Parrot 0.9.1 Released | parrot.org/ | < 1 week to Parrot 1.0!
Set by moderator on 10 March 2009.
dalek rrot: r37290 | cotto++ | trunk/tools/util/gen_release_info.pl:
[tools] update url to svn repo
 00:02
cotto Didn't we used to have a script that generated a graph of commits over time? 00:03
nm. it's on tpf's wiki 00:05
dalek tracwiki: v69 | allison++ | Languages 00:06
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
shorten dalek's url is at xrl.us/bei495 00:07
00:08 AndyA joined
dalek rrot: r37291 | allison++ | trunk (4 files):
[c99] Migrating c99 to svn.parrot.org/languages/c99/trunk.
 00:22
00:52 eternaleye joined 01:02 zpmorgan joined
dalek rrot: r37292 | allison++ | trunk (4 files):
[pod] Relocating Pod parser to
 01:03
01:06 bacek_ joined 01:11 TiMBuS joined
dalek rrot: r37293 | allison++ | trunk (4 files):
[json] Relocating JSON parser to
 01:38
01:49 ujwalic_ joined
dalek rrot: r37294 | allison++ | trunk (4 files):
[pir] Relocating PGE-based PIR parser to
 02:02
Coke_afk msg kid51 deleting and recreating a file to update properties kills things like svn blame and history. 02:05
purl Message for kid51 stored.
kid51 Coke: I realized that too late. Sorry. I was confused about the output of the test. 02:06
Cokee k.
Coke ah, there's my nick.
kid51 I didn't realize that the output of the test had been changed to specifically instruct you what to do to correct the test's failure!
02:07 baest joined
kid51 Coke: I'm not sure that Makefile ever was in the repository, really. When I checked out the last rev before the delete and looked at the Makefile's svn log, all I found was this: 02:13
02:13 omega joined
kid51 $ svn log Makefile 02:13
------------------------------------------------------------------------
r37282 | jkeenan | 2009-03-10 19:01:17 -0400 (Tue, 10 Mar 2009) | 1 line
Add SVN Id tag and svn:keywords property per report in TT 428; mikehh++.
------------------------------------------------------------------------
r37265 | jquelin | 2009-03-10 04:21:44 -0400 (Tue, 10 Mar 2009) | 1 line
removing last remnant bits of befunge language
------------------------------------------------------------------------
02:13 skv joined
kid51 Maybe that's why its properties were so screwy. 02:14
02:31 skv joined 02:36 skv_ joined 02:42 jdv79 left
dalek rrot: r37295 | Util++ | trunk (2 files):
Updated streams.t to match previously changed .pir data-sources (TT#429)
 02:45
02:48 contingencyplan joined 03:02 tetragon joined
dalek rrot: r37296 | allison++ | trunk (4 files):
[ecmascript] Relocating ECMAScript parser to
 03:13
03:16 mikehh joined 03:25 slavorgn joined
Tene Oh, nice. I'm presenting tomorrow and I guess I deleted my scheme implementation I was going to present on... 03:26
No idea when it disappeared.
wayland Ouch! :) 03:29
Not in SVN?
03:42 janus joined
dalek rrot: r37297 | allison++ | trunk (4 files):
[lolcode] Relocating LOLCODE compiler to
 04:04
Tene no, I just had it in a local git repo, apparently not mirrored anywhere. 04:05
04:12 Andy joined
wayland ouch 04:13
04:17 tetragon joined 04:48 HG` joined 05:04 eternaleye joined 05:05 cotto joined 05:15 davidfetter joined
dalek rrot: r37298 | allison++ | trunk (4 files):
[abc] Relocating to the examples directory.
 05:18
05:33 cognominal joined
dalek rrot: r37299 | allison++ | trunk/examples/languages/abc (2 files):
[abc] Directory changes for ABC's new location.
 05:34
rrot: r37300 | allison++ | trunk/examples/languages/abc/Configure.pl:
[abc] A configuration script for ABC, to generate its makefile based on
 05:38
05:42 eternaleye joined
Tene Um, mk_language_shell.pl is broken 05:46
Can't open perl script "/usr/local/lib/parrot/0.9.1-devel/tools/dev/gen_makefile.pl": No such file or directory
when run from a checkout
allison Tene: that would be expecting an installed parrot 05:47
Tene: have you installed parrot?
Tene I've never installed Parrot.
allison Tene: it's safe now
Tene: most of the languages are shifting over to running from an installed parrot 05:48
Tene It doesn't give me good feelings about the safety of developing while there's an installed Parrot when this tool can't tell the difference. :)
allison Tene: ? 05:49
Tene: we decided the default would be an installed parrot, since that's what the greenest users would be using
Tene does the parrot checkout keep to itself and avoid looking at installed Parrot?
allison Tene: yes
Tene 'kay. I'll consider installing.
allison Tene: you can install into a local directory just to experiment 05:50
wayland Tene: Are you on Fedora?
Tene orly?
purl YA RLY.
Tene wayland: yes
wayland ok
allison say, "perl Configure.pl --prefix=/home/tene/tempparrot"
or something like that
wayland I've just submitted a patch that will hopefully update the specfile on Fedora to work with the latest rakudo 05:51
but, while rakudo gets further into the build than without the patch, rakudo still won't build
Tene wayland: I've got a coworker who is considering doing the work to get Parrot into Fedora's default repositories.
05:52 masak joined
wayland Great. I'll send a link to that patch 05:52
two relevant tickets. The main one is: trac.parrot.org/parrot/ticket/270 05:55
But you may also be interested in looking at trac.parrot.org/parrot/ticket/371
Does anyone know why the line "rebless_subclass sub, $P0" would fail with "error:imcc:syntax error, unexpected VAR, expecting '(' ('sub')" when I run "/usr/bin/parrot -o perl6_s1.pbc perl6.pir"? 05:56
Tene wayland: if you could send mail to sweeks@gurulabs.com about those tickets, that would be great. 05:57
wayland I'm expecting it's because it's looking for some file that isn't in the right place, but I'm not sure of that
Tene allison: do I need to make clean after changing the prefix? 05:59
I ran 'make install', but ~/parrot/ is empty... but Configure.pl now claims it might conflict with my existing installed Parrot at ~/parrot ? 06:00
... >< 06:01
It interprets ~/parrot as the literal directory '~' in the current directory.
Fail?
purl Fail is, like, kuvaton.com/bshit/howtonotfailatlifepn3he9.png
06:01 allison_ joined
allison_ Tene: rerunning Configure will overrite the old files 06:02
Tene Yes, I found it. It installed to /home/tene/parrot.svn/~/parrot/ :)
allison_ Tene: with prefix=~/parrot? 06:03
Tene Yes.
Bash only expands ~ if it's the first character in a token.
Should configure be expanding it? 06:04
chromatic Gah, you can only build Pheme now from an installed Parrot.
allison_ chromatic: can change back if you want
chromatic Or at least you can only *configure* it from an installed Parrot.
Tene Related to what I'm running into, I think... 06:05
chromatic I might add a --devel flag to configure to run when you have it checked out into languages/pheme/
allison chromatic: there's one hitch in Pheme, in that somewhere in the implementation it's using a fixed path
chromatic Probably pheme.pir, but that's fixable.
allison chromatic: I replaced all the "langauges/pheme" paths I could find 06:06
Tene chromatic: I found out about an hour ago that I lost my scheme implementation for my presentation tomorrow. Making a new one now. :)
allison chromatic: and it configures and builds fine, but the tests are looking for a specific path
chromatic Could be Test::More. 06:07
allison my guess was "(include_file lib/PhemeTest.pir") was the culprit
chromatic: I added a -I for the installed tools dir to the Makefile invocation of t/harness, which is working 06:08
Tene: the "~" is rather platform-dependent, so Configure shouldn't be handling it 06:10
Tene nods.
allison Tene: but, you can use --prefix=/home/tene/parrot
Tene That's what I did. Seems to work fine.
allison cool
Tene allison: looks like 'make install' doesn't install tools/dev/gen_makefile.pl, which the generated Configure.pl tries to use. 06:17
allison Tene: checking...
Tene it doesn't even create a tools/ 06:18
allison Tene: oh, yes, none of the compilation tools install unless you do 'make install-dev'
Tene Ah.
Thanks. :)
allison++
Tene hides IRC and gets to work. 06:20
06:24 Psyche^ joined 06:34 korshak joined 06:40 korshak left 06:43 eternaleye joined
dalek rrot: r37301 | allison++ | trunk (3 files):
[squaak] Relocating Squaak to the examples directory.
 06:55
06:55 korshak joined
dalek rrot: r37302 | allison++ | trunk/examples/languages/squaak/Configure.pl:
[squaak] Update configure script for new location.
 07:06
rrot: r37303 | allison++ | trunk/examples/languages/squaak/t/harness:
[squaak] Update lib directory for new location.
 07:10
07:11 uniejo joined
Tene Hmm. What's the magic "Don't backtrack beyond this point" in PGE? 07:24
moritz there's : 07:25
and ::
masak and ::: 07:31
moritz and :::: 07:32
no, just kidding
masak ::::P 07:36
Tene ah, the magic ~ stuff 07:38
oh, pmichaud never fixed trying to use ~ before something in []s. 07:39
moritz seems like :( 07:41
Tene Hmm... have the basics working... function calls and such... now I need to add blocks and lambdas and other variable handling stuff 07:45
moritz what are zou doing? a new language? 07:46
s/z/y/
Tene scheme
moritz ah
Tene for a presentation tomorrow.
I had a good scheme impl. a while back, but I apparently deleted it, or something?
It's not here anymore.
korshak JonathanWorthington, Š•Š²Š°Š½Š³ŠµŠ»Šøст 6-Š³Š¾ ŠŸŠµŃ€Š»Š°. 07:51
oops, miss
dalek tracwiki: v70 | allison++ | Languages 07:57
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
shorten dalek's url is at xrl.us/bei6fw
dalek tracwiki: v71 | allison++ | Languages 08:01
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
shorten dalek's url is at xrl.us/bei6gk
Tene got a bit done; sleeping 08:23
09:10 korshak left
dalek rrot: r37304 | fperrad++ | trunk/config/gen (2 files):
[config] move the generation of languages/Makefile. The step gen::languages wil be die.
 09:23
09:31 Psyche^ joined
ujwalic how to use *Parrot_runops_fromc_args_event* in windows 09:33
coredump on windows 09:39
dalek rrot: r37305 | fperrad++ | trunk/config/gen/makefiles/languages.in:
[languages] lolcode, pod, squaak, abc, PIR, ecmascript, dotnet, pheme, c99 & json are gone.
rrot: r37306 | fperrad++ | trunk/t/steps/gen_languages-01.t:
[config] fix test broken in r37304
 09:58
rrot: r37307 | fperrad++ | trunk/config/gen/makefiles/languages.in:
[external languages] add LOLCODE, POD, PIR, EcmaScript, dotnet, Pheme, C99 & JSON
 10:18
10:48 mj41 joined 10:49 Psyche^ joined
nopaste "bacek" at 87.250.243.86 pasted "[PATCH] Replacing few more PMC_int_val with VTABLE_get_integer" (49 lines) at nopaste.snit.ch/15859 10:50
bacek Can anyone check this patch? 10:51
10:57 Psyche^ joined 11:17 Psyche^ joined 11:20 tomyan joined 11:49 tomyan joined 11:53 rdice joined
wayland76 What's our equivalent of lambdabot here? 12:11
12:11 tetragon joined
wayland76 oh, wait, it's purl -- just purl doesn't respond well to the "help" command :) 12:12
cotto rakudo: say hi 12:16
polyglotbot OUTPUT[Could not find non-existent sub hiā¤current instr.: '_block14' pc 53 (EVAL_16:38)ā¤called from Sub '!UNIT_START' pc 18229 (src/builtins/guts.pir:321)ā¤called from Sub 'parrot;PCT;HLLCompiler;eval' pc 950 (src/PCT/HLLCompiler.pir:527)ā¤called from Sub 'parrot;PCT;HLLCompiler;evalfiles' pc 1275
..(src/PCT/HLLCompiler.pir:688)ā¤called from Sub 'par...
cotto rakudo: say 'hi'
polyglotbot OUTPUT[hiā¤]
cotto or not
nm. I was mentally substituting evalbot for some reason. 12:17
12:17 korshak joined
dalek rrot: r37308 | jkeenan++ | trunk/t/tools/install:
Create directory to hold tests of subroutines used in install-related
 12:21
purl programs are good at _implementing_ semantics
Coke masque? 12:24
purl rumour has it masque is figuratively speechless. or COOL WITH BURRITOS EVEN IF MINE DON'T SAY TOMMY HILBURRITO or Masquenfusion on AIM - USE THIS TO FIND HIM, IRC SUCKS or totally in love with warningsToBrowser, but forgetting to turn that off is something I fear. or awake. or Masquenfusion and Euqsam. or DJ Fresh Catnip or pleased with Masque's copy. or (see masque 2) or fond of used things that work. or a herring
Coke seen masque?
purl masque was last seen on #perl 1 days, 17 hours, 10 minutes and 7 seconds ago, saying: HACKING UNSECURED WIRELESS NETWORKS!? The guy is a GENIUS! [Mar 9 19:13:46 2009]
Coke masque 2?
purl well, masque 2 is You sass that hoopy Masque? There's a frood who really knows where his towel is. or hellyeah sysop IIRC (or knows who is) or avocado-powered, baby
Coke msg masque Hey, would it be possible for purl to ignore the other bots in the #parrot discussion? 12:25
purl Message for masque stored.
dalek rrot: r37309 | jkeenan++ | branches/install_tools:
Creating branch to hold work on extracting code from tools/dev/*install*.pl
 12:26
wayland76 btw, just did a svn up and there was a conflict, but it's easily resolveable (the extra comment in install_files needs to go in lib/Parrot/Install.pm) 12:28
12:33 rg joined
Coke did you have a local modification? 12:35
12:36 gryphon joined
dalek tracwiki: v7 | cotto++ | ParrotQuotes 12:45
tracwiki: Tene++ apparently wants to win at arson. 12:46
tracwiki: trac.parrot.org/parrot/wiki/Parrot...ction=diff
rrot: r37310 | coke++ | trunk (2 files):
Fix 'make codetest'.
shorten dalek's url is at xrl.us/bei6vt
13:10 ujwalic joined
ujwalic NCI callbacks not working in Windows 13:12
any one working with parrot on windows 13:21
Coke I test it once every so often. 13:23
but we have some primarily-windows developers, yes. 13:24
13:28 tomyan joined
ujwalic I want to know if NCI callbacks are working in windows .... 13:30
I found no tests for callback .. and how to create a dll using Parrot::Test ... only exe's is possible 13:31
Coke: where/how to enable RUNLOOP_TRACE 13:33
13:37 contingencyplan joined 13:39 tomyan joined 13:50 Andy joined
Coke when you Config parrot, you can add more defines at config time. 14:06
perl Configure.pl --ccflags="-DRUNLOOP_TRACE", perhaps? 14:07
14:21 particle1 joined, Tene joined 14:29 jhorwitz joined
jhorwitz emerges from IRC hibernation 14:29
davidfetter waves to Util 14:43
make[1]: Entering directory `/home/shackle/parrot/trunk/rpmbuild/BUILD/parrot-0.9.1' 14:44
make[1]: *** No rule to make target `perl6'. Stop.
make[1]: Leaving directory `/home/shackle/parrot/trunk/rpmbuild/BUILD/parrot-0.9.1'
error: Bad exit status from /var/tmp/rpm-tmp.9t9jmQ (%build)
Coke If you want perl6, you should be getting rakudo first, not parrot. 14:45
davidfetter i don't want perl6. i want to build the parrot rpms
make rpms causes that error
jhorwitz make rpms shouldn't be trying to build perl6 14:46
davidfetter clearly
davidfetter checks out a clean svn 14:47
14:52 PacoLinux joined
Coke ah. 15:09
I saw the perl6 and assumed that's what you wanted. sorry. =-)
jq just packaged parrot 0.9.1 on mandriva (jquelin.blogspot.com/2009/03/parrot...riva.html) 15:41
masak jq++ 15:42
16:20 tgz joined 16:23 Psyche^ joined 16:24 korshak left 16:27 tgz joined 16:49 Psyche^ joined 16:52 Theory joined 17:02 korshak joined 17:15 Patterner joined, tgz joined 17:27 slavorgn joined
dalek rrot: r37311 | fperrad++ | trunk/languages/pynie (3 files):
[Pynie] re-run mk_language_shell.pl
 18:05
rrot: r37312 | allison++ | trunk/src (2 files):
[io] More detailed error message on failed open.
 18:09
Tene welcomes jhorwitz back to irc. 18:16
jhorwitz looks around cluelessly
Tene pmichaud: is there a reason that '~' in PGE doesn't work with []s? 18:17
18:21 rurban joined
rurban FYI Just fixed the final 64bit pbc issue. patch soon to TT #254 18:22
Coke rurban++
pmichaud Tene: I'm not aware of a reason, no.
is there an easy test case I can look at?
rurban two minor stupid errors
Tene pmichaud: '(' ~ ')' [ <ident> <ident> ] 18:23
moritz pmichaud: I've submitted a Perl 6 bug for that some months ago
pmichaud I'm guessing it's a parsing issue -- I have to run a short errand but will check when I get back.
(about 20 min)
Tene Yes, it's a parse fail. 18:24
Coke tene: are you fixing PGE things? 18:26
dalek rrot: r37313 | allison++ | trunk/examples/languages/squaak/t/harness:
[squaak] Modify squaak's test harness to pass full path to Parrot
purl well, executable is rediculously larger than ever
Coke because have I got a deal for you.
Tene Coke: I'm asking pmichaud again about a bug I harassed him about months ago.
Coke ah well. =-) 18:27
Tene What's your deal?
free car?
free goat?
rurban Infinoid: trac.parrot.org/parrot/attachment/...it-2.patch 18:28
shorten rurban's url is at xrl.us/bei8a7
Coke tene: free tickets.
Infinoid "For sale: car, $FOO or best offer. Comes with free parking tickets." 18:33
dalek rrot: r37314 | fperrad++ | trunk/languages/cardinal (3 files):
[Cardinal] re-run mk_language_shell.pl
 18:34
jhorwitz has lots of $FOO 18:38
rurban I'd need someone for some ptr casting trouble I have on TT#254. Maybe this is fatal on certain compilers 18:39
18:53 allison joined 18:55 davidfetter joined
nopaste "pmichaud" at 72.181.176.220 pasted "PIR code for testing PGE" (11 lines) at nopaste.snit.ch/15861 18:57
"pmichaud" at 72.181.176.220 pasted "possible patch to fix" (23 lines) at nopaste.snit.ch/15862 19:05
"moritz" at 91.10.236.138 pasted "more tests for pmichaud++" (14 lines) at nopaste.snit.ch/15863 19:06
dalek rrot: r37315 | coke++ | trunk (2 files):
Pass t/codingstd/perlcritic.t

  - skip perl6-ish files
  - don't skip files that don't exist.		 19:10
rrot: r37316 | coke++ | trunk/t/codingstd/pod_description.t:
This file isn't run as part of 'make codetest', so there's no reason

it from the skiplist in DefaultTests.pm		 19:27
pmichaud Tene: ~ with brackets should be fixed in r37317 19:29
Tene pmichaud: thank you 19:30
dalek rrot: r37317 | pmichaud++ | trunk (2 files):
[pge]: Fix handling of goal syntax; with help from moritz++ and Tene++
 19:32
rrot: r37318 | coke++ | trunk/t/codingstd/opcode-doc.t:
[t] Test the ops files; before this commit, this file was a noop.
 19:37
rrot: r37319 | coke++ | trunk/t/codingstd/pod_todo.t:
This file isn't run as part of 'make codetest', so there's no reason

it from the skiplist in DefaultTests.pm		 19:42
rrot: r37320 | coke++ | trunk (2 files):
Rename file to match other filenames in this dir.
 19:46
rrot: r37321 | coke++ | trunk/MANIFEST:
regen manifest to track recently added files.
 19:50
19:51 NordQ joined
dalek rrot: r37322 | coke++ | trunk/languages/cardinal/Configure.pl:
Pass t/codingstd/copyright.t
 19:54
rrot: r37323 | coke++ | trunk/t/codingstd/copyright.t:
SKIP -> TODO
 19:58
Coke .rob? where'd he steal that name from? 20:00
PerlJam Where's the rimshot? 20:02
purl Ba-dum TISH!
PerlJam there we go
dalek rrot: r37324 | coke++ | trunk (4 files):
Add docs to pass a test so we can run it in 'make codetest'
 20:07
Coke if you do an svn rename on a file and run t/codingstd/pod_ch*.t , it fails, trying to find the old filename. 20:10
sorr, pod_syn*
Coke wonders, on a 2-core machine, using 'make -jN', what's my optimal N. 20:15
moritz Coke: usually 2 or 3
purl rumour has it usually 2 or 3 is sufficent
dalek kudo: 9844bf0 | (Moritz Lenz)++ | build/PARROT_REVISION:
bump PARROT_REVISION so that we benefit from recent PGE fixes
 20:16
shorten dalek's url is at xrl.us/bei8t3
20:17 barney joined 20:19 rafl joined
dalek rrot: r37325 | coke++ | trunk (2 files):
rename test file;
 20:23
Coke dalek's count is off there; 4 files were touched. 20:26
rg dalek doesn't count move and trac doesn't list a moved and then changed file as modified. 20:30
20:33 ujwalic joined 20:35 bacek joined 20:46 wayland76 joined 20:56 Whiteknight joined 20:57 alvar joined
dalek tracwiki: v4 | coke++ | AbandonedLanguages 20:58
tracwiki: note cola removed
tracwiki: trac.parrot.org/parrot/wiki/Abando...ction=diff
shorten dalek's url is at xrl.us/bei8z9
dalek tracwiki: v72 | coke++ | Languages
tracwiki: cola removed.
tracwiki: trac.parrot.org/parrot/wiki/Langua...ction=diff
shorten dalek's url is at xrl.us/bei82b
dalek rrot: r37326 | coke++ | trunk (9 files):
Remove languages/cola, one of the first languages to target parrot.
21:01 donaldh joined
barney will hear a talk about cola next thursday, phpugffm.de/index.php/next-devdusk-...3,373.html 21:09
21:09 NordQ joined
Coke ... that's not the same cola, surely. 21:10
barney I suppose it's not
Coke checks... nope. =-)
dalek rrot: r37327 | coke++ | trunk/lib/Parrot/Test/Pod/Utils.pm:
Fix a stray PerlFo Copyright notice.
 21:15
moritz what's trac's syntax for code blocks? 21:20
I can't remeber it :(
thought it was {{ code }} but that doesn't work in the preview
GeJ Good morning everyone 21:22
Infinoid try with 3 curlies, moritz
hai GeJ
{{{ code }}} is what I've always used, and has previewed for me properly in the past too 21:23
moritz ah, good
GeJ Heya Infinoid. How's life treating you?
servus moritz
moritz <rant> why do I have to remeber wiki syntax at all for submitting bug reports?</rant> 21:24
mikehh Coke: I was just trying to keep TT#420 up to date
moritz hi GeJ
rg moritz: you don't really. you could use the little buttons over the text field
moritz rg: there are no buttons over the text field 21:25
rg: and perhaps in the distant ffuture they even fix the ticket submissioin via email, then I'll have to remeber it again anway. Suckage. 21:26
Infinoid There are here.
There's even a "code block" button which gives me triple-curlies
I'm looking at the ticket edit page, perhaps the ticket create page is missing them
21:26 korshak left
rg you don't have B I A <world logo> <code block> ... buttons over the text field? 21:26
moritz maybe it requires javascript, but doesn't say so?
Infinoid Probably. 21:27
purl Really? Probably? Are you Certain it's not certain? Are you sure it's unsure? I think you need to look harder.
Infinoid purl sounds like a volunteer
purl Infinoid: sorry...
rg it most likely requires js, otherwise it couldn't modify the text
moritz if it requires js, it should say so in a <noscript> tag.
that's what they are designed for 21:28
Infinoid Maybe they prefer to have a clean interface, I wouldn't blame them
rg well it's just some support for remembering the tags.
Coke mikehh: Just didn't want you to think those failures were unique to your config.
Infinoid It's not really a requirement, just a feature they enable when possible
Coke s/you/folks/
moritz sorry, I#m just in rant mode, and trac is an easy target
Infinoid true 21:29
mikehh Coke: I have been following the other tickets - it was essentially a report as was my TT#428 21:31
21:32 ron joined
ron mad a typ o - must sacrifice US and other press credibility 21:34
Infinoid huh? 21:36
mikehh morotz: it comes out as {{{ code }}} in emails 21:37
moritz: it comes out as {{{ code }}} in emails 21:38
moritz mikehh: I know. That also bothers me. 21:39
mikehh ron: your o confused me :-}
dalek kudo: cd256dc | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 317 files, 7148 passing, 0 failing
 21:44
shorten dalek's url is at xrl.us/bei88f
moritz <meta name="generator" content="vim" /> 21:45
<meta name="author" content="Moritz Lenz" />
<meta name="language" content="en" />
orry
mikehh moritz: when I load the new ticket page it does not seen to use javescript 21:51
moritz: how are you accessing trac
moritz mikehh: I created a new ticket 21:52
mikehh moritz: what browser are you using? 21:53
moritz firefox 3.0
mikehh moritz: ditto - it has the icpns and stuff for moi 21:54
moritz mikehh: without javascript?
mikehh moritz: well no - but I have web dev and it does not show javascript on that page 21:56
rg mikehh: there's plenty of scripts in that page 21:57
moritz mikehh: just look at the HTML source
mikehh moritz: I just loaded the page source and yes I see 21:59
<script type="text/javascript" src="/parrot/chrome/common/js/wikitoolbar.js"></script><script type="text/javascript"> 22:02
I just have a couple of icons on the web dev toolbar that show if there are css or javascript errors and they were greyed out on that page 22:09
Infinoid guess it's well-written javascript then :) 22:10
(I'm using that toolbar too) 22:11
Tene purl: msg allison 'make test' in a generated language tries to run a literal "../../parrot", which doesn't work of course 22:12
purl Message for allison stored.
Tene That's almost big enough for me to send mail to the list.
Almost. :)
22:14 japhb joined
mikehh infinoid: no - I just looked at the error console and there are errors - but the icon was neither a green check or red cross 22:15
Infinoid Ah. Mine's a green check. 22:16
mikehh It comes on on the view tickets page but not on the new ticket page 22:18
gah - now it does - I think I need to reload firefox 22:19
pmichaud for generated languages, I think I would highly recommend rakudo's t/harness 22:22
22:23 kid51 joined
pmichaud (which doesn't depend on Parrot::Test::Harness) 22:23
22:24 Limbic_Region joined
mikehh yeah but it needs to thread 22:26
pmichaud we can make it thread 22:29
that's not all that difficult -- we just need to see if a threadable version of perl's T::H is available and dtrt 22:30
otoh, the way that rakudo's t/harness works is by depending on a fakecutable, so perhaps that's not such a good idea.
cotto Is anyone familiar enough with GC to say if the comment in src/hash.c before parrot_new_hash_x is still relevant? It was originally added in r4834.
Whiteknight, ping 22:37
Whiteknight cotto: pong
dalek kudo: 7edb29f | (Cory Spencer)++ | src/ (2 files):
Implement Any.first in Perl 6

commit aae0fa1c81d0d44ce1c3a1bc6d8fd61cf983a1b9 Author: Cory Spencer <cspencer@sprocket.org> Date: Wed Mar 11 13:28:22 2009 -0700
   Modified fail message for first method.
commit 6ae1a7d4288510a85718cbce4ebfc4918bd5b742 Merge: f2e2a16... 9844bf0... Author: Cory Spencer <cspencer@sprocket.org> Date: Wed Mar 11 13:22:24 2009 -0700
   Merge branch 'master' of git://github.com/rakudo/rakudo into any-list
commit f2e2a16a7e60b5660eba4e8c0694cc7e5397742f Author: Cory Spencer <cspencer@sprocket.org> Date: Wed Mar 11 13:00:10 2009 -0700
   Changed pairs method to "is export", removed pairs sub definition. Added a first method/sub, modified grep method to use @.list instead of $values invocant parameter.		 22:38
kudo: ea32839 | (Ujwal Reddy Malipeddi)++ | Configure.pl:
Fixed conversion of '/' to '\\' inside URLs in Makefile on MSWin32 platform.
shorten dalek's url is at xrl.us/bei9g6
cotto do you have any idea about the comment before parrot_new_hash_x?
shorten dalek's url is at xrl.us/bei9g8
cotto (wrt my question just before I pinged you) 22:39
Infinoid In other words, can Buffers define their own custom mark functions?
cotto or if the Hash* needs to be marked (rather than the container PMC) 22:40
I'd expect that a Hash would always need to have parrot_hash_destroy called on it manually rather than caring about GC. 22:41
Whiteknight cotto, which file? 22:43
cotto src/hash.c
:1055
I don't understand why that code needs to care about GC at all. 22:44
Whiteknight cotto: no, I don't understand it either
I would need to do some digging
cotto ok. I'll barge ahead and see if anything explodes. 22:45
Thanks.
dalek rrot: r37328 | jkeenan++ | branches/install_tools/lib/Parrot/Install.pm:
Extract code from tools/dev/install_files.pl and install_dev_files.pl and place in module for shared use and testing. wayland++
 22:53
Whiteknight pmichaud, ping 22:55
dalek rrot: r37329 | jkeenan++ | branches/install_tools/tools/dev/install_dev_files.pl:
Code has been extracted and placed in lib/Parrot/Install.pm.
 22:57
allison Tene: it should be using the configure value for bin_dir instead 22:58
Tene allison: it isn't. 22:59
allison Tene: was this directly produced by mk_language_shell.pl? (I haven't looked at it yet, just been making conversion directly
Tene Yes. 23:00
I haven't touched the generated configure.pl at all.
allison it's the generated root.in that's likely the problem 23:01
Tene: it's right in the current version of mk_language_shell.pl (generates "PARROT := $(BIN_DIR)/parrot@exe@") 23:02
Tene: ah, it's the shell t/harness that's the problem 23:03
Tene Ah.
allison instead of "compiler => 'foo.pbc'", it needs "exec => ['path/to/parrot', 'foo.pbc']" 23:04
cotto allison, I'd like to simplify the interface for Hash creation. Would there be anything wrong with doing all creation through (an non-static) create_hash and removing the other hash creation functions? 23:05
allison in the patch I submitted for rakudo, I passed in the bin_dir as a command-line option to t/harness from the Makefile
cotto: looking... 23:06
cotto Making the switch in the AddrRegistry PMC doesn't cause any test failures. 23:07
allison cotto: what would you be removing? the static functions seem to be building keys and accessing values, not hash creation 23:08
Tene allison: can we get this fixed in the generated makefile.in ?
nopaste "cotto" at 96.26.202.243 pasted "use create_hash in AddrRegistry" (134 lines) at nopaste.snit.ch/15864
allison Tene: yes, can fix now...
Tene: it will expose a bug (in Perl's test harness libraries) that patrick and I talked about earlier today, but he's putting a fix for that into the compiler tools 23:09
Tene Ah. 23:10
dalek rrot: r37330 | jkeenan++ | branches/install_tools (3 files):
Code touch-ups in new Parrot::Install. Update MANIFEST. Handle 'patch'
 23:11
allison cotto: the general answer is that unification into one function is good, if it doesn't overly increase complexity elsewhere (and might want to wait until after 1.0) 23:14
cotto I'll see what's needed to convert other uses of those functions and how much complexity it adds. Thanks. 23:16
23:22 bacek_ joined
dalek rrot: r37331 | jkeenan++ | branches/install_tools (3 files):
Export/import subroutines only on request.
 23:23
23:24 mire joined
Tene allison: what options did you pass in the Makefile to harness? 23:24
Eh, i can check myself. 23:25
allison Tene: see r37332 23:26
Tene: I modified both the Makefile template and t/harness 23:27
dalek rrot: r37332 | allison++ | trunk/tools/dev/mk_language_shell.pl:
[languages] Allow language shell to set bin/ directory for parrot (for
Tene allison++ :)
allison What you will get from this is an "Unable to open filehandle" warning on the first test
(aside from all tests passing) 23:28
Patrick's fix will take care of this
but, if you want to hide it in the meantime
$S0 = args_str[1] 23:29
if $S0 != '-le' goto not_harness
exit 0
not_harness
Adding that chunk of code in the 'main' routine of <yourlang>.pir will hide it 23:30
(before the call to 'command_line')
dalek rrot: r37333 | jkeenan++ | branches/install_tools/t/tools/install/01-install_files.t:
Skeleton of a test file.
 23:31
ujwalic anyone successful working with NCI callbacks on Windows 23:32
Tene allison: do you mean to use GetOptions in there? 23:34
23:34 Theory joined
ujwalic there are no test to check callbacks ... how to create dlls using Parrot::Test to write callback tests 23:34
allison Tene: that's how it gets the bindir from the Makefile 23:35
Tene: or, what did you mean?
Tene allison: I mean "I can't read". Nevermind. :)
dalek rrot: r37334 | jkeenan++ | branches/install_tools/MANIFEST:
Update MANIFEST.
 23:36
allison Tene: it does complicate the t/harness, unfortunately (from 2 lines to about 20), hopefully its useful knowledge for new language hackers
Tene Looks like it works. Allison++. :)
allison Tene: cool
23:40 TiMBuS joined
NotFound allison: Have you seen TT #436 ? 23:43
allison NotFound: not yet, looking
NotFound: for 1.0, document current practice 23:46
NotFound Ok
allison NotFound: if there are places you think we might want to chage later, file a specific ticket for each question with an [RFC] tag
NotFound And the doubt about if the fields are intended for deprecation or just not documented yet? 23:47
allison NotFound: assume that they're just not documented
NotFound Ok
Then the pdd can be easily fixed :)
But tomorrow, now is late to me :O 23:48
allison NotFound: thanks! :) 23:49

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!