Camelia, the Perl 6 bug

IRC log for #parrot, 2009-08-10

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:02 allison joined #parrot
00:08 dalek parrot: r40473 | allison++ | trunk/lib/Parrot/Docs/Section/Parrot.pm:
00:08 dalek parrot: [html] Draft chapters of the book shouldn't appear in the HTML documentation.
00:08 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40473/
00:32 Andy joined #parrot
00:50 quek joined #parrot
01:08 dalek parrot: r40474 | allison++ | branches/pcc_arg_unify/compilers/imcc/pbc.c:
01:08 dalek parrot: [pcc] Don't make a method call when the method itself is just a vtable
01:08 dalek parrot: function call. (For that matter, don't make methods that are just vtable
01:08 dalek parrot: function calls.)
01:08 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40474/
01:16 dukeleto joined #parrot
01:22 dukeleto hola
01:22 purl hi, dukeleto.
02:02 wayland76 joined #parrot
02:15 quek left #parrot
02:20 mokurai joined #parrot
02:41 janus joined #parrot
03:05 dalek parrot: r40475 | dukeleto++ | trunk/t/codingstd/svn_id.t:
03:05 dalek parrot: [t] Fix svn_id coding standard tests to work when parent directory is a symlink and improve portability
03:05 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40475/
03:06 dukeleto dcommit is haning for me right now, it did 1 out of 2 commits, but now is just sitting there
03:10 dukeleto i fixed it with a sledgehammer
03:12 dalek parrot: r40476 | dukeleto++ | trunk (2 files):
03:12 dalek parrot: [debugger] Add test for adding watchpoint
03:12 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40476/
03:15 mikehh I had a couple of problems yesterday (well Saturday) after a security update from Ubuntu - seems ok now
03:20 donaldh joined #parrot
03:34 mikehh All tests PASS (pre/post-config, smolder, nqp_test, fulltest) at r40476 - Ubuntu 9.04 amd64
03:50 mikehh rakudo (39cc848) builds on parrot r40476 make test/make spectest (up to 27938) PASS - Ubuntu 9.04 amd64
03:54 dukeleto joined #parrot
04:00 dalek close: r92 | Austin++ | trunk/ (18 files):
04:00 dalek close: More progress. PrettyPrinter is showing promise in identifying tree confusion.
04:00 dalek close: review: http://code.google.com/p/close/source/detail?r=92
04:05 john joined #parrot
04:12 allison joined #parrot
05:39 chromatic joined #parrot
05:49 cotto chromatic, https://trac.parrot.org/pa​rrot/wiki/LoritoPrimitives could use some love from someone who knows a little Smalltalk.
05:49 chromatic I'll do my best.
05:52 uniejo joined #parrot
05:52 cotto Thanks.
06:00 eternaleye I like that there are 42 primitives ;D
06:01 treed http://gist.github.com/165025
06:01 treed Is there anyone around who can take a look at that and tell me if that's a bug or if I'm being stupid?
06:02 treed (Short PIR example.)
06:02 chromatic Nothing jumps out at me as wrong.
06:02 treed Regarding P6metaclass.
06:03 treed If I change the name of the subclass, it works fine.
06:04 chromatic That reminds me of something I thought I heard about in a bug report, but I don't remember it at the moment.
06:04 treed I searched bug reports for P6metaclass, but didn't see anything about this.
06:04 treed Kinda like the bug I already filed, though.
06:04 * treed looks that up.
06:04 treed It had to do with subclassing as well.
06:05 chromatic I changed the name to 'Hash' and it worked fine.
06:05 treed Well, it's not subclassing Hash, though.
06:05 cotto Isn't there something tricky about subclassing a C PMC from PIR, or would P6metaclass take care of that?
06:05 treed It seems to have to do subclassing a class with the same name?
06:06 treed I wonder if it would still do weird things if I made a different subclass from parrot;String
06:06 chromatic Oh, now that sounds likely.
06:06 treed Which?
06:06 purl which witch? or important or why that topic branch i commited yesterday doesn't work or presumably where the other things put things
06:06 chromatic Subclassing a class with the same name goes awry.
06:07 treed I'd expect the namespaces to help, though.
06:07 treed hll/namespaces
06:07 treed https://trac.parrot.org/parrot/ticket/835 is the ticket I already had, also about subclassing, but a crash instead
06:09 treed Okay, new example, it is the same name thing.
06:09 treed Gist updated.
06:10 TiMBuS joined #parrot
06:14 treed So, bug? Or what?
06:14 tetragon joined #parrot
06:23 treed Updated again, showing that the problem only exists with P6metaclass.
06:23 treed Direct subclassing has no such problem.
06:29 chromatic Seems like a bug to me.
06:29 treed k
06:29 * treed opens a ticket.
06:30 treed Thanks for confirming that.
06:30 treed Often times I have a hard time knowing whether it's just my ignorance or if something really is a bug.
06:30 treed (It's usually the former.)
06:30 chromatic Try C++ sometime!
06:33 treed I used to do C++.
06:33 treed And ASM.
06:33 treed (So when Tene tells me that PIR is "like ASM", I kinda laugh.)
06:33 treed Kinda like, I guess, but not really.
06:33 chromatic Less crashy.
06:34 treed PASM is obviously the ASM analogue here.
06:34 treed PIR has some syntactic sugar, where ASM had very, very little.
06:34 treed Having to learn calling conventions, for instance.
06:34 treed Doesn't matter to me in PIR.
06:35 chromatic I've used a couple of decent macro assemblers that made life somewhat easier.
06:36 * treed hasn't done ASM much since, um... the mid-90s?
06:36 treed Maybe late 90s.
06:36 treed I did IA-32 a lot in the mid-90s.
06:36 dalek TT #909 created by treed++: Difficulty subclassing through P6metaclass when subclass and parent have ...
06:36 treed I think the 6502 wasn't until 98 or 99 or so.
06:43 dalek TT #885 closed by dukeleto++: parrot_debugger printing registers does not work
06:43 dalek TT #889 closed by dukeleto++: parrot_debugger throws an exception when deleting a breakpoint
06:43 tetragon joined #parrot
06:56 dalek pynie: r85 | allisonrandal++ | trunk/Grammar/Actions.nqp:
06:56 dalek pynie: Cleanup copyright.
06:56 dalek pynie: review: http://code.google.com/p/pynie/source/detail?r=85
07:01 dukeleto joined #parrot
07:21 donaldh joined #parrot
07:25 tetragon_ joined #parrot
07:38 szabgab joined #parrot
08:01 estrabd joined #parrot
08:16 masak joined #parrot
08:27 dalek cardinal: 3bdceec | treed++ | src/parser/grammar.pg:
08:27 dalek cardinal: Small cleanup of grammar.pg
08:28 dalek cardinal: review: http://github.com/cardinal/cardinal/commit​/3bdceec08026e69d6841ae337560fe027537b2a8
08:34 cotto chromatic, no joy with the LoritoPrimitives page?
08:37 cotto I wonder if there are any other similarly designed VMs we could steal from.  It'd be nice to avoid the need for a Smalltalk geek to prevent reinventing this particular wheel.
08:37 cotto (or at least find out what "round" looks like)
08:43 chromatic I haven't broken out my Smalltalk manuals yet.
08:51 iblechbot joined #parrot
09:36 HG` joined #parrot
09:46 bacek joined #parrot
09:48 bacek o hai
09:58 jan joined #parrot
10:05 gaz joined #parrot
10:25 payload joined #parrot
10:39 MoC joined #parrot
11:08 donaldh left #parrot
11:10 tetragon joined #parrot
11:20 donaldh joined #parrot
11:25 Psyche^ joined #parrot
11:30 dalek parrot: r40477 | NotFound++ | branches/auto_attrs (81 files):
11:30 dalek parrot: merge from trunk r40476
11:30 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40477/
11:53 tetragon_ joined #parrot
11:58 tetragon joined #parrot
12:01 tetragon_ joined #parrot
12:03 tetragon joined #parrot
12:05 tetragon_ joined #parrot
12:11 tetragon joined #parrot
12:12 whiteknight joined #parrot
12:13 tetragon_ joined #parrot
12:31 azawawi joined #parrot
12:31 azawawi hi
12:31 azawawi is there a perl 5 library for parrot debugger ?
12:51 whiteknight none that I am aware of
12:52 azawawi so to emulate it atm i need to read/write to the parrot_debugger.exe process, right?
12:55 NotFound azawawi: if you have a parrot interpreter, it already has the debugger inside, you only need a way to start it.
12:55 azawawi i see, thanks
13:04 azawawi left #parrot
13:21 dalek parrot: r40478 | NotFound++ | branches/auto_attrs/lib/Pa​rrot/Pmc2c/PMCEmitter.pm:
13:21 dalek parrot: heuristic to allow non auto_attrs PMC inehriting from auto_attrs ones in corner cases
13:21 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40478/
13:48 ruoso joined #parrot
14:08 AndyA joined #parrot
14:30 Coke joined #parrot
14:31 Coke ~~
14:35 szabgab joined #parrot
14:48 Psyche^ joined #parrot
14:55 Coke Rosellyne Thompson?
14:55 Coke purl, you lazy git.
14:55 purl Coke: i'm not following you...
15:00 kj joined #parrot
15:20 donaldh joined #parrot
15:48 bacek joined #parrot
15:57 Coke wow. I just got an email from her. serendipitous.
16:07 Coke http://www.boingboing.net/2009/​08/10/common-operating-sys.html - sound familiar?
16:23 davidfetter joined #parrot
16:29 MoC joined #parrot
16:39 mokurai joined #parrot
16:51 hercynium joined #parrot
17:03 dalek TT #910 created by coke++: macport for 1.4.0
17:13 darbelo joined #parrot
17:26 einstein joined #parrot
17:29 chromatic joined #parrot
17:46 gaz joined #parrot
18:02 braceta joined #parrot
18:20 mikehh_ joined #parrot
18:22 ash_ joined #parrot
18:24 whiteknight joined #parrot
18:33 Coke Don't forget to consider nominating folks for the board.
18:43 dalek rakudo: bb1c2be | moritz++ |  (2 files):
18:43 dalek rakudo: initial NYI.pm implementation
18:43 dalek rakudo: This commit introduces some code/constant duplication; my attempts to factor it out
18:43 dalek rakudo: into a lexical sub or a lexical variable holding a block were not successful -
18:43 dalek rakudo: there former aren't found (see RT #67868), the latter produce Null PMC access
18:43 dalek rakudo: in invoke()
18:43 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/b​b1c2bef5f5bd5f184df484a98ced69796c4409e
18:44 whiteknight who is nominating who for what board?
18:44 whiteknight and how is who doing it?
18:49 NotFound whiteknight: I've posted now a short report of the auto_attrs branch status,
18:49 whiteknight NotFound: Awesome
18:51 whiteknight Once all PMCs are converted, we don't need to worry any more about inheritance problems, right?
18:51 treed Can an ExceptionHandler be called to handle exceptions that occured while it executes?
18:51 whiteknight treed: what do you mean?
18:51 NotFound whiteknight: as a long as we docuement it and do a deprecation cycle, I suppose yes.
18:51 treed I have a situation where there's exactly one eh pushed onto the stack.
18:52 treed When it gets called, it goes into a loop of itself, never finishing one iteration.
18:52 NotFound treed: yes, this is the excpected behavior.
18:52 treed It occured to me that maybe an exception is getting thrown in the eh itself.
18:52 whiteknight yeah, unfortunately
18:52 ash_ how does one sign up to be a member of the mailing list? (not just a viewer of it?)
18:53 moritz ash_: which mailing list? parrot-dev?
18:53 ash_ yeah
18:53 whiteknight http://lists.parrot.org/ma​ilman/listinfo/parrot-dev
18:53 moritz http://lists.parrot.org/ma​ilman/listinfo/parrot-dev
18:53 treed could "exit 1" be causing an exception to be thrown?
18:53 moritz ah, whiteknight++ was faster
18:53 ash_ was the one i was trying to mail to, but it said i am not a member so it said someone will have ot review my message parrot-dev@lists.parrot.org
18:54 ash_ hmmm i thought i signed up for that, i'll try again (might of used one of my aliased emails :-X)
18:55 dalek rakudo: 1e00038 | moritz++ | src/setting/NYI.pm:
18:55 dalek rakudo: want() is not NYI, it's gone - KyleHa++
18:55 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/1​e000383bf83f89fcb244aeb043f3b5545d86e19
19:01 Coke whiteknight: (board) see the parrot-members mailing list.
19:01 whiteknight I've never even heard of parrot-members
19:01 whiteknight is that new?
19:01 whiteknight or am I retarded?
19:01 treed Huh, it was the exit 1.
19:01 treed How interesting.
19:01 cotto same here
19:02 Coke "that isn't an or question"
19:02 whiteknight okay, so I'm retarded it's new
19:02 Coke the list I think has existed for some time, but it's the only message that's posted there.
19:03 Coke http://lists.parrot.org/pipermail/parrot-​members/Week-of-Mon-20090803/000000.html
19:04 NotFound Parrot members aren't wings and legs?
19:05 * treed wonders if it's also expected that ex['message']= doesn't work for subclasses of Exception?
19:05 treed It doesn't crash, it just fails silently.
19:05 treed The message isn't set.
19:05 Coke I'm coming in late, but yes, 'exit 1' throws an exit exception.
19:06 NotFound treed: there are some unresolved problems on inheriting from exceptions and exception handlers.
19:06 treed Ah, well, that makes sense then.
19:06 treed NotFound: I see.
19:06 NotFound Well, and on a lot of other PMCs X-)
19:06 treed Yeah, I filed such a ticket last night.
19:07 treed Actually, both of my tickets that I filed have to do with inheriting from PMCs.
19:07 treed Is there a ticket for this particular issue? I'd like to note that I'm having trouble with it, too.
19:08 NotFound treed: What tickets?
19:08 purl tickets are good
19:09 ash_ has anyone here had issues with a child class explicitly calling a parent class method? thats what i sent in the mailing list
19:09 treed NotFound: 909 is the one I filed last night
19:10 treed https://trac.parrot.org/parrot/ticket/835
19:10 purl https://trac.parrot.org/parrot/ticket/835 is the ticket I already had, also about subclassing, but a crash instead
19:10 treed Is the other.
19:10 treed Thanks, purl.
19:10 purl my pleasure treed
19:12 NotFound treed: I took at look at 835. Unfortunately, there is no easy solution at this point.
19:12 treed Well, I worked around that one.
19:12 treed But I'm blocking on 909.
19:13 treed It's not a serious block, just prevents classes from having their real name.
19:13 japhb joined #parrot
19:13 treed So that things like obj.class work appropriately.
19:13 NotFound I can't help with that, never look at p6metaclass.
19:14 dalek decnum-dynpmcs: r159 | darbelo++ | trunk/inc/decnum.pasm:
19:14 dalek decnum-dynpmcs: Update inc/decnum.pasm with some missing constants.
19:14 dalek decnum-dynpmcs: review: http://code.google.com/p/decnu​m-dynpmcs/source/detail?r=159
19:14 NotFound treed: a short pir example that triggers the problem might be helpfuk.
19:15 treed On 909?
19:15 treed There's one on there.
19:15 treed One on both, actually.
19:15 treed I try to have something like that before filing a bug, so that I know I'm not just being stupid
19:16 treed well s/know/am pretty sure/
19:16 Tene joined #parrot
19:17 NotFound Don't know what I do, it showed me a stacktrace instead of the pir code in the first attempt :?
19:17 treed Huh.
19:18 treed For me, it says "foo" then "1", then it crashes when attempting to set_string_native
19:18 treed Saying that it can't be found for class String.
19:18 treed Oh, I see what you mean.
19:20 donaldh joined #parrot
19:20 joeri joined #parrot
19:25 dalek rakudo: 9c291ca | (Kyle Hasselbacher)++ | t/spectest.data:
19:25 dalek rakudo: [spectest.data] filetest.t regressed due to spec change (RT #68160)
19:25 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/9​c291ca540e1fbb810dcf5ebabeb58559d40da97
19:27 NotFound treed: Are you sure that 'parrot;String' has the semantic your are trying to use?
19:29 dalek decnum-dynpmcs: r160 | darbelo++ | trunk/aux/gen_decnum_pasm.c:
19:29 dalek decnum-dynpmcs: Add the c file that generates decnum.pasm
19:29 dalek decnum-dynpmcs: review: http://code.google.com/p/decnu​m-dynpmcs/source/detail?r=160
19:35 treed NotFound: It works for FooString.
19:35 * Coke gets linking errors with ICU on an attempted 1.4.0 with macports.
19:36 treed There's no difference at all between foo;FooString and foo;String besides the name
19:36 dalek rakudo: af3efaf | moritz++ | src/ (2 files):
19:36 dalek rakudo: parse feed operators, and die with a NYI message
19:36 dalek rakudo: Note that after this commit you need to clean; a simple 'make' doesn't rebuild
19:36 dalek rakudo: enough of the parser, it seems.
19:36 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/a​f3efaf3698751e0383a27944978cf3142ed53ae
19:36 treed And, yeah parrot;String should have set_string_native
19:36 treed If anything does.
19:37 Coke if FooString is a subclass of String, there's a BIG difference under the covers.
19:37 treed FooString and foo;String are both subclasses of parrot;String
19:37 Coke what's your TT#?
19:37 treed 909
19:38 treed https://trac.parrot.org/parrot/​attachment/ticket/909/test.pir
19:38 Coke hey, trac is dead.
19:38 treed What?
19:38 treed Since when?
19:38 Coke https://trac.parrot.org/parrot => Traceback (most recent call last): File "/usr/lib/python2.5/site-p​ackages/trac/web/api.py", line 367, in send_error
19:38 dduncan joined #parrot
19:38 treed Huh.
19:38 treed Hold on a sec, the PIr file is also in a gist.
19:39 dduncan left #parrot
19:39 * Coke opens a ticket with OSUOSL
19:39 dalek decnum-dynpmcs: r161 | darbelo++ | trunk/ (4 files):
19:39 dalek decnum-dynpmcs: Add tests for both of the logarithm METHODs.
19:39 dalek decnum-dynpmcs: review: http://code.google.com/p/decnu​m-dynpmcs/source/detail?r=161
19:39 NotFound treed: Coke this is what I see a few moments ago when trying to read treed's attachment,
19:39 treed http://gist.github.com/165025
19:39 treed That's the attachment and the results
19:39 Coke ah. p6metaclass ain't parrot. no clue.
19:39 NotFound Looks like an intermitent failure, maybe some resource too busy,
19:40 Coke (ok, it is, but it's not something I ever use in partcl.)
19:40 treed Yeah.
19:40 treed The screwy thing is that it works just fine when subclassing directly.
19:40 treed Tene has previously expressed the opinion that cardinal should use or at least subclass P6object.
19:41 treed But I'm not sure that it provides much that Cardinal needs that we couldn't easily do ourselves.
19:41 treed So I presume he says that from an hll-interop standpoint.
19:41 NotFound treed: there were problems with classes with the same name as core PMCs, and I don't think they are full resolved yet,
19:42 treed NotFound: As I said, it works fine if you just use subclass.
19:42 treed That example shows that with parrot;Integer and foo;Integer.
19:42 treed If I tried to subclass Integer via p6metaclass, it complains about "set_integer_native".
19:54 treed Weeeeeeee
19:54 treed add_attribute() not implemented in class 'Exception'
19:54 treed -_-
19:55 braceta joined #parrot
19:55 * treed gives up for the day.
19:55 mokurai joined #parrot
19:56 Tene joined #parrot
19:57 jan joined #parrot
20:01 Coke
20:04 * l3t0 is currently pushing 221 parrot svn branches to github
20:04 Coke ... there aren't that many branches in svn. are you pushing all the historical ones too?
20:06 NotFound treed: I think I found the root of your problem with exceptions: the throw opcode doesn't allow subclasses, and throws a fatal exception without message,
20:07 treed Huh.
20:07 treed Why doesn't it allow subclasses?
20:07 Coke treed: bug
20:07 treed (This would also explain why "typeof" is returning "Exception".
20:07 treed Ah, nice.
20:08 NotFound Bug disallowing, and bug not setting a clear message
20:08 treed Ticket for this? Or did you just discover it?
20:08 NotFound treed: I just do some checks in the code, don't remember any ticket related.
20:12 Coke sounds familiar.
20:22 dalek parrot: r40479 | NotFound++ | trunk/src/ops/core.ops:
20:22 dalek parrot: [cage] throws a more descriptive exception whrn trying to throw something not accepted as a valid exception, treed++
20:22 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40479/
20:23 braceta joined #parrot
20:27 kesselhaus joined #parrot
20:33 joeri left #parrot
20:35 athomason joined #parrot
20:39 GeJ Good morning everyone
20:40 davidfetter hi GeJ
20:45 mokurai joined #parrot
20:50 mokurai left #parrot
21:06 donaldh_ joined #parrot
21:09 ash_ has anyone here had issues with a child class explicitly calling a parent class method? is there a way of doing that? i can't find an example in the spec
21:10 kesselhaus isn't that language redundancy a bit stupid? 4 implementations for scheme, 3 for lua, 2 for pythons, 2 for lisp... why not put that effort into 1 instead
21:11 l3t0 Coke: seems like I did
21:11 moritz kesselhaus: because that's not how voluntueers work
21:12 moritz some people just want to start their own compilers to try things out
21:12 cotto kesselhaus, people work on what's interesting to them.  There's no way to force effort to be spent efficiently if you're not paying people.
21:13 moritz and given that some languages are more popular, it seems logical that some languages get multiple (partial) coverage
21:13 PerlJam kesselhaus: isn't what you want to work on a bit stupid?  Why not work on someone *else's* project?    ;_)
21:13 PerlJam ;-) even
21:13 kesselhaus i don't think it is a matter if i pay people or not..
21:14 kesselhaus but it makes it all confusing... there is even no status somewhere, so how would one choose?
21:14 moritz kesselhaus: it does. If you pay people, you can tell them what to do
21:17 kesselhaus then these are no real projects... i doubt the quality then
21:17 moritz kesselhaus: you're free to do that
21:17 chromatic What do you think "real projects" means?
21:18 kesselhaus and just makes lots of noise... the noise i already see on sf.net, lots of abandoned 'this is my personal implementation of XYZ'
21:19 mokurai joined #parrot
21:19 donaldh_ kesselhaus: are you offering to coordinate language development and encourage collaboration?
21:20 kesselhaus the thing is, that parrot.org is happy that it can say, 'hey, see all that language projects on top of parrot', but there is nowhere a status of quality or status, nothing about, to what degree the languages are implemented
21:21 l3t0 kesselhaus: things move fast, so it is not that it is impossible, but it is a lot of work to keep track of everything and keep the info up-to-date and useful
21:22 l3t0 the parrot community definitely needs a lot of non-developer volunteers to help our online presence
21:22 PerlJam kesselhaus: how would you measure the degree to which a language was implemented?
21:22 ash_ conformance to a spec if it exists?
21:23 l3t0 Coke: check out http://github.com/leto/parrot/network to see a visualization of each parrot branch (perhaps too many :))
21:24 kesselhaus at least, not just link to a strange github.. but have them plan the releases, and make a statement about it, before they are added to that list?
21:25 PerlJam kesselhaus: er, why?
21:25 kesselhaus e.g. "A self-educational Scheme to PIR compiler implemented in Scheme. Currently not under active development. Takers are welcome."
21:26 donaldh_ is that a good or a bad example?
21:26 donaldh_ seems pretty self explanatory to me
21:26 kesselhaus cool, at least it is listed on the parrots languages site, one of the 4 scheme implementations
21:26 kesselhaus a bad sample i would say
21:28 kesselhaus at least, if parrot should make any kind of success not just as a perl6 host
21:29 chromatic A volunteer working on a volunteer project with volunteer time reflects badly on Parrot?
21:29 cotto kesselhaus, don't forget: 90% of everything is crap
21:29 bacek joined #parrot
21:29 kesselhaus also, on sf.net, what is steme? this on the other side is not even listed on the laguages site
21:29 PerlJam cotto: is that the first 90% or the second 90% ? :)
21:30 moritz PerlJam: both, actually ;-)
21:30 PerlJam chromatic: and someone using perl to store passwords in plain text is enough to make people want to disavow perl altogether.
21:31 cotto kesselhaus, steme is Tene's scheme implementation.
21:31 chromatic Said person should disavow every site with unencrypted logins.
21:36 kesselhaus well, and where can i get a status of that one? you make it more or less a release 1.4.0, but there is nowhere a reference on parrot.org about it. what could i try out?
21:38 braceta left #parrot
21:38 * l3t0 now realizes that only 20 of the 200-something branches that he pushed are currently active. got a little overzealous :)
21:39 l3t0 i just used everything in .git/svn/svn , instead of doing a svn ls to see what is current. lesson learned
21:39 cotto now you know
21:39 purl And knowing is half the battle.
21:40 donaldh_ very astute purl
21:40 donaldh_ for a change
21:43 kesselhaus don't take me wrong. i'm not against opensource, or volunteering work, but i'm against the unnecessary noise produced by such forks or my-implementation-of-xyz
21:44 PerlJam kesselhaus: who deems them unnecessary?
21:46 kesselhaus if i have 4 implementations for scheme on parrot, i personally would for 3 of them
21:46 PerlJam kesselhaus: If predestination is true, then I guess I can see how forking from the One True Implementation would be unnecessary, but I don't think that's really the case in general.
21:47 eternaleye joined #parrot
21:48 chromatic If you'd like to encourage developers to converge on one implementation, feel free.  I'd like to see convergence.
21:48 treed Only one Ruby, though!
21:48 treed Ha-HA!
21:48 PerlJam Sam Ruby?  ;-P
21:48 nperez it seems like the best way to have convergence is to steal ideas/code from your forks :)
21:51 kesselhaus yeah, merge them and then drop the dead ones (and also drop them then from the langauges site)
21:51 PerlJam kesselhaus: so ... are you volunteering to push that along?
21:54 l3t0 kesselhaus: what is the criteria for death? sometimes projects "die" for a few weeks/months then .... zombies!!!
21:54 kesselhaus i'm not a responsible from parrot, but as one, i would probably want to get an active status and progress, to list them, otherwise, i would have lots of abandoned language impl. projects, which shed a strange light on my parrot vm
21:55 chromatic Are you willing to help?
21:56 PerlJam kesselhaus: no one is responsible for parrot except those willing to be so.  Are you willing?
21:56 l3t0 i think it sheds a light that many people are fiddling around with Parrot, which is a good thing. I agree with you, though, that it would be nicer for an end user to be able to quickly find out "what is the furthest along implementation for language X?"
21:56 treed What annotations can I expect to be present in a backtrace?
21:56 treed file, line, sub?
21:56 nperez maybe a simple "Well volunteered" is that needs to be said :)
21:57 l3t0 kesselhaus: patches welcome! :)
21:58 ash_ so, does parrot normally spec things before / after / during implementation?
21:59 cotto ash_, yes
21:59 kesselhaus hehe, no no, i've no knowledge how to write a compiler or interpreter, i'm doing embedded systems development btw., but e.g. as a resposible i would probably introduce something like a status report which must be updated, and if not, remind them to do that, or the project gets unlisted from the language site
21:59 ash_ cotto: woot! thats what i like to hear
22:00 ash_ so.... if there seems to be a piece missing from the spec should i bring it up in here? or hack away at it for a bit then give a failing test that someone else can implement?
22:00 chromatic kesselhaus, you don't have to know how to write a compiler to contact language authors and ask them for status reports and update the Wiki.
22:00 ash_ cause i'd prboably fail miserably at it
22:00 ash_ probably even
22:01 cotto ash_, do what you can.  Bringing it up here or filing a tt (especially with a failing test case) is great.
22:01 l3t0 kesselhaus: I am sure that you could help in desiging/writing the infrastructure for a system like that
22:02 donaldh_ joined #parrot
22:02 ash_ cotto: in the object system there currently isn't a way for you to explicitly call a parent classes version of a function, that seems like a logically useful thing to do right?
22:02 l3t0 ash: i find that creating a trac ticket will get it the most visibility. mentioning it in here and on parrot-dev will make sure everyone sees it
22:02 chromatic We've had SUPER specified a few times, but never implemented correctly.
22:02 l3t0 ash: writing a failing TODO test is always instructive and fun as well
22:03 kthakore joined #parrot
22:03 kthakore hi chromatic
22:03 l3t0 that way, people can run the TODO test on their branch to see if it effected, etc.
22:03 kthakore apparently I have access to sdl.perl.org?
22:04 chromatic I just responded to your mail, kthakore.
22:04 kthakore oh great
22:04 ash_ chromatic: has there been a decided direction for trying?
22:04 ash_ and just implementation goes wrong, or has there been no consensus on the way it should be designed yet
22:05 kthakore chromatic, so webmaster@perl.org is where robert and Ask are?
22:05 chromatic yes.
22:05 chromatic ash_, there was a Super PMC at one time, but it never worked.
22:05 chromatic I'm not sure what the right approach is.
22:06 ash_ chromatic: i know in the perl 6 spec they took the approach of self.Parentclasss::methodname(); that doesn't mean parrot has to but its an example of how it might work. do you think i should send a message in the mailing list? I think it would be a useful thing to figure out and hack at and maybe finish one day :P
22:07 Whiteknight joined #parrot
22:07 chromatic One of the holdups is definitely the syntax for it.
22:07 dalek tracwiki: v3 | cotto++ | LoritoPrimitives
22:07 dalek tracwiki: s/primitive/operation/ - primitive means something different in Smalltalkland
22:07 dalek tracwiki: https://trac.parrot.org/parrot/wiki/Lor​itoPrimitives?version=3&action=diff
22:09 Whiteknight irclogs?
22:09 purl well, irclogs is http://irclog.perlgeek.de/parrot/today or see also: infrared clogs
22:10 ash_ chromatic: well i'll see if i can come up with something useful, was the Super PMC in the repository at one point? i'd like to take a look at it to see what work was done before
22:10 chromatic I think we removed it before 1.0, so look sometime around r27000.
22:12 Whiteknight treed: ping
22:12 treed yo
22:13 Whiteknight treed: are you trying to subclass ExceptionHandler?
22:14 treed I'm trying to subclass Exception.
22:14 treed And, a lot of other things, with varying degrees of success.
22:14 Whiteknight I don't think that's currently possible
22:14 treed But ExceptionHandler not among them.
22:14 treed Are there plans to make it possible?
22:14 treed Seems kinda important.
22:14 chromatic Should work.
22:15 Whiteknight there are at least two attributes in Exception that are not subclassable
22:15 treed On a related topic, how can I get at a backtrace?
22:15 Whiteknight which means attempting to access those will throw an exception (which itself won't be subclassable and throw an exception, etc)
22:15 treed The backtrace opcode doesn't take arguments.
22:15 treed Which is... odd.
22:15 treed Like, even a place to put it.
22:15 treed Whiteknight: Yeah, I discovered that. :-/
22:16 ruoso joined #parrot
22:16 ash_ chromatic: i found a version of it thanks
22:17 ruoso joined #parrot
22:18 Whiteknight the answer is that yes, there is a goal to fix it, but we don't have a complete plan in place yet to do that
22:18 allison joined #parrot
22:21 ruoso_ joined #parrot
22:21 Limbic_Region joined #parrot
22:24 l3t0 ruoso: ahoy
22:24 treed Whiteknight: I see.
22:24 ruoso l3t0, hey
22:25 kid51 joined #parrot
22:25 mokurai joined #parrot
22:25 treed So how can I get the backtrace?
22:26 jimk joined #parrot
22:26 Whiteknight Exception has a .backtrace() method
22:27 treed Aha.
22:27 treed The docs say it's a vtable.
22:27 Whiteknight we've found a typo!
22:27 cotto We could make it a VTABLE function so the docs would be correct.
22:27 treed It's at the end of the list.
22:28 cotto We can never have too many VTABLE slots.
22:28 treed [15:27] [inara:/Users/treed/code/cardinal]% ./cardinal temp.rb
22:28 treed ::in `raise': Not a throwable object (Exception)
22:28 treed Better, but it looks like I have no line or file annotations.
22:29 treed Oh, because raise is an hll_global.
22:29 treed Hm.
22:29 treed I wonder how to get the data for that.
22:30 treed Nah, you'd think it'd still have the annotations available.
22:34 kid51 joined #parrot
22:36 kid51 cotto ping
22:36 Whiteknight Coke: ping
22:37 Whiteknight Coke: unping
22:38 cotto kid51, pong
22:39 kid51 cotto:  Did you get the info you needed re adding a config step?
22:39 rg1 joined #parrot
22:41 cotto yes, although I'm not sure now that a configuration step is necessary.
22:41 kid51 fine by me!  A config step that isn't added is one I don't have to maintain!
22:42 Whiteknight hey kid51
22:42 kid51 Yo
22:42 Whiteknight I might be in NYC on Sept 26th. You be interested in getting together then?
22:43 kid51 Quite possibly.  What day of week is that?
22:44 Whiteknight Saturday
22:44 purl Saturday is excepted, I'm sure
22:44 Whiteknight purl forget Saturday
22:44 purl Whiteknight: I forgot saturday
22:44 kid51 I have one personal commitment that has to be scheduled for some weekend in Sept or Oct, but otherwise am clear.
22:45 kid51 Conversely, there's one weekend I want to go to the Phila area in the coming months.
22:45 dalek decnum-dynpmcs: r162 | darbelo++ | trunk/ (3 files):
22:45 dalek decnum-dynpmcs: Integrate abs.decTest into our test suite.
22:45 dalek decnum-dynpmcs: review: http://code.google.com/p/decnu​m-dynpmcs/source/detail?r=162
22:46 kid51 Oh, damn, I missed the art exhibition which I wanted to see in Phila!
22:47 kid51 So, scratch *that* reason for going to Philly.
22:50 Whiteknight my wife is going up to NYC to see some friends, so I will be there anyway
22:56 kid51 okay, let me think about what we can work out.
23:05 chromatic http://yesistilluseperl.blogspot.com/2009​/08/parrot-even-runs-basic-programs.html
23:12 mokurai joined #parrot
23:20 donaldh joined #parrot
23:25 dalek parrot: r40480 | allison++ | branches/pcc_arg_unify/src/library.c:
23:25 dalek parrot: [pcc] Allow the include path to search rooted at the build directory
23:25 dalek parrot: when a build directory is defined. (PGE needs this at the moment.)
23:25 dalek parrot: review: https://trac.parrot.org/parrot/changeset/40480/
23:28 payload joined #parrot
23:30 dalek cardinal: 6e2543b | (Danius Michaelides)++ |  (2 files):
23:30 dalek cardinal: Map/map! are collect/collect!. Fix for collect! return val.
23:30 dalek cardinal: Signed-off-by: Ted Reed <ted.reed@gmail.com>
23:30 dalek cardinal: review: http://github.com/cardinal/cardinal/commit​/6e2543ba854d086daf4dd15cb790851f28903f1d
23:33 dalek left #parrot
23:33 dalek joined #parrot
23:33 Infinoid msg moritz Thanks for the patch, lemme know if it works.
23:33 purl Message for moritz stored.
23:54 dalek cardinal: c445fca | (Danius Michaelides)++ |  (3 files):
23:54 dalek cardinal: Implement Array#nitems(). Second try.
23:55 dalek cardinal: Signed-off-by: Ted Reed <ted.reed@gmail.com>
23:55 dalek cardinal: review: http://github.com/cardinal/cardinal/commit​/c445fca58a96d72d5a7010f7cab90b1d28a9a03b

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

Parrot | source cross referenced