Camelia, the Perl 6 bug

IRC log for #parrot, 2008-03-03

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:14 wknight8111 joined #parrot
01:37 svnbotl r26183 | jkeenan++ | tcif:
01:37 svnbotl : Parrot::Configure::Options::process_options:  regex binding was incorrect.
01:37 svnbotl : Parrot::Configure::Options​::Test::get_all_options():  To be handled properly by
01:37 svnbotl : Parrot::Configure::Options::process_options(), each element has to begin with
01:37 svnbotl : double hyphens and join key and value with equals sign.
02:26 wknight8111 joined #parrot
02:27 mdiep joined #parrot
02:31 Coke mdiep: ping
02:35 mdiep Coke: pong
02:36 Coke mdiep: you should register to mentor someone for partcl. =-)
02:36 mdiep I'd be happy to mentor someone
02:37 mdiep I almost added myself to the wiki yesterday, but thought there might be better candidates: you, jerry, patrick, chromatic
02:40 mdiep Coke: now that you're on the grant committee, can we get tpf to approve a partcl project? :)
02:42 Coke I will have to avoid answering that question. But grant submittals are always welcome.
02:42 Coke (well, in season.)
02:43 mdiep I was talking about SoC
02:44 Coke grant committee doesn't have anything to do with SoC, methinks.
02:45 c9s joined #parrot
02:45 Coke I'm getting segfaults on tcl on feather. Those should be fixable by someone with a clue.
02:45 mdiep someone at tpf looks at the proposals and decides which they'd like to do. I assumed it was the grant committee.
02:46 mdiep does it segfault under os x?
02:46 Coke ... oh hell yes.
02:47 Coke I cleaned up [lsort] a bit the other day.
02:47 * mdiep svn ups
02:48 Coke (eliminated our copy of quicksort.)
02:50 contingencyplan joined #parrot
02:51 mdiep ld: warning in /opt/local/lib/libicudata.dylib, file is not of required architecture
02:51 mdiep ld: symbol(s) not found for architecture ppc
02:51 mdiep (this is an intel mac)
02:51 mdiep didn't someone fix this a while back?
02:51 Coke haven't used ICU in ages.
02:52 mdiep I'm not using it -- I'm trying to build parrot.
02:52 Coke I don't have icu installed, so I never see that warning. =-)
02:54 mdiep well let's try this again without icu then :)
02:58 Coke (segfault on os x/intel) -btw, with the new gcdebug runcore, you can usually force the segfault on any platform if you find it on one.
02:58 mdiep I don't know how successful I'll be at fixing any gc problems.
02:59 Coke mdiep: I recommend investigating the segfault in tcl_var_subst.t first; that one fails in the same place as most of the segfaults.
02:59 Ademan joined #parrot
03:00 mdiep I see we haven't gotten any faster thru my neglect :)
03:06 Coke -> zzz
03:07 Andy joined #parrot
03:09 kid51 joined #parrot
03:09 svnbotl r26184 | jkeenan++ | trunk:
03:09 svnbotl : Parrot::Configure::Options::process_options:  regex binding was incorrect.
03:09 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26184
03:23 svnbotl r26185 | chromatic++ | trunk:
03:23 svnbotl : [src] Converted all applicable uses of const_string() to CONST_STRING().
03:23 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26185
03:29 * kid51 must sleep
03:29 purl $kid51->sleep(8 * 3600);
03:33 Theory joined #parrot
04:01 svnbotl r26186 | petdance++ | trunk:
04:01 svnbotl : Added appropriate macros for everything the headerizer demands in imcc.y
04:01 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26186
04:04 cotto joined #parrot
04:38 particle joined #parrot
05:00 x___ joined #parrot
05:07 particle joined #parrot
05:22 slightlyoff joined #parrot
05:58 svnbotl r26187 | chromatic++ | trunk:
05:58 svnbotl : [src] Minor formatting improvements, some documentation added.
05:59 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26187
06:40 iblechbot joined #parrot
07:14 Andy joined #parrot
07:20 uniejo joined #parrot
07:21 uniejo joined #parrot
07:43 uniejo joined #parrot
08:11 iblechbot joined #parrot
08:15 svnbotl r26188 | fperrad++ | trunk:
08:15 svnbotl : [emacs]
08:15 svnbotl : - handles block .macro .endm
08:15 svnbotl r26189 | fperrad++ | trunk:
08:15 svnbotl : [Lua]
08:40 svnbotl r26190 | fperrad++ | trunk:
08:40 svnbotl : [Lua]
08:40 svnbotl : MANIFEST
08:40 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26190
08:40 purl : MANIFEST is updated accordingly, and file props are set. yay!
08:47 cosimo joined #parrot
08:51 Ademan joined #parrot
09:19 DarkWolf84 joined #parrot
09:59 AndyA joined #parrot
10:20 ruoso joined #parrot
11:07 nopaste joined #parrot
11:07 wknight8111 joined #parrot
11:24 RichiH joined #parrot
13:34 Coke We could use some mentors for parrot's attempt at SoC; If anyone has a project they are interested in, but don't personally have the time to do all the work on, a signup on the wiki would be most welcome.
13:34 Topic for #parrotis now "Please mentor for SoC | http://parrotcode.org/
13:34 Coke soc?
13:34 purl hmmm... soc is short for [Google's] Summer of Code or Is it June 24th yet? http://code.google.com/summerofcode.html or Security Operations Center or http://www.perl.org/advoca​cy/summerofcode/index.html
13:45 cognominal le soc de la charrue de l'open source pour ceux qui ne sont pas des boeufs...
13:46 Coke french-zactly.
13:53 gryphon joined #parrot
13:59 mdiep joined #parrot
14:06 cognominal oops, wrong channel
14:24 wknight8111 joined #parrot
14:30 dwave joined #parrot
14:32 svnbotl r26191 | fperrad++ | trunk:
14:32 svnbotl : [build] on Win32
14:32 svnbotl : - reverts r26185 (packdump.c is a part of pdump.exe, not of libparrot.dll)
14:32 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26191
14:39 particle andy: src\spf_render.c(99) : warning C4005: 'snprintf' : macro redefinition
14:39 particle c:\usr\local\parrot\trunk\​include\parrot/misc.h(28) : see previous definition of 'snprintf'
14:46 rdice joined #parrot
15:04 iblechbot joined #parrot
15:09 Andy joined #parrot
15:16 svnbotl r26192 | petdance++ | trunk:
15:16 svnbotl : doc patches from ajr
15:16 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26192
15:33 particl1 joined #parrot
15:38 jhorwitz joined #parrot
15:45 Coke win 4
15:49 wknight8111 joined #parrot
16:17 Andy who wants to chase segfaults!
16:23 cognominal I have segfaults  some time after I do syntax errors when I use rakudo interactively but I am busy adding stuff to rakudo
16:23 cognominal on my macbook leopard.
16:24 cognominal but if you want traces or whatever I can give you that.
16:24 cognominal Probably exception handling has bugs.
16:24 Coke cognominal: there's a ticket for that, IIRC.
16:24 cognominal ok
16:25 Coke http://rt.perl.org/rt3/Tic​ket/Display.html?id=49758
16:27 cognominal yea
16:28 cognominal speaking of scope, the interactive mode should not wrap a block around the eval  to behave like pugs
16:28 cognominal so one can declare lexicals and use them on later lines
16:29 cognominal that was bugging me so much for perl 5 that I wrote a RFC about that.
16:30 IllvilJa joined #parrot
16:30 cognominal http://dev.perl.org/perl6/rfc/351.html
16:31 cognominal boy that was almost eight year ago
16:35 particle so, you suggest that "perl6 -e 'my $x = 1' -e 'say $x'" should print 1?
16:36 particle or is this only in interactive mode?
16:37 cognominal that should be in interactive mode. for successive -e, I don't have opinion one way or the other because I never use it.
16:38 cognominal not even sure that rakudo options parser supports it.
16:38 particle i believe it does. that was certainly my intent when i implemented it.
16:39 particle however, i can't remember.
16:39 cognominal I had started rewriting in Perl6 but I got sidetracked by stuff missing in rakudo.
16:39 particle rewriting what in perl 6?
16:40 cognominal Getopt::Obj.
16:40 particle ah
16:42 cognominal jonathan and allision are planning long term : role and mmd... I just want to add basic stuffs missing so I can use rakudo immediately
16:42 cognominal and write an article in the french linux mag.
16:42 particle and Getopt is the subject of the article?
16:42 particle or a tool needed to do your work?
16:43 cognominal I don;t think so. That was just an exercise to better know pir and perl 6
16:43 particle ok
16:43 cognominal but why not it could be.
16:43 cognominal anyway there matter for many articles
16:44 particle indeed.
16:44 particle i need to work on perl6's command line syntax
16:44 cognominal some other french mongueur  has already started with three articles on pir.
16:45 cognominal but my goal is to show that rakudo can be immediately useful. I hope that will be true of the next release.
16:45 particle what's missing from rakudo for you to convert getopt::obj?
16:45 cognominal I don't remember :)
16:46 cognominal but when playing interactively I found some stuff wrong or missing
16:46 cognominal perhaps nothing concerning about getopt::obj after all :)
16:46 jrockway joined #parrot
16:46 cognominal but I got pissed of :)
16:47 particle yes, i wish it was done too :D
16:47 cognominal I sent a first patch yesterday, hoping to send more during the week.
16:47 particle great! looking forward to them
16:48 cognominal also the perl6 pod seems an interesting thing to work on later
16:49 particle please do!
16:49 particle i've got so many things to do now, i can't keep up with them all
16:49 cognominal we need to interlink our docs by annotating them properly
16:50 cognominal I am out of work. I had health problem that make me unable to sleep much and to be productive during the day. Getting better so I can now contribute.
16:50 cognominal And I need to demonstrate my expertise to find some work...
16:51 particle publishing articles on cutting edge software may help with that :)
16:51 cognominal we are publishing article about Perl about every month...
16:52 cognominal The french linux mag is what Dr Dobb's was twenty year ago.
16:52 parrot-poke joined #parrot
16:52 cognominal so we fit right in.
16:52 particle excellent
16:52 cognominal 2 year ago, I did even a april fool joke about a perl6 processor :)
16:53 cognominal http://articles.mongueurs.net/m​agazines/perles/perles-22.html
16:55 particle hey, i can read french! well... some french words. like SUN and SPARC ;)
16:56 cognominal the bibliography is in English, including the last one about Douglas Adams :)
16:57 cognominal what is nice about the magazine is that we can publish the articles on the net  4 month after publication
16:57 particle very nice, indeed
17:09 sjansen joined #parrot
17:43 stephen joined #parrot
17:47 Andy oooo, Patrick has pissed off Acme of Orange: http://use.perl.org/~acme/journal/35798?from=rss
17:50 cognominal paste?
17:50 purl paste is (see: nopaste) or like glue but a little safer to sniff. or http://nopaste.snit.ch:8001/ or http://scsys.co.uk:8001/ anywhere shadowpaste is or mmm, sticky paste or You there! Eating the paste. or <see> 2 girls, 1 paste
17:51 jrockway joined #parrot
17:51 avar Am I to understand that it's on?
17:53 peeps[work] joined #parrot
17:53 cognominal http://scsys.co.uk:8001/13680   # I try to override methods and they are not called when I  write  'my %a = (1,2)' or 'my %a; %a.say'
17:54 cognominal well get_string in Hash.pir is not my own doing but is not called either.
17:55 cognominal also, and probably realted  when a  new 'Hash' is done, how Perl knows if it is the Perl6 class or the Parrot one, I am confused
17:55 rdice joined #parrot
18:00 Coke new 'Hash' is most likely the parrot hash.
18:00 Coke er, the parrot 'Hash'.
18:01 cognominal so TGE generates the wrong code.
18:01 cognominal how to say, I want to create a Perl6 Hash?
18:01 cognominal TGE or actions.pm....
18:02 Coke ... you know what? Nevermind. I have no idea how perl6 works. =-)
18:02 cognominal I am still confused with these object layers
18:03 cognominal jonathan and allison know...
18:03 cognominal guess I need to go back to the sources and the doc
18:04 cognominal btw we have found a tshirt for our French perl5 pumpking, a fan of doctor Who : http://www.thinkgeek.com/ts​hirts/popculture/9f69/zoom/
18:06 Stephen joined #parrot
18:09 particle you can't name a perl 6 object after a parrot object
18:09 particle that's why we have Perl6Str instead of String
18:09 particle call it Perl6Hash
18:09 particle when we start using HLL properly with rakudo/pct/etc then it'll work as you expect
18:11 cognominal ok
18:11 cognominal thx
18:13 Coke particle: what is src/class/Hash.pir then?
18:14 particle ah, ok. that's a parrot Hash, with overridden methods
18:15 particle ...which is what cognominal is using. sorry.
18:15 Coke So perl6 is using two different hashes at the moment?
18:15 particle i thought it was a new file, since it wasn't diff -u format
18:15 particle no
18:16 Coke ah, not hash, no. but string, yes?
18:16 Coke ah. no.
18:16 cotto_ joined #parrot
18:17 cognominal is there other instances of Perl class that have the same name as a Parrot class? I could see how the problem is solved
18:19 cognominal very strange, I had no such problem with the Pair class
18:19 particle cognominal: could you post your changes to src/classes/Hash.pir as an svn diff?
18:19 cognominal on the other hand I did not override
18:21 cognominal http://scsys.co.uk:8001/13684
18:21 cognominal yesterday I sent a patch to parrotbug that would like to see applied
18:22 cognominal this one, at least, does what he is intended to do.
18:23 cognominal also I noted that  %a = 1, 2  works with the Perl 5 relative preceddence wich is a bug.
18:30 particle cognominal: submit a test to pugs for that, or report it as a bug to rakudobug@perl.org
18:30 barney joined #parrot
18:31 cognominal hum, probably pugs has a test for that. I check
18:34 cognominal how can I run the pugs test on rakudo?
18:36 cotto_ joined #parrot
18:36 TimToady t/spec/S03-operators/assign.t is close to correct
18:40 particle make spectest
18:40 purl make spectest is failing more than usual too
18:45 cotto_ how current is pmc_freeze.pod?
18:46 particle what does $Id$ say?
18:46 particle or svn log?
18:46 purl svn log is probably faster for some reason
18:46 cotto_ 2006-07-24
18:46 purl 1975
18:46 particle purl, you moron.
18:46 purl particle: excuse me?
18:47 cotto_ although it doesn't tell me how much the structure of Parrot has changed since then
18:47 particle cotto_: that date doesn't give me much confidence
18:48 particle well, ok. it's design notes for the prototype version of freeze/thaw
18:48 particle so, it's a historical document.
18:48 particle however, there haven't been major pmc changes since then
18:49 particle that will change this month, when we merge the pdd17pmc branch into trunk
18:49 cotto_ so as soon as I figure it out, it will change?
18:49 particle if you're curious how that might affect you, feel free to co the branch and play around
18:49 cotto_ like the universe
18:50 particle the universe changes faster. i suspect it's not an all-volunteer effort ;)
18:54 cotto_ what's special about the pdd17pmc branch?
18:54 amoore joined #parrot
18:57 particle pdd17 is new, and we're implementing it in the branch
18:57 particle pdd17 is the design doc for pmcs
18:57 cotto_ so if I read pdd17, I'll have some idea of what's going on?
18:57 particle yes
18:58 cotto_ thanks
18:58 cotto_ particle++
19:03 cotto_ should I expect the pdd17pmc build to fail?
19:04 diakopter 2006/07/24
19:04 purl 11.9404761904762
19:05 cotto_ purl can't get a date
19:05 purl cotto_: excuse me?
19:08 particle cotto_: no, it should build. however, i get a build error with msvc
19:08 particle it's a gc bug iirc
19:09 particle diakopter: the ircops requested we throttle svnbotl to 1msg/sec
19:09 Andy joined #parrot
19:09 particle think you can manage to modify the startup script etc on feather to use throttling? (i think it's an option on that script)
19:10 particle i mean, i think it's an option on the ircbot script
19:13 diakopter particle: I've tried
19:13 diakopter I tried to add the option, but implemented it poorly/wrongly, I think
19:13 particle ah, hrmm.
19:14 svnbotl r26193 | bernhard++ | trunk:
19:14 svnbotl : #42769: Remove unneeded type names from PIR
19:14 svnbotl : In imcc.y rename 'type_without_typename' back to 'type'.
19:14 svnbotl : Remove deprecation notice, as the deprecated feature is gone.
19:14 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26193
19:24 arcady joined #parrot
19:42 slightlyoff joined #parrot
20:06 nopaste "as" at 82.135.229.130 pasted "as" (1 line) at http://nopaste.snit.ch/12434
20:07 Coke as?
20:07 purl i think as is the tutorial or ActiveState
20:09 schmalbe joined #parrot
20:23 hugues23 joined #parrot
20:29 * barney will miss #ps tomorrow, due to Frankfurt.pm meet
20:30 particle if you have status report, feel free to post it there, or email it to me
20:42 iblechbot joined #parrot
20:56 barney particle: mailed report
20:59 avar joined #parrot
21:09 Tene ack, ps tomorrow.  I'd better get some stuff done.
21:10 * Tene requests that the scheduled time for #ps be mentioned in the #ps topic
21:10 Coke no one in that discussion has ops
21:11 Tene eh, I'll forget about it shortly anyway.
21:11 Coke updated http://www.parrotcode.org/examples/pir.html so all the examples compile again.
21:11 * particle pictures tene in every scene from memento
21:12 particle coke: i think we had some tests for that type of thing
21:12 particle t/examples/*.t
21:12 Coke that file is not checked into the parrot repo.
21:13 particle ah.
21:13 particle yuck.
21:22 Coke particle: you going to reply to gabor's email?
21:23 Coke I'll do it. =-)
21:25 particle just did
21:25 Coke ... ok.
21:40 lichtkind joined #parrot
21:46 lichtkind could someone point me please to a doc that explaines the scheme how is nqp written ?
21:50 gabriel joined #parrot
22:08 Tene particle: I'm not the one who writes patches and then forgets about writing them.
22:11 * spinclad has a suspicion re svnbotl: is its output full-buffered, so delays between lines only delay final output, or line-buffered, so they should work?  must check.  later, sadly.  must be elsewhere soon for meeting.  drive-by posting, no need to get up.  &
22:56 x___ joined #parrot
23:04 Andy OK, back
23:20 Limbic_Region joined #parrot
23:41 kid51 joined #parrot

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

Parrot | source cross referenced