»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
moritz good morning 04:49
sorear good morning moritz 05:01
dalek ecza: b344401 | sorear++ | lib/ (2 files):
ThawBuffer knows its setting
 06:46
ecza: d8852de | sorear++ | lib/ (2 files):
Fix over-eager caching of method handlers, colomon++
sorear phenny: tell colomon the bug you found while trying to enhance "cmp" is fixed now
phenny sorear: I'll pass that on when colomon is around.
dalek rlito: bd673d3 | (Flavio S. Glock)++ | / (15 files):
Perlito5 - add "Javascript3" backend (containers are objects)
 07:30
dalek rlito: ad58fdb | (Flavio S. Glock)++ | / (2 files):
Perlito5 - Javascript3 - fix method names
 07:40
masak morning, #perl6 08:12
sorear o/ 08:15
dalek rlito: bb527ad | (Flavio S. Glock)++ | / (4 files):
Perlito5 - js3 - add make build-5js3
 08:17
cognominal nqp: say(?$/) 09:13
p6eval nqp: OUTPUT«(signal SEGV)»
cognominal keeping at breaking things
FROGGS morning
cognominal morning
masak nqp: ?$/ 09:28
p6eval nqp: OUTPUT«(signal SEGV)»
masak nqp: $/ 09:29
p6eval nqp: ( no output )
masak nqp: say $/.WHAT
p6eval nqp: OUTPUT«Confused at line 2, near "say $/.WHA"␤current instr.: 'panic' pc 19999 (src/stage2/gen/NQPHLL.pir:7315) (src/stage2/gen/NQPHLL.pm:324)␤»
masak hm :)
dalek rlito: 9e72206 | (Flavio S. Glock)++ | / (2 files):
Perlito5 - js3 - fix list interpolation
 09:30
dalek rlito: 5640877 | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js3 - fix "our" variables
 09:55
dalek rlito: 7f2e5dc | (Flavio S. Glock)++ | / (2 files):
Perlito5 - js3 - fix die()
 10:02
felher I guess I am just blind, but where is specified what happens if you C<use> a file that is just a normal file, not a module, not a package? 10:16
tadzik nqp: say($/.WHAT) 10:20
p6eval nqp: OUTPUT«(signal SEGV)»
tadzik nqp: say($/)
p6eval nqp: OUTPUT«Null PMC access in get_string()␤current instr.: '' pc 791 (src/stage2/gen/NQPCORE.setting.pir:363) (src/stage2/NQPCORE.setting:321)␤»
jnthn felher: S11 notes that EXPORT is a lexical installed in UNIT. Thus exports working out is orthogonal to what module/class is actually declared inside the file. 10:23
It's just convention (though a good one) to match file name and path with having a module of that name declared inside the file.
felher jnthn: ah, a UNIT is a compilation unit. And files are compilation units, so they get their EXPORT automatically. 10:25
felher jnthn: thnx a lot :) 10:25
jnthn np :) 10:26
Ulti this is a good talk www.youtube.com/watch?v=tCel6lpDTZI 10:36
felher yeah, the rakudo speekup is really great :) 10:42
Ulti yeah its very noticable, even if I haven't done much p6 programming I've built each release of star for the last couple of years and its gotten a lot better 10:43
felher indeed :) 10:46
leont Last question of this interview is interesting for us too: www.codeguru.com/cpp/article.php/c1...ustrup.htm 10:50
jnthn Time to head to the airport. Back in ~20 hours, in a timezone far, far away. &
leont jnthn: YAPC::Asia? 10:51
doy leont: that seems about equally relevant to perl 5 and perl 6 10:52
(:
dalek rlito: 283f52b | (Flavio S. Glock)++ | / (2 files):
Perlito5 - js3 - fix container stringification
 11:21
rlito: 5ff149c | (Flavio S. Glock)++ | / (2 files):
Perlito5 - js3 - fix warn()
 11:32
dalek rlito: aa057ce | (Flavio S. Glock)++ | t5/01-perlito/19-local.t:
Perlito5 - more tests
 11:49
dalek rlito: 469a34f | (Flavio S. Glock)++ | / (4 files):
Perlito5 - js3 - fix local(), more tests
 12:20
dalek rlito: df13543 | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js3 - add @_ aliasing
 13:51
dalek rlito: 62131aa | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js3 - increment/decrement
 14:36
masak obviously, the NP-oracle looks at prime candidates and says "sorry kid, but you're not the one". 14:55
arnsholt =D 14:55
masak <moritz> öööööööööööööööööl 5~ 14:57
luckily, that means "oil" in German, not "beer" as in Swedish. 14:58
I'm all for bringing up Ronja in the Linux hacker tradition, but beer can probably wait a few more years... :P 14:59
dalek rlito: fdcaf07 | (Flavio S. Glock)++ | / (4 files):
Perlito5 - js3 - impl PUSH - this will be used by tied arrays someday
 16:18
diakopter TimToady: I let it run through base 22 - bases 19-22 aren't anywhere close to as big as 18 16:39
dalek rlito: 6b80f2a | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js3 - more tied methods
 16:42
diakopter bases 3-22: Execution time: 36531.603 s 16:45
so around 10 hours
9.5 of that being base 18 I'm sure 16:46
[Coke] blahs at the partcl nqp conversion effort.
[Coke] tries to figure out the QAST syntax for calling a sub and passing it arguments 17:12
if I want to see what PIR is generated with some QAST, is there a way to do that? 17:17
(in lieu of more docs on QAST)
[Coke] phenny: ask jnthn how one is supposed to invoke a named sub from QAST. QAST::Op.new( :op<call>, :name<invoke> ); complains that it cannot find a sub called "invoke" (There's one at program top level, and one in the actions class, since I don't know where it's looking) 17:26
phenny [Coke]: I'll pass that on when jnthn is around.
GlitchMr github.com/sorear/niecza/commits/master 17:36
this commit history starts to annoy me
Why commit made 5 days ago would be before commit made a day ago? 17:37
[Coke] I'll bite, why does that commit history annoy you?
... because you can commit things whenever you like locally, reorder them, and then push them.
GlitchMr Fix over-eager caching of method handlers, colomon++: sorear authored 11 hours ago
ThawBuffer knows its setting: sorear authored 5 days ago
[Coke] Sure.
GlitchMr Add BigInteger handling in ln: colomon authored a day ago
[Coke] the order the commits are in in history isn't necessarily the order commits are in when you committed them. 17:38
GlitchMr 5 days is shorter period than 1 day or what
[Coke] GlitchMr: stop typing and read what I wrote.
GlitchMr This is... confusing 17:39
[Coke] this is... git
dalek href="https://glitchmr.github.com/perl6-changes:">glitchmr.github.com/perl6-changes: a5ffc5d | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
Actually, it's fourth Thursday, not third.
 17:41
href="https://glitchmr.github.com/perl6-changes:">glitchmr.github.com/perl6-changes: d3761af | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
Upload Perl 6 changes for 2012W38
[Coke] you don't even have to reorder - I do a commit locally. I sit on it for a week. you do a commit and push it today, all at once. I am reminded to push my commit - I rebase, and push. your commit is in first, but mine is older.
dalek href="https://glitchmr.github.com:">glitchmr.github.com: d721657 | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
Placeholder Perl 6 changes for 2012W38 page

Currently, not much was mentioned on this page, but I've made this branch so you can read it and report any bugs with this list if you will find any (like spelling errors or factual errors).		 17:42
href="https://glitchmr.github.com:">glitchmr.github.com: 8179af9 | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
Add more Perl 6 changes to blog article.

Currently, the changes aren't too exciting compared to previous week, but we still have three days - who knows what will happen then... Besides, isn't Rakudo Star release awesome enough - I mean, it brings newest changes to useful and usable distribution of Perl 6 - you know, without installing Panda :-).
href="https://glitchmr.github.com:">glitchmr.github.com: a5ffc5d | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
Actually, it's fourth Thursday, not third.
href="https://glitchmr.github.com:">glitchmr.github.com: d3761af | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
Upload Perl 6 changes for 2012W38
href="https://glitchmr.github.com:">glitchmr.github.com: 9aca3c1 | GlitchMr++ | _posts/2012-09-23-perl-6-changes-2012W38.md:
Merge branch 'perl6-changes'
GlitchMr wow, I think merging is really spammy
moritz backlogs the summary 17:56
sorear good * #perl6
moritz \o sorear
sorear [Coke]: and that's exactly what happened, I forgot to push my second-to-last commit until after colomon++ started doing the only functional changes of the release cycle 17:57
sorear o/ moritz 17:58
colomon sorear: release is tomorrow? 18:05
phenny colomon: 06:46Z <sorear> tell colomon the bug you found while trying to enhance "cmp" is fixed now
colomon afk # but still interested in the answer... 18:09
pmurias [Coke]: re seeing PIR for QAST, I wrote a Perl6 -> QAST dumper which can show you how a bit of Perl 6 code is represented as QAST 19:37
for getting PIR from QAST I would try using QAST::Compiler to get PIRT 19:40
and then maybe the .pir method to get pir from that
[Coke]: also take a look at nqp/t/qast 19:41
moritz pmurias: where's that dumper? 19:42
pmurias moritz: github.com/pmurias/rakudo-js/blob/...dumper.nqp 19:45
it's a bit rough
moritz pmurias: thanks 19:47
GlitchMr perl6: for 1 .. 10 { .say; next + next + next + next } 20:01
p6eval rakudo 7d6aa0, niecza v21-23-gd8852de: OUTPUT«1␤2␤3␤4␤5␤6␤7␤8␤9␤10␤»
colomon rn: say 1 cmp 5 20:12
p6eval rakudo 7d6aa0, niecza v21-23-gd8852de: OUTPUT«Increase␤»
colomon rn: say { foo => 1, bar => 2 } cmp { bar => 2, foo => 5 }
p6eval niecza v21-23-gd8852de: OUTPUT«Same␤»
..rakudo 7d6aa0: OUTPUT«Decrease␤»
colomon So... I've reworked niecza's infix:<cmp> to masak++'s suggestion for ordering hashes, and if I'm not mistaken, rakudo is doing it wrong. (assuming, of course, masak's method is correct.) 20:13
colomon (that is, go through the keys in order and compare their values.) 20:15
sorear colomon: hi. release is tomorrow 20:25
colomon sorear: okay. will try to have more useful stuff checked in by then.
sorear ++colomon 20:39
so I have plans for the next month that involve some moderately serious (ab)use of the CLR assembly naming/resolution system and I'm concerned I might hit a Microsoft/Mono incompatibility 20:46
colomon is ablaze with curiosity.... 20:48
[Coke] pmurias: that test file does have lots of great examples of QAST, but not the nqp they would correspond to, which makes it slightly less useful. I'll see if I can tease out what I need from it, though. 20:51
felher 'night, #perl6 :) 21:13
leont libuv is a bit disappointing 21:40
They don't seem to know the rule #1 of callbacks in C: always include a void pointer for extra state 21:41
sorear how does node work around that? 21:44
leont Ah, I think I found it, there's a public data field in all request objects (that do get passed) 21:52
masak colomon: might well be that rakudo is wrong. 21:54
colomon: I'm still mulling over whether't's a good idea to compare hashes that way.
I'd be much more hesitant to do it with sets, for example. but hashes are not that different. 21:55
sorear o/ masak 21:56
masak \o
<GlitchMr> Why commit made 5 days ago would be before commit made a day ago? 21:59
because... that's what "before" means...?
but yeah, what [Coke] said. if you want your commits ordered *temporally*, you can do that. but that's not the interesting order. 22:00
'night, #perl6 22:01
leont Also, libuv's habit of not declaring variables in examples is profoundly annoying 22:10
MikeFair o/ whiteknight 23:11
n: 2.ln()
p6eval niecza v21-23-gd8852de: ( no output )
MikeFair n: say 2.ln()
p6eval niecza v21-23-gd8852de: OUTPUT«0.69314718055994529␤»
MikeFair sorear: I'm not seeing the problem there... google's answer is: 0.69314718056 23:12

Powered by Cro and the Raku® Programming Language.

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

Thank you!