Camelia, the Perl 6 bug

IRC log for #parrot, 2011-09-22

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:07 dalek Heuristic branch merge: pushed 18 commits to Rosella/commandline by Whiteknight
00:08 dalek rakudo/nom: 8453ed1 | jnthn++ | src/core/Routine.pm:
00:08 dalek rakudo/nom: Add a way to get a set of matching candidates for a multi.
00:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8453ed1c07
00:40 jsut_ joined #parrot
00:58 benabik joined #parrot
00:58 benabik o/
01:00 woosley joined #parrot
01:03 woosley joined #parrot
01:09 cotto aloha, clock?
01:09 aloha cotto: LAX: Wed, 18:09 PDT / CHI: Wed, 20:09 CDT / NYC: Wed, 21:09 EDT / UTC: Thu, 01:09 UTC / LON: Thu, 02:09 BST / BER: Thu, 03:09 CEST / TOK: Thu, 10:09 JST / SYD: Thu, 11:09 EST
01:23 cottoo joined #parrot
01:39 plobsing ~~
01:52 * soh_cah_toa has started on a new github wiki at https://github.com/soh-cah-toa/parrot/wiki
01:56 * cotto prepares to enter THE FUTURE
01:56 soh_cah_toa :)
01:57 cotto soh_cah_toa, why not parrot/parrot?  You have a commit bit.
01:57 soh_cah_toa i didn't want to piss people off at first
01:57 soh_cah_toa it's ok if i move it there?
01:58 cotto I'll be angry if you don't, so it averages out either way.
01:58 soh_cah_toa ha! sure thing
01:58 cotto ;]
01:59 cotto It'll be nice having all our out-of-date and disorganized information in a modern format.
02:00 cotto and if it's organized, even better!
02:00 soh_cah_toa yes, definitely
02:01 benabik Can you check out the github wikis via git?
02:01 soh_cah_toa benabik: yessir
02:02 cotto That's half the benefit.
02:02 benabik So we can just use fun things like git-grep and ack to search them.  Fun.
02:02 soh_cah_toa yup
02:03 soh_cah_toa there, all merged. it's on parrot/parrot now
02:05 cotto I like the future.  It smells nice.
02:05 soh_cah_toa it's refreshing
03:18 soh_cah_toa cotto: what's the url for your blog again?
03:19 cotto reparrot.org
03:19 soh_cah_toa ok, just adding it to the wiki
03:19 cotto er, reparrot.blogspot.com
03:53 dalek parrot: 851a2aa | plobsing++ | / (2 files):
03:53 dalek parrot: fix declaration constyness
03:53 dalek parrot: review: https://github.com/parrot/parrot/commit/851a2aaf17
03:53 dalek parrot: 0818052 | plobsing++ | t/src/extend_vtable.t:
03:53 dalek parrot: cast to keep c++ happy
03:53 dalek parrot: review: https://github.com/parrot/parrot/commit/081805293b
04:18 soh_cah_toa oh crap, what's this sample code submission thing for gsoc?
04:19 soh_cah_toa uh oh, i don't know how i missed this
04:22 soh_cah_toa somehow, the email ended up in my spam folder :(
04:23 cotto soh_cah_toa, email carol and explain.  We can vouch if needed.
04:23 soh_cah_toa good idea. i just figured after i got my "you passed" email, i was done. so long
04:29 soh_cah_toa ugh...it's too late at night to worry about this
04:38 aloha joined #parrot
04:57 JimmyZ joined #parrot
05:02 JimmyZ looks like parrot is broken on win32
05:04 cotto JimmyZ, that's good.
05:04 cotto We need to break the build more often.
05:04 cotto JimmyZ, do you have a fix?
05:05 cotto If not, I'll make sure to look into it tomorrow at $dayjob.
05:05 JimmyZ https://gist.github.com/1234080
05:06 cotto no shortage of warnings there
05:06 JimmyZ cotto: I don't know how to fix it. I couldn't find the problem
05:07 cotto JimmyZ, ok.  I'll make some noise and dig a bit tomorrow.
05:07 JimmyZ cotto: thanks
05:09 cotto JimmyZ, thanks for reporting
05:21 SHODAN joined #parrot
05:24 dalek parrot: fe4e1b9 | jimmy++ | src/platform/win32/ (3 files):
05:24 dalek parrot: quick fixes for win32
05:24 dalek parrot: review: https://github.com/parrot/parrot/commit/fe4e1b9fe7
05:24 JimmyZ cotto: got a quick fixes for window :)
05:31 cotto That's it?
05:31 cotto JimmyZ, how did that work before?
05:34 dalek parrot: 230a03b | jimmy++ | src/pmc/mappedbytearray.pmc:
05:34 dalek parrot: oops, forgot another one fix
05:34 dalek parrot: review: https://github.com/parrot/parrot/commit/230a03b722
05:34 preflex_ joined #parrot
05:35 cotto JimmyZ, does that fix everything?
05:35 JimmyZ cotto: I think kill_threads kill some include from io_threads.h
05:37 cotto JimmyZ, that's plausble
05:37 cotto *plausible
05:38 JimmyZ cotto: yes, this one https://github.com/parrot/parrot/commit/a5b5​0107f50da6ce68053545e5e0983a0ee14ec0#diff-17
05:39 cotto JimmyZ, ok
05:39 JimmyZ gcc--, which give me bad errors
05:40 cotto I'd be impressed with any compiler that guessed that you were missing an include.
05:43 moritz how do ByteBuffer and MappedByteArray mix?
05:43 moritz .oO( do they blend? )
05:53 cotto moritz, I think I see where you're going.
05:55 cotto you can use MappedByteArray.'get_string' (the method, not the VTABLE) with an offset, length and encoding.
05:56 sorear cotto: java ides can do that
05:56 cotto sorear, I don't see the connection.
05:57 sorear re. 22:40 < cotto> I'd be impressed with any compiler that guessed that you were  missing an include.
05:57 cotto sorear, ah.  that'd gone out of scope.
05:57 cotto with an ide I'd be less surprised.  They're pretty dwimmy and magical.
05:58 JimmyZ where, gcc give me 'syntax error', not  'DWORD is not pre-defined'
05:58 JimmyZ s/where/well/
05:58 cotto That's an important hint.
05:58 cotto but lta
05:58 moritz well, C syntax is quite complicated
05:58 sorear it's hard to be awesome in a language defined with macros
05:59 JimmyZ clang is better with error reporting?
06:00 JimmyZ re java, java can import missing package by mouse in eclipse ;)
06:01 cotto When PaFo is a multi-billion dollar company, PIR will have a similar ide.
06:01 JimmyZ and closed soure :(
06:02 cotto JimmyZ, like eclipse?  wink.
06:02 sorear cotto: wow, you're cynical.
06:03 cotto alternately, we could make an eclipse plugin
06:03 wagle joined #parrot
06:03 sorear "When PaFo is a multi-billion dollar company, we'll still be writing PIR directly."
06:04 JimmyZ When PaFo is a multi-billion dollar company, parrot removed PIR :)
06:04 cotto sorear, I kid.  PIR needs to die a fiery death, just as soon as we have something better.
06:04 sorear Fire is too good for PIR.
06:05 cotto sorear, you're prefer continued life-support?
06:10 wagle joined #parrot
07:00 mj41 joined #parrot
07:06 rfw joined #parrot
07:35 schmooster joined #parrot
08:14 lucian joined #parrot
08:47 moritz more like a demontor's kiss :-)
08:52 moritz feature request: I'd like to be able to write ByteBuffer objects to a FileHandle
08:52 moritz I can already read them with read_bytes
08:53 moritz it seems I can only write String PMCs
09:24 dalek rakudo/nom: 5a5f068 | moritz++ | / (4 files):
09:24 dalek rakudo/nom: implement basic, cheating Buf
09:24 dalek rakudo/nom:
09:24 dalek rakudo/nom: The "cheat" part is that it is actually a Buf8, not something
09:24 dalek rakudo/nom: that can store arbitrary integers
09:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5a5f068f09
09:24 dalek rakudo/nom: f516563 | moritz++ | src/core/IO.pm:
09:24 dalek rakudo/nom: IO.read
09:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f516563aad
09:24 dalek rakudo/nom: 1186484 | moritz++ | src/core/Buf.pm:
09:24 dalek rakudo/nom: simplify some Buf code, jnthn++
09:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1186484e13
09:30 honza joined #parrot
09:37 dalek rakudo/nom: be999e7 | moritz++ | src/core/IO.pm:
09:37 dalek rakudo/nom: IO.write(Buf)
09:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/be999e79d0
09:54 dalek rakudo/nom: 56df3d9 | moritz++ | src/core/Buf.pm:
09:54 dalek rakudo/nom: Buf.gist, eqv for Buf
09:54 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/56df3d93eb
09:56 dalek rakudo/nom: 5a7752c | moritz++ | NOMMAP.markdown:
09:56 dalek rakudo/nom: update NOMMAP
09:56 dalek rakudo/nom:
09:56 dalek rakudo/nom: Buf is now "good enough" (I hope) to be removed from the punch list,
09:56 dalek rakudo/nom: and the segfault in mandelbrot-color is long gone
09:56 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5a7752cb23
10:24 cotto joined #parrot
10:28 dalek rakudo/nom: 909cc3c | moritz++ | src/ (2 files):
10:28 dalek rakudo/nom: parse prefix:<~^>, and implement it for Buf
10:28 dalek rakudo/nom:
10:28 dalek rakudo/nom: also add Buf.elems
10:28 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/909cc3c689
10:52 dalek rakudo/nom: 9a2f7fc | jnthn++ | src/Perl6/Metamodel/NativeHOW.pm:
10:52 dalek rakudo/nom: Make Int ~~ int and various other things on native type objects not explode.
10:52 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9a2f7fcac7
10:52 dalek rakudo/nom: 1e27868 | jnthn++ | src/Perl6/Metamodel/NativeHOW.pm:
10:52 dalek rakudo/nom: NativeHOW should just compose MROBasedTypeChecking, not do its own thing.
10:52 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1e27868704
10:52 dalek rakudo/nom: 6a20116 | jnthn++ | src/Perl6/Actions.pm:
10:53 dalek rakudo/nom: Basic support for an argument with trait_mod:<is>.
10:53 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6a20116466
10:53 dalek rakudo/nom: 6eee0ca | jnthn++ | src/ (3 files):
10:53 dalek rakudo/nom: Stub in a few sized natives. No associated functionality yet.
10:53 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6eee0caab4
11:11 dalek rakudo/nom: 60e9ad7 | jnthn++ | src/core/natives.pm:
11:11 dalek rakudo/nom: Fix copy-paste fail noticed by JimmyZ++.
11:11 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/60e9ad7b4d
11:32 lucian joined #parrot
11:32 dalek rakudo/nom: e5fab7f | jnthn++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
11:32 dalek rakudo/nom: Fix generic instantiation bug that led to over-sharing of dispatchee lists.
11:32 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e5fab7fdf6
11:35 Psyche^ joined #parrot
11:48 whiteknight joined #parrot
11:48 whiteknight good morning, #parrot
11:49 moritz good morning whiteknight
11:57 woosley joined #parrot
12:03 dalek rakudo/nom: 2466044 | moritz++ | src/core/IO/Socket.pm:
12:03 dalek rakudo/nom: IO::Socket.read (returns Buf)
12:04 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2466044f25
12:04 whiteknight hello moritz, how are you doing today?
12:04 moritz whiteknight: great, I've had fun with ByteBuffer yesterday night
12:04 moritz ... and I still have
12:06 moritz NotFound++
12:06 whiteknight real fun, or the sarcastic kind of not-fun?
12:06 moritz some of the interfaces could need a bit polishing, but all in all I managed to do everything I wanted
12:06 moritz whiteknight: real fun.
12:06 whiteknight good
12:06 whiteknight ByteBuffer was a very nice addition
12:07 moritz whiteknight: rakudo/nom now has a Buf type, which wraps ByteBuffer
12:07 moritz it is. It would just be nice if it were integrated with more subsystems
12:07 whiteknight it is relatively newish. But yes, improved integration would be nice
12:07 whiteknight suggestions welcome
12:07 moritz for example FileHandle can read bytes into a ByteBuffer, but you can't directly print a ByteBuffer to a FileHandle
12:08 whiteknight oh, that is a very good point
12:08 moritz same for Socket
12:08 whiteknight okay, noted. I might be able to play with that this weekend
12:08 moritz (Socket doesn't have any ByteBuffer integration that I could see)
12:08 moritz whiteknight: want a ticket?
12:08 whiteknight yeah, sure.
12:08 whiteknight moritz++
12:14 Coke does reparrot need to be added to the parrot planet?
12:15 giwi joined #parrot
12:15 Coke "build broken on win32" is often the same as "build is broken with g++".
12:15 whiteknight what's reparrot?
12:15 whiteknight oh, cotto's blog?
12:16 whiteknight I'll fire up a windows build now to see what happens
12:16 moritz whiteknight: http://trac.parrot.org/parrot/ticket/2195
12:18 whiteknight thanks!
12:20 dalek rakudo/nom: 42abf87 | jnthn++ | t/spectest.data:
12:20 dalek rakudo/nom: Run S12-methods/multi.t.
12:20 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/42abf8785b
12:23 Coke someone should setup dalek to track the 2 planet svn repos.
12:25 allison joined #parrot
12:27 JimmyZ joined #parrot
12:28 benabik joined #parrot
12:30 dalek TT #2195 created by moritz++: TODO: integrate ByteBuffer better with FileHandle and Socket
12:30 dalek TT #2195: http://trac.parrot.org/parrot/ticket/2195
12:33 whiteknight Coke: windows build goes fine for me here. extend_vtable.t fails some subtests
12:33 dalek rakudo/nom: 8ba1a33 | moritz++ | src/core/IO/Socket (2 files):
12:33 dalek rakudo/nom: [IO] accept Cool args; $.listen is now Bool
12:33 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8ba1a3375c
12:41 bluescreen joined #parrot
12:51 dalek rakudo/nom: 4152d21 | moritz++ | LHF.markdown:
12:51 dalek rakudo/nom: update NOMMAP
12:51 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/4152d21cf5
13:14 Coke whiteknight: I wasn't the original reporter. I suspect that the recent g++ build fixes fixed the original issue in time for you to test clean.
13:38 whiteknight Coke: okay, thanks
13:51 dalek rakudo/nom: 393309c | moritz++ | src/core/Buf.pm:
13:51 dalek rakudo/nom: non-spec Buf.append_inplace hack for bbkr++
13:51 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/393309cbec
13:53 bluescreen joined #parrot
14:18 NotFound ~~
14:19 NotFound moritz: yes, I've been thinking abou using ByteBuffer in several internals, but lacked the time to play with that ideas,
14:30 dalek nqp: 7595e8b | jnthn++ | src/PAST/NQP.pir:
14:30 dalek nqp: Add some NQP ops.
14:30 dalek nqp: review: https://github.com/perl6/nqp/commit/7595e8b601
14:30 dalek nqp: ca60d32 | jnthn++ | src/NCI/CallBuilder.nqp:
14:31 dalek nqp: Start to sketch out NQP/6model level NCI handling. Rather than do everything itself, the Perl 6 NativeCall library can be built on top of this.
14:31 dalek nqp: review: https://github.com/perl6/nqp/commit/ca60d32a1d
14:32 benabik Hm.  Low-level P6 NCI handling.  Expect a lot of people trying to figure our our NCI stuff soon...
14:33 benabik By default we build with libffi if it exists, right?
14:34 NotFound benabik: that depends on how you measure "exists"
14:35 NotFound As usual, in some systems you need the devel package, in addition to the bare library.
14:35 benabik NotFound: Can be found by auto::ffi?
14:35 benabik *libffi
14:36 NotFound libffi-dev in Debian, for example.
14:36 benabik Right.  But if it's there, we use it.
14:36 NotFound Yeah
14:37 benabik Mostly I'm wondering if Parrot built automatically by NQP/Rakudo will have libffi or if we're going to have people complaining once jnthn++ finishes this NativeCall stuff.  :-D
14:37 NotFound I never build without it these days.
14:37 benabik I probably don't either…  Appears to be a base part of the 10.7 SDK.
14:39 bluescreen joined #parrot
14:51 jsut joined #parrot
15:04 dalek nqp: 4c6d236 | jnthn++ | / (3 files):
15:04 dalek nqp: Install NCI::CallBuilder.
15:04 dalek nqp: review: https://github.com/perl6/nqp/commit/4c6d236801
15:14 moritz NotFound: I still found ByteBuffer to be quite useful
15:18 lucian joined #parrot
15:18 alester joined #parrot
15:31 woosley left #parrot
15:38 dalek rakudo/nom: aa5e9b6 | jnthn++ | src/Perl6/Actions.pm:
15:38 dalek rakudo/nom: Generate lexicals of the correct primitive type for holding arguments.
15:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/aa5e9b68d4
15:38 dalek rakudo/nom: bd4af3e | jnthn++ | src/ (2 files):
15:38 dalek rakudo/nom: Start passing nativeness info down towards the binder.
15:38 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/bd4af3e962
15:55 dalek winxed: df7cebe | NotFound++ | winxedst1.winxed:
15:55 dalek winxed: bareword int, float, string and var and partial class operator supoort in 'multi' modifier
15:55 dalek winxed: review: https://github.com/NotFoun​d/winxed/commit/df7cebe4fb
16:07 dalek rakudo/nom: a1032e0 | Coke++ | t/spectest.data:
16:07 dalek rakudo/nom: run fudged test, track failures
16:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a1032e046e
16:39 Coke The word of the day is.... STABULOSO.
16:43 benabik SABULOSO: No results found
16:43 mj41 joined #parrot
16:44 * benabik finds it ridiculous that he has three "Search in Google" items in his context menu...
17:19 fperrad joined #parrot
17:59 dukeleto ~~
17:59 whiteknight hello duke
18:02 davidfetter joined #parrot
18:03 dukeleto whiteknight: how goes it this morning?
18:03 whiteknight as good as ever, I suppose. you?
18:10 dukeleto whiteknight: getting back on the productivity wagon, I suppose
18:10 dukeleto whiteknight: i was a bit sick for a few weeks
18:10 whiteknight oh nice, it's a fun wagon to be on
18:10 whiteknight sick? oh noes!
18:11 nine whiteknight: kill_threads merged yet?
18:11 whiteknight nine: yessir
18:11 whiteknight yesterday morning
18:12 whiteknight I don't know what kind of havok it is going to wreak on the gsoc_threads branch if you try to merge from master now
18:12 nine whiteknight: good. Since I'm stuck quite firmly with that build error, I'll just try to merge further. Maybe it will go away magically...
18:13 whiteknight nine: Yeah, I haven't had a chance to look at it any more closely myself. I suspect there is probably something stupid that is missing, like needing to do some kind of obscure renumbering step
18:14 whiteknight I don't remember if that branch added any new opcodes or new pmc types. If one of those got updating and numbering wasn't redone, there could be a mixup
18:14 nine whiteknight: it does add new ops
18:15 nine that's why I'm trying to make bootstrap-ops which unfortunately depends on having working ops...
18:15 whiteknight unfortunately, something like "make bootstrap-ops" requires you to have a working Parrot in your build directory first, to redo the ops. So trying to fix that is going to involve some complicated bootstrapping, some tequila, and some cursewords
18:15 whiteknight :)
18:15 whiteknight nine: I'll try to tackle that tonight. It's likely to be a multi-step process
18:15 nine whiteknight: that'd be awesome :)
18:16 whiteknight go to master branch and build parrot, checkout the gsoc_threads branch without touching anything and make bootstrap-ops, etc
18:16 nine I think I tried exactly that. Ended up with different build errors...
18:16 whiteknight oh fun
18:16 whiteknight I might not have enough tequila
18:17 nine I'm usually not one to give up easily :)
18:17 whiteknight I do have plenty of cursewords though
18:17 dukeleto nine: good to hear.
18:18 dukeleto EINSUFFICIENTTEQUILA is a sad, sad error condition.
18:18 whiteknight dukeleto: yes, but one we can catch and handle
18:22 dukeleto Nice to see https://github.com/parrot/parrot/wiki
18:25 dukeleto so, we are supposed to have a linkable M0 implementation in C in a month for 3.9.
18:25 dukeleto We need to get on that.
18:25 dukeleto whiteknight: what are your priorities for the next month?
18:26 dukeleto whiteknight: in terms of parrot core
18:26 whiteknight dukeleto: party hard!
18:26 whiteknight I've got some GC stuff I want to play with. I've got some PCC stuff to dig at too. I think between those I can free up a few % performance
18:26 whiteknight IMCC, packfiles, debugger
18:27 whiteknight I want to help nine with threads too
18:46 cotto_work ~~
18:48 Coke dukeleto: parrot doesn't provide tequila, only mash.
18:52 nine Seems like bootstrap-ops alone is not enough. I built again from master, checked out green_threads and modified the makefile to not rebuild ops2c for bootstrap-ops. With that I got bootstrap-ops to rebuild core_ops.c ops.h and opsenum.h before failing. But even with that I get the same error.
18:52 dukeleto nine: what exactly are you trying to accomplish?
18:53 nine dukeleto: update the gsoc_threads branch to (semi-)current master
18:53 nine dukeleto: getting this error after the merge: http://paste.scsys.co.uk/144054
18:54 dukeleto nine: that might be from some things recently being removed from the default pmc
18:55 nine dukeleto: the master I merged with is from 2011-09-14
18:55 dukeleto nine: i think that is before the kill-useless-defaults merge, then
18:55 nine yep
18:59 dukeleto nine: how exactly are you performing the merge?
19:01 nine dukeleto: well a naive git merge master followed by a long git mergetool session and some guessing what the right result should look like
19:07 dukeleto nine: are you trying to merge gsoc_threads into master, or the other way around?
19:08 nine dukeleto: the other way around
19:09 dukeleto nine: create a new branch from master, then try to merge gsoc_threads into that and see if you get the same conflicts/build errors
19:09 dukeleto nine: if you hit a roadblock
19:09 dukeleto nine: if you think you are close with the way you did it now, press on. But I don't know what is up with that set_attr_str error
19:10 dukeleto nine: i have a hunch that plobsing++ could help you complete the merge. He has scary ninja powers.
19:11 nine dukeleto: actually I think the merge was quite ok. But as I ran out of options, I'll just give it a try to merge the other way around.
19:14 dukeleto nine: i see. it isn't the conflicts but the build errors. I point towards asking on parrot-dev
19:16 nine Can't hurt in any case. Writing right now
19:16 dukeleto nine++
19:17 tadzik I thought nine already posted that to parrot-dev
19:17 nine tadzik: not yet
19:18 tadzik maybe I misremember then
19:18 nine tadzik: only the intention of working on threads but not the build error
19:18 tadzik a'riht
19:20 dmalcolm joined #parrot
19:26 dalek rakudo/nom: 4c67ac6 | jnthn++ | src/binder/bind. (2 files):
19:26 dalek rakudo/nom: Tentatively start to refactor the binder so it can work with natively typed arguments and parameters. This doesn't do much besides change the API used by a couple of low-level bits.
19:26 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/4c67ac65aa
19:26 dalek rakudo/nom: 78198e7 | jnthn++ | src/binder/bind.c:
19:26 dalek rakudo/nom: Extensive refactor of parameter binding to work towards supporting natively typed parameters/arguments.
19:26 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/78198e72a0
19:26 dalek rakudo/nom: e089c70 | jnthn++ | src/binder/bind.c:
19:26 dalek rakudo/nom: Get binder to handle being passed natively typed values and binding them to natively typed lexical registers. Note that the code-gen isn't really up to scratch for using this yet, but the binder part now seems to work.
19:26 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e089c70f2a
19:26 dalek rakudo/nom: ff86674 | jnthn++ | src/Perl6/Actions.pm:
19:26 dalek rakudo/nom: Ensure we generate correct code for natively typed parameter lookups when they're used.
19:26 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ff866742e9
19:48 dalek rakudo/nom: 8943be1 | moritz++ | src/core/Buf.pm:
19:48 dalek rakudo/nom: infix:<~>(Buf, Buf)
19:48 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8943be1d6c
19:50 dalek nqp: c4f9df4 | jnthn++ | src/pmc/stable.pmc:
19:50 dalek nqp: Give us a bit more space in the type cache IDs.
19:50 dalek nqp: review: https://github.com/perl6/nqp/commit/c4f9df4bdd
19:50 dalek nqp: 732b4f3 | jnthn++ | src/6model/ (8 files):
19:50 dalek nqp: Extend storage spec support so a REPR can describe the things it can/canny box.
19:50 dalek nqp: review: https://github.com/perl6/nqp/commit/732b4f37b9
19:50 dalek nqp: ce48590 | jnthn++ | src/6model/reprs/P6opaque.c:
19:50 dalek nqp: Eliminate an unused variable, and the warning that goes with it.
19:50 dalek nqp: review: https://github.com/perl6/nqp/commit/ce48590c6d
19:50 benabik Thought on nine's error.  The merge is probably attempting to merge the bootstrapped ops.  Should merge just the .ops files and then may be able to bootstrap.
19:52 dalek rakudo/nom: 943dcd1 | moritz++ | src/core/Buf.pm:
19:52 dalek rakudo/nom: [Buf] remove append_inplace hack; ordinary concatenation performs nearly equally well, so this was a case of premature optimization. Go figure.
19:52 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/943dcd1fce
19:54 benabik nine: You might want to try to redo your merge from master.  Ensure the result has the bootstrapped code from master and merge the source appropriately.
19:54 dukeleto benabik: do you think the git peeps would take a patch that implemented git cherry-pick --abort ?
19:55 benabik dukeleto: How is that different than git reset --hard?
19:55 dukeleto benabik: sugar
19:55 dukeleto benabik: i usually get nervous when I have to git reset. git rebase has an --abort, why not cherry-pick ?
19:56 benabik Does cherry-pick have a --continue?
19:56 dukeleto benabik: also, i signed up to attend the Git Together this year
19:56 benabik Generally --abort is on ops with --continue
19:56 dukeleto benabik: no, it doesn't. but it has --no-commit
19:56 benabik dukeleto: That's awesome.  It tends to be west coast, so I don't go.
19:57 dukeleto benabik: it is the 2 days directly after the gsoc mentor summit
19:57 benabik dukeleto: Awesome, but I don't get to go to that either, I think.  :-P
19:57 dukeleto benabik: so i will just chill at google for 4 days straight, conferencing.
19:57 dukeleto benabik: touche. Only mentors. But you could be a mentor next year...
19:57 benabik dukeleto: Awe-mazing.
19:58 cotto_work That could work.  I know a guy.
19:58 cotto_work I wish I could go to the VMIL workshop without giving the ACM money.
19:59 benabik I might try to be a student again, if that's kosher…  If everything goes right I'll be done with Thesis but have nothing to do until Fall.
19:59 * dukeleto is knee deep in arcane command-line options to bioinformatics programs that evovled like slime molds
19:59 dukeleto benabik: being a student again sounds great too.
19:59 dukeleto cotto_work: yeah, i am allergic to giving people money. Especially rich people.
20:00 cotto_work On the other hand, I really want to get some face time with Lars Bak and steal some ideas for M0.
20:05 dukeleto cotto_work: could be worth it.
20:06 cotto_work yeah
20:08 cotto_work dukeleto: have you gotten anywhere integrating gitster++'s changes to the git workflow?
20:16 dukeleto cotto_work: nope
20:16 dukeleto cotto_work: all my paying jobs have been keeping me busy :)
20:17 dalek rakudo/nom: 57dc7ad | jnthn++ | / (2 files):
20:17 dalek rakudo/nom: Bump to latest nqp/6model.
20:17 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/57dc7ad953
20:17 dalek rakudo/nom: 5af26ae | jnthn++ | src/binder/bind.c:
20:17 dalek rakudo/nom: Handle case where natively typed parameters are passed boxed arguments. Binding works if the thing passed knows how to unbox itself to the correct native type.
20:17 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5af26ae125
20:18 dukeleto cotto_work: most of the gitster++ change was that I was referring to "index" which evidently would confuse git core devs
20:20 dalek winxed: 211f690 | NotFound++ | / (4 files):
20:20 dalek winxed: update NEWS and generated files
20:20 dalek winxed: review: https://github.com/NotFoun​d/winxed/commit/211f690a0a
20:21 dukeleto cotto_work: but git_workflow.pod gives slightly wrong directions for people working in forks versus having a commit bit. That needs to be fixed.
20:22 dalek parrot: 5d88ac2 | NotFound++ | ext/winxed/ (2 files):
20:22 dalek parrot: update winxed snapshot to 211f690a0a to allow testing new multi features
20:22 dalek parrot: review: https://github.com/parrot/parrot/commit/5d88ac2e9b
20:24 cotto_work what branch is nine working in?
20:24 tadzik gsoc_threads
20:26 whiteknight did he create a new branch from that, or a fork?
20:26 tadzik I think he's trying to merge between it, and master
20:28 cotto_work I think it should be straightforward (if hacky) to fix, once I figure out what's going on.
20:28 cotto_work ah. green_threads, not gsoc_threads
20:28 cotto_work but not in parrot/parrot.git
20:29 cotto_work what's his github username?
20:29 cotto_work niner
20:33 mj41 joined #parrot
20:33 cotto_work seen nine
20:33 aloha nine was last seen in #parrot 1 hours 15 mins ago saying "tadzik: only the intention of working on threads but not the build error".
20:34 tadzik this one an hour ago? Damn, time flies
20:37 soh_cah_toa joined #parrot
20:37 cotto_work generated code is nice because you can do horribly hacky things to it and once it works, all the hacks magically go away.
20:43 cotto_work that branch needs a bit more love than I expected
20:57 dalek parrot: 56865e6 | soh_cah_toa++ | tools/release/auto_release.pl:
20:57 dalek parrot: Removed -j switch when executing 'make'. Also got rid of the Env module in favor of just using the %ENV hash.
20:57 dalek parrot: review: https://github.com/parrot/parrot/commit/56865e6572
21:01 dukeleto Having to maintain instances of your own software sure does give one a does of self-medicine.
21:01 soh_cah_toa :)
21:04 dukeleto dose, even.
21:12 soh_cah_toa msg whiteknight i started moving our ugly wiki over to github. take a look: https://github.com/parrot/parrot/wiki it's purdddddyyyy :)
21:12 aloha OK. I'll deliver the message.
21:15 dalek nqp: 0751cd8 | jnthn++ | tools/build/Makefile.in:
21:15 dalek nqp: Fix build for various platforms.
21:15 dalek nqp: review: https://github.com/perl6/nqp/commit/0751cd8dac
21:16 dalek rakudo/nom: 33e65ba | jnthn++ | tools/build/NQP_REVISION:
21:16 dalek rakudo/nom: Use a build of NQP that isn't broken on some platforms.
21:16 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/33e65ba625
21:17 benabik soh_cah_toa: the release script might want to take a -j option to pass to make and a --test-jobs option to use as TEST_JOBS.  (Random thought)
21:18 dukeleto benabik: useful random thought
21:18 dukeleto soh_cah_toa++ # wiki-moving
21:18 benabik soh_cah_toa++ # General awesomeness.
21:19 soh_cah_toa ha!
21:19 benabik soh_cah_toa: I still need to get biggs to drink that "beer" we bought for you at YAPC.
21:19 soh_cah_toa ha! x2
21:19 soh_cah_toa that's funny
21:19 benabik He said he would, but somehow hasn't gotten around to it.
21:20 soh_cah_toa boy, i wonder why... ;)
21:26 dalek nqp: 76afaff | tadzik++ | tools/build/Makefile.in:
21:26 dalek nqp: [Makefile.in] Fix dependencies
21:26 dalek nqp: review: https://github.com/perl6/nqp/commit/76afaffa6f
21:29 estrai joined #parrot
21:47 GodFather joined #parrot
21:58 simcop2387 joined #parrot
22:10 dalek rakudo/nom: 952b7bb | jnthn++ | src/binder/multidispatch. (2 files):
22:10 dalek rakudo/nom: Note presence of native types in the multi-dispatcher candidate graph; not doing anything with them yet.
22:10 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/952b7bb084
22:26 dmalcolm joined #parrot
22:32 nbrown joined #parrot
22:37 dalek nqp/kill-useless-defaults: 7595e8b | jnthn++ | src/PAST/NQP.pir:
22:37 dalek nqp/kill-useless-defaults: Add some NQP ops.
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/7595e8b601
22:37 dalek nqp/kill-useless-defaults: ca60d32 | jnthn++ | src/NCI/CallBuilder.nqp:
22:37 dalek nqp/kill-useless-defaults: Start to sketch out NQP/6model level NCI handling. Rather than do everything itself, the Perl 6 NativeCall library can be built on top of this.
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/ca60d32a1d
22:37 dalek nqp/kill-useless-defaults: 4c6d236 | jnthn++ | / (3 files):
22:37 dalek nqp/kill-useless-defaults: Install NCI::CallBuilder.
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/4c6d236801
22:37 dalek nqp/kill-useless-defaults: c4f9df4 | jnthn++ | src/pmc/stable.pmc:
22:37 dalek nqp/kill-useless-defaults: Give us a bit more space in the type cache IDs.
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/c4f9df4bdd
22:37 dalek nqp/kill-useless-defaults: 732b4f3 | jnthn++ | src/6model/ (8 files):
22:37 dalek nqp/kill-useless-defaults: Extend storage spec support so a REPR can describe the things it can/canny box.
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/732b4f37b9
22:37 dalek nqp/kill-useless-defaults: ce48590 | jnthn++ | src/6model/reprs/P6opaque.c:
22:37 dalek nqp/kill-useless-defaults: Eliminate an unused variable, and the warning that goes with it.
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/ce48590c6d
22:37 dalek nqp/kill-useless-defaults: 0751cd8 | jnthn++ | tools/build/Makefile.in:
22:37 dalek nqp/kill-useless-defaults: Fix build for various platforms.
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/0751cd8dac
22:37 dalek nqp/kill-useless-defaults: 76afaff | tadzik++ | tools/build/Makefile.in:
22:37 dalek nqp/kill-useless-defaults: [Makefile.in] Fix dependencies
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/76afaffa6f
22:37 benabik Oh, bah.  it's just a merge.
22:37 dalek nqp/kill-useless-defaults: 9ed7428 | benabik++ | / (12 files):
22:37 dalek nqp/kill-useless-defaults: Merge branch 'master' into kill-useless-defaults
22:37 dalek nqp/kill-useless-defaults:
22:37 dalek nqp/kill-useless-defaults: Conflicts:
22:37 dalek nqp/kill-useless-defaults: src/PAST/NQP.pir
22:37 dalek nqp/kill-useless-defaults: tools/build/PARROT_REVISION
22:37 dalek nqp/kill-useless-defaults: review: https://github.com/perl6/nqp/commit/9ed74283d1
22:59 whiteknight joined #parrot
23:05 whiteknight good evening, #parrot
23:16 whiteknight NotFound: ping
23:19 whiteknight msg NotFound take a look at https://gist.github.com/1236313 . Get a null pmc access on line 5. It's looking up the variable foo as a global instead of using the local value
23:19 aloha OK. I'll deliver the message.
23:21 dalek Rosella/commandline: e923be2 | Whiteknight++ | / (2 files):
23:21 dalek Rosella/commandline: Add CommandLine to VERSION. cleanups for test_template utility
23:21 dalek Rosella/commandline: review: https://github.com/Whiteknig​ht/Rosella/commit/e923be2bae
23:21 dalek Rosella/commandline: 6b933f4 | Whiteknight++ | src/u (4 files):
23:21 dalek Rosella/commandline: Update test_all_lib to use CommandLine. Cleanups and fixes
23:21 dalek Rosella/commandline: review: https://github.com/Whiteknig​ht/Rosella/commit/6b933f49bc
23:21 dalek Rosella/commandline: 53590c5 | Whiteknight++ | src/utilities/test_all_lib.winxed:
23:21 dalek Rosella/commandline: +some comments for test_all_lib
23:21 dalek Rosella/commandline: review: https://github.com/Whiteknig​ht/Rosella/commit/53590c5fbd
23:21 dalek Rosella/commandline: 6cd4601 | Whiteknight++ | src/utilities/ (2 files):
23:21 dalek Rosella/commandline: Convert mk_winxed_header to the CommandLine library
23:21 dalek Rosella/commandline: review: https://github.com/Whiteknig​ht/Rosella/commit/6cd4601ce3
23:21 dalek Rosella/commandline: c627fa3 | Whiteknight++ | / (7 files):
23:21 dalek Rosella/commandline: Add in stub test files for the commandline classes. Fix up the test_class.winxed.template with some improvements. A few other misc fixes
23:21 dalek Rosella/commandline: review: https://github.com/Whiteknig​ht/Rosella/commit/c627fa321c
23:46 GodFather joined #parrot
23:50 benabik joined #parrot

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

Parrot | source cross referenced