Camelia, the Perl 6 bug

IRC log for #parrot, 2008-08-19

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:06 davidfetter joined #parrot
00:08 bacek joined #parrot
00:09 AndyA joined #parrot
00:33 dalek Andrew Whitworth | pdd27mmd_tasklist:
00:33 dalek link: http://www.perlfoundation.org/pa​rrot/index.cgi?pdd27mmd_tasklist
00:34 Whiteknight I should have registered on the wiki as "Whiteknight"
00:34 Whiteknight meh, internet pseudoanonyminity isn't all it's cracked up to be
01:11 rurban_ joined #parrot
01:14 particle1 joined #parrot
01:19 kid51 joined #parrot
01:19 dalek r30316 | jkeenan++ | scriptconfigure:
01:19 dalek : Add additional test of file-based configuration.  Begin to document file-based configuration in Configure.pl POD.
01:19 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30316
01:20 Zaba_ joined #parrot
01:53 dalek r30317 | Whiteknight++ | trunk:
01:53 dalek : [core] Add some function-level documentation in src/inter_call.c for functions that I understand. There are still some functions that are undocumented, however.
01:53 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30317
01:55 bacek seen pmichaud
01:55 purl pmichaud was last seen on #parrot 2 hours, 28 minutes and 56 seconds ago, saying: it's probably about time for me to do another :-)
02:03 Tene http://www.austinlug.org/node/293
02:04 dalek r30318 | Whiteknight++ | trunk:
02:04 dalek : [core] Add some function-level documentation in src/library.c for functions that I understand. There are still some functions that are undocumented, however.
02:04 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30318
02:32 dalek r30319 | pmichaud++ | trunk:
02:32 dalek : Rakudo and PCT NEWS updates for 0.7.0 release.
02:32 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30319
02:36 Tene orite, NEWS.
02:50 dalek r30320 | tene++ | trunk:
02:50 dalek : NEWS updates for Cardinal.
02:50 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30320
02:55 dalek r30321 | jkeenan++ | scriptconfigure:
02:55 dalek : In configuration files, the sections now begin with a single '=' sign rather
02:55 dalek : than a double '==' (more like POD).  Add explanation of file-based
02:55 dalek : configuration to Configure.pl.  Modify tests to work with single '=' sign.
02:55 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30321
02:57 dalek r30322 | pmichaud++ | trunk:
02:57 dalek : Another NEWS update for Rakudo.
02:57 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30322
03:52 StephenPollei joined #parrot
04:16 Andy joined #parrot
04:17 Andy ping pmichaud
04:17 purl I can't find pmichaud in the DNS.
04:34 pmichaud pong
04:41 pmichaud (off to bed, back later)
05:00 Theory joined #parrot
05:18 jkva joined #parrot
05:34 Psyche^ joined #parrot
05:53 tetragon joined #parrot
06:26 uniejo joined #parrot
06:28 ank joined #parrot
06:51 tetragon joined #parrot
07:01 dalek allison@perl.org | hackathon_yapc_eu_2008:
07:01 dalek link: http://www.perlfoundation.org/parro​t/index.cgi?hackathon_yapc_eu_2008
07:13 tetragon joined #parrot
07:27 masak joined #parrot
07:32 tetragon joined #parrot
08:08 desertmax joined #parrot
08:27 tetragon joined #parrot
08:35 Zaba joined #parrot
08:59 krunen joined #parrot
09:10 rurban_ joined #parrot
09:18 Whiteknight joined #parrot
09:20 tetragon joined #parrot
09:40 tetragon joined #parrot
10:27 tetragon joined #parrot
10:28 desertmax joined #parrot
10:33 iblechbot joined #parrot
10:46 dalek r30323 | pmichaud++ | trunk:
10:46 dalek : [rakudo]: spectest-progress.csv update: 121 files, 2220 passing tests
10:46 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30323
10:49 dalek r30324 | pmichaud++ | rakpbc:
10:49 dalek : temporary branch for Test.pbc impl to avoid interference with Aug 2008 release
10:49 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30324
10:57 AndyA joined #parrot
11:07 tetragon joined #parrot
11:13 tetragon joined #parrot
11:22 dalek r30325 | coke++ | trunk:
11:22 dalek : [docs] RT #58078, update Solaris's status.
11:22 dalek : Courtesy Andy Dougherty
11:22 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30325
11:25 AndyA joined #parrot
11:27 tetragon joined #parrot
11:33 dalek r30326 | julianalbo++ | trunk:
11:33 dalek : parrot_debugger now loads the source automatically when launched with pir or pasm argument
11:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30326
11:37 Whiteknight joined #parrot
11:51 cognominal joined #parrot
11:54 bacek joined #parrot
12:15 bacek evening everyone
12:25 dalek r30327 | fperrad++ | trunk:
12:25 dalek : [Lua]
12:25 dalek : - fix compiler (broken by r30297 : PCT/POST)
12:25 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30327
12:30 pmichaud good evening, bacek
12:30 bacek pmichaud: hey! It should be morning in US :)
12:31 pmichaud it is morning in US
12:31 dalek r30328 | pmichaud++ | rakpbc:
12:31 dalek : [pct]: Make anonymous PAST::Block nodes be :anon in PIR.
12:31 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30328
12:31 dalek r30329 | fperrad++ | trunk:
12:31 dalek : [Lua]
12:31 dalek : update NEWS for 0.7.0
12:31 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30329
12:31 bacek pmichaud: when you'll have spare time, can you pleas review my impl of StrPos in #57936?
12:33 bacek I don't like get_string/etc methods, but can't find other ways to not break everything.
12:35 dalek r30330 | fperrad++ | trunk:
12:35 dalek : [PLATFORM]
12:35 dalek : - update for MinGW gcc 3.4.5
12:35 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30330
12:38 dalek r30331 | pmichaud++ | rakpbc:
12:38 dalek : [rakudo]: running load_bytecode on a precompiled perl6 file runs the mainline
12:38 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30331
12:40 dalek r30332 | pmichaud++ | rakpbc:
12:40 dalek : [abc]:  Correct svn:ignore properties on src/
12:40 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30332
12:46 smash joined #parrot
12:46 * smash hello everyone
12:49 dalek r30333 | coke++ | trunk:
12:49 dalek : [t] lua is now passing all tests.
12:49 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30333
12:50 masak rakudo: my $a; given 5 { when 5 { $a = "should be this" }; $a = "but is this" }; say $a
12:50 polyglotbot OUTPUT[but is this␤]
12:50 masak when blocks do not exit out of the surrounding given if they match
12:51 masak is this a known shortcoming?
12:51 jonathan masak: yes
12:52 jonathan when doesn't throw a control exception at the end yet
12:52 masak ah -- exceptions
12:52 purl exceptions are a special case of continuations.
12:52 masak then I guess it's not so trivial to fix
12:53 masak otherwise I might have had a go at it
12:54 jonathan no, it's non-trivial, waiting on pm's exception changes in PCT
12:54 dalek r30334 | coke++ | trunk:
12:54 dalek : [docs] actually show verbose output, and pick a file that actually exists as a sample.
12:54 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30334
12:54 dalek r30335 | coke++ | trunk:
12:54 dalek : [docs] remove pointer to language no longer in the repository
12:54 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30335
12:58 masak will there be a parrot release today?
13:01 jonathan Yes.
13:01 jonathan I believe so.
13:03 pmichaud we probably want/need to implement .leave for given/when
13:04 AndyA joined #parrot
13:04 masak ah, ok
13:10 dalek r30336 | coke++ | trunk:
13:10 dalek : [t] revert r24988; no test harnesses support --master, and at least one was reporting it as an invalid option during 'make test' in languages.
13:10 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30336
13:16 Andrew_ joined #parrot
13:17 dalek r30337 | rgrjr++ | RELEASE_0_7_0:
13:17 dalek : Starting release 0.7.0
13:17 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30337
13:19 gryphon joined #parrot
13:23 dalek r30338 | coke++ | trunk:
13:23 dalek : [t] Sigh. unrevert last commit, as --master is defined elsewhere. Add a note to the next person eager to rip out code.
13:23 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30338
13:24 smash (running fulltest on branches/RELEASE_0_7_0/)
13:24 pmichaud uh oh
13:24 pmichaud does rgrjr's 0.7.0 branch include Coke's unreverted commit ?
13:25 smash not it does not
13:25 pmichaud why not?
13:26 smash it seems rfrjt copy from branch was before coke's unrevert
13:26 pmichaud that's my point
13:26 pmichaud 30336 committed a change
13:26 pmichaud 30337 is a copy
13:26 pmichaud 30338 reverted 30336
13:26 pmichaud which means that 30337 (the release) has a change that needs to be reverted
13:26 smash it seems so
13:28 smash is rgrjr on irc ?
13:32 pmichaud I sent a note, and coke++ did as well.
13:32 pmichaud hopefully rgrjr will catch them :-)
13:33 smash i'm sure he will
13:34 Zaba joined #parrot
13:44 rdice joined #parrot
13:45 dalek r30339 | rgrjr++ | RELEASE_0_7_0:
13:45 dalek : [NEWS] Make style consistent with recent releases.
13:45 dalek :    + Move "Specification" before "Implementation".
13:45 dalek :    + Fix capitalization of Cardinal, Pipp, and Tcl bullets.
13:45 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30339
13:56 Wknight8111 Is that 7_0 release branch going to be merged back into trunk, or is it going to be a snapshot?
13:57 pmichaud release branches are usually snapshots
13:57 pmichaud and normally they go in tags/, not branches/
13:57 pmichaud but yes, you're correct that the files need to be merged to trunk
13:57 * Wknight8111 learns something new every day
13:58 pmichaud I suspect rgrjr is using a branch to prepare the release, will snapshot it into tags/, and then merge into trunk/
13:58 smash pmichaud: i'm guessing the branch is still going to be merged, and only after that the tag will be created
13:58 smash exactly
13:59 pmichaud I tend to prefer doing all of the release work in trunk, myself
13:59 pmichaud it doesn't take that long
14:02 Zaba joined #parrot
14:08 * pmichaud wonders if it's really safe to be working in trunk again
14:10 dalek r30340 | rgrjr++ | RELEASE_0_7_0:
14:10 dalek : [t] Sigh. unrevert last commit, as --master is defined elsewhere. Add a
14:10 dalek : note to the next person eager to rip out code.  [per coke, merged r30338
14:10 dalek : from trunk.]
14:10 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30340
14:10 smash hmm, some tests failling besides coding standards
14:11 pmichaud nopaste?
14:11 clunker3 http://pasta.test-smoke.org/ or http://paste.husk.org/ or http://nopaste.snit.ch:8001/ or http://rafb.net/paste or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/
14:11 purl i heard nopaste was at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or App::Nopaste or tools/dev/nopaste.pl
14:19 nopaste "smash" at 213.205.93.24 pasted "make fulltest output for release 0.7.0 branch" (2536 lines) at http://nopaste.snit.ch/13836
14:20 pmichaud the error is in one of the alternate runcores -- we often release even if that's the case
14:21 * smash nods.
14:25 smash running tests on windows now
14:25 particle1 pmichaud: yes, it's safe to work in trunk. chip used this method for releases, btw
14:25 pmichaud I meant wrt the fact that some patches haven't made it into the release branch yet
14:26 pmichaud as long as rgrjr doesn't try to resync trunk with the release branch, we're probably okay
14:26 pmichaud er, vice-versa
14:27 pmichaud but okay, based on that I'll go ahead and put my changes into trunk
14:27 davidfetter joined #parrot
14:32 masak rakudo: if 1 { .say }
14:32 polyglotbot OUTPUT[Null PMC access in find_method()␤current instr.: '_block18' pc 82 (EVAL_13:33)␤called from Sub '_block11' pc 31 (EVAL_13:17)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:481)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1078
14:32 polyglotbot ..(src/PCT/HLLCompiler.pir:610)␤called from Sub 'parrot;PCT::HLLCompiler;...
14:33 Tene rakudo: if 1 { say $_ }
14:33 polyglotbot OUTPUT[␤]
14:33 dalek r30341 | pmichaud++ | trunk:
14:33 dalek : merge rakpbc branch into trunk
14:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30341
14:33 dalek r30342 | pmichaud++ | rakpbc:
14:33 dalek : Remove obsolete rakpbc branch.
14:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30342
14:34 gryphon joined #parrot
14:34 masak Tene: ah, in your case a null check is inserted in PIR code
14:35 * masak enjoys his new tool 'pirdiff' for checking these things
14:35 Tene Is $_ not getting set to the result of the conditional?
14:35 pmichaud not yet.
14:35 masak ah, and it will work once it is? good.
14:35 pmichaud oh, and in this case I don't think it gets set to the result of the conditional.
14:36 masak no, it doesn't
14:36 pmichaud i.e., it's not supposed to do so.   To make $_ set to the result of the conditional, use
14:36 pmichaud if 1 -> $_ { say $_ }
14:36 pmichaud although I don't think that works yet either :-)
14:36 masak aha
14:37 pmichaud I just need to make a few PCT changes to enable that
14:37 masak so, `if 1 { .say }` should be semantivally equal to `.say`
14:37 pmichaud yes
14:37 pmichaud from S04:
14:37 pmichaud However, use of $_ with a conditional statement's block is not considered sufficiently explicit to turn a 0-ary block into a 1-ary function, so both these methods use the same invocant:
14:38 pmichaud if .haste { .waste }
14:39 masak mm
14:39 Wknight8111_ joined #parrot
14:39 masak how do I run PIR code directly, again?
14:39 pmichaud add a 'load_bytecoe "perl6.pbc"' near the top of the first sub
14:39 pmichaud er, load_bytecode
14:39 masak pmichaud: gracias
14:39 pmichaud getting standalone pir to work is pretty close now
14:40 pmichaud I'm almost to the point of getting precompiled .pm files to work, but src/builtins/io.pir needs some refactoring
14:42 masak cool!
14:44 desertmax_ joined #parrot
14:44 masak re the Null PMC Access of `if 1 { .say }`: seems that the if statement is turned into a .sub in PIR code, and this sub goes to fetch $_ from its outer block, where it was created. for some reason, this fails, although creating it anew in that .sub works.
14:46 kj joined #parrot
14:48 pmichaud $_ is initialized to null in any Routine
14:48 masak Tene's version survives because there an unless_null check is made in the PIR
14:48 masak I'll check why it's done in the `say($_)` case but not in the `.say` case
14:48 pmichaud correct, I suspect there's a missing :viviself parameter on the .method
14:49 pmichaud yes, that's it
14:51 pmichaud $ ./parrot perl6.pbc -e 'if 1 { .say }'
14:51 pmichaud
14:51 pmichaud now works
14:51 pmichaud (committing)
14:51 masak wohoo!
14:51 masak http://rt.perl.org/rt3/Tic​ket/Display.html?id=57860
14:52 pmichaud r30343
14:52 pmichaud of course, there's a good argument to be made that it should throw an exception
14:53 pmichaud (use of uninitialized value at ... )
14:53 masak yes.
14:53 dalek r30343 | pmichaud++ | trunk:
14:53 dalek : [rakudo]: Fix .method call when $_ uninitialized (RT #57860, masak++ Tene++)
14:53 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30343
14:53 particle1 that argument should be made in the test suite :)
14:53 pmichaud it will do that as soon as I get Failure updated
14:53 masak particle1++
14:54 masak pmichaud: confirmed. works here too, now. thanks!
14:54 particle pmichaud: does test_summary include the time to run the tests?
14:54 pmichaud particle: not currently
14:54 particle i have a feeling that will be an important metric soon :)
14:55 pmichaud heh
14:55 pmichaud well, it should speed up dramatically soon, also :-)
14:55 pmichaud I also want to update test_summary to report passing todo's
14:55 pmichaud maybe I'll do that now -- right now there's a bit too much commotion for me to be able to focus on the .pbc stuff
14:55 pmichaud (around the house, that is)
14:56 pmichaud oh, wait, I supposedly already have code to do that
14:56 pmichaud wonder why it's not working....
14:56 particle heh
14:57 dalek r30344 | coke++ | RELEASE_0_7_0:
14:57 dalek : [tcl] temporarily TODO some failing tests for the release.
14:57 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30344
14:57 pmichaud does time-of-test-suite need to be a new column in spectest-progress.csv ?
14:58 AndyA joined #parrot
15:01 particle hrmm. only if that file is run from the same platform every time
15:02 particle otherwise it doesn't make sense there
15:02 pmichaud good point
15:02 particle should i expect failures in spectest_regression?
15:02 pmichaud not at the moment
15:03 particle ooh, i get a bonus test in S29-list/reverse
15:03 particle oh, they're all bonus tests, not failures
15:04 rurban coke: yesterday I also pasted some cygwin tcl failures for inexact floats. http://nopaste.snit.ch/13835
15:04 particle t\spec\S06-signature\named-parameters.rakudo
15:04 particle t\spec\S16-io\basic-open.rakudo
15:04 particle t\spec\S29-list\reverse.rakudo
15:04 nopaste "smash" at 81.84.255.161 pasted "make test output for release 0.7.0 branch on windows with strawberry perl" (15 lines) at http://nopaste.snit.ch/13838
15:05 rurban For opengl on cygwin I fixed one more thing besides #57006. the header detection logic is also wrong. But I guess this will not applied anyway today.
15:07 Theory joined #parrot
15:07 nopaste "rurban" at 212.183.62.244 pasted "57006-opengl-cyg.patch" (133 lines) at http://nopaste.snit.ch/13840
15:09 donaldh joined #parrot
15:10 rurban I'll do some short surf now and be back in the evening.
15:12 Zaba joined #parrot
15:27 Andy joined #parrot
15:41 md3 joined #parrot
15:42 md3 hi guys, I have a question about "stacklessness" of parrot
15:42 md3 I would like to implement language which can handle many (millions) of green threads
15:42 md3 would that be possible with parrot?
15:43 tewk I don't see why not.
15:43 japhb_ rurban: I have to leave now, but before I go:  Please work things out to consensus with donaldh, and any other Win32 users you can find, re: OpenGL.  No more fighting patches, please!  :-)
15:44 tewk But You would probably have to write your own thread and concurrency_schedular classes/pmcs.
15:44 md3 tewk: must any of the green thread have a stack?
15:44 md3 hmm
15:44 md3 scheduler is not the problem
15:45 md3 and is there a way how to tell parrot to allocate function call frames on heap?
15:45 md3 this would require a specific GC as well probably, but that's not a problem as well
15:46 tewk We have a new concurrency_scheduler that supports asynchronous taks, but I'm not sure if they have all the green thread support you would want.
15:46 jonathan md3: Parrot allocates contexts (which are its equivalent to call frames) on the heap.
15:46 md3 jonathan: that's great!
15:47 donaldh japhb_: at the very least, I think we'll need to probe installed packages on cygwin and report conflicts
15:52 jonathan md3: However, I don't think anyone has tried to do anything as thread-heavy as you are suggesting with Parrot yet.
15:55 md3 jonathan: Parrot is very temptating, so I guess sooner or later somebody will try to experiment with it
15:55 Zaba joined #parrot
15:57 jonathan If you're willing to stumble on some bugs here and there, then certainly - please do experiment! :-)
15:58 dalek r30345 | rgrjr++ | RELEASE_0_7_0:
15:58 dalek : [release] Clean up "make fulltest" (see list discussion).
15:58 dalek : * config/gen/makefiles/root.in:
15:58 dalek :    + (fulltest):  Run codetest instead of codingstd_tests.
15:58 dalek : * docs/project/release_manager_guide.pod:
15:58 dalek :    + So we don't need to say "failures are OK" here.
15:58 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30345
15:59 md3 well, surely I will! I don't foresee much free time for that, at least not the following couple months, but who knows ;-)
16:00 jonathan :-)
16:01 rurban The recent condigstd fixes fixed my remaining cygwin errors now. 0 FAILED. languages 72.80% (954 failed) lua mainly
16:03 rurban japhb_: I mainly have to fix your design, not donaldh's patches :)
16:04 rurban The problem with opengl is that the header parsing should be coupled to the libs selection. headers belong to libs. If you add sigs from headers which are not used we'll get into troubles.
16:05 rurban cygwin has two GL options, so we have the trouble.
16:05 donaldh rurban: agreed.
16:06 donaldh I'm not sure we get to choose between the GL options if both are installed. I.e. one set of includes preempt the others in the default search path.
16:07 dalek r30346 | rgrjr++ | RELEASE_0_7_0:
16:07 dalek : Update version numbers etc. for Parrot release 0.7.0 "Severe Macaw".
16:07 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30346
16:09 donaldh rurban: is that fixable?
16:11 rurban rigth now I made the simple userchoice: DISPLAY=bla + freegklu installed => freeglut, else try native winapi GLUT, else no opengl. better would be a new configure option: ---gl=
16:11 rurban --gl=freeglut,none,GLUT
16:12 donaldh that seem reasonable.
16:12 rurban my current patch works if both are installed now. I just remove the freeglut headers from the search path if GLUT is used
16:12 rurban but I have to leave now for about 2hrs. bye
16:13 donaldh bye
16:59 jkva_ joined #parrot
17:06 dalek r30347 | julianalbo++ | trunk:
17:06 dalek : quick fix for 'script' debugger command
17:06 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30347
17:08 moritz re
17:10 rurban_ joined #parrot
17:15 jkva_ left #parrot
17:33 Casan joined #parrot
17:34 Andy joined #parrot
17:55 ruoso joined #parrot
18:01 Wknight8111 there's a #ps today, right?
18:03 barney joined #parrot
18:08 NotFound Wknight8111: at least I don't see any message cancelling it.
18:09 dalek r30348 | julianalbo++ | trunk:
18:09 dalek : add an assertion that prevents segfault RT#58044
18:09 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30348
18:09 Wknight8111 Thanks, NotFound. I couldn't remember what time it was supposed to be, and I showed up early
18:10 jonathan eek, I need to go out...
18:10 * jonathan pre-reports
18:11 jonathan done
18:11 jonathan may make it for the end of the meeting/questions if it drags on enough ;-)
18:12 NotFound joined #parrot
18:13 NotFound I've found a bug... in konversation
18:20 dalek r30349 | allison++ | pdd27mmd:
18:20 dalek : [pdd27mmd] Apply a patch from Andrew Whitworth changing the name of
18:20 dalek : 'Parrot_mmd_sort_candidate_list' to 'Parrot_mmd_sort_manhattan'. Resolves RT
18:20 dalek : #58088.
18:20 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30349
18:22 Andy joined #parrot
18:23 dalek allison@perl.org | pdd27mmd_tasklist:
18:23 dalek link: http://www.perlfoundation.org/pa​rrot/index.cgi?pdd27mmd_tasklist
18:26 pmichaud #ps in 4.
18:28 moritz pmichaud, particle: I just filled out the gsoc final evaluation
18:28 pmichaud moritz++  Auzon++
18:29 moritz I think there's nothing controversial in there, but if you want to know and/or discuss what I wrote I can send you my answers
18:29 pmichaud I'm glad to comment if you want my opinion, but otherwise I think it's fine
18:31 pmichaud moritz:  re   cloning behavior of Array
18:31 pmichaud if I have
18:31 moritz re comment, no need
18:31 pmichaud class Foo { ... };   my $a = Foo.new;    my $b = [42];   $b[1] = $a;
18:31 pmichaud do we expect $a to be cloned or not?
18:31 allison joined #parrot
18:32 moritz no
18:32 pmichaud what about with
18:32 moritz $b[1] is just a normal container, and assignment generally doesn't clone
18:32 pmichaud my $a = Integer.new;  my $b = [42];   $b[1] = $a;
18:32 moritz on scalars
18:32 moritz do you mean Int?
18:33 pmichaud er, yes, Int
18:33 moritz Int is an immutable type
18:33 moritz so it doesn't make a difference
18:33 dalek r30350 | coke++ | trunk:
18:33 dalek : [tcl] remove useless use of filename in a void context. Also eliminate a chunk of inadvertently repeated textinadvertently repeated text
18:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30350
18:34 pmichaud okay, works for me.
18:35 moritz what's more interesting
18:35 purl more interesting is 2) Rather than moving a piece, you may place an opponent's captured piece back on the board at an empty space, as a new piece for you.
18:35 moritz assigning an '4 but Foo'
18:35 moritz where Foo has attributes
18:35 jhorwitz joined #parrot
18:35 Wknight8111 purl forget more interesting
18:35 purl Wknight8111: I forgot more interesting
18:40 dalek r30351 | moritz++ | trunk:
18:40 dalek : [rakudo] added two S02 tests to spectest_regression, s1n++
18:40 dalek : Patch Courtesy by Jason Switzer (RT #58094)
18:40 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30351
18:44 DietCoke joined #parrot
18:44 DietCoke left #parrot
18:49 smash joined #parrot
18:50 smash hello everyone
18:51 Wknight8111 hello smash
18:58 dalek r30352 | allison++ | pdd27mmd:
18:58 dalek : [pdd27mmd] Apply a patch from Andrew Whitworth to merge
18:58 dalek : 'mmd_create_builtin_multi_meth2' into 'mmd_create_builtin_multi_meth'. Resolves
18:58 dalek : RT #58090.
18:58 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30352
19:04 Wknight8111 I would have committed those changes myself, but i wasn't entirely certain that I was doing the right thing
19:05 allison Wknight8111: that's okay. Once I've reviewed and applied the patch, and run all the tests, committing it is a few keystrokes extra.
19:06 allison Wknight8111: I'm in the middle of replying to your email question now.
19:06 Wknight8111 excellent.
19:08 particle ah crap! i missed parrotsketch
19:08 particle darned $work!
19:08 pmichaud particle:  you can paste your report now -- I'll watch.  :-)
19:09 particle :P
19:09 particle i want to talk about merging tewk's work to trunk
19:09 dalek r30353 | rgrjr++ | RELEASE_0_7_0:
19:09 dalek : tagged release 0.7.0
19:09 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30353
19:09 particle allison, coke, any objections?
19:10 particle joined #parrot
19:12 allison I haven't looked at it yet
19:12 allison but, I trust your code review
19:14 particle ok
19:15 particle that'll be my "copious free time" task
19:15 particle the in-laws arrive at midnight.
19:15 particle the in-laws leave on labor day.
19:18 allison I'll try building the branch tonight
19:20 particle rgrjr++
19:21 smash rgrjr++ # release
19:27 Tene pmichaud: there are resumable exceptions in parrot right now, mostly.
19:28 Tene pmichaud: the last issue is with some awkwardness about the handlers not being managed properly before resuming, as the expectation was that after a handler exited, it would either exit or call other handlers and then exit.
19:28 Tene I can details more in a bit, teaching again.
19:30 Tene I keep getting distracted with other interesting tasks when I try to work on a proposal or documentation for this.
19:30 Tene Also, I don't have a specific goal for what I'm trying to produce when talking about architecture changes.  That makes it a bit harder.
19:31 Tene I'll try again tonight.
19:37 allison joined #parrot
19:45 gryphon joined #parrot
19:50 jonathan <@moritz> assigning an '4 but Foo' # clones the Int before applying the role
19:52 moritz jonathan: we were talking about the case my $a = []; my $b = 4 but Foo; $a[0] = $b; # does the last assignment clone the stuff in Foo?
19:58 nopaste "tewk" at 155.97.237.62 pasted "OSX i386 gsoc_nci_001 build error, anyone seen this before" (5 lines) at http://nopaste.snit.ch/13847
20:04 allison tewk: looks suspiciously like a dylib compiled for the wrong architecture
20:08 particle joined #parrot
20:09 tewk Its not my machine, I'll have to look deeper,
20:09 particle irc log?
20:09 purl irc log is http://irclog.perlgeek.de/parrot/
20:11 rurban kid51, here?
20:11 rurban the parrot perms for cpan are missing and most of theme belong to former parrot uploaders.
20:12 particle i hate that dance.
20:12 smash particle: true
20:12 rurban I just wrote to modules@cpan and p5p so that they also can fix it
20:13 rurban Theree are missing for rgrjr: Parrot::Configure::Options::Test::Prepare Parrot::Harness::DefaultTests  Parrot::Harness::Options  Parrot::Harness::Smoke  Parrot::Pmc2c::Attribute  Parrot::Pmc2c::PMC::PrintTree  Parrot::SearchOps  Parrot::Test::Util::Runloop
20:14 rurban owners please add rggrjr as co-maint for these. kid is the owner of most
20:15 rurban PARROTRE should also be able to fix that as I see on pause
20:17 particle joined #parrot
20:17 particle no_index:
20:17 particle directory:
20:17 particle - lib
20:18 particle stupid computers. why are you dropping my network connection?
20:18 Tene It's slippery!  Stop soaping your packets!
20:25 tewk jit isn't even enabled by default on OSX
20:33 dalek Bob Rogers | Parrot:
20:33 dalek link: http://www.perlfoundation.o​rg/parrot/index.cgi?parrot
20:33 rgrjr joined #parrot
20:34 Topic for #parrotis now http://www.parrotcode.org | Last Release: 0.7.0 "Severe Macaw"
20:37 Whiteknight joined #parrot
20:41 dalek rurban@cpan.org | Parrot:
20:41 dalek link: http://www.perlfoundation.o​rg/parrot/index.cgi?parrot
20:59 rurban smash: How's the debian package going?
21:01 smash rurban: haven't started packaging it yet
21:02 rurban I've got testing failures when accessing the parrot dll at first, aargh
21:02 japhb_ rurban, donaldh: I just backlogged.  It sounds like you two have a decent plan -- as long as donaldh's "include path priority" problem is dealt with (so we never get the include path trying to outthink our decision), I'm happy.  I just want to see the end of this, so we can get on to more -Ofun.
21:02 japhb_ :-)
21:04 rurban japhb_: I believe --gl is not good enough. --gl-include and --gl-libs should be it. On other platforms with gl headers in /usr/include/GL and /usr/local/include/GL we have the same problem. We have to find matches somehow as a proper autoconf would do it.
21:05 rurban for now my cygwin patch is okay, but the next release we should solve the problem not to parse all headers, only the hedares which match the given libs
21:06 rurban Ok, I just gave up on packaging for cygwin today. It will be tomorrow.
21:06 rurban bye!
21:09 dalek r30354 | Whiteknight++ | gsoc_pdd09:
21:09 dalek : [gsoc_pdd09] update to trunk from r30310
21:09 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30354
21:10 japhb_ rurban: I'm fine with doing what you say.  For historical info ... I did not do this originally, because at the time I started, there was no way to pass this kind of information between Configure steps, only from Configure to Makefile.  Hopefully this has changed.
21:10 allison joined #parrot
21:11 dalek r30355 | allison++ | trunk:
21:11 dalek : [pdd] Adding more details on signatures and the "find next" feature to the
21:11 dalek : multi dispatch PDD.
21:11 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30355
21:14 dalek r30356 | Whiteknight++ | pdd09gc:
21:14 dalek : Creating a new GC-related branch to work in. Will merge/remove branches/gsoc_pdd09 in time.
21:14 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30356
21:38 gmansi joined #parrot
21:39 donaldh joined #parrot
22:19 dalek r30357 | Whiteknight++ | pdd27mmd:
22:19 dalek : [PMC] Add a basic get_iter vtable method for the multisub pmc. Needs tests. Also, added a quick stub and some comments for a get_iter method.
22:19 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30357
22:31 iblechbot joined #parrot
22:32 dalek mestre.smash@gmail.com | pdd27mmd_tasklist:
22:32 dalek link: http://www.perlfoundation.org/pa​rrot/index.cgi?pdd27mmd_tasklist
22:33 dalek r30358 | smash++ | pdd27mmd:
22:33 dalek : [branches/pdd27mmd]: Replace 'mmdvtregister' and 'mmdvtfind' opcodes with 'add_multi' and 'find_multi'
22:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30358
22:34 dalek mestre.smash@gmail.com | pdd27mmd_tasklist:
22:34 dalek link: http://www.perlfoundation.org/pa​rrot/index.cgi?pdd27mmd_tasklist
22:35 Limbic_Region joined #parrot
22:59 tetragon joined #parrot
23:04 GeJ purl: nopaste?
23:04 purl nopaste is at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or App::Nopaste or tools/dev/nopaste.pl
23:04 GeJ merci
23:06 dalek r30359 | Whiteknight++ | pdd27mmd:
23:06 dalek : [PMC] Add a basic implementation of the get_pmc vtable method for the multisub pmc. needs testing
23:06 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30359
23:06 dalek r30360 | rgrjr++ | trunk:
23:06 dalek : Add missing "=over".
23:06 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30360
23:17 Zaba_ joined #parrot
23:17 dalek r30361 | Whiteknight++ | pdd27mmd:
23:17 dalek : [PMC] add stubs for get_pmc_keyed and get_pmc_keyed_string vtable methods. Add basic implementation of a get_iter method, which relies on these two vtable methods being implemented.
23:17 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30361
23:23 dalek r30362 | rgrjr++ | trunk:
23:23 dalek : [RELEASE] Merge from /branches/RELEASE_0_7_0/:
23:23 dalek : revision: 30346;  author: rgrjr
23:24 dalek :    Update version numbers etc. for Parrot release 0.7.0 "Severe Macaw".
23:24 dalek : revision: 30345;  author: rgrjr
23:24 dalek :   [release] Clean up "make fulltest" (see list discussion).
23:24 dalek :   * config/gen/makefiles/root.in:
23:24 dalek :      + (fulltest):  Run codetest instead of codingstd_tests.
23:24 dalek :   * docs/project/release_manager_guide.pod:
23:24 dalek :      + So we don't need to say "failures are OK" here.
23:24 dalek : revision: 30344;  author: coke
23:24 dalek :   [tcl] temporarily TODO some failing tests for the release.
23:24 dalek : revision: 30339;  author: rgrjr
23:24 dalek :   * NEWS:  Make style consistent with recent releases.
23:24 dalek :      + Move "Specification" before "Implementation".
23:24 dalek :      + Fix capitalization of Cardinal, Pipp, and Tcl bullets.
23:24 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=30362
23:34 jonathan rgrjr++ # release
23:38 * smash later all
23:57 tetragon_ joined #parrot

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

Parrot | source cross referenced