jnthn .tell pmurias Ah, makes sense (the local register thing, and why a temporary was needed). The leak fix looks good. 12:10
yoleaux2 jnthn: I'll pass your message to pmurias.
jnthn nine: Yeah, updating the website was something I didn't feel like spending my weekend doing. Will get to it soonish. :)
TheLemonMan m: sub prefix:<++> (int $x is rw) { $x }; my int $x = 4; say ++$x; 13:06
camelia rakudo-moar 668dc5: OUTPUT«5␤»
TheLemonMan but if I replace 'int' with 'Int' it works 13:07
[Coke]_ m: multi sub prefix:<++> (int $x is rw) { $x }; my int $x = 4; say ++$x; 13:11
camelia rakudo-moar 668dc5: OUTPUT«5␤»
[Coke] my response is "doctor it hurts when I do this", but perhaps someone else can fix it. 13:12
why are you trying to override the builtin prefix ++
?
TheLemonMan need to check some codegen detail 13:14
[Coke] ... of what happens when you override a builtin? 13:15
*try to*
looks like both int and Int prefix ++ are defined as multi subs, btw; I thought the difference might have to do with a difference in sigs there. 13:17
TheLemonMan [Coke], I'm doing some more testing wrt NQP's #304 13:19
[Coke] hokay. 13:24
how do we edit docs on rakudo.org? 13:46
are there html files somewhere? a wordpress install?
jnthn [Coke]: Think it's wordpress and pmichaud is the one to ask for access, iirc 13:57
TheLemonMan hmm, the roast scripts keep saying 'Looks like you planned XX tests, but ran YY', how am I supposed to know what's wrong ? 14:21
[Coke] count the tests. 14:24
how are you running the tests?
is this with rakudo? 14:25
TheLemonMan [Coke], yes, apparently it stops at a block that's marked with 'skip' 14:27
[Coke] what command line are you using to run the test(s)
TheLemonMan perl6 <path-to-the-test> (using the harness6 has the same effect) 14:28
[Coke] in general, you can't do that.
perlpilot [Coke]: I can give you an account on rakudo.org if you don't already have one. 14:29
[Coke] you need either 'make t/spec/path/to/test.t' or 'prove -v -e t/fudgeandrun t/spec/path/to/test.t'
perlpilot: I have no idea if I do. it's not under my normal email addy if it's there.
perlpilot looks
[Coke] the tests are meant to be fudged first (see t/spec/fudge) before being run.
so, can you give me an -exact- command line you're using? 14:30
are you using a prebuilt perl6? the one in a build directory?
if prebuilt, you're going to have to jump through more hoops
perlpilot [Coke]: nope. What email address do you want to use?
[Coke] spec tests are designed to be run out of a build directory. 14:31
perlpilot: thanks. Are wordpress logins per-site ? 14:33
or is this going to somehow impact other wordpress logins I have?
perlpilot I'm not really sure. I've seen it both ways. 14:34
TheLemonMan [Coke], that was it, I forgot to preprocess the tests 14:36
[Coke] TheLemonMan: excelelnt, gald it's working
*glad
[Coke] wishes to come work for the perl 6 collective. blah. 15:33
jdv79 having a p6 related job would be pretty sweet 15:54
lizmat waves from Cluj 16:16
jnthn lizmat: Hope you had a nice journey :) 16:17
lizmat yeah, well, the first 100km into Rumania wasn't too nice to look at...
the rest was a lot better :-) 16:18
there being no highways in this area, makes you see things all the better :-)
jnthn :-) 16:19
lizmat m: my $b = <a b c d>.Bag; my $m = <a b c d>.Mix; say $b.ACCEPTS($m) # I wonder if this is correct ? 16:35
camelia rakudo-moar 668dc5: OUTPUT«True␤»
jnthn I don't immediately see why it'd be a problem? 16:36
It is smartmatch... :)
lizmat ok
:-)
jnthn If === came out the same we'd be upset :) 16:37
But afaik a mix is just a bag with Num weights, and 1 ~~ 1e1 comes out alright
dalek kudo/nom: 933e9a6 | lizmat++ | src/core/Baggy.pm:
Make Baggy.ACCEPTS(Baggy) about 25x faster

For a 4 element Bag, that is. Larger bags should be even faster still.
  - don't do (<+) twice
  - check number of keys first
  - run it off the .WHICH keys, so we don't need to call .WHICH
  - rewrite in nqp ops		 16:38
lizmat m: my $b = <a b c d>.Bag; my $m = <a b c d>.Mix; say $b === $m
camelia rakudo-moar 668dc5: OUTPUT«False␤»
moritz do we have a style guide for code inside the rakudo repo anywhere? 18:01
timotimo "use the style that's fastest" :P 18:02
moritz we do seem to use uncuddled else (newline between } and else) pretty consistently
and four-space indents
[Coke] match the style in the section you're in, I think, is the closest we have. 18:07
we spent a LOT of time in parrot trying to coerce code to our coding style, time I think was ultimately wasted. I wouldn't worry too much about it. 18:08
moritz ok 18:09
I merely ask because the core infrastructure best practise thingy suggests we document the contribution process, including coding style
I guess "match what you see" is good enough for us
japhb TimToady had a list of personal preferences for Perl 5 code (which I default my editor to use), but I don't recall him making one for Perl 6 code. And it's likely that Rakudo's source would violate some of that anyway for numerous reasons (history, bootstrapping, performance-critical hacks, etc.) 18:14
[Coke] I would love if it said "no tabs, 4 spaces" in there somewhere. :) 18:22
timotimo no stabs, 4 paces 18:24
jnthn Uncudled else, 4-space indents, stick with _ over - in names of any code written in NQP (grammar, actions, MOP), be consistent with what you see existing in the code that you're editing. 18:29
timotimo how bad is it that i keep cuddling elses? 18:38
i can make an effort if people want
jnthn I don't feel strongly on it, though I do find it makes the code easier to scan 18:39
(I didn't used to do it myself until I saw it done commonly here and found it a little easier on the eye.)
timotimo ill give it a think
jnthn Short of tabs vs spaces though, I'm generally not inclined to reject code contributions on little stylistic things. :) 18:40
dalek kudo/nom: 560170f | LemonBoy++ | src/Perl6/Actions.nqp:
Reject signatures with both slurpy parameters and default values.
 18:44
kudo/nom: 3af93c4 | jnthn++ | src/Perl6/Actions.nqp:
Merge pull request #844 from LemonBoy/defv

Reject signatures with both slurpy parameters and default values.
ast: 5b544b2 | LemonBoy++ | S06-signature/defaults.t:
Tests for parameter default values.
 18:45
ast: 7f6e44f | jnthn++ | S06-signature/defaults.t:
Merge pull request #144 from LemonBoy/defaults

Reject default values for slurpy parameters.
TheLemonMan \m/
timotimo cool 18:47
dalek ast: f5f4ece | LemonBoy++ | S02-types/native.t:
Unfudge some passing tests.

Also, silence a deprecation warning by using is-approx instead of is_approx.		 18:54
ast: 1b9ba1a | jnthn++ | S02-types/native.t:
Merge pull request #143 from LemonBoy/unfudge-rt127813

Unfudge some passing tests.
jnthn wonders if he'll get this blog post finished in time for it to sneak into the weekly :) 22:19
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2016/08/22/...from-cluj/ 22:21
nope :-)
jnthn: unless it's done now ?
jnthn It is, I'm already pasting it into wordpress :) 22:22
timotimo ooooh
lizmat ok, I'll add it then!
timotimo we have Dofgooding! 22:23
jnthn 6guts.wordpress.com/2016/08/22/con...ng-part-1/ 22:24
timotimo yay blog 22:28
jnthn lizmat: Thanks for letting it sneak in :) 22:31
lizmat jnthn: you're welcome 22:32
jnthn Also, lizmat++ for the weekly :)
lizmat it does mean you're at the end of this week's list, rather than at the top of next week's list
jnthn I hope to get another post in that series done later this week anyways :) 22:34
lizmat jnthn++ :-) 22:37
time to get some sleep
so good night, #perl6!
jnthn Me also 22:38
'night o/
jdv79 thanks for the post jnthn 23:44

Powered by Cro and the Raku® Programming Language.

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

Thank you!