Camelia, the Perl 6 bug

IRC log for #parrot, 2008-10-03

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:09 AndyA joined #parrot
00:19 petdance joined #parrot
00:22 Tene pmichaud: your tests are passing.  I need more failures.  :)
00:23 Tene going to do the :: split on the plane now.
00:23 Tene Then look into caller hll
00:25 particle yay
00:25 dalek john.haltiwanger@gmail.com | PMC Documentation Guidelines:
00:25 dalek link: http://www.perlfoundation.org/parrot/i​ndex.cgi?pmc_documentation_guidelines
00:26 Khisanth joined #parrot
00:27 dalek john.haltiwanger@gmail.com | PMC Documentation Guidelines:
00:27 dalek link: http://www.perlfoundation.org/parrot/i​ndex.cgi?pmc_documentation_guidelines
00:28 dalek john.haltiwanger@gmail.com | PMC Documentation Guidelines:
00:28 dalek link: http://www.perlfoundation.org/parrot/i​ndex.cgi?pmc_documentation_guidelines
01:29 dalek r31576 | Whiteknight++ | trunk:
01:29 dalek : [Config] Add file-level and function-level documentation for config/gen/platform/win32/stat.c
01:29 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31576
01:31 petdance joined #parrot
01:43 dalek r31577 | Whiteknight++ | trunk:
01:43 dalek : [Config] Add file-level and function-level documentation for config/gen/platform/win32/misc.c
01:43 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31577
01:53 Khisanth joined #parrot
02:17 tetragon joined #parrot
02:19 Zaba_ joined #parrot
02:25 dalek r31578 | Whiteknight++ | trunk:
02:25 dalek : [Config] Add file-level and function-level documentation for compilers/imcc/debug.c
02:25 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31578
02:57 ab5tract joined #parrot
03:24 ab5tract joined #parrot
03:41 ab5tract Tene: what determines the number passed to the plan method in the cardinal tests 'array.t' and 'hash.t' ?
04:04 Tene ab5tract: the number of tests that you run in the file.
04:04 Tene ab5tract: the reason we pass that up fron is so that we can see when a test doesn't fully complete, or dies early.
04:04 ab5tract k. i was picking up on that pattern. never written for tap before
04:06 ab5tract working on hash methods atm.
04:08 Tene Fantastic!
04:08 Tene I'd really love to get more tests. :)
04:09 ab5tract i'm curious: when is 'is' more appropriate that "puts 'ok #{test_num}"?
04:09 Tene I'm meeting with tewk sometime soon to hack on cardinal together.  It will be nice to have a list of things to do.
04:09 Tene ab5tract: you want to use 'is'
04:09 Tene the 'puts ok' stuff is left over from before we could load external libraries.
04:10 ab5tract okay cool!
04:10 ab5tract so instead of puts "1..3" i do plan 3
04:10 Tene Using the test library, I'm going to add things to it to track how the test goes and track 'skip' and 'todo' and such.
04:10 ab5tract and require 'test' in every test
04:10 Tene Right.
04:12 Tene Yes.
04:14 ab5tract okay, i'll finish up a couple and post them in a gist, you can tell me if it looks alright
04:14 Tene Excellent!
04:14 purl EGG-see-lent!
04:26 ab5tract http://gist.github.com/14502
04:28 ab5tract oops, there was an error
04:28 ab5tract fixed now
04:30 Tene for the ones where you check against true or false, use 'ok' or 'nok' instead
04:30 Tene is preferred
04:31 Tene in delete.t, you have this line:
04:31 Tene is a.delete "c"
04:32 Tene Is that right?  it looks suspicious to me.
04:33 Tene I'll be back in a while and I'll commit all of this for you.  :)
04:36 ab5tract Tene: perhaps you expect parentheses?
04:44 Khisanth joined #parrot
04:47 ab5tract nm i see what you mean.
04:48 ab5tract Tene: should i write a file for every single method or include synonyms in the first nym i come across?
04:56 Bzek joined #parrot
05:02 Zaba joined #parrot
05:07 Khisanth joined #parrot
05:08 Zaba_ joined #parrot
05:25 ab5tract Tene: how to test "should raise TypeOfError" ?
05:32 Tene ab5tract: I lean towards including synonyms, but it's not a big deal either way.
05:32 Tene ab5tract: I haven't dealt with exceptions yet... lemme see how that's traditionally checked in other places.
05:32 Tene Or, if you have a suggestion, we could implement that.
05:32 ab5tract cool
05:33 ab5tract well, bacon is a tiny library. it is to rspec what nqp is to perl6.
05:33 ab5tract almost there but way simpler
05:33 Tene Oh, interesting.
05:33 Tene link?
05:34 ab5tract http://www.ruby-forum.com/topic/137928
05:35 ab5tract there is no web site or rdocs
05:36 Tene That seems worth pursuing.
05:38 ab5tract its got tap :)
05:38 Tene The only issue is how much we need to implement before we can use it.
05:39 ab5tract right.
05:39 Tene I downloaded it, looking at it now...
05:39 ab5tract also i was appreciating the simplicity of #is
05:40 Tene I need to figure out what :symbols are.
05:40 ab5tract i mean, it should be easy enough to add a 'bork' method
05:40 ab5tract to test.rb
05:40 ab5tract well, assuming rescue is implemented...
05:40 Tene bacon requires exceptions, looks like
05:41 ab5tract yeah
05:41 Tene rescue isn't implemented, but it will be soon.
05:41 ab5tract okay
05:41 Tene I'm going to add exception handler support to PCT directly, so everyone gets it for cheap.
05:41 ab5tract excellent!
05:41 * purl plays air guitar
05:42 Tene Also it re-opens classes, which we don't support yet.
05:44 particle1 joined #parrot
05:45 Tene It doesn't look too far off, though.
05:45 Tene It's a good goal to work for.
05:45 Tene Thanks for the pointer! :)
05:46 ab5tract no doubt. i'm happy to contribute
05:47 ab5tract i've been following parrot since september 10 2001
05:47 ab5tract :)
05:47 ab5tract comments are implemented correct?
05:48 Tene They should be.  I've seen a little bit of weird behavior, but I think I have it all worked out.
05:48 Tene If you run into a weird situation with comments not parsing, please tell me.
05:48 Tene I really need to work on getting the parser to give more descriptive error messages in appropriate places.
05:51 ab5tract make test fails
05:52 ab5tract in languages/cardinal
05:52 Tene Yes, there are a few tests that don't pass yet.
05:52 Tene Things I still need to implement.
05:53 ab5tract okay. i guess i meant "make test fails and it doesn't look like my fault" ;)
05:53 * ab5tract knows the nature of the "benchmark failure"
05:54 Tene Those are features that require a bit more work and thought than the others, so I've been tackling the easy ones first. :)
05:57 ab5tract how do i submit a patch?
05:57 Tene So far the only person working on cardinal with commit rights is me, so you just get a patch to me however you like, but it looks like dmknopp is getting a commit bit any day now, so it might be better to post patches to the list.  Check out docs/submissions.pod
05:58 Tene Well, we could really track patches wherever we like
06:00 Tene The parrot policy is that if your patches are good after a while, we can get you a commit bit if you fill out some paperwork and mail it in and such.  If you don't want to deal with that, we could look into moving cardinal out of parrot's repository to make it more accessible.
06:01 Tene It's really up to whoever feels strongly enough about it to express an opinion.
06:02 ab5tract hmm
06:02 ab5tract im coming from git being sort of my first vcs i've used in production
06:02 Tene My preference would probably be to have a separate tracker for patches and bugs and to keep cadinal in the parrot repo, at least for now.
06:02 Tene I only use git-svn.  I can't stand using straight svn.
06:03 ab5tract haven't tried it.
06:03 Tene I want to keep cardinal in the parrot repo for now because of the significant changes going on in PCT as it develops.  Once PCT is more stabilized, I'd be more comfortable separating it out.
06:03 ab5tract no reason not to stay inside of parrot however
06:03 Tene Also, I like having cardinal in the standard parrot releases, too.
06:03 ab5tract pct is still moving? i didn't realize
06:04 Tene Yeah.  I'm working on patches for it right now.
06:04 ab5tract does that mean we might be able to implement functionality (at some level) using the ruby grammar?
06:04 Tene I don't understand the question.
06:05 ab5tract sorry, i meant "coding functionality in ruby instead of pir"
06:05 ab5tract not that i have anything against pir
06:05 ab5tract in fact i cant wait to try coding it
06:06 Tene Yes, that's certainly an option for some things.  Standard library sort of things, once we have enough base functionality.
06:07 ab5tract awesome
06:08 ab5tract at that point there might be a chance that rubinius' code will come in handy
06:08 ab5tract as they are trying to "write ruby in ruby"
06:09 ab5tract so whats the proper tap idiom for checking for an exception?
06:09 Tene tap just defines an output standard.
06:09 ab5tract or how do you want to see it implemented in test.rb?
06:10 Tene Lemme look around to see how other implementions do it.
06:10 ab5tract right, but ok and nok are part of that
06:10 ab5tract right?
06:10 purl right is, like, wrong. or more likely to not have problems with torturing people
06:10 Tene They're pretty common.  I'm basically duplicating perl's Test::More API.
06:13 Tene I don't think we should worry about it for now until I get the exception stuff in PCT sorted out.
06:14 ab5tract okay cool
06:15 ab5tract i just commented out a mimic of the test
06:16 ab5tract Tene: did you say you're grappling with :symbol?
06:16 Tene Yeah, I don't know what those are.
06:17 ab5tract i may be somewhat wrong, but i think it is basically a special array of strings in a namespace
06:18 ab5tract being a symbol means there is no string stuff attached
06:18 ab5tract it just points directly to its own memory space
06:18 Zaba joined #parrot
06:19 ab5tract and so it comes in handy with hashes, since the symbols memory space becomes a pointer for the memory space of its value
06:19 ab5tract again, i may be wrong on some details, but that is the gist of it
06:23 ab5tract did that make any sense? :)
06:23 Tene Hmm.  Not quite.  I'll read it again in the morning. :)
06:24 dalek r31579 | tene++ | hllmagic:
06:24 dalek : [P6object]
06:24 dalek : * Store protoobjects in the right HLL namespace.
06:24 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31579
06:24 ab5tract basically the cruby interpretor has some extra cycles that occur in the interpreting of strings
06:25 ab5tract :symbols avoid that
06:25 ab5tract so they become almost like constants
06:25 ab5tract then again, maybe i should try explaining again in the morning
06:25 ab5tract if you don't have the hummingbird book, i definitely recommend it
06:27 Ademan joined #parrot
06:29 uniejo joined #parrot
06:34 ab5tract oh and there is always _why's http://poignantguide.net
07:05 * Tene sleeps.
07:06 moritz good morning
07:10 ab5tract motning moritz :)
07:11 moritz how far is the cardinal implementation?
07:11 moritz I mean if I find a ruby program in the wild, what are my chances to run it with cardinal?
07:12 Tene moritz: 0
07:12 ab5tract hehe
07:12 ab5tract maybe this one: hello_world.rb ;)
07:13 ab5tract does puts 'hellow world' work Tene
07:13 ab5tract ?
07:13 Tene yes
07:14 Tene moritz: similar to rakudo approximately 6-9 months ago, iirc.
07:14 Tene moritz: check out 'make test' in languages/cardinal.
07:16 moritz a few parse failures
07:16 Tene yeah
07:18 Tene There are still a few pretty basic parts missing.
07:23 moritz dpkg++
07:23 moritz I just removed the kernel that I'm running...
07:24 moritz "if you know exactly what you're doing, and are prepared to hose your system, say 'No'"
07:24 cotto just copy the one you're running from memory
07:32 ab5tract good night, happy hacking
07:35 Tene Programming while cats rub their faces on your face is hard.
07:36 moritz programming is hard, let's go shopping
07:39 apeiron joined #parrot
07:40 Tene will there be kittens with sharp claws at the mall?
07:41 moritz probably not, but you can't know without writing tests first ;)
07:52 dalek r31580 | cotto++ | trunk:
07:52 dalek : [pipp] implement freeze/thaw (working) and get_iter (not so much)
07:52 dalek : also add debugging symbols to pipp_hash.o
07:52 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31580
08:01 iblechbot joined #parrot
08:05 johbar joined #parrot
08:10 particle joined #parrot
08:24 masak joined #parrot
08:24 dalek r31581 | tene++ | hllmagic:
08:24 dalek : [p6object]
08:24 dalek : * Split new class names on :: and create them in the right HLL.
08:24 dalek : * This only works on classes that have a namespace.  Include a workaround for those that don't.
08:24 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31581
08:27 Tene purl: msg pmichaud What's the right behavior when asked to create a class that doesn't already have a namespace?  Do we need to make the newclass opcode smart enough to understand an array of strings representing a namespace?
08:27 purl Message for pmichaud stored.
08:42 dalek r31582 | rurban++ | trunk:
08:42 dalek : [cage] remove extra / in "src/cpp//Makefile"
08:42 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31582
08:50 tomyan joined #parrot
08:53 Tene moritz: still up?
08:53 purl The Canadian Dollar
08:53 moritz Tene: sure, it's 11am here ;)
08:53 Tene Oh, right.
08:54 Tene Is this appropriate material for parrotblog: http://pleasedieinafire.ne​t/~tene/imccblogpost.html ?
08:55 moritz Tene: I'm not involved wiht parrotblog, but probably yes ;)
08:55 Tene You're not?  Huh.  I'm curious about why I thought you were.
08:56 moritz I've published on rakudo.org
08:56 moritz maybe you confused that
08:56 Tene That sounds likely.
08:58 Tene Hmm.
08:59 Tene No, it's not the same type of post as the rest of the posts there.  I'll post to use.perl.org and try to harass Coke? into aggregating that into planet.parrotcode.org.
09:00 kj joined #parrot
09:00 dalek r31583 | kjs++ | trunk:
09:00 dalek : [pirc/new] More work in progress.
09:00 moritz not the same, but still interesting. And some people like diversity ;)
09:00 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31583
09:04 Tene Hmm.  Okay.
09:07 Tene There, posted.  If it's inappropriate I expect someone to complain to me.
09:07 moritz (ask-for-forgiveness-rather-​than-permission-principle)++
09:08 Tene (generate problems to motivate other people to make decisions for you)++
09:14 cotto joined #parrot
09:15 jan_ joined #parrot
09:28 rurban joined #parrot
09:39 barney joined #parrot
09:52 azawawi joined #parrot
09:53 azawawi hi
09:56 azawawi is there a ready parrot op code to get the current process id 'getpid()' ?
10:07 dalek r31584 | bernhard++ | trunk:
10:07 dalek : [Pipp] Add another test of fot-loop.
10:07 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31584
10:09 barney s/fot/for/
10:12 dalek r31585 | rurban++ | trunk:
10:12 dalek : [library] use Storable 2.13; protect against too old Storable-2.09, which cannot overload stringify
10:12 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31585
10:14 dalek r31586 | bernhard++ | trunk:
10:14 dalek : [Pipp] another for-loop test
10:14 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31586
10:24 dalek bernhard.schmalhofer@gmx.de | Pipp:
10:24 dalek link: http://www.perlfoundation.​org/parrot/index.cgi?pipp
10:25 dalek bernhard.schmalhofer@gmx.de | Pipp:
10:25 dalek link: http://www.perlfoundation.​org/parrot/index.cgi?pipp
10:35 bacek joined #parrot
10:42 dalek bernhard.schmalhofer@gmx.de | PHP OO for Perl 6 programmers:
10:42 dalek link: http://www.perlfoundation.org/parrot/i​ndex.cgi?php_oo_for_perl_6_programmers
10:44 iblechbot joined #parrot
11:00 bacek joined #parrot
11:05 bacek_ joined #parrot
11:50 bacek_ evening
11:55 tetragon joined #parrot
11:59 Zaba_ joined #parrot
12:33 dalek r31587 | bernhard++ | trunk:
12:33 dalek : [doc] Point to the Wiki-page for splint
12:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31587
12:44 pmichaud purl: msg Tene  -- every class in Parrot has a namespace.  Just use .get_namespace() on the class object to get it.
12:44 purl Message for tene stored.
12:44 pmichaud purl: msg Tene -- oh, you're talking about creating the class.  Simply do get_namespace() and then create a class from that.
12:44 purl Message for tene stored.
12:45 pmichaud argggh, can't type this morning.
12:53 * bacek cast teleport Pm to Australia
12:53 bacek It's late night here :)
12:54 bacek And much close to South Pole, so you can finish LTM overnight
12:56 bacek pmichaud: any particular task for "seasoned developers"?
13:09 pmichaud bacek: not at the moment.
13:09 pmichaud lots of internal refactorings taking place right now
13:10 bacek pmichaud: ok.
13:12 * bacek requesting LTM :)
13:12 pmichaud protoregexes are probably more important and useful than ltm at the moment.
13:17 petdance joined #parrot
13:17 bacek pmichaud: anyway, I can't help with this refactorings :)
13:48 gryphon joined #parrot
13:50 dalek r31588 | rurban++ | pdd30_install:
13:50 dalek : [languages] various minor fixes for make install
13:50 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31588
13:54 masak rakudo: class A { method x { self++ } }; A.new.x
13:54 polyglotbot OUTPUT[increment() not implemented in class 'A'␤current instr.: 'postfix:++' pc 15873 (src/gen_builtins.pir:9912)␤called from Sub 'parrot;A;x' pc 169 (EVAL_12:75)␤called from Sub '_block11' pc 33 (EVAL_12:16)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:480)␤called
13:54 polyglotbot ..from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc...
13:54 masak rakudo: class A { method x { self++ }; method increment { say "OH HAI" } }; A.new.x
13:54 polyglotbot OUTPUT[increment() not implemented in class 'A'␤current instr.: 'postfix:++' pc 15873 (src/gen_builtins.pir:9912)␤called from Sub 'parrot;A;x' pc 169 (EVAL_12:75)␤called from Sub '_block11' pc 33 (EVAL_12:16)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:480)␤called
13:54 polyglotbot ..from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc...
13:54 masak why doesn't that last one work?
13:55 pmichaud I think it's looking for an increment vtable function
13:55 pmichaud i.e., it's internal to Parrot
13:55 masak that's a bit unintuitive
13:55 masak at least to me
13:55 pmichaud when it's done right, ++ will be looking for the .succ method
13:55 pmichaud (See S03.)
13:55 masak I see.
13:56 masak so, is this worth filing a bug report over?
13:57 PerlJam Hmm.  does that mean someone can accidently change the meaning of ++ be creating a .succ method in their class?
13:57 pmichaud they change the meaning of ++ for that class, yes.
13:57 PerlJam s/be/by/
13:58 PerlJam That seems action-at-a-distancey
13:58 pmichaud well, one can change the meaning of prefix:?  by creating a .true method
13:58 pmichaud there's a fair bit of that going on in Perl 6
13:58 PerlJam Is there some level of warnings that will let the programmer know they're doing something like this?
13:59 pmichaud that might be added.
13:59 pmichaud S03:  Increment and decrement on non-<Str> types are defined in terms of the .succ and .pred methods on the type of object in the Scalar container. More specifically,
13:59 pmichaud ++$var --$var
14:00 pmichaud are equivalent to
14:00 pmichaud If the type does not support these methods, the corresponding increment or decrement operation will fail. (The optimizer is allowed to assume that the ordinary increment and decrement operations on integers will not be overridden.)
14:00 pmichaud $var.=succ $var.=pred
14:01 PerlJam I mean, I thought the whole idea of the twigils was to let you know the "scope of influence" ... it seems like methods that will change the behavior of built-in ops should have some sort of sigil/twigil
14:01 rurban joined #parrot
14:05 pmichaud or perhaps be all caps, or have some other sort of signifier, maybe.  Anyway, that's a p6l issue.  :-)
14:06 pmichaud or a #perl6 one :-)
14:11 Tene pmichaud: if I say: p6meta.new_class('OMG:LOL::WTF') without also doing a: .namespace ['OMG';'LOL';'WTF'] with some contents in it, should that succeed?
14:11 Tene That is happening a couple of times in p6object.t, or at least it looked so last night
14:11 pmichaud right now I'm working on refactoring the tests a bit
14:12 pmichaud but yes, it should succeed.  creating a class should not depend on defining a namespace
14:12 pmichaud (as it does not in Parrot, either.)
14:12 pmichaud note, however, that namespaces get created even if there's no .namespace directive for them
14:13 Tene get_root_namespace was returning null
14:13 jan joined #parrot
14:13 Tene pmichaud: specifically, if you remove the contents of the wtf::lol namespace I added last night, that test I added tails.
14:14 pmichaud okay.  but the answer is ultimately that creating a namespace should not be a prerequisite
14:14 pmichaud indeed, for many dynamic languages it cannot be.
14:14 pmichaud I also think we might not be able to use the HLL from the parrotclass
14:15 pmichaud i.e., if I do     .register(someclass, 'name'=>'XYZ::Foo')   then I want someclass to be registered in my HLL namespace, not the HLL namespace of someclass.
14:18 dalek r31589 | rurban++ | pdd30_install:
14:18 dalek : [perl6] add CP to Makefile
14:18 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31589
14:44 rurban finally the cygwin 0.7.1 release is ready, uploaded.
14:50 hercynium joined #parrot
14:53 uniejo joined #parrot
15:09 davidfetter joined #parrot
15:22 Theory joined #parrot
15:27 dalek r31590 | allison++ | trunk:
15:27 dalek : [docs] Edit the Release Manager Guide with new instructions for adding the
15:27 dalek : release announcement to the website, now that the main page for parrotcode.org
15:27 dalek : redirects to parrot.org.
15:28 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31590
15:37 pyrimidine joined #parrot
16:08 jhorwitz joined #parrot
16:31 ab5tract joined #parrot
16:37 ruoso joined #parrot
16:39 particle pmichaud: ping
16:41 Theory joined #parrot
17:06 jhorwitz particle: you arriving in pittsburgh in time for friday's festivities?
17:06 particle festivities? no idea, lemme check my schedule
17:06 particle is there an ml i should join?
17:06 jhorwitz just the wiki.  :)
17:06 particle ah
17:07 particle probably not :(
17:07 jhorwitz bah
17:07 particle leave seattle 1p, arrive pit 11:37p
17:08 jhorwitz tell me what beer you like and i'll drink it for you.  ;-)
17:08 particle damn.
17:11 Ivatar joined #parrot
17:23 gryphon joined #parrot
17:41 Tene pmichaud: so always use caller's .hll unless there's a :hll(...) passed?
17:43 ab5tract what's up in pittsburg?
17:44 jhorwitz pittsburgh perl workshop next weekend
17:46 jhorwitz there's a pre-workship friday night social (read beer)
17:48 ab5tract nice. sounds cool. is it like a hackathon or more like conference with talks and whatnot?
17:50 jhorwitz it's a small 2-day conference.  there's going to be a perl6 hackathon stuck in there somewhere as well. http://pghpw.org
17:57 pmichaud particle: pong
18:05 rdice joined #parrot
18:25 particle pmichaud: still about?
18:26 pmichaud yes
18:26 particle ok, i think i have what i need for my patch, but good to know you're around if i need
18:26 particle in conversations now, will be hacking in ~15m
18:26 pmichaud okay, I should be here a while.  I'm working on hll tests for p6object.
18:27 Tene pmichaud: so I should always use the caller's .HLL unless there's a :hll(...) passed in?
18:27 pmichaud Tene: that's what I'm thinking.
18:27 NotFound joined #parrot
18:27 pmichaud in general p6object wants to be acting on behalf of the caller.
18:28 * Tene nods.
18:28 Tene I'll wait for you to finish your tests.  I'll go make pancakes instead.
18:29 pmichaud it may be a little while -- I'm also doing a fair bit of refactoring and updating.
18:29 pmichaud the p6object test suite has gotten a bit bigger than I initially imagined it would.  :-)
18:29 pmichaud I'll be doing commits soon, though.
18:30 sjansen joined #parrot
18:30 Tene Refactoring is another good reason for me to hold off and make pancakes.
18:30 * particle wants a pancake
18:31 Tene particle: but I'm making them with raspberries and chocolate chips
18:31 particle i like peanut butter and chocolate chips
18:31 particle but i'll take raspberries
18:31 Tene I have peanut butter too.
18:32 peepsalot joined #parrot
18:33 Tene Wait, peanut butter in the pancakes or on the pancakes?
18:33 particle peanut butter chips, actually
18:33 particle in
18:33 Tene Interesting idea.
18:33 Tene I'll have to try that.
18:34 particle great for hiking, no syrup needed
18:35 * jhorwitz is getting hungry
18:35 * particle too
18:35 * pmichaud just had lunch.  Yum.
18:35 particle but i've promised to have pork today in support of the bailout bill
18:36 pmichaud ...which passed already.  :-)
18:36 particle yes, and now we all enjoy pork
18:43 jhorwitz this bill wasn't very kosher
18:50 particle definitely not blessed by a rabbi
18:51 donaldh joined #parrot
18:53 particl1 joined #parrot
18:55 particle left #parrot
18:58 dalek r31591 | julianalbo++ | trunk:
18:58 dalek : pirric: add ISA function
18:58 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31591
18:59 ab5tract next year financial analysts will be urging us to privatize government programs to pay off our debt
19:00 ab5tract and "stimulate the economy"
19:01 * ab5tract digresses
19:02 jhorwitz next thing you know, parrot will have it's own foundation.  ;-)
19:02 ab5tract haha
19:02 jhorwitz ab5tract: u in philly?
19:02 ab5tract around it, why?
19:03 ab5tract are you whois'ing me :P
19:03 * jhorwitz whois's liberally
19:03 jhorwitz i'm in KOP
19:03 ab5tract right on, i'm holed up in WC
19:04 jhorwitz the world needs more acronyms
19:04 ab5tract TWNMA
19:04 ab5tract :)
19:05 particl1 TLB
19:05 * jhorwitz rolls eyes
19:05 purl You rolled a 11, jhorwitz
19:05 ab5tract haha
19:05 particl1 TLB is short for TLA++
19:05 ab5tract oh goodness there is a good joke i'm missing out on in there isn't there
19:19 johbar joined #parrot
19:26 cotto you and your perl strings
19:50 confound I just moved out of philly
19:51 ab5tract word? i didnt realize i'm so close to some parrotheads
19:52 * particl1 moved out of philly 3+ years ago
19:52 davidfetter any word on rick's?
19:52 particl1 didn't realize i started a trend
19:53 * davidfetter always hits the reading terminal market when in philly
20:05 petdance joined #parrot
20:05 dalek r31592 | cotto++ | trunk:
20:05 dalek : [pipp] make iterators work, fix resizing, enable more tests
20:05 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31592
20:28 particle joined #parrot
20:45 pyrimidine joined #parrot
20:55 dalek r31593 | allison++ | pdd27mmd:
20:55 dalek : [pdd27mmd] Bringing the pdd27mmd branch up-to-date with trunk r31592.
20:55 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31593
20:55 peeps[work] joined #parrot
21:07 particle joined #parrot
21:32 Whiteknight joined #parrot
21:38 iblechbot_ joined #parrot
22:00 rdice joined #parrot
22:33 dalek r31594 | pmichaud++ | hllmagic:
22:33 dalek : [p6object]:  Refactor tests in p6object.t .
22:33 dalek : This new version does more complete testing, and also makes it
22:33 dalek : easier to test how P6object performs from a different .HLL namespace.
22:33 dalek : Some of the old-style tests are still in the code (commented out)
22:33 dalek : and need refactoring into the new style.
22:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31594
22:37 dalek r31595 | julianalbo++ | trunk:
22:37 dalek : pirric: refactor flow and error control
22:37 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31595
22:45 TiMBuS joined #parrot
22:47 tetragon joined #parrot
22:50 dalek r31596 | julianalbo++ | trunk:
22:50 dalek : pirric: fix FOR var increment
22:50 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31596
22:51 dalek r31597 | allison++ | pdd27mmd:
22:51 dalek : [pdd27mmd] Don't recursively call the same vtable function on the same
22:51 dalek : PMC.
22:51 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31597
23:01 dalek r31598 | cotto++ | trunk:
23:01 dalek : [pipp] PIRify boolean.t, refactor test-pmc target to run all PMC test files
23:01 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31598
23:11 dalek r31599 | julianalbo++ | trunk:
23:11 dalek : pirric: add STEP option to FOR and fix error messages
23:11 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31599
23:11 dalek r31600 | Whiteknight++ | trunk:
23:11 dalek : [IMCC] Write some function-level documentation for compilers/imcc/instructions.c
23:11 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31600
23:11 Whiteknight IMCC is the worst. It's so convoluted, and so much of it is undocumented
23:14 Whiteknight As your next president, I will clean up IMCC, cut subsidies to convoluted code, and make Parrot's register allocator work for you, the people!
23:14 * Whiteknight is getting into politics too much
23:15 NotFound Whiteknight: I'll give you my vote if you promise to subsidize retro basics ;)
23:16 Whiteknight I will, especially QBasic!
23:16 s1n pmichaud: ping
23:16 Whiteknight But let us not forgot such favorites as GWBasic, or MSBasic
23:17 Whiteknight I was actually half thinking about writing a cobol compiler for parrot, although I'm not sure I want to commit a crime against humanity at such a young age :)
23:18 s1n Whiteknight: you could do worse.... ada
23:18 NotFound Whiteknight: I was tinking about a PL/M compiler, but decided that a Basic will be a greater service to humanity (specially the Dijkstra part of it)
23:20 Whiteknight I would love to revive a bunch of cool old languages, PL/I, Fortran, Algol, etc
23:20 Whiteknight one day I'll actually have the time to do it
23:20 NotFound Fortran is already too live
23:20 s1n yeah fortran didn't technically die
23:22 NotFound If someone wants to play with the compiler tools, PL/M may be not a difficult task. The hard parte is to find manuals for it.
23:23 Whiteknight Yeah, documentation is slim. And good luck getting any tech support on it
23:23 Whiteknight What is BCG?
23:23 purl BCG is Byte Code Generator
23:23 Whiteknight thanks purl, very prompt
23:23 NotFound (PL/M is a simplified PL/1 for 8 bit computers, for those that never heard of it)
23:23 Whiteknight There are a lot of undocumented functions in BCG
23:24 Whiteknight I'm a young'un here, and I've heard of it
23:24 Whiteknight and PIRC! For shame, there are so many undocumented functions!
23:24 NotFound I suppose there are few people here not interested in languages ;)
23:25 s1n i am but lack the time to work on one, christmas vacation is coming up though :)
23:26 NotFound My dream is to write an INTERCAL compiler that implements the ABSTAIN instructions by dynamically modifying the bytecode at runtime.
23:26 NotFound Well, maybe "my nightmare" is a better description.
23:28 Whiteknight INTERCAL? Good luck
23:31 NotFound PLEASE ABSTAIN
23:35 Limbic_Region joined #parrot
23:38 dalek r31601 | Whiteknight++ | trunk:
23:38 dalek : [Core] Add function-level documentation for C<src/atomic/gcc_x86.c>. I may have gotten some things wrong, GAS syntax always gets me turned around.
23:38 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=31601
23:50 particle1 joined #parrot
23:53 Robrt joined #parrot
23:57 Robrt FYI, svn.perl.org upgraded to svn 1.5
23:57 particle1 \o/ robrt++
23:58 Robrt Sorry it took so long

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

Parrot | source cross referenced