www.parrot.org/ | Parrot 1.4.0 "Mundo Cani" Released! | For 1.5: Remove Deprecated Features | Planet Parrot planet.parrotcode.org/
Set by moderator on 4 August 2009.
allison japhb: I'm comfortable with collecting module metadata rather than building our own distribution system 00:02
mberends japhb: for Rakudo I hope to build out the 'use' command towards S11, and a CLI tool is definitely part of that process. 00:05
00:09 bacek joined
bacek good morning #parrot 00:09
00:10 tetragon_ joined
mberends good morning bacek (it's morning in London, 1:12am) 00:12
l3t0 bacek: good localtime() 00:13
dalek cnum-dynpmcs: r143 | darbelo++ | trunk/src/pmc/decint.pmc:
Integer-ification of METHOD return values. Copypasting this time as SUPER()
bacek mberends: 10:13am here :)
l3t0 it's about beer:30 in PDX
japhb clock?
purl japhb: LAX: Fri 5:13pm PDT / CHI: Fri 7:13pm CDT / NYC: Fri 8:13pm EDT / LON: Sat 1:13am BST / BER: Sat 2:13am CEST / IND: Sat 5:43am IST / TOK: Sat 9:13am JST / SYD: Sat 10:13am EST /
bacek l3t0: Mate, It's always a good time for a beer in Australia! :)
mberends l3t0: beer, wine, port, all floats my boat :) 00:14
japhb l3t0: half past beer? What do you usually have after the beer?
chromatic Hangover.
purl when you've partied too hard and your head feels like you've been coding asm for weeks on end.
l3t0 japhb: another beer and perhaps some chips 00:17
00:17 quek joined 00:18 patspam joined
mikehh the only possibility I can see that breaks test 16 in t/examples/pir.t is r40442 00:29
cottoo chromatic, typo in your blog post: cople 00:30
mikehh it breaks at the first line in pir_16.pir - .include "../../examples/pir/levenshtein.pir"
chromatic Thanks, ccoottoooo
mikehh abd it passed at r40439 00:31
and
cottoo why the long me? 00:32
;)
chromatic You have an extra O
japhb "the long me" ... reminds me of "the long now" 00:33
chromatic Paper Mario: The Ten-Thousand Year Branch
japhb Just as long as the VCS branches don't last that long .... 00:34
chromatic I never finished that one. I think the Goomba merged it at the end.
cotto I don't see any extra o 00:35
not_cottoOO O RLY?
purl NOWAI!
bacek tempted to put /languages/pir back to .parrot... 00:36
00:38 bkuhn joined 00:40 darbelo joined
dukeleto is running codingstd_tests for the first time. wow, it is slow 01:08
cotto It's not bad considering that it looks through all files of various types. 01:10
but yeah, don't hold your breath
dalek rrot: r40446 | dukeleto++ | trunk/docs/embed.pod:
[embed] Correct function signature in the docs for Parrot_str_to_cstring
 01:11
chromatic idly wonders how to use git format-patch with git-svn. 01:12
dukeleto chromatic: what are you trying to do? 01:13
01:13 [1]Limbic_Region joined
dukeleto git diff --no-prefix will create a diff that normal patch can read 01:14
chromatic I have changes I don't want to dcommit yet, but I want to share them in patch form. 01:15
dukeleto chromatic: interesting. can you use git svn diff ? 01:16
chromatic I don't see that as a git-svn command. 01:17
dukeleto me neither. huh 01:18
chromatic: please document what you find on the git-svn wiki page, 'cause I would like to know this as well. gotta jet for some $food
01:19 quek left
mikehh dukeleto: what system are you running codetest on? 01:19
01:20 Limbic_Region left, quek joined
chromatic git format-patch remotes/trunk 01:22
or whichever branch 01:23
mikehh I managed to run make -j9 codetest TEST_JOBS=5 in 70 secs 01:24
most of the time spent in perlcritic.t 01:25
about 101 seconds for just make codetest 01:31
dalek rdinal: 58f7557 | (Danius Michaelides)++ | (5 files):
Added support for index,count and range versions of []=. Tests included. Second attempt.
 01:35
tracwiki: v11 | chromatic++ | git-svn-tutorial 01:38
tracwiki: Added local/repo change trick
tracwiki: trac.parrot.org/parrot/wiki/git-sv...ction=diff
01:39 treed joined 01:40 TiMBuS joined
mikehh it seems that the -j option doesn't seem tohave any effect on the make (test) just the TEST_JOBS 01:41
cotto mikehh, that's because make test goes through a perl-based test harness. Make can only start one test harness process, so the harness needs to be responsible for parallelizing tests. 01:47
Coke I think "git diff trunk" works. 01:50
chromatic Not on a branch, not for me. 01:57
02:03 kid51 joined
cotto we need to get bacek_at_work to write up his git-svn workflow on the wiki 02:04
chromatic I'll put it on my task list, then you dare him to do it. 3) Profit! 02:05
02:08 davidfetter joined
chromatic Yay, spammer in TT. 02:08
dalek rdinal: 9533a4d | treed++ | (2 files):
Fix for x in y parsing. Fixes issue #2.
 02:25
02:25 wayland76 joined
nopaste "mikehh" at 90.209.117.193 pasted "PATCH for examples_tests FAILure at r40445 (t/examples/pir.t)" (13 lines) at nopaste.snit.ch/17498 02:34
02:35 janus joined
mikehh make examples_tests PASSes after the patch 02:35
chromatic +1 on the patch 02:37
Will apply now, if no one beats me to it.
dalek rrot: r40447 | chromatic++ | trunk/t/examples/pir.t:
[t] Fixed t/examples/pir.t #16 with patch from mikehh.
 02:41
mikehh chromatic: I sent my CLA via email to legal@parrot.org 02:46
02:47 bacek joined
bacek hi again 02:53
purl oh, you're back!
bacek purl: annoying girl...
purl bacek: huh?
wayland76 purl, purl is also an annoyling girl 02:54
purl okay, wayland76.
wayland76 purl?
purl wayland76?
wayland76 Phooey. It was worth a go :) 02:55
cotto purl, purl?
purl ♬ beat on the bot // beat on the bot // beat on the bot with a baseball bat// oh yeah ♬ or pretty stupid, aren't you or an annoyling girl
bacek purl: purl?
purl ♬ beat on the bot // beat on the bot // beat on the bot with a baseball bat// oh yeah ♬ or pretty stupid, aren't you or an annoyling girl
kid51 must sleep 03:03
purl $kid51->sleep(8 * 3600);
03:16 l3t0_ joined 03:17 particle1 joined 03:20 donaldh joined 03:21 satrac joined 03:59 Andy joined
mikehh All tests PASS (pre/post-config, smolder, nqp_test, fulltest) at r40447 - Ubuntu 9.04 amd64 04:18
dukeleto git format-patch is pretty darn cool 05:12
mikehh: i am using os x 05:16
is there interest in adding function signatures to the huge list of functions at the bottom of docs/embed.pod ? 05:21
i am running the parrot_debugger from within gdb. scary. 06:11
Ryan52 debugging debuggers sounds like fun. 06:18
dalek rdinal: 235d413 | treed++ | t/range (2 files):
Clean some of the range tests.
 06:19
rdinal: f3e45ff | treed++ | src/classes/ (3 files):
Make all usages of 'nil' use the same hll_global. Fixes #26.
dukeleto Ryan52: it is. i just fixed stuff! 06:32
i found a very interesting bug that I will report soon
sometimes we are told that we have string registers when there are none, so REG_STR returns a null byte instead of a Parrot_str and things go boom! 06:34
06:40 dduncan joined 06:41 dduncan left 06:49 joeri joined 07:21 donaldh joined 07:28 nperez joined
nperez ping 07:28
dukeleto pong 07:29
nperez I am interested in implementing select() in rakudo, and I was curious if that functionality even exists in parrot 07:30
dukeleto nperez: remind my of the definition of select() in the perl6 spec ? 07:31
nperez dukeleto, couldn't tell you, this is all discovery at the moment 07:32
dukeleto nperez: do you mean like select on an IO handle or select the array function ?
nperez select on multiple IO handles
dukeleto nperez: at first glance, there does not seem to be a "select" keyword in PIR 07:35
you may want to look at examples/pir/io.pir for some intro to how Parrot does IO
nperez I am going to guess that implementing the c-based select(2) in parrot will be non-trivial 07:36
dukeleto nperez: and the source is in src/io/*.c
nperez acknowledged 07:37
dukeleto nperez: non-trivially fun
nperez we'll see
mberends yes, but the guys who might do it were discussing it recently: irclog.perlgeek.de/parrot/2009-08-03#i_1362532
nperez fascinating 07:38
looking in io/socket_api.c #148: return PIO_POLL(interp, pmc, which, sec, usec); 07:40
ack can't find PIO_POLL defined anywhere
ahh, wasn't high enough in the directory hierarchy 07:42
looks like Tene followed through on adding poll as a method to Socket.pmc 07:52
dukeleto nperez: are there any tests? 07:58
nperez haven't checked. i was checking rakudo for any prelim work using that method 07:59
doesn't look like there is any
08:00 quek left
nperez it seems like there needs to be a generic poll beyond just sockets 08:00
but it is now 3am CDT and I need sleep 08:01
dukeleto there is a note that both types of select() were removed from the spec at the bottom of perlcabal.org/syn/S32/IO.html
nperez heh 08:02
removed.. 08:03
so what is to take the place of them? 08:04
i mean, at least select(2) 08:05
08:05 Khisanth joined
nperez unless it is to live in IO::POSIX 08:06
mberends nperez: please keep at it, even if it will take patience. you may have to connect with the right person at the right time to move this forward, but this is the way lots of parts were made. 08:14
dalek rrot: r40448 | dukeleto++ | trunk (5 files):
[debugger] Add ability to assign to registers
 08:18
08:19 joeri joined 08:39 bacek joined
bacek o hai 08:40
cotto nobody's here 08:41
These are not the parrots you're looking for. 08:42
bacek cotto: oh shi... It's a pity. I've got another beautyful idea how to break stuff... :) 08:53
cotto You get a lot of those. 09:02
bacek At least 2 all the time.
BTW, what about git-svn?
cotto Oh. It'd be nice if you could write up your workflow on the wiki page. 09:03
(I personally haven't started using it yet, but I support making it easier to learn.)
bacek It's all described in man git-svn. But I can try to write small "workflow" on wiki 09:04
dukeleto bacek: that would be awesome 09:06
bacek Where is wiki page for git on parrot?
dukeleto trac.parrot.org/parrot/wiki/git-svn-tutorial 09:07
purl, git is trac.parrot.org/parrot/wiki/git-svn-tutorial
purl ...but git is teh new svn or perl5.git.perl.org/perl.git or at git-scm.com/...
bacek dukeleto: thanks
dukeleto i am keeping a github mirror up, but I haven't pushed all the branches, just trunk 09:08
cotto time for sleeps before I get stupid enough to try to commit something
night
dukeleto evidently you can push a branch called "svn/foo" to github, but then the web interface borks itself, so I am waiting to hear back from them
cotto: night
bacek cotto: commit but no dcommit 09:10
09:15 wayland76 joined
bacek cotto: my git-svn workflow on wiki now. 09:24
dalek tracwiki: v12 | bacek++ | git-svn-tutorial 09:26
tracwiki: Tracking svn branches
tracwiki: trac.parrot.org/parrot/wiki/git-sv...ction=diff
10:26 Whiteknight joined 10:34 wayland76 joined 11:01 darbelo joined 11:20 donaldh joined 11:35 bacek joined
mikehh just ran make cover - claims we have 64.6% total coverage 11:43
bacek mikehh: it's bad... We had 66.4 couple weeks ago 11:47
tapir2.ro.vutbr.cz/cover/cover-resu...4/c_cover/
mikehh there are too many files with 0%
bacek tests welcome! :)
mikehh I'll try and figure some out - not sure about some of the stuff :-} 11:50
bacek src/events.c can be ignored for now
All functions from src/tsq.c can be deprecated and removed. Roadmap/RFC ticket for ThreadSafeQueue PMC should be added to replace them 11:53
For all src/pmc/*.pmc with low coverage we need more tests... 11:55
mikehh the report is similar to yours - but quite a few more 0%'s
bacek There is more <10% files 11:56
I'm just giving my ideas about what I'm >90% sure :)
dalek rrot: r40449 | NotFound++ | branches/auto_attrs (76 files):
merge from trunk r40448
 12:00
tracwiki: v13 | bacek++ | git-svn-tutorial 12:17
tracwiki: Added links to "Git Book" and "Git magic"
tracwiki: trac.parrot.org/parrot/wiki/git-sv...ction=diff
tracwiki: v14 | bacek++ | git-svn-tutorial 12:24
tracwiki: Fix git command line args
tracwiki: trac.parrot.org/parrot/wiki/git-sv...ction=diff
kudo: 93f2fce | moritz++ | README:
[README] document TEST_JOBS, KyleHa++
 12:30
12:31 mokurai left 12:55 quek joined 13:00 JimmyZ joined
dalek kudo: c13df6b | moritz++ | t/spectest.data:
three more passing test files (KyleHa++ for nil.t)
 13:10
13:16 Khisanth joined, bkuhn joined 13:27 MoC joined 13:48 darbelo_ joined 13:54 quek left
Whiteknight good morning #parrot 14:05
bkuhn ahoy hoy, Whiteknight. 14:09
Are there any PCT experts about?
Whiteknight I'm not an "expert", but I can try to help
bkuhn I'm looking for a way to get more debugging output on the HLLCompiler class.
I'm working on pynie.
And I'm currently attempting to add better multi-line string support, which is a bit tricky. 14:10
14:10 darbelo joined
bkuhn I have a syntax error, and I'm wishing to get more output about the parser's traversal to git some hints about how to fix the rules. 14:10
14:10 cogno joined
dalek rrot: r40450 | whiteknight++ | trunk/docs/book (2 files):
[book] Add two new directories here for the two new books: the PCT book and the Modules book
 14:10
bkuhn The problem is --target=parse, which I thought would DWIM, doesn't exactly.
Because it gives me nothing when there's a syntax error anywhere in the code. 14:11
input code for target lang, I mean.
I'm new to all the PCT/HLLCompiler Class stuff, so hints would be welcome. 14:12
I'd like to just watch the parser traverse, show me which productions it's trying, etc.
Whiteknight hmmm, that is quite a good question and I don't know the answer to it
but pmichaud definitely would
bkuhn Ok, I'll but pmichaud next time he's around. It's no hurry.
Whiteknight your best bet might be to add some debug statements yourself to the parser to keep track of things 14:13
bkuhn yeah, that's what I was going to do if I didn't get a better answer. ;)
s/but/bug/ back there, but I think that was clear. ;)
dalek rrot: r40451 | whiteknight++ | trunk/docs/book/pct (0 files):
[book] move some of the PCT-related files from draft to the new PCT folder
 14:14
rrot: r40452 | whiteknight++ | trunk/docs/book/pct (0 files):
[book] copying two introductory paragraphs to the PCT book folder, because I think they (or a version of them) would be helpful there
 14:20
rrot: r40453 | whiteknight++ | trunk/docs/book/pct (0 files):
[book] rename the files in this folder to account for the new ordering of things
14:45 tetragon joined 14:47 Psyche^ joined 15:10 quek joined 15:13 JimmyZ joined
dalek rrot: r40454 | whiteknight++ | trunk/docs/book/pct/ch03_compiler_tools.pod:
[book] overhaul much of the PCT chapter to not rely on stuff from the PIR book and also to be more clear
 15:14
15:16 HG` joined 15:20 donaldh joined 15:21 Khisanth joined 15:27 quek left 15:29 Khisanth joined 15:41 dukeleto joined
dukeleto good localtime() 15:46
mikehh I am getting a svn server error - 15:50
svn: Server sent unexpected return value (304 Not Modified) in response to GET request for '/parrot/!svn/bc/40451/trunk/docs/book/pct/ch03_nqp.pod' 15:53
dukeleto interesting 15:58
which command did you type ?
and which version of svn ? 15:59
mikehh svn update
purl At revision 666.
mikehh svn, version 1.5.4 (r33841) 16:02
dukeleto is it reproducable ? 16:03
mikehh compiled Aug 7 2009, 02:02:06 - I think there was a security update this morning on Ubuntu 9.04 amd64
MoC I get that too, if I use the commandline svn (on Windows), with Tortouise it works.
svn 1.6.3. 16:04
dukeleto i can do an svn up but I am using svn 1.4.4 (!) 16:16
mikehh I don't know what went wrong but I managed a clean checkout 16:18
It is still giving an error in one of my parrot directories but is fine in another 16:25
I'll just delete that onbe and work with the clean co 16:26
JimmyZ syn? 16:27
purl Go, and SYN no more
dukeleto latest svn rev has manifest problems: gist.github.com/164451 16:32
dalek kudo: 2db9166 | pmichaud++ | src/setting/ (2 files):
metaclass objects currently match Object, not Any
 16:34
kudo: eda5f51 | pmichaud++ | t/harness:
Instruct the harness to ignore exit code errors for now.
kudo: ce7d95d | pmichaud++ | :
Merge branch 'master' of git@github.com:rakudo/rakudo
16:38 kid51 joined
moritz Infinoid: we're missing dalek over in #perl6 - care to give it a small punch? ;-) 16:39
16:46 Andy joined
Whiteknight Infinoid: ping 16:52
cotto dukeleto, for future reference tools/dev/mk_manifest_and_skip.pl updates MANIFEST 17:16
dukeleto cotto: yeah I know, it doesn't work for me 17:17
cotto: interesting. now it does
dukeleto: but thanks for the heads up
i just used git-svn successfully for the first time to commit 17:18
dalek rrot: r40455 | dukeleto++ | trunk/docs/debugger.pod:
[debugger] Bring documentation up to date with reality
 17:19
rrot: r40456 | cotto++ | trunk/MANIFEST:
[MANIFEST] update manifest after recent changes to docs/book
dukeleto cotto: it doesn't work with git-svn, that is it 17:20
$ perl tools/dev/mk_manifest_and_skip.pl
can't open giant lock (/Users/leto/.svk/lock): Permission denied
it actually borks the file as well if you are using git-svn
cotto sounds like a bug 17:24
purl bzzzzzzzzzzzzzz...
cotto It should at least fail gracefully.
dukeleto cotto++ 17:25
i will take a look
cotto: should I just exit if .git exists ? perhaps with an inspirational message 17:27
cotto That'd be an improvement over what we have now.
dukeleto cotto: ok, on it
dalek rrot: r40457 | dukeleto++ | trunk/tools/dev/mk_manifest_and_skip.pl:
[tools] Fix mk_manifest_and_skip.pl so it doesn't bork the manifest when run under git-svn
 17:42
17:56 chromatic joined
Whiteknight cotto++ (thanks for the MANIFEST fixes) 18:13
purl msg Infinoid let me know if you're available tomorrow to hack at all 18:24
purl Message for infinoid stored.
chromatic www.dwheeler.com/essays/releasing-f...tware.html 18:34
dalek kudo: 5667dca | (Kyle Hasselbacher)++ | tools/autounfudge.pl:
[autounfudge.pl] fix typo
 18:59
19:20 donaldh joined 19:32 bacek joined 20:01 mokurai joined 20:12 dukeleto joined 20:13 mokurai joined 20:27 davidfetter joined 20:43 Tene joined 20:48 smash joined
smash hello everyone 20:48
dukeleto smash: hola 20:52
21:46 mikehh joined 21:49 Whiteknight joined 21:52 mokurai joined
dalek ose: r90 | Austin++ | trunk/ (26 files):
Grammar refactor continues. Single <declaration> is working
 21:54
Whiteknight Austin: ping 22:08
22:10 braceta joined 22:11 braceta joined
dalek cnum-dynpmcs: r144 | darbelo++ | trunk/ (3 files):
Convert 'pow' decTests to PIR. Tests failures caused by the library were left
 22:12
Whiteknight all the close tests fail for me 22:24
22:39 rg1 joined 22:44 bacek joined
bacek good morning 22:52
purl And good moroning to you, bacek.
dalek cnum-dynpmcs: r145 | darbelo++ | /:
Replace����/trunk/t/add.t

   The big add.t cleanup is here:
  - Removed all add.t test failures caused by library bugs
  - Removed various test relating the subtleties of exact and inexact zeros.
  - Removed tests that depended on not rounding at assignment time.
  - Removed tests involving invalid input.		 23:02
23:05 mokurai joined
darbelo Hmm, it looks like dalek isn't skipping over the "Replace" lines in the "Changed Paths" part of the google-code RSS feed. 23:12
Does anyone know who dalek's keeper is? 23:13
bacek darbelo: Infinoid 23:17
purl rumour has it Infinoid is Mark Glines <mailto:mark@glines.org> or likes shiny things or the master of the universe
23:17 Limbic_Region joined
darbelo Ah. 23:17
msg Infinoid It looks like dalek isn't skipping over the "Replace" lines in the "Changed Paths" part of the google-code RSS feed. See irclog.perlgeek.de/parrot/2009-08-08#i_1377212 for an example. 23:19
purl Message for infinoid stored.
23:21 donaldh joined
mikehh All tests PASS (pre/post-config, smolder, nqp_test, fulltest) at r40457 - Ubuntu 9.04 amd64 23:36
23:39 Whiteknight joined
Whiteknight hello #parrot 23:51
23:57 mokurai 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!