Camelia, the Perl 6 bug

IRC log for #parrot, 2008-10-13

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:04 Andy joined #parrot
00:06 TiMBuS joined #parrot
00:09 AndyA joined #parrot
00:47 particle joined #parrot
00:52 Auzon joined #parrot
01:47 dalek r31918 | chromatic++ | trunk:
01:47 dalek : [t] Revised several tests to use unique temporary PIR, PASM, and PBC files.
01:47 dalek : This avoids spurious and accidental race conditions which cause weird failures
01:47 dalek : during parallel testing.
01:47 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31918
01:49 Ontolog joined #parrot
02:11 particle joined #parrot
02:27 ab5tract joined #parrot
02:42 davidfetter joined #parrot
03:46 Theory joined #parrot
03:52 dmknopp left #parrot
04:10 magnachef__ joined #parrot
04:14 MariachiElf joined #parrot
04:30 TiMBuS joined #parrot
04:57 Andy joined #parrot
05:42 AndyA joined #parrot
05:43 uniejo joined #parrot
05:52 uniejo joined #parrot
05:56 TiMBuS joined #parrot
07:04 dalek r31919 | fperrad++ | trunk:
07:04 dalek : [Lua]
07:04 dalek : - add conditional skip
07:04 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31919
07:10 iblechbot joined #parrot
07:57 barney joined #parrot
08:01 magnachef__ joined #parrot
08:13 dalek bernhard.schmalhofer@gmx.de | Pipp:
08:13 dalek link: http://www.perlfoundation.​org/parrot/index.cgi?pipp
08:31 ejs joined #parrot
08:52 tomyan joined #parrot
09:09 GeJ joined #parrot
09:12 dalek bernhard.schmalhofer@gmx.de | Pipp:
09:12 dalek link: http://www.perlfoundation.​org/parrot/index.cgi?pipp
09:14 isop joined #parrot
09:14 dalek bernhard.schmalhofer@gmx.de | Pipp:
09:14 dalek link: http://www.perlfoundation.​org/parrot/index.cgi?pipp
09:17 dalek bernhard.schmalhofer@gmx.de | Pipp:
09:17 dalek link: http://www.perlfoundation.​org/parrot/index.cgi?pipp
09:19 AndyA joined #parrot
09:26 TiMBuS joined #parrot
09:34 pancake joined #parrot
09:35 magnachef__ joined #parrot
09:44 dalek r31920 | bernhard++ | trunk:
09:44 dalek : [Pipp] write name of the proto regexes as identifiers, not as a string
09:44 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31920
09:50 cotto barney, ping
09:52 barney hi cotto
09:52 cotto what'd be a good way to get familiar with the rest of Pipp's code?
09:52 cotto is there some moderately easy feature or bug that needs work?
09:53 kj joined #parrot
09:53 barney Implementing functions, filling in the stubs generared by fperrard, should be straightforward
09:54 cotto that'd be a good and useful place to start
09:55 cotto what about the pct code?
09:55 barney http://kevin.vanzonneveld.net/te​chblog/article/phpjs_licensing/  could be a guide
09:56 barney the pct is fairly straightforward, the grammar is in grammar.pg and the rules for generating PAST is in actions.pm
09:56 barney I ususally write corresponding Perl 6 code and look at the PAST generated by Rakudo
09:57 barney --target=past is our friend
09:57 cotto so it shouldn't be too hard to grok and extend
09:57 cotto yes
09:57 barney not at all
09:58 cotto that's good to know.  I never got anywhere trying to turn the source code into a useful spec.
09:58 cotto er, Zend PHP's source code
09:59 Bzek joined #parrot
09:59 barney Editing the Wiki about the State of the implementation is also an useful thing
09:59 barney The Roadsend source code might be more approachable, if you know Scheme (which i don't really)
09:59 cotto that should give me enough to start playing with
10:00 cotto nope, no scheme for me
10:00 barney I'm just fiddling with public class members.
10:00 Ademan joined #parrot
10:01 barney Finding out whether PAST supports protected members will be next
10:01 cotto for PHP's "OO" support?
10:01 barney Yes.
10:01 cotto I can't imagine that anything designed for perl6 wouldn't handle PHP, but I haven't tried yet.
10:02 barney AFAIK Perl 6 has not protected members, which is good
10:03 cotto ok.  It's 0300 here and I need sleep.  Thanks for the answers.
10:03 cotto night
10:03 barney Gute Nacht
10:03 purl Gute nacht, barney
10:22 gaz joined #parrot
10:22 bacek joined #parrot
10:33 dalek r31921 | bernhard++ | trunk:
10:33 dalek : [Pipp] Class names don't need to be split on '::'.
10:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31921
10:40 dalek r31922 | bernhard++ | trunk:
10:40 dalek : [Pipp] use correct class name in test case.
10:40 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31922
10:46 dalek r31923 | bernhard++ | trunk:
10:46 dalek : [Pipp] Paamayim Nekudotayim is not allowed in class names.
10:46 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31923
11:01 bacek joined #parrot
11:14 dalek r31924 | bernhard++ | trunk:
11:14 dalek : [Pipp] Put class methods into an immediate block, as seen on Rakudo.
11:14 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31924
11:24 bacek joined #parrot
11:48 dalek r31925 | bernhard++ | trunk:
11:48 dalek : [Pipp] Start to put member declaration into the symbol table.
11:48 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31925
12:16 xiaoyafeng joined #parrot
12:22 iblechbot joined #parrot
12:23 magnachef__ joined #parrot
12:53 dalek bernhard.schmalhofer@gmx.de | Pipp:
12:53 dalek link: http://www.perlfoundation.​org/parrot/index.cgi?pipp
13:09 gryphon joined #parrot
13:17 Andy joined #parrot
13:26 Zaba joined #parrot
13:30 PacoLinux joined #parrot
13:34 Tene Oogh.  This is way too early to be awake.
13:45 jhorwitz joined #parrot
13:47 dmknopp joined #parrot
14:11 TiMBuS joined #parrot
14:36 Andy joined #parrot
14:47 tetragon joined #parrot
14:48 omega joined #parrot
15:04 selenamarie joined #parrot
15:18 coke joined #parrot
15:34 sjansen joined #parrot
15:45 Zaba joined #parrot
16:00 selenamarie left #parrot
16:05 Lorn joined #parrot
16:15 dalek r31926 | jhorwitz++ | trunk:
16:15 dalek : update NCI signatures for recent mod_parrot changes
16:15 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31926
16:15 tetragon joined #parrot
16:16 particle joined #parrot
16:18 ruoso joined #parrot
16:34 Lorn joined #parrot
16:43 barney joined #parrot
16:50 Theory joined #parrot
17:19 Coke who did the recent :: -> '; ' changes?
17:25 Tene pmichaud and me.
17:25 Tene I would have updated tcl, but, it's not in the tree anymore.
17:25 Tene Jerk.
17:25 Tene ;)
17:26 Coke Can you peek at http://rt.perl.org/rt3/Tic​ket/Display.html?id=59870 ?
17:26 Tene I'd be glad to help you with it, though.
17:26 Coke danke.
17:26 Tene Sure.
17:26 Tene except a bit later.  teaching now.
17:26 Coke no rush, I'z at work.
17:31 Tene Wait, is tcl in the tree at all anymore?
17:31 Tene Ah, no it's not.
17:31 Coke nope.
17:31 Tene Where can I get some partcl to look at?
17:31 Tene Oh, attachment
17:31 purl attachment is probably reality.
17:31 Coke you don't need it for that ticket.
17:32 Coke but if you want to play, here: http://code.google.com/p/partcl/wiki/PartclSource
17:32 Coke partcl?
17:32 purl partcl is probably tcl on parrot or being discussed on lily (see lily) in -partcl. or ask coke for details
17:32 Coke no, partcl is tcl on parrot
17:32 purl okay, Coke.
17:32 Coke partcl is also http://code.google.com/p/partcl
17:32 purl okay, Coke.
17:50 Coke if anyone would like to try out the new partcl arrangement, I'd certainly appreciate it. =-)
17:53 ab5tract joined #parrot
17:58 Theory joined #parrot
18:13 Ivatar joined #parrot
18:14 Zaba joined #parrot
18:36 peepsalot joined #parrot
18:49 ab5tract joined #parrot
18:55 PacoLinux Coke: I cant access to partcl svn, insists in ask for a user/passwd
18:58 PacoLinux r:~/parrot/languages$ svn checkout https://partcl.googlecode.com/svn/trunk/ tcl -> Authentication realm: <https://partcl.googlecode.com:443> Google Code Subversion Repository -> Password for 'paco':
19:03 cotto PacoLinux, try svn checkout http://partcl.googlecode.com/svn/trunk/ partcl-read-only
19:03 cotto that one worked for me
19:04 Coke PacoLinux: "what cotto said". See: http://code.google.com/p/partcl/source/checkout
19:04 PacoLinux cotto: thx, this method works :)
19:04 Coke I'll update the notes on my copy because I think they're wrong.
19:29 donaldh joined #parrot
19:39 tetragon t/native_pbc/header.t doesn't like my 64-bit build.  The build's wordsize is too large
19:48 Infinoid I'm seeing that too
19:49 tetragon I've filed #59876 over that
19:49 tetragon (and #59874 over my recent 32-bit failure)
19:57 Infinoid do we have parallel testing yet?  I've been out of the loop for a few months.
19:59 Infinoid hmm
19:59 Infinoid #   Failed test 'I-reg shl and PMC shl are consistent'
19:59 Infinoid #   at t/op/bitwise.t line 513.
19:59 Infinoid # Exited with error code: [SIGNAL 11]
19:59 tetragon I just occasionally do a couple builds.  It's been a while since last did a set, but I'm taking a bit of a break from a talk I'm preparing and getting JIT working with Intel OS X
20:00 Infinoid cool
20:00 tetragon No failures or crashes for me on t/op/bitwise.t on either 32- or 64- bit builds
20:01 Infinoid awesome, a DOD crash
20:01 tetragon The first JIT stackframe is the wrong length for Intel OS X to like it
20:02 tetragon Most of my current crashes are from the pmc tests
20:02 nopaste "Infinoid" at 96.238.213.50 pasted "DOD crash in t/op/bitwise.t on linux/x86_64" (66 lines) at http://nopaste.snit.ch/14281
20:04 tetragon Does test 54 of t/pmc/complex.t crash on you?
20:05 Infinoid yes
20:05 tetragon It always segfaults on my 64-bit build, and either aborts or segfaults on my 32-bit build (depending upon how it's invoked)
20:05 Infinoid I think I discussed that with pmichaud last week, and he has a ticket for it
20:06 Infinoid I took a brief stab at it but couldn't figure out the problem
20:07 Infinoid tetragon: http://rt.perl.org/rt3/Tic​ket/Display.html?id=59630
20:11 nopaste "tetragon" at 206.248.132.107 pasted "Different failures of t/pmc/complex.t 54 on 32-bit OS X builds" (203 lines) at http://nopaste.snit.ch/14282
20:15 magnachef__ joined #parrot
20:15 Infinoid the difference seems to be whether inter_call.c:388 gets around to calling Parrot_Confess or not
20:16 Infinoid crashes at more or less the same spot, either way
20:16 bacek joined #parrot
20:17 Coke what do folks think about adding a USAGE attribute to Subs and other invokables that was used as the exception message when invoked with a wrong number of args?
20:17 Coke would any other HLLs find that useful?
20:17 tetragon And this behaviour is consistent.  It always trips the assertion if I run it from the command line or prove (non-verbose).  It always segfaults if I use prove -v
20:21 dalek r31927 | cotto++ | trunk:
20:21 dalek : [pipp] * get_repr implementation and tests
20:21 dalek : * remove unused macro
20:21 dalek : * change table ATTR to hash_table
20:21 dalek : * add debugging symbols back to src/pipp_hash.o
20:21 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31927
20:22 davidfetter joined #parrot
20:30 Coke PacoLinux, cotto: I updated http://code.google.com/p/partcl/wiki/PartclSource
20:31 johbar joined #parrot
20:49 grim_fandango joined #parrot
21:32 bacek joined #parrot
21:54 ruoso joined #parrot
22:23 TiMBuS joined #parrot
22:24 Zaba joined #parrot
22:33 particle1 joined #parrot
22:41 tetragon joined #parrot
23:41 nopaste joined #parrot
23:57 bacek_ joined #parrot

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

Parrot | source cross referenced