Camelia, the Perl 6 bug

IRC log for #parrot, 2012-03-02

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:05 whiteknight good evening, #parrot
00:08 bubaflub joined #parrot
00:18 whiteknight joined #parrot
00:36 benabik joined #parrot
01:22 dukeleto whiteknight: howdy
01:22 whiteknight hello dukeleto
03:11 alester joined #parrot
03:11 cotto ~~
03:17 benabik ~~
03:45 bacek_at_work aloha, humans
03:48 cotto hio bacek_at_work
03:48 aloha (parrot/parrot) Issues opened : 722 (Properties related VTABLEs are deprecated) by bacek : https://github.com/parrot/parrot/issues/722
03:50 bacek_at_work cotto, hello
04:05 cotto bacek_at_work, for gh #722, how bit of a patch would rakudo and nqp need?
04:06 bacek_at_work cotto, none. No one overrides props VTABLEs. And changing of props ops are in #351
04:07 cotto rakudo has at least one use of VTABLE_getprop (in src/binder/multimethod.c)
04:08 cotto so that'll need a patch before those vtable slots can go away
04:08 cotto 6model does quite a bit more with it
04:12 benabik nqp has a variety of uses of VTABLE_{get,set}prop.
04:12 benabik Would we just be removing the VTABLE in favor of non-overridable behavior?
04:13 cotto my understanding is that the ops would stay and the vtable slots would go away
04:14 cotto so yes
04:14 benabik We could #define the VTABLES to refer to the function instead for a transition period.
04:15 benabik Or whatever.
04:33 bacek_at_work yes. Idea is to provide non-overridable prop functions and get rid of vtables. Functionality will stay.
04:36 benabik Isn't that removing a layer of indirection?  Isn't the solution _adding_ layers of indirection?  ;-)
04:37 cotto Fine.  We can throw in a couple extra pointer dereferences if it'll make you happy.
05:22 slavorg joined #parrot
05:25 woosley joined #parrot
05:50 johbar_work joined #parrot
06:37 woosley left #parrot
06:55 alvis_ joined #parrot
07:38 mj41 joined #parrot
07:49 alvis joined #parrot
08:06 PacoAir joined #parrot
08:29 davidfetter joined #parrot
08:55 alvis joined #parrot
10:18 fperrad joined #parrot
11:45 seme hi gentleman... can anyone give me a hand understanding how to integrate PIR with the grammar files?  It is used by pynie to handle the python indentation requirements but I've been experimenting with it and I don't think it is working
11:46 seme I haven't been able to find much documentation or examples to show how this is used so perhaps if you could point me toward some I could try to understand it myself first :)
11:47 moritz I remember running into the same kind of problem
11:48 moritz seme: are you using nqp-rx?
11:49 seme well I'm trying to take the work done on pynie and puffin and fit it into the structure and form that the squaak language uses... pynie has several functions to handle the indent issues in pir and it looks like it calls those in the Grammar.pm file... such as <.indent_zero>
11:49 seme it doesn't appear to be working though
11:50 seme I had to change it to {{ indent_zero }} synxtax just to get things to run/compile but it still doesn't appear to actually be calling the pir code
11:50 seme that may not have been the right thing to do
11:51 moritz which grammar engine do pynie and puffin use?
11:52 seme not sure what you mean... Pynie::Grammar is PCT::Grammar and it is written in the parrot grammar engine, PGE, I think.
11:52 seme is that what you mean?
11:52 moritz yes
11:53 moritz it means it doesn't use the newer NQP-rx or NQP, with which I'm a bit familiar :/
11:53 seme for instance the TOP token is written token TOP { <.indent_zero> <file_input> {*} }
11:54 seme I see...
11:54 seme right and the squaak tutorial is based on the HLL::Grammar... is that the same as the NQP-rx or NQP?
11:55 moritz that can be either NQP or NQP-rx (though I suspect it's the latter)
11:56 moritz anyway, in the parrot repo, compilers/pge/PGE/Regex.pir has some examples that implement built-in rules in PIR
11:56 seme ooh I see... I didn't realize the distinction... hrm... the documentation is based on the PGE::Grammar and the squaak tutorial uses the HLL::Grammar :)
11:58 seme for a new language which grammar should be used?
11:58 moritz well, on the surface PGE, NQP and NQP-rx look quite familiar, because the all implement Perl 6 regexes to some degree
11:59 moritz hm, good question
11:59 moritz NQP-rx seems like an obvious choice, because it's much newer than PGE, and bundled with parrot
12:00 moritz and NQP is also an obvious choice, because it comes with 6model, and that can be quite convenient if you want to implement your own object system
12:04 seme oddly enough the Actions are written (it would appear as nqp) and the grammar file gets parsed using the PGE/Perl6Grammar generator
12:05 seme ok I need to do some research now and get an understanding of the differences between the PGE/Perl6Grammar NQP-rx and NQP
12:06 seme the squaak tutorial appears to use the NQP-rx generator but in the doc is says it uses perl6 rules...
12:06 moritz yes, NQP-rx implements perl 6 rules
12:07 seme ok got it
12:07 seme and if you are using the PGE/Perl6Grammar generator is that then the same thing?
12:08 moritz well, it's only mostly the same on the surface
12:09 moritz the conventions of how acall to a grammar rule maps to PIR is different between those implementations, I think
12:09 bluescreen joined #parrot
12:18 mj41 joined #parrot
12:25 bluescreen joined #parrot
12:28 davidfetter joined #parrot
12:56 bubaflub joined #parrot
12:56 JimmyZ joined #parrot
13:26 mtk joined #parrot
13:35 dngor_ joined #parrot
13:35 contingencyplan_ joined #parrot
13:35 contingencyplan joined #parrot
15:44 Psyche^ joined #parrot
16:30 alester joined #parrot
16:39 dngor joined #parrot
16:41 dukeleto ~~
17:02 dukeleto this looks like a useful general technique: http://mweissbacher.com/blog/2012/02/1​8/finding-differences-in-javascript-pr​ograms-with-the-help-of-parse-trees/
17:04 dmalcolm joined #parrot
17:05 dalek Heuristic branch merge: pushed 98 commits to rakudo/match-refactor by jnthn
17:18 dalek Heuristic branch merge: pushed 92 commits to parrot/m0 by leto
17:51 dalek parrot/threads: 60efbf6 | nine++ | src/ (2 files):
17:51 dalek parrot/threads: Block GC while scheduling a task on another thread
17:51 dalek parrot/threads:
17:51 dalek parrot/threads: Scheduling may cause the garbage collector to kick in. If we schedule on
17:51 dalek parrot/threads: another thread, this would start the other interpreter's GC from the
17:51 dalek parrot/threads: wrong thread where it would find foreign PMCs on the stack. A situation
17:51 dalek parrot/threads: we really want to avoid.
17:51 dalek parrot/threads: review: https://github.com/parrot/parrot/commit/60efbf6f8c
17:53 dukeleto this looks interesting: http://swdc.se/dyncon2012/
17:56 nine 2500 SEK entry fee
17:56 moritz so slightly more than 300€
17:57 dukeleto whoa
17:57 dukeleto didn't see that
17:57 dukeleto madness
17:57 moritz that's expensive-ish for an open source conference, but dead cheap for corporate conferences
17:57 dalek rakudo/nom: 93104f1 | jnthn++ | src/ (5 files):
17:57 dalek rakudo/nom: Only need multi handling stuff in Routine; don't make every other block pay the price of the few extra attributes.
17:57 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/93104f1c96
17:57 dalek rakudo/nom: 99bc8b3 | jnthn++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
17:57 dalek rakudo/nom: Fix over-sharing of proto thunks, which caused mis-dispatches in some cases.
17:57 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/99bc8b320c
17:59 nine No info, just two sessions, one of which may be somewhat interesting. No idea why one should go there :)
18:02 dngor joined #parrot
18:02 moritz what I'd really like to visit is the "emergent languages" track on OSCON
18:03 moritz ah wait, that was last year. This year they are part of "The Strange Loop"
18:03 moritz oh, *emerging
18:03 moritz http://emerginglangs.com/
18:03 travis-ci joined #parrot
18:03 travis-ci [travis-ci] parrot/parrot#87 (m0 - bae9ca3 : Jonathan "Duke" Leto): The build passed.
18:03 travis-ci [travis-ci] Change view : https://github.com/parrot/par​rot/compare/e66f175...bae9ca3
18:03 travis-ci [travis-ci] Build details : http://travis-ci.org/parrot/parrot/builds/781222
18:03 travis-ci left #parrot
18:12 dalek parrot/threads: f29b241 | nine++ | src/pmc/scheduler.pmc:
18:12 dalek parrot/threads: Keep a list of foreign tasks
18:12 dalek parrot/threads:
18:12 dalek parrot/threads: We don't know if the user keeps references to the tasks he starts. So we
18:12 dalek parrot/threads: have to assume he doesn't and do it ourselves.
18:12 dalek parrot/threads: review: https://github.com/parrot/parrot/commit/f29b241933
18:13 dalek rakudo/nom: e46bfea | moritz++ | src/ (4 files):
18:13 dalek rakudo/nom: Merge remote branch 'origin/match-refactor' into nom
18:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e46bfea35d
18:14 dalek Heuristic branch merge: pushed 150 commits to nqp/qbootstrap by jnthn
18:20 nine Any idea why my parrot would not start in gcc? http://paste.scsys.co.uk/184893
18:21 tadzik huh, you didn't really mean gdb, did you?
18:21 Coke what is gcc? I assume it's not the gnu c compiler...
18:21 nine ah gdb of course
18:21 nine classic ENOCOFFEE
18:22 nine Surprisingly gdb likes the executable much more
18:38 dalek rakudo/nom: c3b4d54 | moritz++ | t/spectest.data:
18:38 dalek rakudo/nom: run operator overloading workout.t
18:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/c3b4d54e6c
18:43 cotto ~~
18:47 dalek rakudo/nom: b6c09b4 | moritz++ | t/spectest.data:
18:47 dalek rakudo/nom: run S06-operator-overloading/imported-subs.t
18:47 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b6c09b4310
19:07 alvis cotto: hello. when i try to install the parrotbug_service module, i get:
19:07 alvis cotto: "parrotbug API resource This version is not compatible with Drupal 7.x and should be replaced".
19:08 cotto alvis: ok.
19:08 alvis cotto: any ideas?
19:08 cotto I think that's because I haven't tested it with drupal 7 yet
19:08 alvis cotto: ah, ok.
19:09 alvis cotto: i'll reinstall 6 and try that over the weekend.
19:09 cotto alvis: no need.  we'll be running on 7 on parrot.org
19:09 mj41 joined #parrot
19:11 alvis cotto: ok, i'll try fiddlin' with it (without much success i'm sure :) 'til i hear further from you.
19:11 cotto I should have a few spare tuits today
19:12 alvis great! just let me know then.
19:12 PacoAir joined #parrot
19:50 dalek rakudo/nom: ced04f7 | (Felix Herrmann)++ | / (2 files):
19:50 dalek rakudo/nom: enable Configure.pl to pass options to Parrot's make
19:50 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ced04f710a
19:50 dalek rakudo/nom: d4dc7d7 | moritz++ | src/Perl6/Grammar.pm:
19:50 dalek rakudo/nom: fix regression introduced in 6588dd39, fixes bug #111492
19:50 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d4dc7d7d95
19:52 cotto alvis: I have the code semi-working but there seems to be a permission issue that I don't have time to dive into until tonight
20:35 pjcj joined #parrot
21:23 cotto alvis: parrotbug-service is working on drupal 7 now (though not on 6)
21:24 cotto you'll need to grant anonymous users the permission to use the parrotbug service to create github issues and to configure the various settings under Configuration -> parrotbug service
21:28 cotto POSTing something like {"title":"bug test","body":"issue body"} to your-site/api/parrotbug should be enough to post a test issue
21:30 cotto once the Services endpoint is set up and has the parrotbug resource enabled
21:44 perlite_ joined #parrot
22:20 dalek rakudo/macros3: fca3907 | masak++ | / (7 files):
22:20 dalek rakudo/macros3: implemented quasi quotes and macros
22:20 dalek rakudo/macros3:
22:20 dalek rakudo/macros3: This works:
22:20 dalek rakudo/macros3:
22:20 dalek rakudo/macros3: - Macro declarations
22:20 dalek rakudo/macros3: - Calling a macro (using `macro()` and `macro` and operators)
22:20 dalek rakudo/macros3: - Quasi quotes
22:20 dalek rakudo/macros3: - Variable lookup from within the quasi quote
22:20 dalek rakudo/macros3:
22:20 dalek rakudo/macros3: This doesn't, yet:
22:21 dalek rakudo/macros3:
22:21 dalek rakudo/macros3: - Returning a macro parameter instead of a quasi
22:21 dalek rakudo/macros3: - Variable lookup from within a macro parameter
22:21 dalek rakudo/macros3: review: https://github.com/rakudo/rakudo/commit/fca39072d2
22:27 dalek rakudo/nom: dc6b0ef | jnthn++ | src/Perl6/Actions.pm:
22:27 dalek rakudo/nom: For sub-signatures using [...] we should constrain to Positional.
22:27 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/dc6b0efb91
22:27 dalek rakudo/nom: 79c8e19 | jnthn++ | src/Perl6/Actions.pm:
22:27 dalek rakudo/nom: Fix the do-we-have-a-subsig test so it doesn't drop 0-arity subsignatures.
22:27 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/79c8e19b4b
22:40 dalek rakudo/nom: 4f62718 | jnthn++ | src/core/traits.pm:
22:40 dalek rakudo/nom: Eliminate old hack from the bad old days when we ran traits again at startup.
22:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/4f62718a0f
23:08 dalek rakudo/nom: 715aed6 | jnthn++ | src/core/operators.pm:
23:08 dalek rakudo/nom: Fix ::('blah') lookups to also fall back to GLOBAL (tadzik++ for noting the bug).
23:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/715aed6db3
23:09 mj41 joined #parrot
23:11 cotto alvis: all changed needed for parrotbug-service to work with drupal 7 have been pushed
23:34 whiteknight joined #parrot
23:43 whiteknight good evening, #parrot

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

Parrot | source cross referenced