Parrot 5.0.0 "Johnny Five Alive" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 23 January 2013.
01:10 Reini joined 01:36 tv joined 02:40 Reini joined 03:00 kid51 joined 03:38 sivoais joined 04:08 Reini joined 04:33 Eddward joined
cotto allison: that response to whiteknight++ was delightful. Thanks for framing the discussion like that. It's a much better expression of how I've been thinking about this whole situation. 05:05
allison cotto: thanks :) 05:06
PerlJam url? 05:24
cotto lists.parrot.org/pipermail/parrot-d...07352.html 05:29
PerlJam thanks. 05:30
nice. 05:31
allison I've got a build failure in nqp_ops.c (using just the sixparrot branch, and stock rakudo+nqp), did I catch the middle of a transition? 05:37
cotto allison: possibly. nqp was broken for a bit yesterday 05:40
And someone mentioned (but I haven't reproduced) an issue where nqp head and parrot head don't play nice. 05:41
allison it looks like explicitly specifying the master branch of nqp and restarting kicks it off in the right direction 05:44
06:03 Reini joined
allison so, NQP seems to use no float params, almost no string params, and a tiny number of int params (mostly for the opt_flag on optional parameters) 06:12
might be able to whack out dispatch complexity that way 06:13
cotto do floats get much use? 06:14
allison not sure, it depends on where Rakudo is choosing to do optimizations 06:15
it is using slurpy some, not much but some 06:17
it's using named more than slurpy 06:18
it doesn't appear to use :multi at all 06:29
sorear isn't 99% of NQP using :call_sig ? 06:32
I don't know how much you've been kept in the loop, but make sure you have the currently active branch of nqp (nom?) 06:34
allison sorear: I'm using whatever rakudo pulls in by default 06:45
sorear: and, I was hoping it mostly used :call_sig, so I could rip out everything else :)
sorear: I see :call_sig as part of what's generated, but not internal to nqp 06:49
sorear: this is master branch of nqp
07:07 Mike-PerlRecruiter_ joined
cotto Blogging is hard. 07:24
I should probably say that after I'm done. 07:27
allison cotto: aye, blogging is not my favorite activity 07:34
cotto: I occasionally think "I should blog about that"
cotto: I even start posts
cotto: actually, I still have an unfinished draft post from when you stepped up as chief architect :) 07:35
cotto: one about my electric car too
(I think both from when I bought it, and another one from when it finally died)
but, you can see how often I actually post 07:36
like, twice a year for UDS, mainly
cotto allison: ditto.
I'm sure that once I log in to copy/paste the post, I'll see several half-ready posts that are now hopelessly obsolete.
allison yeah 07:38
so, I think mmd and :multi and MULTI are totally removable 07:39
cotto I... um... 07:40
wow
allison definitely more work than dropping an op, but worth it if it significantly speeds up dispatch 07:43
cotto Is that assuming that we'll be using hand-hacked ops?
i.e. no opsc
allison dispatch for things like the i_i variant of add is a simpler case than the manhattan distance stuff 07:45
(and we never really *properly* did manhattan distance anyway) 07:46
it's a bit of a flip for some of the core pmcs, but it's actually only returning them to an earlier state
oof, I should sleep 07:49
more tomorrows
cotto 'night
question for the room: what should Parrot's release cycle look like in the future? I don't have an answer either way. 08:08
just something to think about 08:09
moritz fwiw rakudo fares well with a monthly release cycle 08:53
I don't really see the point in the development/supported release distinction
cotto moritz: yeah. That's an implication of the current plan that I didn't realize was true until you mentioned it just now. 08:56
moritz (I haven't seen the point of it in the past either, fwiw)
cotto moritz: it would have been a good idea if Parrot were like Perl 5 at the time it was adopted. 08:57
moritz cotto: agreed 08:58
cotto Don't give me a time machine or I'll go mad just trying to decide what to fix. 09:00
I'd probably end up going back in time to stop you from giving me a time machine. 09:01
;)
09:02 bouncy joined
moritz :-) 09:27
09:45 Reini joined
dalek kudo/nom: 5a12919 | jnthn++ | tools/build/NQP_REVISION:
Bump to an NQP with a build fix.
 09:46
10:00 woolfy joined 10:01 Liz joined
dalek rrot: 5896d97 | dukeleto++ | / (2 files):
Add a sort method to FixedFloatArray
 10:03
rrot: 26cdda8 | dukeleto++ | t/pmc/fixedfloatarray.t:
Add some tests for FFA.sort
rrot: ea40261 | dukeleto++ | src/pmc/fixedfloatarray.pmc:
Fix a subtle bug where sort would not work correctly on systems where sizeof(INTVAL) != sizeof(FLOATVAL)
rrot: 8c9a371 | dukeleto++ | src/pmc/fixedfloatarray.pmc:
Hopefully fix FFA.sort() bug, caused by my questionable casting. @nwellnhof++ #929
rrot: 94e7f64 | dukeleto++ | src/pmc/fixedfloatarray.pmc:
Eschew the use of == on FLOATVALs, @pmichaud++
rrot: 9fe9dfd | dukeleto++ | / (2 files):
Update some copyrights
rrot: e79967d | dukeleto++ | / (3 files):
Merge branch 'ffa_sort'
10:22 travis-ci joined
travis-ci [travis-ci] parrot/parrot#811 (master - e79967d : Jonathan "Duke" Leto): The build was broken. 10:22
[travis-ci] Change view : github.com/parrot/parrot/compare/6...9967dc122d
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/4765235
10:22 travis-ci left
dalek p: c83b4b5 | jnthn++ | src/QRegex/NFA.nqp:
s/our/my/ in a place things needn't be our.
 10:32
p: d89a658 | jnthn++ | src/QAST/Compiler.nqp:
Toss dead code-path.
p: a746f77 | jnthn++ | src/QRegex/Cursor.nqp:
Performance/portability tweaks to Cursor.
p: 749768c | jnthn++ | src/ (2 files):
Fix nqp::isinvokable to know invocation spec.
 11:20
p: 28961ee | jnthn++ | src/NQP/World.pm:
Update a comment.
 12:11
p: da26f1c | jnthn++ | src/NQP/World.pm:
Replace one use of nqp_get_package_through_who.

This op really needs to go away.
p: cf72dc1 | jnthn++ | src/NQP/Actions.pm:
Eliminate final nqp_get_package_through_who use.

This also means that Foo::Bar can now work on nqp-jvm.
12:52 he_ joined 13:57 PacoAir joined 13:58 Reini joined 14:08 bluescreen joined 14:11 uvtc joined 14:21 Psyche^ joined 14:42 awwaiid joined, bluescreen_ joined 15:00 benabik joined 15:32 alester joined 15:39 dmalcolm joined 17:52 uvtc left 17:57 xcombelle joined 18:12 sivoais joined 18:35 zby_home joined 18:59 Reini joined 19:07 Mike-PerlRecruiter_ joined 20:00 Reini joined 20:22 benabik joined 21:20 zby_home joined 21:44 benabik joined 21:49 donaldh joined 22:06 Reini joined
dalek p: c9bc4a2 | jnthn++ | t/nqp/59-nqpop.t:
Add nqp::split tests; donaldh++.
 22:13
23:37 Reini joined 23:48 kid51 joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!