#parrot Parrot 4.11.0 "[All together - Happy Birthday Lovebird]" --with-threads | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 11 January 2013.
00:18 patspam joined 01:06 woosley joined 01:11 Mike-PerlRecruiter_ joined
kid51 seen benabik? 01:38
aloha benabik was last seen in #perl6 1 hours 51 mins ago joining the channel.
01:58 benabik joined 02:50 kid51_ joined 06:31 benabik joined 09:42 Psyche^ joined 10:39 woolfy joined 11:03 bouncy joined 11:21 woolfy joined 11:28 whiteknight joined 11:34 Liz joined 11:35 woolfy joined 11:48 woolfy joined, Liz joined 11:49 Liz joined, woolfy_ joined 11:52 Liz_ joined 13:11 Mike-PerlRecruiter_ joined 13:39 mtk joined 13:57 PacoAir joined
dalek p: 81d9937 | (Paweł Murias)++ | t/nqp/59-nqpop.t:
Add a test for nqp::existskey.
13:59
14:07 contingencyplan joined 14:32 Liz joined, woolfy joined 14:34 bluescreen joined 14:48 woolfy joined, Liz joined 14:57 Liz joined 14:58 woolfy joined 15:08 woolfy joined, Liz joined 15:16 Liz joined, woolfy joined 15:17 awwaiid joined 15:36 woolfy joined 15:38 Liz joined 15:39 woolfy joined
Coke should have all the pugs tests fudged and summarizable properly again in a day or so. 15:41
Most of the damage was from inclusion of Test::Util, which doesn't work on pugs, but referencing it anywhere (even in a block) caused the entire test to abort. 15:42
moritz Coke: most of the time you can just move the 'use' statement into the block where throws_like is used 15:45
Coke: and then do a #?pugs skip in front of the whole block
EWRONGCHANNEL btw :-) 15:46
Coke whoops.
and yes, that's what I did.
rurban thanks anyway :) 15:54
16:47 alester joined 16:49 dmalcolm joined 19:02 davidfetter joined 19:28 tuxit joined 20:15 Khisanth joined 20:31 Khisanth joined 20:40 Khisanth joined 20:45 Liz_ joined 20:47 woolfy joined 20:48 Liz joined
dalek p: 7c8021d | jnthn++ | src/ (2 files):
Replace a couple of pir op uses with nqp ops.

Just some assorted ones where we already have the NQP op defined.
21:14
p: 4e51b4c | jnthn++ | src/QAST/Operations.nqp:
Add an nqp:: op API for working with SCs.

It replaces a rather non-uniform mix of v-table calls, method calls and pirops. Probably there are a few more needed, but this covers the majority.
p: aa64ea7 | jnthn++ | src/stage0/ (9 files):
Update bootstrap so we can use the SC ops in NQP.
p: ff51c75 | jnthn++ | src/ (2 files):
Start using the nqp:: ops for SC handling.
p: 9815308 | jnthn++ | t/serialization/0 (3 files):
Start updating SC tests to use new ops.
21:19 Khisanth joined 21:24 Liz joined 21:34 woolfy joined 21:35 Liz joined 22:24 Liz joined 22:38 bluescreen joined 22:49 woolfy joined 22:50 Liz joined
dalek p: fb2b87e | jnthn++ | src/how/NQPAttribute.pm:
Toss some dead code.
22:54
p: caa866f | jnthn++ | src/how/ (5 files):
Various cleanups and simplifications in MOP code.
p: 8319c39 | jnthn++ | src/how/NQPModuleHOW.pm:
Eliminate a now-unrequired hack.
p: 4084b68 | jnthn++ | src/how/ (4 files):
A further round of MOP code simplifications.

We've relied in various places on methods that Parrot's PMCs provide, but that it's not reasonable to expect a ground-up 6model runtime to offer. In general, the only things that define internal methods are the KnowHOW and KnowHOWAttribute, from which everything else is built. By keeping the MOP simpler, it means we can keep the runtime leaner and write more of NQP in NQP. This isn't the entire set of changes we need, but many of them.
23:21 woolfy joined, Liz joined 23:24 aloha joined