parrot
13:47 rurban_ joined #parrot
13:47 dalek tracwiki: v1 | whiteknight++ | WhiteknightTasklist
13:47 dalek tracwiki: +a quick tasklist for myself. All the cool kids are doing it.
13:47 dalek tracwiki: http://trac.parrot.org/parrot/wiki/White​knightTasklist?version=1&amp;action=diff
13:50 rurban left #parrot
13:50 rurban_ is now known as rurban
13:59 kid51 joined #parrot
14:05 dalek parrot/nqp_pct: 9ab0d4b | bacek++ | compilers/pct/src/POST/Compiler.pm:
14:05 dalek parrot/nqp_pct: Rewrite POST::Compiler.pir(POST::Op) to nqp.
14:05 dalek parrot/nqp_pct: review: https://github.com/parrot/parrot/commit/9ab0d4b5f4
14:05 dalek parrot/nqp_pct: b67fb0b | bacek++ | compilers/pct/src/POST/ (13 files):
14:05 dalek parrot/nqp_pct: Rebootstrap PCT
14:05 dalek parrot/nqp_pct: review: https://github.com/parrot/parrot/commit/b67fb0b9ac
14:05 dalek parrot/nqp_pct: a4db63e | bacek++ | compilers/pct/src/POST/Compiler.pm:
14:05 dalek parrot/nqp_pct: Rewrite POST::Compiler.hll_pir to nqp.
14:05 dalek parrot/nqp_pct: review: https://github.com/parrot/parrot/commit/a4db63e29b
14:05 dalek parrot/nqp_pct: d804e66 | bacek++ | compilers/pct/src/POST/ (13 files):
14:05 dalek parrot/nqp_pct: Rebootstrap PCT
14:05 dalek parrot/nqp_pct: review: https://github.com/parrot/parrot/commit/d804e6679e
14:06 dalek parrot/nqp_pct: 851b54b | bacek++ | / (65 files):
14:06 dalek parrot/nqp_pct: Merge branch 'master' into nqp_pct
14:06 dalek parrot/nqp_pct: review: https://github.com/parrot/parrot/commit/851b54bec5
14:11 ambs joined #parrot
14:15 dalek parrot/imcc_cleanups: ca855f2 | Whiteknight++ | / (4 files):
14:15 dalek parrot/imcc_cleanups: implement the test for Parrot_api_serialize_bytecode_pmc. Fix some ARG modifiers to allow nulls, since we have logic in there to handle nulls
14:15 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/ca855f2ca1
14:22 Patterner left #parrot
14:48 Patterner joined #parrot
14:51 PerlJam left #parrot
14:53 dalek tracwiki: jimmy++ | continuation-closure-passing.ps.gz attached to LoritoCPSTutorial
14:53 dalek tracwiki: implementing a CPS code generator for ML
14:53 dalek tracwiki: http://trac.parrot.org/parrot/attachme​nt/wiki/LoritoCPSTutorial/continuation​-closure-passing.ps.gz&amp;action=diff
14:53 dalek tracwiki: jimmy++ | stal-tic.pdf attached to LoritoCPSTutorial
14:53 dalek tracwiki: <span class="icon">�</span>http://www.cs.cornell.edu/talc/papers/stal-tic.pdf</a>
14:53 dalek tracwiki: http://trac.parrot.org/parrot/attachment/wiki/​LoritoCPSTutorial/stal-tic.pdf&amp;action=diff
14:53 PerlJam joined #parrot
14:53 PerlJam left #parrot
14:54 mberends joined #parrot
14:54 PerlJam joined #parrot
15:03 dalek nqp-rx/nom: c759804 | jonathan++ | src/NQP/Actions.pm:
15:03 dalek nqp-rx/nom: Get basic case of multi-subs in place. Just handles lexical case and multis in nested scopes are probably busted still.
15:03 dalek nqp-rx/nom: review: https://github.com/perl6/nqp-rx/commit/c759804e85
15:04 dalek nqp-rx/nom: 46a10cd | jonathan++ | t/nqp/51-multi.t:
15:04 dalek nqp-rx/nom: Replace old multi-dispatch tests. Multi-method dispatch is now in a separate test file, added yesterday. 51-multi.t will now just test multi subs. Needs more tests, but does cover arity based dispatch and that the proto functions as it should - being called first then delegating to the multi-dispatcher.
15:04 dalek nqp-rx/nom: review: https://github.com/perl6/nqp-rx/commit/46a10cde6e
15:25 dalek blizkost: bca8b6c | qiuhw++ | Configure.pl:
15:25 dalek blizkost: The warning about "Perl is not built as a dynamic library" was actually never reported, fix it.
15:25 dalek blizkost:
15:25 dalek blizkost: And the newest parrot_config don't have a 'cc_hasjit' key if it's empty.
15:25 dalek blizkost: review: https://github.com/jnthn/b​lizkost/commit/bca8b6c0dc
15:42 dalek parrot: 48da21a | nwellnhof++ | t/src/embed/strings.t:
15:42 dalek parrot: [t] Fix t/src/embed/strings.t with MSVC
15:42 dalek parrot: review: https://github.com/parrot/parrot/commit/48da21aff0
15:42 dalek parrot/nqp_pct: 1f02017 | jimmy++ | compilers/pct/src/POST/Node.pm:
15:42 dalek parrot/nqp_pct: added missing =cut
15:42 dalek parrot/nqp_pct: review: https://github.com/parrot/parrot/commit/1f020173a3
16:12 mikehh I am getting a failure in t/pmc/float.t (g++ --optimize) - Failed test:  102 - comparison ops: cmp_p_p: equality (passes g++ without --optimize and gcc - Ubuntu 10.10 i386)
16:26 cogno joined #parrot
16:44 theory joined #parrot
16:45 kid51 mikehh: Unable to reproduce that on an older Debian, with all other arguments same as yours
16:47 mikehh kid51: looking at it - it works with g++ (no --optimize) and gcc with or without --optimize
16:49 mikehh I am using gcc/g++ 4.5.1
16:54 dalek parrot/imcc_cleanups: bd67e17 | Whiteknight++ | / (2 files):
16:54 dalek parrot/imcc_cleanups: add temporary .c files to .gitignore. Also, unlike temporary files used by tests
16:54 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/bd67e176b1
16:54 dalek parrot/imcc_cleanups: 255a8d8 | Whiteknight++ | / (5 files):
16:54 dalek parrot/imcc_cleanups: move -G commandline argument parsing from IMCC into the frontend
16:54 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/255a8d8f91
16:54 dalek parrot/imcc_cleanups: 9c5ae86 | Whiteknight++ | compilers/imcc/ (2 files):
16:54 dalek parrot/imcc_cleanups: remove -w argument processing from IMCC. It's handled in the frontend
16:54 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/9c5ae86565
16:54 dalek parrot/imcc_cleanups: 020e292 | Whiteknight++ | compilers/imcc/main.c:
16:54 dalek parrot/imcc_cleanups: do not Parrot_destroy in IMCC
16:54 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/020e29232c
16:54 dalek parrot/imcc_cleanups: 2442dea | Whiteknight++ | compilers/imcc/main.c:
16:54 dalek parrot/imcc_cleanups: small cleanup
16:54 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/2442dea1dd
16:54 dalek parrot/imcc_cleanups: b523125 | Whiteknight++ | compilers/imcc/ (2 files):
16:54 dalek parrot/imcc_cleanups: remove two dead functions
16:54 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/b523125f30
16:54 dalek parrot/imcc_cleanups: b606d3d | Whiteknight++ | compilers/imcc/ (4 files):
16:54 dalek parrot/imcc_cleanups: Add in some notes to tell where various IMCC call-in functions are being used. Consolidate the register_compilers function
16:54 dalek parrot/imcc_cleanups: review: https://github.com/parrot/parrot/commit/b606d3defa
16:58 kid51 me: g++ --versiong++ (Debian 4.3.2-1.1) 4.3.2
16:58 dalek parrot: 147e2d8 | Whiteknight++ | DEPRECATED.pod:
16:58 dalek parrot: add two deprecation notices
16:58 dalek parrot: review: https://github.com/parrot/parrot/commit/147e2d8fcc
16:59 ambs left #parrot
16:59 ambs joined #parrot
17:41 contingencyplan joined #parrot
17:42 kid51 clock
17:42 kid51 time
17:57 whiteknight aloha: clock
17:57 whiteknight aloha: clock?
17:57 aloha whiteknight: LAX: Sun, 09:57 PST / CHI: Sun, 11:57 CST / NYC: Sun, 12:57 EST / UTC: Sun, 17:57 UTC / LON: Sun, 17:57 GMT / BER: Sun, 18:57 CET / TOK: Mon, 02:57 JST / SYD: Mon, 04:57 EST
17:59 kid51 left #parrot
18:15 mikehh left #parrot
18:24 whiteknight left #parrot
18:27 cotto good morning
18:27 ambs left #parrot
18:27 ambs joined #parrot
18:30 plobsing msg whiteknight: if there is not going to be a PASM compreg, perhaps we should deprecate PASM altogether.
18:30 aloha OK. I'll deliver the message.
18:35 nwellnhof joined #parrot
18:38 cotto plobsing, that'll mean translating a lot of pasm tests.
18:38 plobsing cotto: I've done that a number of times already. Most tests in PASM don't depend on their language of implementation.
18:39 plobsing and those that do usually have duplicate PIR tests to ensure constructs work in both languages
18:41 plobsing but my main argument is that, if we are removing the visibility of PASM to hosted languages, we should take it away altogether to keep everyone on equal footing.
18:43 dalek tracwiki: v33 | cotto++ | CottoTasklist
18:43 dalek tracwiki: http://trac.parrot.org/parrot/wiki/Cot​toTasklist?version=33&amp;action=diff
18:56 cotto plobsing, M0 will need some kind of struct-like functionality.  Do you think it'd be sane to base M0 structs on bytes?
18:59 * cotto will think
19:09 mikehh joined #parrot
19:12 Kristaba left #parrot
19:13 Yuki`N joined #parrot
19:15 workbench joined #parrot
19:23 Kristaba joined #parrot
19:30 plobsing cotto: I think you should consider what kinds of elements M0 needs in those structs. I suspect it will consist of only a small set of types. M0 shouldn't need a fullblown ManagedStruct.
19:30 mikehh t/pmc/float.t - Failed test:  102 with g++ configured with --optimize (passes without and with gcc) on both Ubuntu 10.04 i386 and Ubuntu 10.10 i386 (g++ 4.4.3 and 4.5.1)
19:32 dalek TT #1971 created by rgrjr++: Oplib paths are truncated in PBC files
19:32 dalek TT #1971: http://trac.parrot.org/parrot/ticket/1971
19:33 cogno left #parrot
19:33 cogno joined #parrot
19:37 simcop2387 left #parrot
19:39 kid51 joined #parrot
19:41 mikehh left #parrot
19:43 cogno left #parrot
19:43 cogno joined #parrot
19:48 dalek TT #1972 created by rgrjr++: Can't compile with TRACE_PACKFILE set
19:48 dalek TT #1972: http://trac.parrot.org/parrot/ticket/1972
19:48 dalek TT #1973 created by rgrjr++: pbc_merge fails silently
19:48 dalek TT #1973: http://trac.parrot.org/parrot/ticket/1973
19:52 dukeleto ~~
19:57 Coke left #parrot
19:57 simcop2387 joined #parrot
19:57 ucz joined #parrot
20:02 dukeleto http://parrots4parties.com/
20:03 simcop2387 left #parrot
20:03 simcop2387 joined #parrot
20:05 cogno left #parrot
20:05 cogno joined #parrot
20:10 simcop2387 left #parrot
20:10 simcop2387 joined #parrot
20:12 mikehh joined #parrot
20:12 Yuki`N left #parrot
20:21 dalek TT #1974 created by rgrjr++: Fixing GMP check on 32-bit system exposes test failure
20:21 dalek TT #1974: http://trac.parrot.org/parrot/ticket/1974
20:22 dukeleto http://trac.parrot.org/par​rot/wiki/LoritoCPSTutorial looks useful
20:26 perlite_ joined #parrot
20:26 kid51 left #parrot
20:29 perlite left #parrot
20:29 perlite_ is now known as perlite
20:30 cogno left #parrot
20:32 dalek parrot: 3c1e33b | dukeleto++ | t/dynoplibs/trans.t:
20:32 dalek parrot: [t][TT #1974] Fix GMP check and test skips in t/dynoplibs/trans.t, rgrjr++
20:32 dalek parrot: review: https://github.com/parrot/parrot/commit/3c1e33b113
20:37 ucz left #parrot
20:37 dalek TT #1974 closed by dukeleto++: Fixing GMP check on 32-bit system exposes test failure
20:37 dalek TT #1974: http://trac.parrot.org/parrot/ticket/1974
20:50 dukeleto the docs for test_more.pir were never updated to describe the new todo => "blarg" feature
20:52 dalek parrot: 7c15738 | dukeleto++ | NEWS:
20:52 dalek parrot: [t] More NEWS updates
20:52 dalek parrot: review: https://github.com/parrot/parrot/commit/7c15738da7
21:13 dukeleto This is still a good read: http://www.oreillynet.com/onlamp/blog/200​8/01/what_the_perl_6_and_parrot_hac.html
21:15 mikehh dukeleto: your last commit missewd a branch around the skip - runs the tests and then tries to skip the tests
21:15 mikehh missed
21:19 mikehh fixed, shall I commit it?
21:24 mberends left #parrot
21:26 dalek parrot: c2077d1 | mikehh++ | t/dynoplibs/trans.t:
21:26 dalek parrot: branch around skipped tests so they don't get skipped if they have been already tested
21:26 dalek parrot: review: https://github.com/parrot/parrot/commit/c2077d10d3
21:27 dukeleto mikehh: thanks for fixing that
21:27 dukeleto mikehh++
21:28 jan joined #parrot
21:38 fperrad left #parrot
21:45 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#3141) fulltest) at 2_11_0-1163-gc2077d1 - Kubuntu 10.10 amd64 (g++-4.5 with --optimize)
21:46 dukeleto plobsing: any reason you assigned the Oplib TT to me?
21:47 plobsing nope. I suspect you get auto-assigned tasks tagged 'tools'
21:47 rurban_ joined #parrot
21:48 plobsing I suspect cotto or bacek would be better people for the task, seeing as there is an ops2c bug.
21:50 rurban left #parrot
21:50 rurban_ is now known as rurban
21:50 dukeleto plobsing: yeah, i don't know that I want to be auto-assigned tools tasks
21:54 plobsing yes, you appear as the owner of the 'tools' component in the Trac Admin page. Not sure how to change that.
22:19 bacek_at_work ~~
22:20 bacek_at_work plobsing, which bug?
22:29 cotto ohai
22:32 cotto dukeleto, did you get a test from rgrjr?
22:34 cotto ah.  I see he filed a bug.  It's nice to see something from him.
22:36 ambs left #parrot
22:40 dukeleto cotto: he filed 4 trac tickets this morning, iirc
22:43 nwellnhof left #parrot
22:48 dukeleto cotto: http://6guts.wordpress.com/2011/01/16/nqp-and​-6model-big-steps-forward-on-parrot-and-jvm/
22:48 cotto bacek_at_work, #1971
22:49 cotto dukeleto, thanks.  I'll read that as soon as I'm caught up with the bug queue
22:49 bacek_at_work cotto, it's not related to "tools", "ops2c" or whatever. It's some stupidity in loading libraries.
22:49 dukeleto 6model Overview: https://github.com/jnthn/6mo​del/blob/master/overview.pod
22:52 spreadsheet joined #parrot
22:52 spreadsheet Hello
22:53 spreadsheet So, if my understanding is correct, with other VMs like Java, you will need an additional runtime to run a program written in another languages besides Java, even though it's compiled JVM bytecode
22:53 spreadsheet With Parrot, will I need additional runtimes to run say, a Python program?
22:53 spreadsheet If I do, then is the runtime smaller than a regular Python interpreter?
23:01 dukeleto spreadsheet: parrot has a runtime, and all languages that run on parrot use that runtime
23:01 spreadsheet Ahh ok
23:01 spreadsheet dukeleto: if I wanted to have an interactive prompt, would that be possible?
23:01 dukeleto spreadsheet: yep
23:02 dukeleto spreadsheet: rakudo perl 6, (perl 6 on parrot), has an interactive prompt when you run it with no arguments
23:02 spreadsheet ok
23:02 dukeleto spreadsheet: also, welcome to #parrot
23:02 spreadsheet Thanks
23:02 dukeleto spreadsheet: how did you hear about Parrot?
23:03 spreadsheet dukeleto: I heard about it because I heard about Perl 6
23:03 spreadsheet And rakudo perl
23:03 dukeleto spreadsheet: yeah, that is what happened to me too :)
23:03 * cotto didn't know that
23:03 dukeleto spreadsheet: and to answer your question about bytecode, Parrot has it's own kind of bytecode, similar to the JVM's
23:04 spreadsheet So does it have as good scaling support, that is, for big websites, as Python or JVM?
23:04 spreadsheet Well Python, JVM is more mature
23:04 dukeleto cotto: i started out by getting a commit bit to the perl 6 spec test suite, and it only took me less than a day to run into a Parrot bug. The rest is history...
23:04 dukeleto spreadsheet: you should think of Parrot as mostly a set of compilers and a runtime for languages to sit on
23:05 dukeleto spreadsheet: you don't exactly write a website in Parrot or JVM
23:05 spreadsheet Well I know that
23:05 dukeleto spreadsheet: what do you mean by "scaling support" ?
23:05 spreadsheet I mean, say I have a Python website. It takes a lot of traffic. Then I have a Rakudo Perl site... would it be the same stability?
23:05 vmspb joined #parrot
23:06 dukeleto spreadsheet: define stability
23:06 dukeleto spreadsheet: currently Rakudo Perl 6 is still getting huge performance improvements, so I wouldn't run a production website with it yet. But that is me.
23:07 spreadsheet not crashing because of memory problems, having memory leaks, performance
23:07 dukeleto spreadsheet: see above
23:07 spreadsheet Ok
23:07 dukeleto spreadsheet: both parrot and rakudo are getting lots of performance optimizations as we speak
23:07 dukeleto spreadsheet: there is a new branch that will be merged post parrot 3.0 (due on tuesday) that makes Rakudo a lot faster and use less memory
23:08 spreadsheet dukeleto: if I had a GTK+ or Qt wrapper for Parrot, would I be able to use it with any language, like with the JVM? JVM libraries are able to be used by any language
23:08 dukeleto spreadsheet: that is the goal. I would say "yes, in theory", but there could be some rough edges still
23:09 dukeleto spreadsheet: we have an opengl wrapper and postgresql wrapper in parrot, but they might need some love
23:09 spreadsheet So, in the future, does the Parrot community see Parrot as being the main way to run dynamic languages on Unix-like OS?
23:10 dukeleto spreadsheet: we would like to think so. But there are lots of other VMs for dynamic languages, such as Rubinius and NekoVM
23:10 dukeleto spreadsheet: but Parrot is the only dynamic VM that wants to support all dynamic languages
23:11 dukeleto spreadsheet: to my knowledge, everyone else makes some restrictions on how you implement the dynamic language. With Rubinius, you must use Ruby, for example.
23:11 dukeleto spreadsheet: with Parrot, we try to make everything pluggable, so it is up to the language author to decide
23:11 dukeleto spreadsheet: we give some tools, but any tools can be used, in theory :)
23:11 cotto I'd like to see a generation of dynamic language implementations that target Parrot but are mostly self-hosted.
23:12 dukeleto cotto: yes
23:12 cotto Some language cultures aren't going to want to get away from their language of choice, and who's to say that they're wrong.
23:13 dukeleto cotto: the perl 6 peeps sure do like to dogfood their compilers :)
23:14 Topic for #parrot is now Parrot 2.11.0 Released | http://parrot.org | Log: irclog.perlgeek.de/parrot/today | Goals: write GettingStartedWithPlumage, review html documentation, test HLLs, review deprecations | only critical fixes and documentation updates until after 3.0.0
23:15 Coke joined #parrot
23:16 spreadsheet left #parrot
23:38 Kristaba left #parrot
23:41 allison joined #parrot
23:42 PacoLinux left #parrot
23:46 KatrinaTheLamia left #parrot
23:51 allison left #parrot
23:56 KatrinaTheLamia joined #parrot

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

Parrot | source cross referenced