Camelia, the Perl 6 bug

IRC log for #parrot, 2008-04-13

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 teknomunk joined #parrot
01:14 PerlJam It could be my imagination, but I it seems like parrot is building faster than it used to.
01:18 chromatic_away It's not.
01:19 chromatic_away It's not your imagination, anyway.
01:25 Debolaz Yay, parrot repo finally packed.
01:26 tetragon joined #parrot
01:39 guru left #parrot
01:42 paula35 joined #parrot
02:05 Debolaz Is t/stm/llqueue known to coredump on the singlethreaded case?
02:15 afbach joined #parrot
02:46 Andy joined #parrot
03:32 AndyA joined #parrot
03:47 dalek r26952 | rgrjr++ | trunk:
03:47 dalek : * DEPRECATED.pod:
03:47 dalek :    + Deprecate stack ops.
03:47 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26952
04:24 dalek r26953 | rgrjr++ | trunk:
04:24 dalek : * t/pmc/complex.t:
04:24 dalek :    + ($fp_equality_macro):  Choose registers carefully so that we don't
04:24 dalek :      need save/restore.
04:24 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26953
04:58 Psyche^ joined #parrot
06:35 dalek r26954 | duff++ | trunk:
06:35 dalek : [rakudo] Add given and for statement modifiers.
06:35 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26954
08:53 barney joined #parrot
09:37 Senaka joined #parrot
09:37 Senaka seen chromatic_away
09:37 purl chromatic_away was last seen on #parrot 8 hours and 19 minutes ago, saying: It's not your imagination, anyway.
09:37 Senaka seen chromatic
09:37 purl chromatic was last seen on #parrot 12 hours and 4 minutes ago, saying: tetragon, jonathan made a change to assign the other day which made it work.
09:38 Senaka need some help in getting the C++ build fixed
09:38 Senaka anyone around?
09:39 Senaka the issue is with the include/parrot/datatypes.h include
09:39 Senaka in here, as you see, the definition of data_types is conditional
09:40 Senaka and in the situation where it is defined as an extern, there is no implementation that is being linked to the final library. Thus, a build failure in C++
09:41 Senaka left #parrot
09:41 Senaka joined #parrot
09:44 Senaka left #parrot
10:10 bwk joined #parrot
10:11 avar joined #parrot
10:34 Ademan joined #parrot
11:06 Ivatar joined #parrot
11:35 spinclad purl, message Senaka isn't the implementation defined in src/global_setup.c ?  it's not static, so should be externally visible, no?
11:35 purl Message for senaka stored.
11:51 kid51 joined #parrot
11:53 kid51 purl seen piper?
11:53 purl piper was last seen on purl 2 days and 4 hours ago, saying: <private message>  [Apr 11 00:39:18 2008]
12:08 itz joined #parrot
12:09 itz I'm getting an internal gcc error with gcc 4.1.2 (debian) and parrot .. is there a recommended gcc?
12:10 dalek r26955 | kjs++ | trunk:
12:10 dalek : [pdd29] add a few more things so i won't forget.
12:10 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26955
12:23 iblechbot joined #parrot
12:23 kj joined #parrot
12:29 kid51 itz:  You're using it.  Can you paste the error?
12:30 kid51 purl nopaste
12:30 purl i guess nopaste is at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://paste.husk.org/ or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or don't bother me while I'm eating
12:32 kj messages kj
12:32 kj messages help
12:32 purl To leave a message, say in channel or privmsg purl "msg <nickname> MESSAGE FOR J00".  To read your messages, privmsg purl "messages".  To erase your messages, privmsg purl "messages erase". or Delivery Not Guaranteed!
12:33 kj messages
12:38 nopaste "itz" at 62.3.198.42 pasted "% make perl6 [snip] /home/stev" (18 lines) at http://nopaste.snit.ch/12687
12:38 itz its actually "make perl6" rather than parrot itself
12:41 Senaka joined #parrot
12:41 Senaka left #parrot
12:48 kid51 itz:  I have not seen that error before ... but then again I haven't done make perl6 very often.  I suggest filing bug report to parrotbug@parrotcode.org.  [BUG] in subject line.  Attach what you pasted as email attachment, preferably ending in .txt.
13:02 ambs joined #parrot
13:25 rdice joined #parrot
13:26 tetragon joined #parrot
14:13 Senaka joined #parrot
14:13 Senaka hi all
14:13 Senaka spinclad: ping
14:14 Senaka seen spinclad
14:14 purl spinclad was last seen on purl 2 hours and 38 minutes ago, saying: <private message>
14:14 Senaka Errors on C++ build: error:imcc:syntax error, unexpected $end, expecting '\n' ('�')
14:14 Senaka in file 'runtime/parrot/library/parrotlib.pir' line 190
14:15 * Senaka I have no idea how to get this straightened
14:42 * Coke yawns
14:42 Senaka Coke: ping
14:42 Coke seen chromatic?
14:42 purl chromatic was last seen on #parrot 17 hours and 9 minutes ago, saying: tetragon, jonathan made a change to assign the other day which made it work.
14:42 Coke Senaka: yes?
14:42 Senaka Errors on C++ build: error:imcc:syntax error, unexpected $end, expecting '\n' ('�')
14:42 Senaka in file 'runtime/parrot/library/parrotlib.pir' line 190
14:43 * Senaka Needs some help...
14:43 * Senaka Doesn't understand how to get this sorted
14:43 Coke what were you running?
14:44 Senaka perl Configure.pl --cc=g++ --link=g++; make
14:44 Senaka machine: Ubuntu 7.10
14:44 Senaka i386
14:45 Coke have you opened a ticket? =-)
14:45 Senaka hmm... not yet... if have fixed some issues in the C++ build. this seems bit tricky. so wanted to sort all and put a patch at once
14:46 Senaka is this an obvious mistake or something? or is it tricky?
14:46 Coke which step was running when you got that error?
14:47 Coke msg chromatic (pmc_class.h) reconfigure; if the dep is from an include step, the build process sees it and adds the dependency for you.
14:47 purl Message for chromatic stored.
14:47 Senaka let me paste it
14:48 Senaka ./parrot -o runtime/parrot/include/parrotlib.pbc runtime/parrot/library/parrotlib.pir
14:48 Senaka error:imcc:syntax error, unexpected $end, expecting '\n' ('�')
14:48 Senaka in file 'runtime/parrot/library/parrotlib.pir' line 190
14:48 Senaka make: *** [runtime/parrot/include/parrotlib.pbc] Error 1
14:50 Senaka is Jens Rieks around?
14:50 Coke Senaka: you're using a big shotgun.
14:51 Coke it's not a problem with that file.
14:51 Senaka Coke: what could it be?
14:51 ambs purl: seen merlyn
14:51 purl merlyn was last seen on #moose 23 days and 17 hours ago, saying: ... http://methodsandmessages.vox.com/lib​rary/post/the-year-of-smalltalk.html  [Mar 20 15:09:20 2008]
14:51 Coke if you're having trouble compiling a PIR file that compiles on every other platform and you're using a new compiler... there's probably a problem with the parrot you've built.
14:52 Senaka I see... so you mean this has something to do with the Parrot build
14:53 ambs joined #parrot
14:53 Coke yes.
14:53 Coke and if you've made local changes to get it to build...
14:54 Senaka Coke: ok I did tweak two places in source to get the build running.... but those are erraneous
14:54 Senaka I think that fixes to those should do the trick
14:54 Coke I'm doing a build now on feather (linux) with your config line to see what explodes.
14:55 Senaka perhaps you might need the diff of my modifications too
14:55 Senaka how can I send it?
14:55 Coke docs/submissions.pod
14:56 Coke open a new ticket for getting it to build with g++, attach the patch as one file, the buildlog as another.
14:56 Senaka hmm ok
14:56 Senaka will do...
14:56 Coke ... and if you changed src/builtin.c , that could definitely affect your ability to parse PIR.
14:57 Coke so let's get the build changes vetted first, and go from there.
14:57 Senaka Coke: yes I think that's a good step forward
14:58 * Coke -> away for a bit
15:08 ruz joined #parrot
15:08 Senaka Coke: with the latest svn up I even wont get that far
15:10 jan joined #parrot
15:17 Senaka msg Coke raised issue regarding C++ build. Please check http://rt.perl.org/rt3//Publ​ic/Bug/Display.html?id=52854
15:17 purl Message for coke stored.
15:17 Senaka left #parrot
15:21 ambs hmms, nice
15:21 ambs purl++
15:34 Andy joined #parrot
15:43 barney joined #parrot
15:47 paco joined #parrot
15:51 jonathan itz: The file that is produced and that the copiler attempts to produce is pretty large...are you running with some memory limits or similar?
16:21 zarchne joined #parrot
16:23 itz jonathan: nope
16:23 jonathan Hmm, OK.
16:23 jonathan Compiler version?
16:23 itz ah it works under bash
16:23 itz not zsh (my default shell)
16:24 itz maybe some different resource limit in zsh
16:24 itz gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)
16:24 Piper joined #parrot
16:24 Piper Hi there.  I am Piper.  I am now publicly logging this channel.  If you don't want to be logged, please leave now.
16:25 ambs Piper: spammer!
16:25 purl go fuck yourself with a spoon! er, no, make that a fork!!
16:26 Ivatar joined #parrot
17:00 dalek r26956 | fperrad++ | trunk:
17:00 dalek : [uuid]
17:00 dalek : - add methods type() & variant()
17:00 dalek : - and tests
17:00 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26956
17:08 Senaka joined #parrot
17:08 Senaka ping
17:08 * Senaka Tries to draw attention to C++ build
17:09 Senaka left #parrot
17:09 Senaka joined #parrot
17:18 Senaka left #parrot
17:54 pmichaud pong
17:55 pmichaud oh, he left.  never mind.
18:10 Theory joined #parrot
18:14 tetragon joined #parrot
18:18 guru joined #parrot
18:22 Debolaz pmichaud: Ping.
18:22 pmichaud pong
18:22 Debolaz pmichaud: I found the first apparent roadblock for using rakudo for my own personal projects, lack of file IO.
18:22 Debolaz Err.. not complete lack of it.
18:22 Debolaz open and close works.
18:23 particle and reading via prefix:= works
18:23 pmichaud jonathan's been working on it, but I haven't looked into it in great detail yet.  (more)
18:23 pmichaud my suggestion would be to identify the appropriate tests for what you need, and/or file a report to rakudobug@perl.org
18:24 jonathan I will have tuits to do more.
18:24 jonathan Any hints on exactly what you want?
18:24 pmichaud if there's a bug report I'm pretty sure it will get worked on quickly
18:24 pmichaud we're definitely focused on features that we know people are wanting to use :-)
18:25 Debolaz jonathan: Well, being able to do $fh.readline would go a long way
18:25 jonathan Very much so. I should probably learn where the bug tracker is though. <embarrassed look>
18:25 rdice pmichaud, and we love you for it. :-)
18:26 dalek r26957 | particle++ | trunk:
18:26 dalek : [gettext] capitalize config setting so a #define is autogenerated
18:26 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26957
18:26 dalek r26958 | particle++ | trunk:
18:26 dalek : [gettext] define some gettext configuration macros
18:26 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26958
18:27 lichtkind joined #parrot
18:27 dalek r26959 | particle++ | trunk:
18:27 dalek : [gettext] setup gettext in parrot main
18:27 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26959
18:27 Debolaz pmichaud: Btw, I set up a git mirror of the parrot repository. Mostly for my own benefits, but anyone is free to use it. :-)
18:28 moritz URL?
18:29 pmichaud jonathan:  bug tracker is the 'perl6' queue at rt.perl.org
18:29 jonathan Debolaz: .readline() method on a filehandle is certainly in the spec, right?
18:30 moritz it is (see http://svn.pugscode.org/pu​gs/docs/Perl6/Spec/IO.pod )
18:30 jonathan OK, good.
18:30 moritz it's a bit confusing because it's a builtin but not in S29
18:31 dalek r26960 | particle++ | trunk:
18:31 dalek : [gettext pmc] add i18n wrappers to exception messages in FixedPMCArray
18:31 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26960
18:31 Debolaz moritz: http://git.andersberle.com/?p=parrot.git
18:31 moritz pmurias asked the same about mkdir in #perl6 ;-)
18:31 moritz Cannot get remote repository information.
18:31 moritz Perhaps git-update-server-info needs to be run there?
18:31 moritz that's what I get when I try a git-clone
18:31 Debolaz moritz: That's the gitweb for it.
18:32 moritz stupid me
18:32 Debolaz moritz: git://git.andersberle.com/parrot.git is the raw clone url.
18:32 moritz thakns
18:33 Debolaz Updated once an hour.
18:34 moritz I'd wish the blead perl git repo was updated as often ;)
18:34 moritz but probably perforce is a bit uglier to import, and less often used
18:34 jonathan Debolaz: See r26961.
18:34 dalek r26961 | jonathan++ | trunk:
18:34 dalek : [rakudo] Implement readline method on IO class.
18:34 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26961
18:35 markus joined #parrot
18:35 Debolaz That was.. fast. :o
18:35 cout joined #parrot
18:35 moritz Debolaz: I guess it's the same as prefix:<=>
18:36 jonathan moritz: Where abouts is the IO stuff spec'd out in more details? I didn't see it in the synopses...
18:36 moritz jonathan: dunno if there's more than IO.pod in the pugs repo
18:37 Debolaz jonathan: http://feather.perl6.nl/syn/S16.html
18:37 Debolaz Draft stuff.
18:39 ambs particle: now we just need to check why gettext is not correctly detected on some systems.
18:39 jonathan Ah, that's more readable...
18:41 jonathan Hmm, that needs a little review. :-)
18:41 jonathan our Bool method getc (IO $self: *@LIST)
18:41 jonathan Returns the next character from the input stream attached to IO
18:41 jonathan our Bool? :)
18:41 Debolaz http://git.andersberle.com/?p=parrot.git;a=commi​tdiff;h=1980a6936b9037c958c1101a87fcb7bbf682ec0e # And there it is synced so I can use it. :-)
18:42 shorten Debolaz's url is at http://xrl.us/bjb2p
18:42 pmichaud readline is probably in the IO synopsis instead of S29
18:43 jonathan pmichaud: I'm reading it now.
18:44 pmichaud source of IO synopsis is http://svn.pugscode.org/pu​gs/docs/Perl6/Spec/IO.pod
18:45 pmichaud (at least, I _think_ that's the source)
18:45 pmichaud I agree with TimToady's comment that we should wikify the synopses somewhere :-)
18:45 pmichaud either that or let's move them all into the pugs repo or somewhere with a liberal commit policy
18:45 moritz pmichaud: and it should have structure that's easily linkable
18:46 pmichaud moritz: proposals for linkable structures are welcome, but I don't know that we should wait for one :-)
18:47 moritz pmichaud: I'm fine if there's a possibility to link to each section or subsection (by name) and line number additionally
18:47 pmichaud the problem with "by name" is that the names might change :-
18:47 pmichaud :-)
18:47 pmichaud or sections may get refactored
18:48 pmichaud I agree very much with "easily linkable", but I don't want us to end up in the early stages of saying "we can't refactor the specs because X, Y, and Z depend on the current linkages"
18:49 pmichaud later after there's more stability to the specs, yes.
18:50 moritz I've already told TimToady that broken links are no excuse not to refactor things (and that I'll try to fix as many smartlinks as possible)
18:50 pmichaud I think we're in agreement.  :-)
18:51 ambs joined #parrot
18:51 jonathan I need to go to the shops for food and, importantly, beer. Back later.
18:54 paco joined #parrot
19:03 pmichaud afk # soccer
19:25 dalek r26962 | fperrad++ | trunk:
19:25 dalek : [eclectus]
19:25 dalek : - use petite on win32
19:25 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26962
19:31 rotty seen fperrad?
19:31 purl I haven't seen 'fperrad', rotty
19:45 avar joined #parrot
21:50 dalek r26963 | jonathan++ | trunk:
21:50 dalek : [rakudo] First cut at eval. Sucks because it doesn't allow you to use variables in the lexical scope the code is compiled in. Yet.
21:50 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=26963
22:02 Limbic_Region joined #parrot
22:16 jonathan pmichaud: Well, if there is no way to runtime-enumerate the lexicals, I'll just have to implement one. ;-)
22:16 jonathan I already did set_outer
22:16 jonathan So we can dynamically set the outer of the compiled code.
22:16 jonathan Which is one half of the problem.
22:17 jonathan Anyway, I'm going to get sleep at a sensible time tonight, so I'll leave looking into that for another day.
22:50 tetragon joined #parrot
23:38 guru joined #parrot
23:59 anna30 joined #parrot

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

Parrot | source cross referenced