Parrot 1.9.0 "Blue-fronted Amazon" released! | parrot.org | Priorities: Add deprecations for 2.0; test platforms; check with HLL implementors | Roadmap: icanhaz.com/parrotroadmap | Latest modified TT's: icanhaz.com/parrotbugs
Set by moderator on 9 January 2010.
00:11 tetragon joined
kid51 bacek_at_work: pong 00:39
00:47 davidfetter joined
cotto_work This looks like a valuable data point when considering how we want to use LLVM: renesd.blogspot.com/2010/01/unladen...-work.html 00:50
00:53 abqar joined 00:57 cconstantine joined
kid51 biaf 00:57
Whiteknight cotto_work: excellent link 01:02
those are the kinds of results I would expect, some work loads will work with JIT and some wont
01:09 abqar_ joined 01:11 mtk joined, mtk left
cottoo yeah, good find cotto_work 01:13
;)
davidfetter nom nom nom 01:18
purl I eat your head!
01:24 cognominal joined 01:25 abqar joined
dalek rrot: r43421 | chromatic++ | branches/tt389_fix/t (6 files):
[t] Fixed tests which relied on the deprecated behavior of :methods accessible
 01:28
01:36 chromatic joined 01:42 cconstantine joined
GeJ chromatic: is tt389_fix a WIP that should be left alone for now or do you accept build failure reports? 01:57
chromatic It's a WIP. I'm trying to fix PGE right now. 01:58
GeJ Okay then. 01:59
While I'm on lunch break, any branch that would like to get a freebsd smoke report? 02:01
chromatic pmc_freeze_cleanup is close to merging. 02:02
GeJ checkout in progress. Thanks. 02:04
02:04 ZeroForce joined
chromatic seen pmichaud 02:19
purl pmichaud was last seen on #parrot 6 days, 9 hours, 25 minutes and 46 seconds ago, saying: mikehh: what sort of "something happens" are you looking for? [Jan 5 16:51:26 2010]
chromatic I suppose I could also: 02:21
pmichaud, ping
02:31 Andy joined 02:52 japhb joined 02:57 nopaste joined 02:58 plobsing joined 03:27 nopaste joined 03:42 nopaste joined
chromatic Hm. NameSpace#export_to() 03:52
04:02 nopaste joined 04:51 JimmyZ joined
chromatic msg pmichaud I've almost fixed TT #389 (no :method in namespace), but I have PGE/TGE problems; I could use some help. 04:52
purl Message for pmichaud stored.
JimmyZ is waiting someone who could take a look at his ticket. 04:58
dalek rrot: r43422 | chromatic++ | branches/tt389_fix/runtime/parrot/library/PGE/Perl6Grammar.pir:
[PGE] Made PGE build after fixing TT #389. Instead of looking up a method by

object as its first argument, use normal method dispatch and avoid the manual lookup.		 05:01
05:15 aardvark joined 05:40 cognominal joined 05:56 preflex joined 06:13 bacek joined 07:04 uniejo joined
dalek rdinal: 9c99a89 | fperrad++ | setup.pir:
- update the URL of Smolder
 07:31
rdinal: dab2076 | fperrad++ | plumage/cardinal.json:
Plumage description

  - move to ports/plumage
07:41 aninhumer joined 07:44 iblechbot joined
dalek rrot-plumage: 4eed381 | japhb++ | metadata/ (2 files):
[METADATA] New metadata for cardinal and forth languages, courtesy of fperrad++
 07:47
treed odd timing 07:50
japhb ur?
purl japhb: q[merlyn]'s Unix Review columns, at www.stonehenge.com/merlyn/UnixReview/ or just a hop skip and a jump away from babylonia or short for URog or a lame way to say you're or Updating a public calendar automatically (Jul 07); Export behavior, not data (May 07); The Replacements (Mar 07); Show me your references (Jan 07); Formatting reports with Template Toolkit (Nov 06) or rong
japhb treed what was odd? 07:51
aninhumer Would it be possible to have the docs/ directory hosted in a wiki, and merge with svn occasionally? It seems quite complicated to submit edits at the moment 07:58
chromatic aninhumer, would you find it easier to edit and submit pull requests on a GitHub mirror? 08:00
aninhumer That would probably be as easy yes 08:02
chromatic, Does one exist that I've missed? 08:03
chromatic github.com/leto/parrot 08:04
aninhumer Ah yes, it was the first result... 08:06
Nonetheless, perhaps that should be added to the Developers page on the website? 08:07
chromatic dukeleto, ping
Let's see what he thinks about that; if he's willing to merge doc changes back in, that sounds reasonable. 08:08
treed japhb: The timing. 08:13
That I pushed his changes to cardinal and then you pushed his changes to plumage.
japhb ah! *chuckle* I didn't realize you had pushed his stuff. I thought that was him. 08:14
(er, doing the pushing, I mean) 08:15
08:43 barney joined
treed Ah, no. 08:44
He has his own repo and does pull requests to me.
09:07 payload joined 09:10 he joined 09:35 AndyA joined 09:56 payload joined
dalek tracwiki: v2 | cotto++ | TestingProfiling 10:10
tracwiki: more thoughts on dealing with profile size
tracwiki: trac.parrot.org/parrot/wiki/Testing...ction=diff
tracwiki: v3 | cotto++ | TestingProfiling
tracwiki: trac.parrot.org/parrot/wiki/Testing...ction=diff
11:08 cconstantine joined 11:18 mikehh_ joined 11:22 mikehh joined 11:23 allison joined 11:45 cconstantine joined
dalek rrot: r43423 | mikehh++ | branches/one_make (3 files):
set svn properties
 11:50
aninhumer I've made a github fork here: github.com/aninhumer/parrot 11:53
I've made a few doc fixes, and some fleshing out of subs.pod that was requested inline
Subject to dukeleto's opinion on merging doc changes
(I suppose this is essentially a pull request) 11:54
11:55 bluescreen joined
mikehh there are a bunch of post-config test failures in one_make branch related to pmc2cutils 12:05
dalek rrot: r43424 | mikehh++ | branches/one_make/tools/dev/checkdepend.pl:
fix codetest failures - pod syntax and trailing spaces
 12:07
12:10 bluescreen joined 12:17 payload joined
nopaste "mikehh" at 81.149.189.7 pasted "one-make branch pre-config test failures" (81 lines) at nopaste.snit.ch/19252 12:42
mikehh sorry that should be post-config 12:44
13:00 bluescreen joined 13:06 ruoso joined 13:11 kid51 joined, bacek joined 13:27 iblechbot joined
Coke msg aninhumer it would be easier for most parrot committers if there was a patch we could apply. 13:30
purl Message for aninhumer stored.
Coke I see three commits on that branch, checking... 13:31
aninhumer Coke, fair enough, I was just suggesting that docs could be treated differently to code in terms of ease of contribution 13:33
Coke still have the same licensing issues atm. 13:34
hurm. looking at github.com/aninhumer/parrot/commit/...56d75cdb69 ; how can I get that in a straight patch format?
but yes, having an easier way for folks to contribute docs would be a good thing. 13:35
someone please mention that at parrotsketch today.
aninhumer I'm not really that familiar with git, I'm looking up how roll a patch from my local git copy now
moritz aninhumer: git-format-patch d3a0c2114055f06862d5558df351b156d75cdb69
Coke don't worry, I'm applying these 3 commits.
moritz or git show $commit
Coke ah. guess I could have done that as well. be nice if github gave me a way to do that. :| 13:36
aninhumer well thanks for future reference moritz
I suppose github assumes you have a local copy for anything advanced
13:37 payload joined
Coke aninhumer: Thanks, applied. 13:41
aninhumer I'll submit patches in future, if that's easier for everyone 13:42
dalek rrot: r43425 | coke++ | trunk/docs/user/pir/intro.pod:
Fixed nesting of pod sections
 13:45
rrot: r43426 | coke++ | trunk/docs/pmc/subs.pod:
Changed incorrect opcode in example, Removed duplicate section.
rrot: r43427 | coke++ | trunk/docs/pmc/subs.pod:
Added a section on defining subs, and included info on most (all?)

Courtesy: aninhumer++
Coke aninhumer: perhaps dukeleto can easily take git forks and push them back upstream. 13:52
13:52 JimmyZ joined
Coke but in general, we're stuck in an svn world. 13:52
moritz hopes that changes at some point
szbalint yeah 13:53
svn--
bacek O hai 14:00
dalek rrot: r43428 | bacek++ | branches/gc_encapsulate/src/gc/gc_private.h:
Add bunch of functions into GC_Subsystem.
 14:01
rrot: r43429 | bacek++ | branches/gc_encapsulate/src/gc/gc_ms.c:
Add stubs for new allocation functions into GC MS.
Coke but, however we get patches, thanks for supplying them. =-)
allison will not be on #ps today. I might not be either. 14:02
dalek rrot: r43430 | bacek++ | branches/gc_encapsulate/src/gc/gc_ms.c:
Implement allocation/deallocation of PMC and STRING in GC MS.
rrot: r43431 | bacek++ | branches/gc_encapsulate/src/gc/api.c:
Migrate src/gc/api.c to use new API for PMC and STRING allocations.
14:04 aninhumer joined 14:11 payload joined
dalek rrot: r43432 | bacek++ | branches/gc_encapsulate (13 files):
Bulk commit:

  - Shuffle functions around.
  - Add explicit Memory_Pools* parameter to many functions.
  - Mark cutted corners with C++ comments and XXX markers.
Everything is broken as usual.		 14:19
14:32 cognominal joined, smash joined
smash hello everyone 14:32
Coke hulk? 14:39
purl Hulk SMASH!!!
Coke ~~
smash Coke: hey :) 14:41
14:46 plobsing joined
Essobi WEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE 14:50
dalek nxed: r360 | julian.notfound++ | trunk/winxedst0.cpp:
refactored break and continue in stage 0
 15:04
nxed: r361 | julian.notfound++ | trunk/winxedst1.winxed:
several refactors, fixes, minor improvements and new predefs in stage 1
 15:09
15:11 Psyche^ joined 15:23 bubaflub joined 15:50 payload joined 16:29 plobsing joined 16:35 sri joined 16:59 theory joined
Coke wonders if a punycode library function would be helpful. 17:20
Coke just waits for someone to write it in perl6. 17:21
OH. we could start rewriting our library stuff in NQP.
Coke will postpone efforts there until after one_make is done. 17:22
17:32 fperrad joined
dukeleto msg chromatic ping 17:41
purl Message for chromatic stored.
Coke ... that's odd.
dukeleto aninhumer: did I hear my name? 17:42
Coke: what is this talk about pushing git forks upstream?
Coke dukeleto: yes. if people fork your parrot github repo, what's the best way for them to get patches back to the svn repo?
I just did a git clone to steal some of aninhumer's commits by hand.
I'd rather avoid doing thaten masse. 17:43
*that en masse
tewk_ python, perl5 and ruby all don't have jits right?
dukeleto Coke: best way is for parrot to use git
Coke: but there are 17 other more complicated ways 17:44
17:44 ruoso joined
cotto_work tewk_, python has unladen swallow (llvm-based) but it's not in cpython yet 17:49
tewk_ cotto_work, yeah no one has a jit in there reference impl, almost everyone has an experimental jit with verying degrees of completeness. 17:50
perl5 doesn't have a jit effort underway that I know of.
cotto_work That seems to be the case.
tewk_ besides parrot
cotto_work interesting data point there 17:51
(java does, of course)
tewk_ and we threw our jit out for the short term.
dukeleto tewk_: why do you ask? 17:52
tewk_ yeah I classify java and c# vms in a different class than dynamic/scripting languages.
dukeleto, PhD research, we are working on performance and parallelization of dynamic languages, (particularly PLT Scheme), so I try to keep current on comparative efforts in other projects. 17:53
cotto_work jruby gets its jit from the jvm 17:54
not sure how you'd count that
same with jython 17:57
Coke dukeleto: in that case I'll just ask folks to submit trac tickets with patches. 18:03
tewk_ cotto_work, yeah I consider that cheating, :) They get it almost for free, but have to spend huge efforts to *efficiently* map their dynamic languages to java or atleast jvm bytecodes. 18:04
They also give up tail recursion and first class continuations. 18:05
18:09 bubaflub joined, bubaflub left, bubaflub joined
dukeleto tewk_: cool 18:25
18:25 chromatic joined
dukeleto chromatic: mornin' 18:26
18:26 davidfetter joined
mikehh #ps? 18:29
purl it has been said that #ps is a status meeting for parrot core committers held every Tuesday at 18:30 UTC in #parrotsketch
mikehh like now 18:30
chromatic hello
#ps time 18:32
18:40 gaz joined 18:51 jjore joined 19:10 ZeroForce joined
cotto_work 2.3 will be a deprecation point, right? 19:15
chromatic Yes. 19:16
cotto_work OK. It should probably be starred in release_manager_guide.pod then. 19:17
19:18 hercynium joined 19:29 payload joined
Coke I thought I had done that already. 19:35
19:38 AndyA joined 19:41 joeri joined
Coke cotto_work: fixed. 19:47
dalek rrot: r43433 | coke++ | trunk/docs/project/release_manager_guide.pod:
mark all known upcoming supported releases
 19:52
20:05 bubaflub left
mikehh Coke: one_make branck r43407 removes a generated dependancy (include/parrot/enums.h from config/gen/parrot_include.pm) that seems to be needed 20:14
20:26 bluescreen joined
Coke enums.h is no longer created by parrot_include.pm 20:29
(but anything that SHOULD depend on enums.h probably doesn't)
(are you referring to the removal of enums.h from the source_files list? 20:30
sorry. enums.pasm now correctly depends on enums.h, but: 20:31
nothing new depends on enums.pasm, and it should.
20:34 Zak joined
Coke also, read my commit message. :) 20:34
mikehh lib/Parrot/Pmc2c/PCCMETHOD_BITS.pm is generated from enums.h and is required by tests and the build 20:35
Coke aha. Thank you! 20:37
mikehh++
I was just focusing on the .pasm files, missed that one. good catch. 20:38
(wonder why I was even still able to build.)
mikehh probably did not do a make realclean
Coke ok. know how I'll fix it, adding to the pile. 20:42
20:48 AndyA_ joined 20:50 AndyA joined
GeJ Good morning everyone! 20:55
cotto_work hi GeJ 20:56
GeJ Hello cotto. 20:57
20:59 bacek joined 21:04 jsut joined 21:06 AndyA_ joined
Coke /msg purl jsut? 21:06
21:07 nopaste joined 21:37 Andy joined 21:52 nopaste joined 22:01 zak_ joined 22:02 kjeldahl joined 22:08 dduncan joined
aninhumer Regarding trac.parrot.org/parrot/ticket/487 are there any cases where s/vtable method/vtable function/ wouldn't be valid (within docs/)? Or does a bulk replace defeat the point of the newbie ticket? 22:09
dduncan does parrotsketch happen here or in another channel? 22:11
aninhumer in #parrotsketch
PerlJam dduncan: it happens on #parrotsketch :)
dduncan okay
dukeleto <rant>
I search for "parrot nci" and links to parrotcode.org OLD docs come up, and no links to the new docs. This really SUCKS. 22:12
</rant>
I just keep missing #ps. Darn.
22:21 zak__ joined
PerlJam dukeleto: you know ... you could write some docs for parrot NCI :) 22:22
22:22 ash_ joined
dukeleto PerlJam: that is not the issue. The issue is that people searching for docs get horribly, horribly outdated ones, and the new ones are no where to be found 22:25
PerlJam: but yes, I do need to add better docs for NCI
this is something we should fix before 2.0
PerlJam Is NCI on the radar for 2.0? 22:26
davidfetter pleads for some NCI love
that being necessary for PL/Parrot :) 22:27
22:29 ash_ joined, ash_ left 22:32 nopaste joined
dukeleto NCI *works* (mostly) but is ill-documented 22:34
some exotic function signatures may not be supported yet
PerlJam dukeleto: I haven't played with NCI in ages; what's "new" about it? 22:35
dukeleto PerlJam: not much, that I know of. I don't think it has changed much recently 22:36
Coke dukeleto: yes, the search sucks. there's not much we can do about this. 22:38
(aside from insure that the old stuff is nuked)
If we're just talking us, though, you can use the special parrot google searcher that someone put together, which might have better results. 22:39
(nope, that also hits parrotcode.rog)
(see www.parrot.org/'s google search box) 22:40
dukeleto when i search for "parrot nci" i want docs.parrot.org/parrot/latest/html/...l.pod.html to come up
Coke dukeleto: ... talk to google? 22:41
dukeleto parrotcode.org needs to die
Coke get people to link to the real docs?
dukeleto Coke: did you take snarkiness pills today? ;)
Coke dukeleto: no, did you take jerk pills?
dukeleto Coke: no, I don't have to, I produce jerkotine naturally
Coke: we are agreeing. Let's agree to agree. 22:42
Coke I am unable to find the last email thread I had with the parrotcode admins that covered this topic. I'll see if I can dig it up this weke. 22:43
dukeleto Coke: thanks! who are "the parrotcode.org admins" ? 22:44
Coke perl.org
purl perl.org is, like, home of all things perl and very rarely way horked
Coke ah, there it is. 22:45
dukeleto coke++
22:45 zak_ joined
Coke robert was just going to do a single redirect: "i.e. m!www.parrotcode.org/.*! => www.parrot.org/" 22:45
but could do something more complicated if we gave him a mapping.
I failed to give any such mapping after 11/17/09, when I hinted I might.
so, lazy me, let's just have any request forward to the top level of parrot.org. does that work for you? 22:46
dukeleto Coke: can you CC me or get me into that loop? I will try to help, instead of only complain.
Coke ok.
dukeleto?
purl hmmm... dukeleto is mentoring a few peeps. can't remember everyone. sure.
dukeleto Coke: that is something
Coke: jaleto@gmail.com
dukeleto is pretty darn good at complaining
Coke sent. 22:48
thanks in advance.
at this point, just having everything forward from pc.org to the top level of p.org seems fine with me.
dukeleto Coke: that is better than the current situation, I agree
i don't want people to be getting really old docs when 2.0 comes out. That would hurt us a lot. 22:49
tewk_ 23:00
23:11 theory joined 23:14 tedkat joined 23:16 redbrain joined, bacek joined 23:31 kid51 joined 23:57 nopaste joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!