Camelia, the Perl 6 bug

IRC log for #parrot, 2008-10-10

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:03 jhorwitz chromatic: you sound as lost as i was.  :(
00:04 dalek r31844 | cotto++ | trunk:
00:04 dalek : [pipp] switch to ATTRs, add more newly passing tests
00:04 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31844
00:09 AndyA joined #parrot
00:11 chromatic Yeah, I'm lost.
00:12 chromatic What *should* happen: IMCC should see that 'name' is a string, should allocate a register for that string, and then should turned the keyed access into a constant key access, creating a constant key containing an integer which contains the number of the appropriate string register.
00:12 chromatic As far as I can tell, that all works.
00:13 chromatic I *can't* tell right now if that constant key changes somehow, or if the string register gets overwritten somehow.
00:27 petdance joined #parrot
00:37 chromatic That might be a good task for Tene, who has free time in hotel rooms.
00:38 pmichaud can someone familiar with pipp try it out in the hllmagic branch and tell us what's broken and what's not?
00:38 jhorwitz chromatic: when i was debugging, i put "say" statements around the hash assignment in eval.pir and stopped at Parrot_say_s.  if you examine the string arg, you can see the strstart buffer is suddenly out of bounds
00:39 chromatic Yeah, it's not a string there.
00:39 jhorwitz odd.
00:40 chromatic Maybe something stomped S2 somewhere.
00:45 jhorwitz joined #parrot
00:48 pmichaud jhorwitz: what form of interpolated namespace do you want/need?
00:48 pmichaud (I forget exactly)
00:49 jhorwitz subroutines.  e.g. ::($mr_handler_module)::handler()
00:50 jhorwitz i meant "my", but "mr. module" works well too.  ;-)
00:51 pmichaud okay if I require sigils on there for now?
00:52 pmichaud (might not need them, but might.)
00:52 jhorwitz sure -- it's all internal to mod_perl6, so no biggie.
00:53 jhorwitz so it would be &::($foo)::bar() or something like that?
00:53 pmichaud yes.  maybe.
00:53 pmichaud I'm still trying to see how STD.pm parses that.
00:54 jhorwitz less sigils == better beer.  ;-)
00:55 Andy joined #parrot
00:56 pmichaud okay, looks like it's a term:name
00:56 pmichaud then <longname>
01:10 TiMBuS joined #parrot
01:28 grim_fandango joined #parrot
01:32 pmichaud afk, soccer
02:26 Lorn joined #parrot
02:34 mdiep joined #parrot
03:05 Theory joined #parrot
03:19 MariachiElf joined #parrot
03:32 Bzek joined #parrot
03:53 tetragon joined #parrot
04:09 dalek r31845 | pmichaud++ | trunk:
04:09 dalek : [rakudo]:  remove perl.t test for now until it can be re-fudged.
04:09 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31845
04:16 Andy joined #parrot
04:43 dalek r31846 | pmichaud++ | trunk:
04:43 dalek : [rakudo]:  Adjust $?NS handling.
04:43 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31846
05:04 Ontolog joined #parrot
05:08 Tene I'm updating c99 in the hllmagic branch.
05:20 MariachiElf Where can I read something about how to create bindings for C++ class libraries?
05:21 MariachiElf Specifically I'm trying to gauge order of magnitude of difficulty for me to comprehend and successfully complete bindings for QT/KDELibs
05:21 confound hmm
05:21 dalek r31847 | tene++ | hllmagic:
05:21 dalek : [c99]
05:21 dalek : * NameSpace updates
05:21 dalek : * Fix a grammar bug
05:21 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31847
05:21 confound is there a reason it would be different from doing the same with C libs?
05:22 * confound knows nothing
05:22 Tene MariachiElf: It's mostly a bunch of busy work, I don't think it's much different from C bindings.
05:24 confound MariachiElf: my guess is that the difficulty would mostly be in slogging through all of it :)
05:25 MariachiElf Well the C++ code has types and class libraries and dynamic dispatch and such
05:26 MariachiElf So that's where my concern is.. a QtWindow is subclassed to a KDEWindow
05:26 confound it'd be interesting.  kde-libs has lots of generic functionality
05:26 MariachiElf So I just don't know anything about OO in Parrot :)
05:27 * MariachiElf can just get his brain wrapped around C bindings.
05:27 MariachiElf confound: ANd ultimately I think very useful
05:27 confound Maybe.  It might be hard to make them, uh, well, I was going to say "Perlish"
05:28 * MariachiElf is ultimately trying to build up to a HyperCard/SuperCard/MetaCard clone based on KDE
05:29 MariachiElf confound: Yeah - that's where my concerns are too -- combine that with that I'm really just a a consumer of languages (not a designer)
05:29 confound I don't know what those things are
05:29 confound It's not necessarily impossible, I just think it's a lot of work.
05:29 confound I mean, kde does a lot of stuff.
05:29 dalek r31848 | tene++ | hllmagic:
05:29 dalek : [eclectus]
05:29 dalek : * remove depracated n_ ops
05:29 dalek : * A few NS updates
05:29 dalek : * Tests don't pass... I don't much understand eclectus architecture...
05:29 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31848
05:30 MariachiElf and I have serious doubts -- but with enough persistence -- or discovering the right people -- I think it's doable
05:30 confound mostly I think it's the persistence :)
05:31 MariachiElf confound: The xCard products are what Visual Basic always wanted to be -- simple programming for non-programmers to be very productive by providing very high level commands
05:32 MariachiElf confound: It's basically like Flash with JavaScript -- but the syntax of the languages is basically real English
05:32 Tene Uses some English words for operators and has a syntax that happens to look like English in most of its typical forms.
05:32 confound oh
05:32 confound like applescript!
05:33 Tene Yeah.
05:33 dalek r31849 | tene++ | hllmagic:
05:33 dalek : [hq9plus]
05:33 dalek : * NameSpace updates
05:33 dalek : * All tests pass
05:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31849
05:33 MariachiElf confound: So instead of x = 5; y = 2; z = x+y;  -- you'd writesomething like  Set x to 5; Set y to 2; Add x + y; put it into z;
05:33 MariachiElf confound: HyperCard was an Apple Product -- yes very much like AppleScript
05:34 MariachiElf confound: The only current vendor I know selling a product like this is a company called Revolution -- www.runrev.com
05:35 MariachiElf confound: There Enterprise edition has 1601 commands
05:35 confound or like COBOL!
05:35 * confound hides
05:35 MariachiElf confound: PLus you can add your own commands through shared library extensions
05:35 MariachiElf confound: hehe - well I understand that COBOL was supposedly easy to code
05:36 Tene MariachiElf: I think that's a really great idea.  I'd love to help you with it however I can.
05:36 dalek r31850 | tene++ | hllmagic:
05:36 dalek : [json]: Single namespace update... need to check if this works in trunk
05:36 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31850
05:36 MariachiElf Tene: Well quite frankly I'm not sure where to start
05:37 ab5tract joined #parrot
05:37 Tene MariachiElf: you should read kjs' PCT tutorial on http://parrotblog.org/ and start on making a compiler for the language.
05:37 MariachiElf Tene: I'm thinking the scripting language is probably the easiest and most straight forward to work on -- but --- exactly what I was looking for
05:37 MariachiElf ab5tract: Hey there
05:38 ab5tract yo MariachiElf whats happenin
05:38 MariachiElf ab5tract: Tene just offered his assistance with ParrotTalk :)
05:38 ab5tract other than xTalk taking over the world ;)
05:38 ab5tract awesome
05:38 Tene I can certainly help with any PCT-related stuff.
05:38 Tene Less so on the C++ bindings, but I can look around for you.
05:39 MariachiElf Did you guys know that parrotblog.org doesn't work for the web page (only www.parrotblog.org)
05:39 Tene I didn't know that.  I'm not sure who to talk to about that.
05:39 Tene Maybe Coke or allison...
05:40 MariachiElf Tene:  It's either the web server isn't configured to accept the domain or the domain doesn't resolve to the web servers IP....
05:40 * MariachiElf checks.
05:40 ab5tract my little pet project for learning parrot is http://esolangs.org/wiki/Tree
05:40 MariachiElf ping parrotblog.org
05:40 MariachiElf ping: unknown host parrotblog.org
05:40 purl I can't find parrotblog.org in the DNS.
05:40 ab5tract erm s/parrot/pct/
05:41 pmichaud Tene/others:  I'm thinking of allowing strings/ints/nums directly into a PAST tree, to avoid the need to create PAST::Val nodes for them.  Any comments?
05:42 Tene Hmm.
05:42 Tene Interesting idea.
05:42 Tene They would be completely equivalent to a PAST::Val node?
05:42 pmichaud yes
05:43 Tene Seems like a decent enough idea.
05:43 Tene Does anyone know if languages/json works in trunk?
05:43 Tene or languages/eclectus?
05:44 pmichaud one could still do PAST::Val nodes for typing reasons or other purposes, but it seems like it'd be nicer to put strings/ints/nums directly into the tree than to have to do PAST::Val.new(:value(...)) all over the place.
05:44 Tene nodnod
05:44 particle pmichaud: will they still have attributes?
05:44 pmichaud no
05:44 particle like, what the parse tree was?
05:44 particle hrmm
05:44 dalek r31851 | tene++ | hllmagic:
05:44 dalek : [markdown]
05:44 dalek : * NS updates
05:44 dalek : * All tests pass
05:44 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31851
05:44 pmichaud I think that's not commonly needed for PAST::Val
05:45 pmichaud here's an example
05:45 particle well, i can see it as an optimization
05:45 pmichaud I'm working on namespaces, and at some point in the compiler I need to do
05:45 pmichaud PAST::Op.new( :pirop('join'), PAST::Val.new( :value('::') ), $( $<list> ) )
05:46 pmichaud the PAST::Val node is a constant, it doesn't appear in the source, so there's no need for attributes on it
05:46 pmichaud it seems like it would be better to write
05:46 pmichaud PAST::Op.new( :pirop('join'), '::', $( $<list> ) )
05:46 MariachiElf ab5tract: Tree is pretty :)
05:47 particle hrmm -- i'm looking for downsides...
05:47 particle i like that it's less syntax
05:47 pmichaud same for other places where the compiler has constants that aren't really part of the source.
05:48 particle so, there's no way to do :returns()
05:49 pmichaud correct -- if it needs to be a specific type, one has to use PAST::Val for that.
05:49 pmichaud (I'm not eliminating PAST::Val, just providing a convenient shortcut where the type doesn't really matter.)
05:49 particle ok, that seems a fine optimization
05:50 pmichaud I just see lots of PAST::Val in actions.pm that look like they might not be all that necessary
05:50 particle actions.pm makes me cry a little sometimes
05:51 particle it's *almost* pretty to read
05:51 confound haha
05:51 pmichaud this would make it prettier, definitely.
05:51 particle yes. +1
05:51 confound particle: ewilhelm said you thought last night was chaos
05:51 confound particle: I thought it went pretty well.
05:51 particle the ~25-1 ratio was a little tough
05:52 particle glad you had a different idea
05:52 confound maybe Ijust happened to be in a corner where people were enjoying themselves
05:52 particle yes, you were all alone, right? ;)
05:53 particle oot! plane in 14 hours. need sleep before work tomorrow.
05:53 particle g'nite all
05:53 pmichaud plane in 12 hrs for me :-)
05:53 particle i land at 11:30pm
05:53 confound particle: pfft
05:54 particle i have a layover in atlanta
05:54 confound particle: enjoy pgh
05:54 pmichaud I get there a bit earlier :-)
05:54 dalek r31852 | tene++ | hllmagic:
05:54 dalek : [pheme]: NS updates, doesn't pass tests
05:54 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31852
05:54 particle karma tene
05:54 purl tene has karma of 274
05:54 particle that number sucks. commit again!
05:54 Tene 2*137
05:54 purl 274
05:54 pmichaud I could've gone through atlanta or charlotte, but those seemed out of the way
05:54 Tene I'm quite fond of it, actually.
05:54 pmichaud I'm getting to where I look fairly closely at total travel time :-)
05:55 Tene I'm going to Atlanta in a couple of weeks.
05:55 Tene Do we know anyone there?
05:55 particle i don't
05:55 Tene pmichaud: where you going?
05:55 pmichaud pittsburgh perl workshop
05:55 pmichaud same place as particle and jhorwitz
05:55 particle bedward &
05:58 Tene seeya, particle
05:58 dalek r31853 | tene++ | hllmagic:
05:58 dalek : [squaak]
05:58 dalek : * NameSpace updates
05:58 dalek : * Remove obsolete n_ opcodes
05:58 dalek : * Tests loop forever
05:58 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31853
05:59 Tene 3*3*5*5
05:59 purl 225
05:59 Tene karma tene?
05:59 purl tene has karma of 275
05:59 Tene 5*5*11
05:59 purl 275
05:59 Tene Lame.
06:01 Tene I'm out of trivial updates, though. :(
06:01 pmichaud are we stuck on some things in hllmagic?
06:03 Tene Check the recent log messages... I'm not sure if these test failures are because of the branch or not.  I need to verify them in trunk.
06:03 Tene Are we fixing everything before merging, or merging and leaving it up to HLL authors to fix?
06:03 pmichaud fixing what we can before merging.
06:03 MariachiElf Tene:  How can I find kjs' tutorial?
06:04 pmichaud MariachiElf: parrotblog.org
06:04 pmichaud look for "tutorial" in the bottom right somewhere
06:04 Tene Yeah, in the right column... start here: http://www.parrotblog.org/200​8/03/targeting-parrot-vm.html
06:05 MariachiElf Ok
06:05 pmichaud Tene: I don't know that we have to fix _every_ hll before merging back.
06:05 MariachiElf I was looking to find kjs' blog posts
06:05 pmichaud oh yes, I was going to write a message to the mailing list about the hllmagic branch... will do that now.
06:05 Tene Fantastic
06:14 pmichaud sent.
06:18 Tene There's still plenty of stuff in runtime/parrot/library to convert, too.
06:19 pmichaud ...that uses PCT or PGE?
06:19 pmichaud I'm not wanting to convert all libraries...just the things that are calling PCT/PGE .
06:19 Tene No, just namespaces
06:19 Tene It'll be good to convert them, but not necessary or very relevant
06:20 Tene Just nice for consistency to get rid of things like: new "Getopt::Obj"
06:20 pmichaud I don't see that as part of hllmagic branch...just general code cleanup.
06:20 Tene Right.
06:23 chromatic Hm, what's the mailing address for Rakudo bugs?
06:24 MariachiElf Anyone have any thoughts on what I should call this language?   I've been throwing ParrotTalk around but I could use some fodder for thought. :)
06:24 Tene purl: rakudobug?
06:24 purl rakudobug is mailto:rakudobug@perl.org
06:24 chromatic grazie
06:25 * MariachiElf is at the part of the tutorial where you generate the language shell files and I just realized I'm really going to have to call it something.
06:25 Tene purl++
06:25 MariachiElf Tene: That's just wrong on so many levels ;)
06:26 chromatic MariachiElf, monkeybutter
06:26 chromatic Or RegencyElf, if you like it old school.
06:26 Tene Oh, the language shell script.  I need to fix it for this branch.
06:26 MariachiElf chromatic: That's better :)
06:26 ab5tract Parakeet
06:26 purl Parakeet is in explorer mode
06:27 ab5tract since its all the power of Parrot
06:27 * moritz suggests "zwoelf". In German "elf" also stands for the number 11, and 12 is "zwoelf" ;)
06:27 dalek r31854 | tene++ | hllmagic:
06:27 dalek : [mk_language_shell]: NameSpace update
06:27 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31854
06:27 moritz might be a bit hard to pronounce for the casual English speaker, though ;)
06:27 ab5tract karma Tene
06:27 purl tene has karma of 276
06:28 ab5tract MariachiElf: do you want something that plays along with the them of Talk
06:28 ab5tract s`them`theme`
06:29 Tene 2*2*3*23
06:29 purl 276
06:29 Tene Boring
06:29 MariachiElf ab5tract: I was thinking so since folks who are familiar with it will get it, and there aren't really any good parallels outside of the xTalk family
06:29 Tene Talk?  I thought these languages were *Card named.
06:29 MariachiElf Tene: The products and environments are Card named
06:30 Tene Ah.
06:30 MariachiElf Tene: The scripting language is typically "Talk" named
06:30 uniejo joined #parrot
06:30 chromatic branmacmuffin
06:31 moritz TalkingToes
06:32 MariachiElf chromatic: I'd feel like I constantly be going to the bathroom if I spent that much time with a BranMcMuffin :)
06:32 chromatic We once had a dwarf in a game named Stumpy Wizzleteats.
06:33 ab5tract haha
06:33 MariachiElf Why would anyone call a game Stumpy Wizzleteats --- I mean a dwarf maybe ... but a game?! ;)
06:34 chromatic It made sense at the time.
06:34 MariachiElf chromatic: hehe
06:34 ab5tract its sort of too bad Squaak is taken in this instance actually
06:34 * MariachiElf had a hard time even spelling Wizzleteats
06:35 ab5tract Chirp?
06:35 ab5tract ChirpChirp?
06:35 dalek r31855 | pmichaud++ | trunk:
06:35 dalek : [rakudo]:  Bring namespace parsing into closer alignment with STD.pm.
06:35 dalek : This also adds the 'parse_name' method to Perl6::Compiler, and allows
06:35 dalek : us to eliminate some inline PIR from actions.pm.
06:35 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31855
06:35 ab5tract or, in one of my favorite interpolations, replace 'ch' with 'x'
06:35 ab5tract xirp
06:36 MariachiElf Wow there's actually a WikiPedia entry for xTalk
06:36 MariachiElf http://en.wikipedia.org/wiki/XTalk
06:38 MariachiElf ParrotScript
06:38 MariachiElf ParrotTalk
06:38 pmichaud there's a wikipedia entry for almost everything.  Ironically, there's not an entry for "Almost everything".  :-)
06:39 ab5tract XIRP - xtalk if running parrot
06:39 MariachiElf ab5tract: That's amazing
06:39 Tene and it sounds like "chirp"
06:39 ab5tract xirp xirp :D
06:40 MariachiElf xirp xirp does have an attraction
06:40 Tene I like it.
06:42 MariachiElf It looks like"Runtime Revolution's product (the only commercial one left I'm aware of) uses TranScript
06:43 Tene 'xirp' doesn't turn up much on google.
06:43 MariachiElf That was my next test
06:43 MariachiElf ok xirp it is
06:44 pmichaud alas, xirp.org and xirp.com are already taken :-P
06:44 MariachiElf It of course could always change but it's easy to type
06:44 MariachiElf I'm not so ambitious at this moment to even think about hosting a project page
06:44 Tene xirp.net isn't
06:45 moritz and xirpcode.org is also free ;)
06:45 ab5tract parotfoundation/xirp/indexmcgi :)
06:45 * MariachiElf is going to feel lucky if he can get his hands on a language spec and get through the tutorial
06:46 pmichaud when naming a project I always check for domain name availability :-)
06:46 MariachiElf Keep in mind cirp is just the scripting language
06:46 MariachiElf the real power is the "Card" part
06:46 MariachiElf I love this example:
06:46 MariachiElf if the last word of the first line of card field 1 contains "hello" then say "Okay"
06:46 pmichaud ah, so it should be called "picard"?   "Make it so!"
06:47 MariachiElf Is there a particularly talkitive bird out there?
06:48 pmichaud When I was little I used to see a big yellow bird on television that talked an awful lot.
06:48 MariachiElf lol
06:48 MariachiElf I don't know if that bird was ever classified
06:48 MariachiElf Kind of like an overgrown chicken
06:49 masak joined #parrot
06:50 MariachiElf Well one step at a time
06:51 MariachiElf I'll worry about the name for the environment once I actually have a language to put in it
06:51 MariachiElf :)
06:51 * MariachiElf is trying to actually have the problem before trying to solve it.
06:52 MariachiElf Holy Moly: http://www.creysoft.com/xtalk/
06:52 MariachiElf One of those has got to have some sort of published langauge spec
07:03 ab5tract MariachiElf: maybe this creysoft character may know where we can find a spec
07:04 MariachiElf Many moons ago -- like 199 I was on a list that was trying to do an Open Source xTalk
07:04 MariachiElf I recognize many of the names from that list
07:04 MariachiElf http://www.mail-archive.com/opencard@metacard.com/
07:05 MariachiElf I remember someone saying they had the HyperCard spec
07:22 ab5tract so is it really gonna be called xirp?
07:23 ab5tract the scripting language
07:23 purl the scripting language is mu or http://foldoc.doc.ic.ac.uk/foldoc/f​oldoc.cgi?query=scripting+language
07:23 ab5tract purl is great
07:26 * masak thinks purl is stupid
07:28 Tene karma masak?
07:28 purl masak has karma of 30
07:28 Tene purl thinks masak is okay
07:28 purl OK, Tene.
07:31 masak Tene: see?
07:31 Tene Okay, maybe a little.
07:31 Tene purl: lolcode?
07:31 purl i think lolcode is HAI CAN HAS STDIO? VISIBLE "HAI WORLD!" KTHXBYE or at http://lolcode.com/
07:31 Tene purl: purl?
07:31 purl or a masochist or a sphincter
07:31 chromatic joined #parrot
07:31 masak actually, I sense there's some sort of "Three Laws of IRC-Robotics", one of which purl breaks
07:32 Tene purl: three laws?
07:32 purl tene: i don't know
07:33 masak something like "a bot should only speak when spoken to, or when there's undeniable cause (such as an external factor changing)"
07:33 Tene Ah, yes.
07:34 viklund joined #parrot
07:35 dalek r31856 | chromatic++ | trunk:
07:35 dalek : [GC] Cleaned up some signedness and conversion warnings in GC.
07:35 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31856
07:35 iblechbot joined #parrot
07:35 ab5tract purl: version
07:35 purl ab5tract: sorry...
07:35 ab5tract hehe
07:36 MariachiElf gnite all
07:36 dalek r31857 | chromatic++ | trunk:
07:36 dalek : [library] Fixed exception message when unable to load PostgreSQL library (Hans
07:36 dalek : Dieter Pearcey, RT #59742).
07:36 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31857
07:40 moritz msg chromatic has masak's CLA arrived already?
07:40 purl Message for chromatic stored.
07:53 cosimo joined #parrot
07:56 TiMBuS joined #parrot
08:02 particle1 joined #parrot
08:04 particle joined #parrot
08:13 cotto irc log?
08:13 purl i heard irc log was http://irclog.perlgeek.de/parrot/
08:22 cotto has the pidigits test of t/examples/shootout.t been failing for anyone else?
08:29 cotto hmmm.  it seems to pass when run separately with prove
08:52 omega joined #parrot
09:07 barney joined #parrot
09:12 rafl joined #parrot
09:14 * barney is checking out the hllmagic branch
09:22 cotto joined #parrot
09:36 tomyan joined #parrot
09:47 kj joined #parrot
09:47 particle joined #parrot
09:52 GeJ joined #parrot
10:00 Theory joined #parrot
10:28 particle joined #parrot
10:38 donaldh joined #parrot
10:53 tomyan joined #parrot
11:09 gaz joined #parrot
11:27 bacek joined #parrot
11:56 bacek joined #parrot
12:50 tetragon joined #parrot
12:51 omega joined #parrot
13:08 gryphon joined #parrot
13:11 slavorgn joined #parrot
13:29 jhorwitz joined #parrot
13:40 Lorn joined #parrot
13:46 ruoso joined #parrot
14:05 particle moritz, masak has commit rights as of a few days ago
14:06 moritz oh, I didn't know that
14:06 particle same time as dmknopp
14:06 moritz masak++ # welcome our new parrot overlord
14:06 masak thank you.
14:06 masak I will try to weild my powers wisely and with good judgement
14:07 masak I guess the commit rights are tied to my email address?
14:07 moritz to your bitcard account
14:07 masak ah ok
14:07 masak good thing I had one, then
14:09 grim_fandango joined #parrot
14:14 coke joined #parrot
14:14 * Coke hurls http://code.google.com/p/p​artcl/issues/detail?id=67
14:14 Andy joined #parrot
14:14 Coke (in case anyone is bored. =-)
14:15 * Coke pokes mdiep
14:15 Coke mdiep++ for his recent commit to partcl!
14:15 * mdiep stirs
14:16 mdiep it was a dirty hack, but it passes the test case :)
14:16 masak does Parrot overlordship entail any liabilities, such as attending the Tuesday #parrotsketch meetings?
14:17 moritz it's recommended, but probably not required
14:17 Coke it's not required. =-)
14:17 Coke it's encouraged, but even folks on teh board have been known to miss meetings.
14:17 * Coke whistles innocently.
14:17 masak heh.
14:17 Coke mdiep: that ticket has another bit of wrongly failing code for you to fix.
14:18 masak I'll certainly maximize my attendance, out of interest if nothing else.
14:18 particle who let mdiep back in here?
14:19 particle did i say that aloud?
14:19 particle hi mdiep!
14:19 mdiep it was Coke. it's his fault.
14:19 particle you at aaps?
14:19 mdiep hi jerry. :)
14:19 particle er, aapl?
14:19 mdiep yes I am
14:19 particle i hope your black turtlneck fits well
14:20 mdiep it does! it keeps me nice and toasty. honestly, I don't know why more people don't wear them.
14:21 mdiep how's life up in the rainy city?
14:21 mdiep Coke: which ticket?
14:21 purl somebody said which ticket was this now?
14:21 particle it's, well, the rainy season has begun
14:22 * Coke hurls http://code.google.com/p/p​artcl/issues/detail?id=67
14:22 mdiep I have a friend who I think is moving up that way. I'll probably fly up sometime soon after he does.
14:22 jhorwitz particle: 70s and sunny here, and supposed to be in pittsburgh too.  so get over here!
14:26 Coke particle: I assigned a ticket in the partcl queue to you.
14:26 particle fabulous, 50s and cloudy here
14:27 particle coke: i noticed that. i may look at it someday, too :)
14:27 particle mdiep: it'd be good to see you, drop me a line when you know more
14:27 * Coke i sad that he gave up on travelling to the PPW.
14:28 Coke now that he knows that there will be parroteers there.
14:28 mdiep particle: will do :)
14:29 mdiep Coke: issue 67 is a parrot bug, from the looks of things
14:29 mdiep it bus errors in $P0 = mod a, b
14:30 Coke can you make the same error happen sans tcl?
14:33 mdiep not off the bat. though parrot says that 1% 12135435435 == -2147483647, which isn't right
14:35 mdiep oh, I see. something else is happening.
14:38 jan joined #parrot
14:41 dalek r31858 | fperrad++ | trunk:
14:41 dalek : [WMLScript] pdd27mmd merge
14:41 dalek : - use modulus instead of cmodulus
14:41 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31858
14:56 johbar joined #parrot
15:04 Debolaz joined #parrot
15:07 Coke mdiep++
15:07 Coke I am not in here often, but i wonder if having the partcl commit feed in here would be a good thing or not.
15:08 * moritz wouldn't mind
15:08 * masak neither
15:09 dalek r31859 | fperrad++ | trunk:
15:09 dalek : [WMLScript] pdd27mmd merge
15:09 dalek : - fix typo (MUTLI)
15:09 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31859
15:11 mdiep I don't expect we'd flood the channel by any means
15:11 apeiron joined #parrot
15:16 * Coke points at http://code.google.com/p/partcl/feeds if someone with a bot wants to set it up.
15:17 mdiep the bot works off the repo itself
15:19 moritz mdiep: then http://partcl.googlecode.com/svn/
15:23 * mdiep is off to work
16:01 * Coke vaguely wonders how to track how parrot is going off into the weeds.
16:06 tetragon joined #parrot
16:29 Coke ahhhh. looks like it's garbage collection.
16:30 Coke I have a bit of code that takes 32s to run with -G; takes 1m48s without
16:34 Coke seen chromatic?
16:34 purl chromatic was last seen on #parrot 9 hours, 59 minutes and 52 seconds ago, saying: It made sense at the time.
16:37 Coke opbots, names
16:38 Coke opbots, trust cotto
16:38 clunker3 But I do not trust you Coke
16:38 slavorg Ok
16:38 slavorgn But I don't trust you there, Coke
16:38 Coke irc?
16:41 dalek r31860 | fperrad++ | trunk:
16:41 dalek : [WMLScript] pdd27mmd merge
16:41 dalek : - fix WmlsInvalid (is_equal & cmp)
16:41 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31860
16:54 barney joined #parrot
17:16 barney branches/hllmagic/t/stm/basic_mt_4.pir hangs on Linux heist 2.6.24-19-generic
17:17 Tene barney: does it hang in trunk?
17:18 barney No
17:19 Tene Seems to work fine for me in hllmagic
17:30 Tene I have a vague suspicion that I know what's going on with chromatic's recursive error handler...
17:31 Tene I won't be here enough to fully debug it until tonight, though, and I doubt I'll remember by then.
17:34 Tene So if someone wants to remember for me and harass me tonight, that would be great.
17:37 Infinoid Coke, mdiep: dalek runs off of the SVN::Web RSS feed, but we have hacked in a couple of custom item tags for things like the branch name, I don't remember what else.
17:38 Infinoid it may or may not be compatible, but it's worth a shot if someone wants to have a play.  the svn-feed stuff is a botnix plugin written in perl, available on request.
17:38 Infinoid (I don't think it should be too difficult to get it working.)
17:48 Lorn joined #parrot
17:49 dalek r31861 | bernhard++ | trunk:
17:49 dalek : [Pipp] Unbreak some tests by no longer using the 'n_*' opcodes.
17:49 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31861
17:52 Tene pmichaud: if pipp is unused enough that the n_* opcodes in there were just noticed, I vote we merge hllmagic today
17:53 barney k
17:56 rdice joined #parrot
17:58 particle if you merge in the next 15m, i'll have it before my flight :)
17:58 Tene particle: you want it before your flight?
17:59 particle actually, it might unblock my is export() work
17:59 Tene Okay.
17:59 Tene I'll merge now.
18:01 Tene A few conflicts... cleaning up...
18:11 particle tick tick tick :)
18:14 Tene Ack!
18:14 Tene I'm hurrying!
18:14 Coke Grep!
18:14 Infinoid Sed!
18:15 Coke sed what what?
18:15 hercynium joined #parrot
18:15 * Tene done with conflicts, running basic tests
18:16 particle tene++ in anticipation
18:16 Tene You have a couple of minutes for me to compile and run some rakudo tests?
18:18 particle sure
18:18 Tene looks like rakudo builds okay...
18:22 Tene committing
18:24 Tene Huh... weird response...
18:24 dalek r31862 | tene++ | trunk:
18:24 dalek : Merge hllmagic branch into trunk.
18:24 dalek : Mostly looks okay.
18:24 dalek : If anyone runs into problems fixing the last few issues with their languages, address issues to Tene or pmichaud.
18:24 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31862
18:24 Tene RA layer request failed: MERGE request failed on '/parrot/trunk': MERGE of '/parrot/trunk': 200 OK (https://svn.perl.org) at /usr/bin/git-svn line 453
18:24 particle tene++ building now
18:24 Tene looks like it went through okay...
18:24 particle yep, just do svn up to clean up
18:25 Tene So, I aim to have HLL interop working by the end of the weekend.
18:27 Coke why is there commented out code in, e.g. compilers/nqp/bootstrap/nqp.pir ?
18:27 particle coke: it's the pir code in nqp
18:27 Tene not all of it
18:27 purl not all of it is, like, fast, but much of it is
18:27 Tene purl: forget not all of it
18:27 purl Tene: I forgot not all of it
18:28 Tene particle: look at line 86
18:28 particle tene: cotto++ did good work with phparray for hll interop, you can take a look at that
18:29 Tene Eh?  What about phparray needs to interoperate?
18:29 particle if you want to take an array from another language and make it into a phparray
18:29 Tene Huh.
18:29 * particle needs to pack the laptop and hit the road
18:30 nopaste "tene" at 97.117.75.168 pasted "spectest_regression failures after hllmagic merge" (18 lines) at http://nopaste.snit.ch/14266
18:30 Tene Have a nice flight, particle.
18:30 particle tene++ thanks!
18:30 Tene s/nice/productive (measured in commits)/
18:30 particle &
18:42 dalek r31863 | fperrad++ | trunk:
18:42 dalek : [Markdown]
18:42 dalek : - update CREDITS
18:42 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31863
19:03 particle1 joined #parrot
19:07 dalek r31864 | bernhard++ | trunk:
19:07 dalek : [docs] Quiet POD complaints.
19:07 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31864
19:10 Lorn joined #parrot
19:57 particle joined #parrot
19:57 pmichaud hello from chicago o'hare :-)
19:59 Coke pmichaud: ho
20:00 Coke I am again disappoitned I won't see you guys this weekend.
20:00 dalek r31865 | bernhard++ | trunk:
20:00 dalek : [Pipp] Get rid of some 'to_number' method calls,
20:00 dalek : pass all but one test.
20:00 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31865
20:00 pmichaud I'm disappointed also.
20:00 pmichaud but hey, you're only an irc channel away!
20:01 Coke I just figure in person I could trick you into working on tcl! =-)
20:01 selenamarie joined #parrot
20:01 pmichaud yes, you might be able to do that.
20:02 Coke odds are slightly better than on IRC, anyway. =-)
20:02 Coke when are you getting to PPW?
20:02 pmichaud flight arrives 6:30ish, I think.
20:02 pmichaud I don't know how long it will take to get from airport to hotel
20:06 * pmichaud files a bug he found on the plane.
20:08 Coke pmichaud: is that because $P0 is getting modified in the first one?
20:09 Coke (nope)
20:10 pmichaud no, it's just that opcode.
20:10 pmichaud (and possibly related shifting opcodes -- I didn't try them all.)
20:10 Coke odd that no core tests are failing. =-)
20:12 pmichaud argggh, it's going to be tough to work on the plane when spectest_regression is failing in trunk.  :-(
20:14 pmichaud was it passing in the branch before the merge?
20:18 Coke chromatic++
20:18 Coke chromatic++
20:22 particle joined #parrot
20:24 selenamarie joined #parrot
20:36 selenamarie joined #parrot
20:37 dalek r31866 | fperrad++ | trunk:
20:37 dalek : [WMLScript] pdd27mmd merge
20:37 dalek : - add some VTABLE modifier
20:37 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31866
20:38 Tene null pmc access in get_parrotclass
20:41 dalek r31867 | pmichaud++ | trunk:
20:41 dalek : [p6object]: Handle case of looking up a non-existent class (e.g., for 'isa').
20:41 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31867
20:42 pmichaud okay, r31867 seems to resolve everything except S05-transliteration/with-closure.t
20:43 pmichaud and I'm not sure that's related to hllmagic.
20:43 Tene I'll look into it.
20:43 pmichaud still 10 minutes until my flight :-)
20:44 pmichaud (hll interoperability) -- fwiw, my intermediate step isn't to get hll interop working -- simply to enable individual languages to use the .HLL directive
20:44 Tene I'll make a branch and start testing that today, yes?
20:44 pmichaud hll interop is a lot more effort.
20:45 pmichaud sure, that'd be great.  I'm sure you'll run into quite a few problems because PCT isn't exporting its symbols yet.
20:45 Tene For what definition of 'interop'?
20:45 pmichaud 'interop'  being able to have programs in one HLL load modules and talk to code written in other HLLs
20:45 pmichaud I'm just wanting to get HLL mapping of parrot types to work.  :-)
20:56 peepsalot joined #parrot
20:59 pmichaud plane ... bbl
20:59 pmichaud tene++   # merging hllmagic back into trunk
21:00 pmichaud I'm glad that's merged -- it'll be easier to work on namespaces in rakudo on the plane.
21:00 Tene particle++ # asking me to merge
21:00 pmichaud well, fortunately the spectest_regression problems were a fairly easy fix.  :-)
21:00 Tene They looked like they would be
21:01 pmichaud I almost have interpolated namespaces parsing in rakudo... just need to do some testing
21:02 pmichaud anyway, flight.  back later!
21:22 bacek joined #parrot
21:26 dalek r31868 | moritz++ | trunk:
21:26 dalek : [rakudo] remove S29-hash/delete.t from spectest_regression. Its syntax has
21:26 dalek : changed, and is not yet supported. I'll open a TODO ticket.
21:26 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31868
21:27 cotto cool.  It looks the svn::web interface is back.
21:28 moritz yes. Infinoid and I fixed it.
21:30 cotto moritz++
21:30 cotto Infinoid++
21:42 Infinoid its sort of half-back.
21:42 Infinoid the RSS feed never really stopped working (that's what dalek watches)... we just got the patch view working too
21:42 Infinoid but if you try to browse the filesystem, it's still broken...
21:45 dalek r31869 | fperrad++ | trunk:
21:45 dalek : [WMLScript] pdd27mmd merge
21:45 dalek : - bitwise methods are now VTABLE
21:45 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31869
21:46 Limbic_Region joined #parrot
21:52 dalek r31870 | fperrad++ | trunk:
21:52 dalek : [WMLScript] pdd27mmd merge
21:52 dalek : - fix typo
21:52 dalek : - All tests PASS on linux
21:52 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31870
22:08 jonathan joined #parrot
22:10 dalek joined #parrot
22:12 rdice joined #parrot
22:21 TiMBuS joined #parrot
22:23 Whiteknight joined #parrot
22:33 Zaba_ joined #parrot
22:34 selenamarie joined #parrot
22:39 davidfetter joined #parrot
22:44 Limbic_Region !seen particle
22:44 clunker3 particle was last seen on #parrot 4 hours, 13 minutes and 10 seconds ago, saying: &
22:50 tetragon joined #parrot
22:53 Theory joined #parrot
23:01 johbar joined #parrot
23:01 davidfetter !clunker3
23:02 davidfetter !seen clunker3
23:02 clunker3 I'm right here!
23:42 bacek rakudo: my %h; say %h{"foo"}
23:42 polyglotbot OUTPUT[Use of uninitialized value␤␤]

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

Parrot | source cross referenced