🦋 Welcome to Raku! raku.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: colabti.org/irclogger/irclogger_log/raku
Set by ChanServ on 14 October 2019.
00:02 Kaiepi left 00:03 Kaiepi joined 00:05 wildtrees left 00:16 warriors joined
squashable6 🍕 vrurg++ wrote a comment on “Add support for .rakumod extension”: github.com/rakudo/rakudo/pull/3246...-544024841 00:35
00:40 go|dfish left 00:43 lelf left 01:00 lucasb left
squashable6 🍕 tony-o++ wrote a comment on “Add support for .rakumod extension”: github.com/rakudo/rakudo/pull/3246...-544033699 01:06
🍕🍕🍕 First contribution by tony-o++! ♥
AlexDaniel “The password you entered is too long.” 01:36
yeah… right :'(
it was just 64 symbols 01:37
ah, 25 max, nice!
02:02 benjif left, benjif joined 02:09 rindolf3 joined
rindolf3 hi all, sup? 02:09
i've been up all night 02:10
chatting and scheming
guifa Ugh… I can’t seem to get an indirect reference to work to avoid cyclic compile problems =\ Looks like I’m going to just have to combine everything into one big mega class file =\ 02:29
rindolf3 guifa: ah :( 02:37
02:58 benjif left
squashable6 🍕 vrurg++ opened pull request “Add --force-rebuild command line option”: github.com/rakudo/rakudo/pull/3248 03:09
🍕 vrurg++ edited pull request “Add --force-rebuild command line option”: github.com/rakudo/rakudo/pull/3248 03:10
03:10 benjif joined
squashable6 🍕 vrurg++ requested a review on pull request “Add --force-rebuild command line option”: github.com/rakudo/rakudo/pull/3248 03:11
03:53 marcusr left 03:54 marcusr joined 04:13 cinch_ joined 04:14 cinch left
discord6 <Aearnus> there's a hackathon happening at uni this weekend, I've been racking my brain for Raku ideas haha 04:21
05:16 leonardus joined
leonardus hi 05:16
rindolf3 leonardus: meow 05:27
05:30 warriors left 05:31 go|dfish joined 05:34 delimited joined 05:41 go|dfish left 05:48 delimited left, delimited joined 05:56 delimited left 05:57 delimited joined 06:33 sena_kun joined
squashable6 🍕 Altai-man++ opened pull request “Test fake-infix adverbs on a construct”: github.com/perl6/roast/pull/588 06:47
🍕 niner++ wrote a comment on “Add support for .rakumod extension”: github.com/rakudo/rakudo/pull/3246...-544108210 06:48
06:55 benjif left 06:56 benjif joined 07:05 benjif left
squashable6 🍕 Altai-man++ opened pull request “Test return constraints for role implementors”: github.com/rakudo/rakudo/pull/3249 07:11
Geth doc/master: 4 commits pushed by Altai-man++, (Juan Julián Merelo Guervós)++ 07:27
07:35 sena_kun left 07:37 pmurias joined
squashable6 🍕 jonathanstowe++ wrote a comment on “Add support for .rakumod extension”: github.com/rakudo/rakudo/pull/3246...-544112654 07:58
🍕🍕🍕 First contribution by jonathanstowe++! ♥
08:05 go|dfish joined 08:16 chloekek joined 08:42 pmurias left 08:51 pat_js joined
pat_js hi I was wondering why √ is not defined in my perl6 08:52
p6: say √9 08:53
camelia 5===SORRY!5===
Argument to "say" seems to be malformed
at <tmp>:1
------> 3say7⏏5 √9
Bogus postfix
at <tmp>:1
------> 3say 7⏏5√9
expecting any of:
infix
infix stopper
postfix
08:56 cpan-raku left 08:57 cpan-raku joined, cpan-raku left, cpan-raku joined 08:58 go|dfish left
AlexDaniel squashable6: status 09:02
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in 1 day and ≈2 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/36b06aa519b04b2b25...2abd46c1a0
AlexDaniel pat_js: github.com/rakudo/rakudo/wiki/save...ASCII#---2 09:03
pat_js AlexDaniel: Thank you, I did not know that wiki page (tbh. I haven't ever been here, was lurking a lot in the old #perl6 quite a while) 09:05
AlexDaniel pat_js: yeah, I try to keep all unicode ideas in the same place :) 09:06
also, on top of the page there's some criteria for adding new ops 09:07
√ satisfies ① but nothing else
arguably most mathy ops do only that 09:08
greppable6: sqrt
greppable6 AlexDaniel, 353 lines, 54 modules: gist.github.com/a0ff4f141cdc746f96...4e7fb62e14
AlexDaniel now why is the output broken… 09:12
09:14 pmurias joined
AlexDaniel ok, it's because of this: gist.github.com/Whateverable/a0ff4...-L151-L152 09:19
now how did that happen…
ah! Found it 09:31
09:42 pmurias left
chloekek I used √ as an infix operator in Haskell once. 09:55
p6: sub infix:<√>($x, $y) { $y ** (1 / $x) }; say 2 √ 9; say 3 √ 125; 09:56
camelia 3
4.999999999999999
chloekek But then again I've also used bookmαrk instead of bookmark as an identifier to avoid a shadowing warning. 09:57
Geth whateverable: 9ac71dca89 | (Aleks-Daniel Jakimenko-Aleksejev)++ | xbin/Greppable.p6
Make greppable not choke on \r\n

  colabti.org/irclogger/irclogger_lo...-10-19#l50
10:11
10:12 greppable6 left 10:14 greppable6 joined
AlexDaniel greppable6: sqrt 10:14
greppable6 AlexDaniel, 359 lines, 54 modules: gist.github.com/1d8f93deb8bc5af938...ec03074e67
AlexDaniel yaay
10:24 sena_kun joined
cpan-raku New module released to CPAN! Inline::Perl5 (0.42) by 03NINE 10:34
11:07 lelf joined
lelf p6: with [[1],[2,3]] {say gather {.permutations».take for $_}; say gather {(.take for .permutations) for $_}} 11:09
camelia (1 2 3 3 2)
((1) (2 3) (3 2))
lelf ^ why results are different?
sena_kun m: [1,2].permutations>>.say; .say for [1,2].permutations; 11:13
camelia 1
2
2
1
(1 2)
(2 1)
sena_kun a golf, though I don't have an idea why they are different 11:14
m: say ((1, 2), (2, 1))>>.say; 11:16
camelia 1
2
2
1
((True True) (True True))
sena_kun m: say ((1, 2), (2, 1))>>.say; .say for ((1, 2), (2, 1));
camelia 1
2
2
1
((True True) (True True))
(1 2)
(2 1)
11:26 chloekek left
fingolfin3 Hello! Would you like to enable math signs support ∧ and ∨ with ASCII equivalents && and || ? 11:38
11:44 chloekek joined
sena_kun squashable6, status 11:48
squashable6 sena_kun, 🍕🍕 SQUASHathon is in progress! The end of the event in 1 day and ≈0 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
sena_kun, Log and stats: gist.github.com/526f6f0eeb7da55960...60317fb9be
12:01 chloekek left 12:02 CIAvash joined
fingolfin3 sena_kun I'm still a beginner and not particularly good at development. Just noticed a missing feature and suggested to adding it :) 12:03
sena_kun fingolfin3, I am not a core developer either... But I would rather fix real bugs than adding more unicode. Or write more batteries. Or write docs. Or tutorials or blog posts. Too many things that are orders of magnitude more helpful than adding more unicode. But that's just my opinion. 12:06
CIAvash lelf: I think it's because of nodality, ». is not the same as for or map, see docs.perl6.org/language/traps#Usin...changeably and docs.perl6.org/language/operators#...op_%3E%3E.
fingolfin3 sena_kun Probably, the creators of the language do not agree with you, since they have already added a lot of unicode. 12:12
sena_kun fingolfin3, probably they don't. as I said, it is just my personal opinion. 12:14
12:15 warriors joined 12:17 fingolfin3 left 12:24 lucasb joined 12:26 go|dfish joined 12:40 pmurias joined 12:48 pmurias left
squashable6 🍕 Altai-man++ edited pull request “Test return constraints for role implementors”: github.com/rakudo/rakudo/pull/3249 12:50
12:50 pmurias joined 13:01 go|dfish left
AlexDaniel fingolfin3: re ∧ and ∨: thanks for the interest! It's a lot of work, more than it seems at first glance (e.g. they need to be versioned, so that you can only use them on newer language versions) 13:23
tellable6 AlexDaniel, I'll pass your message to fingolfin3
AlexDaniel fingolfin3: we had some not nice experience with other unicode stuff (e.g. unicode minus) which was a real pain to support properly and maintain, so it's unlikely we'll be adding more ops any time soon 13:24
tellable6 AlexDaniel, I'll pass your message to fingolfin3
AlexDaniel fingolfin3: maybe in the future, but right now we have some other things we should focus on :)
tellable6 AlexDaniel, I'll pass your message to fingolfin3
AlexDaniel sena_kun: that's not just your opinion, you're actually right :) 13:26
I'm one of the users who uses primarily unicode ops, and I can't see myself using ∧ and ∨ 13:27
so who's going to use them, really?
and then we'll find out that performance of ∧ and ∨ is different from performance of && ||, and someone will spend hours trying to fix that and test it 13:28
warriors didnt unicode ops contribute to the demise of apl 13:32
not really sure why rakudoers are such a fan of them
apl reincarnation J (which is a lot of fun) consciously decided to be all ascii and drop the weird symbols 13:34
AlexDaniel warriors: it depends on which operators we're talking about 13:35
warriors: ‘’ “” and such are just better
(because you can nest things without extra toothpicks)
visually they're perfectly readable, and you don't have to use them if you don't want to 13:36
warriors yes, but they are not on my keyboard, i need to use special emacs mode to convert regular ascii sequences to those
which is .. in my opinion a needless layer on mental effort that we dont need 13:37
AlexDaniel warriors: you can enable compose key and you'll be able to type a bunch of stuff anywhere in your system 13:38
warriors in the case of quotes you are right, quotes will never become api
but any ops, who risk becoming api, is trouble
AlexDaniel you don't have to, but I'm much happier being able to type special symbols, emoji and other stuff directly :)
warriors: ok, what about say 2³² 13:39
warriors 2^32
AlexDaniel m: say 2^32
camelia one(2, 32)
AlexDaniel m: say 2**32
camelia 4294967296
AlexDaniel you used the wrong op :P
unicode version is less ambiguous 13:40
warriors yea sorry
i am used to ^ being the one
AlexDaniel ∞ being Inf is also pretty obvious
warriors it is easier to read, but harder to type
AlexDaniel 3 key presses with compose key, 4 key presses without 13:41
13:41 finanalyst_ joined
lelf is there a shorted way to write (@x».pred) aka (@x X- 1) ? 13:41
AlexDaniel compose 8 8 vs Shift + I n f
warriors and if rakukings and rakuqueens insist on them, you should also spend a lot more time educating people how to configure their systems to type them
AlexDaniel which is not horrible, default input methods really suck and IMO people should improve the defaults 13:42
lelf: are you golfing? :)
m: my @a = < 1 2 3 4 >; @a»--; say @a
camelia [0 1 2 3]
AlexDaniel lelf: I don't think it'll get any shorter than that 13:43
warriors: so yeah, we do that too: docs.perl6.org/language/unicode_entry
AlexDaniel shower & 13:44
lelf AlexDaniel: golfing, but only a little :) thanks! 13:45
warriors didnt know about this page, should be promoted more :) 13:46
thanks for sharing
AlexDaniel lelf: if you're not simply making your code smaller but actually golfing, then here's a page with some hints: github.com/AlexDaniel/raku-golf-cheatsheet 13:51
13:52 rindolf3 left
cpan-raku New module released to CPAN! ORM::ActiveRecord (0.0.8) by 03GDONALD 13:56
14:08 finanalyst_ left 14:11 Tirifto joined
cpan-raku New module released to CPAN! Failable (0.1.0) by 03KAIEPI 14:27
AlexDaniel sena_kun: did you register for hacktoberfest too? :) 15:00
sena_kun: you have 4 pull requests, that's exactly a t-shirt I think
squashable6: status 15:01
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈20 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/ba6b10a075623dd4f2...3783cb8b70
sena_kun AlexDaniel, I did not, though sent some patches to roast. I am living in a middle of nowhere and it is always a hell when dealing with anything related to mailing. :|
AlexDaniel sena_kun: hacktoberfest.digitalocean.com/ click the button, maybe it'll even count them 15:02
sena_kun lemme see...
>You submitted Update schedule to perl6/advent <- this is just cheating. :P 15:04
AlexDaniel hehe 15:05
last year I didn't do enough pull requests, it was weird
to be fair I wasn't doing any specifically for hacktoberfest :P
sena_kun AlexDaniel, to be honest, I'd rather be happier to see the PRs merged (after a review, ofc) than getting yet another t-shirt. ;) 15:06
AlexDaniel sena_kun: will you do more if I review and merge them? :)
sena_kun AlexDaniel, I planned to do more anyway, after today's $dayjob, but sure. :) 15:07
vrurg AlexDaniel: if it's about the roast PRs I plan to work on it tomorrow or Monday.
AlexDaniel squashable6: status 15:08
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈20 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/d5427c1eefa0df50c2...b4c9417702
AlexDaniel vrurg: ↑ it's better if squashathon stuff is accepted while the squashathon is going :)
vrurg Too bad. I'm not having much time as of yesterday. :( 15:09
Should be less loaded tomorrow.
AlexDaniel I'll review, these are not that hard 15:10
squashable6 🍕 AlexDaniel++ wrote a comment on “Test Parameter subclassing”: github.com/perl6/roast/pull/585#is...-544157632
🍕 AlexDaniel++ closed issue “Parameter subclasses don't initialize attributes”: github.com/rakudo/rakudo/issues/2829
🍕 AlexDaniel++ merged pull request “Test Parameter subclassing”: github.com/perl6/roast/pull/585
vrurg AlexDaniel: thanks! 15:11
vrurg is afk again.... 15:12
AlexDaniel 6c: say 1 «*« |cis(1).reals x 2
committable6 AlexDaniel, gist.github.com/1a1986987733e8cd6f...04331828df
AlexDaniel c: 2018.10-119-g18ded174e say 1 «*« |cis(1).reals x 2 15:13
committable6 AlexDaniel, ¦2018.10-119-g18ded174e: «Cannot test this commit (Commit exists, but an executable could not be built for it)»
AlexDaniel c: 18ded174e say 1 «*« |cis(1).reals x 2
committable6 AlexDaniel, ¦18ded17: «Cannot test this commit (Commit exists, but an executable could not be built for it)»
AlexDaniel c: 18ded174e^^^ say 1 «*« |cis(1).reals x 2
committable6 AlexDaniel, ¦18ded174e^^^: «Cannot modify an immutable Num (0.5403023058681398)␤ in block <unit> at /tmp/4a95_MefBY line 1␤␤ «exit code = 1»»
AlexDaniel I see
so this one was actually a regression
squashable6 🍕 AlexDaniel++ wrote a comment on “Slip and hyper operation…”: github.com/rakudo/rakudo/issues/24...-544158224 15:14
AlexDaniel c: 18ded174e^^^ use Test; lives-ok { |<1 2> >>xx>> 2 } 15:15
committable6 AlexDaniel, ¦18ded174e^^^: «not ok 1 - ␤# Failed test at /tmp/IJ2lSBLt_z line 1␤# Cannot modify an immutable IntStr (1)␤ «exit code = 1»»
AlexDaniel c: 18ded174e^^^,HEAD use Test; lives-ok { |<1 2> >>xx>> 2 }
committable6 AlexDaniel, ¦18ded174e^^^: «not ok 1 - ␤# Failed test at /tmp/V_Q1StvvXn line 1␤# Cannot modify an immutable IntStr (1)␤ «exit code = 1»» ¦HEAD(e45bb34): «ok 1 - ␤»
AlexDaniel sena_kun: shouldn't we test the values too?
like, what if it lives but gives garbage output? 15:16
squashable6 🍕 AlexDaniel++ wrote a comment on “Test hyperop results in a slip”: github.com/perl6/roast/pull/587#is...-544158581 15:17
AlexDaniel c: 2015.12,HEAD use Test; my Pair $p; is-deeply ($p .= new :key<foo> :value<bar>), :foo<bar>.Pair, ‘foo’ 15:18
committable6 AlexDaniel, ¦2015.12: «===SORRY!===␤Unknown QAST node type NQPMu␤ «exit code = 1»» ¦HEAD(e45bb34): «ok 1 - foo␤»
AlexDaniel so this one doesn't even parse
sena_kun AlexDaniel, I have an impression that results are already tested in numerous places, and we paid attention to "they are not immutable" part. 15:19
squashable6 🍕 AlexDaniel++ wrote a comment on “Test fake-infix adverbs on a construct”: github.com/perl6/roast/pull/588#is...-544159255 15:21
sena_kun in a similar manner one can argue that we need to test it with e.g. all possible numbers, because otherwise we don't guarantee that e.g. 100_000 after 99_999 successful numbers works.
OTOH I don't feel too opinionated on this and can add values as well 15:22
AlexDaniel I'd feel better if at least some values were tested too
sena_kun nods 15:23
squashable6 🍕 Altai-man++ wrote a comment on “Test fake-infix adverbs on a construct”: github.com/perl6/roast/pull/588#is...-544159673 15:24
AlexDaniel meh, not sure if metamodel stuff should be in roast, but we already have tests for it 15:26
so, fine!
pretty cool code too
squashable6 🍕 AlexDaniel++ wrote a comment on “Test runtime-created classes being attributes”: github.com/perl6/roast/pull/586#is...-544160140 15:28
🍕 AlexDaniel++ merged pull request “Test runtime-created classes being attributes”: github.com/perl6/roast/pull/586
🍕 AlexDaniel++ closed issue “Smartmatch and type chec…”: github.com/rakudo/rakudo/issues/2607
sena_kun design.perl6.org/S12.html#Custom_Meta-objects mentions MOP-related things
>These are for the standard Perl 6 Meta-Object Protocol, but other MOPs may define other traits
so MOP belongs to roast, imho
squashable6 🍕 AlexDaniel++ wrote a comment on “Test fake-infix adverbs on a construct”: github.com/perl6/roast/pull/588#is...-544160240 15:29
🍕 AlexDaniel++ merged pull request “Test fake-infix adverbs on a construct”: github.com/perl6/roast/pull/588
🍕 AlexDaniel++ closed issue “Unknown QAST node type N…”: github.com/rakudo/rakudo/issues/1500
AlexDaniel oh, design subdomain is up!
rba++ # I guess you did it? :)
15:31 sena_kun left
AlexDaniel sena_kun: well, I don't think we thought through the idea of different compilers, partial compliance and stuff like this 15:31
tellable6 AlexDaniel, I'll pass your message to sena_kun
AlexDaniel sena_kun: with my limited knowledge, I can totally imagine a compiler that implements almost everything but MOP, perhaps for some good reasons too. But arguably there are many features that can potentially be “skipped” but various implementations, and we don't yet know how this will work out 15:32
tellable6 AlexDaniel, I'll pass your message to sena_kun
rba AlexDaniel: jw 15:33
AlexDaniel sena_kun: fudging is ok but very LTA in my opinion
tellable6 AlexDaniel, I'll pass your message to sena_kun
15:36 sena_kun joined 15:42 haxmeister joined
haxmeister :P 15:43
Tirifto Hello, moving from #perl6! 15:57
AlexDaniel o/
haxmeister For other options, go to rakudo.org/how-to-get-rakudo/#Insta...Star-Linux 15:58
that's a dead link
on perl6intro
AlexDaniel ouch! 16:03
Tirifto So apparently ‘⇒’ doesn't go for ‘=>’ in Raku? :P 16:05
AlexDaniel Tirifto: yeah, currently not! 16:06
haxmeister that would be cool though
AlexDaniel Tirifto: same for -> and →, <-> and ↔
lizmat github.com/rakudo/rakudo/wiki/save...from-ASCII
guifa sub infix:<⇒> ($a,$b) { Pair.new: $a,$b } # Now it does 16:07
AlexDaniel Tirifto: interestingly, we were just discussing some unicode ops before you joined :) See colabti.org/irclogger/irclogger_lo...-10-19#l67 16:08
Tirifto lizmat: Wow, that's awesome there's been discussion on so many! :D 16:12
lizmat Tirifto: and it's a Wiki, so if you have any additional ideas, please don't hesitate to name them :-) 16:13
haxmeister trying to make good decisions here, installing raku on opensuse
lizmat Tirifto: no guarantee that they will ever be implemented, though,
afk&
AlexDaniel haxmeister: what's the latest version there? 16:15
haxmeister: seems like we do have some opensuse packages here: github.com/nxadm/rakudo-pkg/releases
haxmeister 2019.07-2.1 16:16
yes rakudo installs
but trying to understand getting zef
guifa DONE. *finally* got LanguageTags to support the T-extension semantics which required a massive reorganization of code because cyclic =\
squashable6 🍕 llelf++ wrote a comment on “Implement ¿? and ¡! as alternative to ""”: github.com/rakudo/rakudo/pull/3218...-544167086 16:17
🍕🍕🍕 First contribution by llelf++! ♥
🍕 niner++ wrote a comment on “Windows NativeCall regression ”: github.com/rakudo/rakudo/issues/32...-544167113
🍕 niner++ closed issue “Windows NativeCall regression ”: github.com/rakudo/rakudo/issues/3247
haxmeister because the zef instructions don't include installing globally nor is there information about running zef to manage raku for the system
guifa but now you can do $tag.extensions<t>.hybrid-locale and get back a fully fledged language tag :D
squashable6 🍕 AlexDaniel++ wrote a comment on “Implement ¿? and ¡! as alternative to ""”: github.com/rakudo/rakudo/pull/3218...-544167383 16:18
🍕 AlexDaniel++ wrote a comment on “Add tests for R#3035”: github.com/perl6/roast/pull/584#is...-544167462 16:19
haxmeister squashable6: a bot or person? 16:21
squashable6 haxmeister, ALRIGHT, LET'S DO IT!!!
haxmeister I C
/ignore squashable6
AlexDaniel haxmeister: it's squashathon time, which is why it's so loud :) 16:22
haxmeister: we're writing tests and closing testneeded tickets this time
16:22 cpan-raku left
haxmeister AlexDaniel: it is actually good to see so much work happening.. not mad here 16:23
16:23 cpan-raku joined, cpan-raku left, cpan-raku joined
AlexDaniel haxmeister: some distros have zef packaged: repology.org/project/perl6-zef/versions 16:23
haxmeister AlexDaniel: I have like 10 channels moving in irssi.. so the bot sends me back to check responses
AlexDaniel haxmeister: but not opensuse, unfortunately!
haxmeister AlexDaniel: does zef install packages in a way that allows it to do system wide installs on multi-user systems? 16:25
AlexDaniel haxmeister: not by default, and I haven't tried making it do that 16:28
haxmeister ok
AlexDaniel nine: nice! Thank you
Tirifto AlexDaniel, lizmat: Thank you, I'll see about adding new ideas if I can think of them! I definitely like ⇒ & the other arrows, ∧ & the other logic symbols, subscripts and the bag delimiters which I'm not familiar with, but they look awesome. I hope they'll make it in along with the other improvements someday! :-)
AlexDaniel Tirifto: it's also possible to implement most of them in a module 16:29
Tirifto: it's just that nobody did that yet 16:30
a lot of these suggestions don't really have to be in core, I think
greppable6: ≠|≤|≥
greppable6 AlexDaniel, 368 lines, 29 modules: gist.github.com/3f54ffdf48140ddfcc...344f8014cb 16:31
AlexDaniel let's see if that works…
almost 16:32
here's a rendered version: gist.github.com/AlexDaniel/1b8d02e...f1eb9f9b3e
guifa had moderate success making ¿ ? a circumfix for truthiness testing and ¡ ! for negating, probably woudl need to tinker deeper in the P6 grammar to get it to work well
Geth whateverable: 47013e95a5 | (Aleks-Daniel Jakimenko-Aleksejev)++ | xbin/Greppable.p6
Skip .pdf files

Otherwise we get a lot of binary data in the gist which breaks rendering.
16:37
16:38 greppable6 left
nine releasable6: status 16:38
releasable6 nine, Next release in ≈2 hours. 6 blockers. Unknown changelog format 16:39
nine, Details: gist.github.com/df4b0784176b07a550...eb6865d679
Tirifto AlexDaniel: ¡False!, but they might make the core nicer! :-) 16:40
16:40 greppable6 joined
AlexDaniel greppable6: ≠|≤|≥ 16:41
this should work now
greppable6 AlexDaniel, 368 lines, 29 modules: gist.github.com/69c89dfd35a61098ea...2c80ab1f0a 16:42
AlexDaniel nine: yeah, we'll get there, not in 2 hours but we'll get there :)
Tirifto AlexDaniel: What exactly is greppable6 telling us here? 16:49
guifa Is there a best practice for renaming a module? Not ideal, but its best for long term before usage increases too much? Make it a new module and deprecate the old one or just upload all the new stuff to github and stub the old name for a little while with a deprecation notice or?
squashable6 🍕 vrurg++ wrote a comment on “Add tests for R#3035”: github.com/perl6/roast/pull/584#is...-544171306 16:50
AlexDaniel Tirifto: in this case it's all uses of ≠≤≥ ops in all modules 16:51
Tirifto Oh, nice!
AlexDaniel guifa: I don't think there's a guide or anything like that, and I haven't heard of any best practices for this. But, one think to keep in mind is that modules are installable by whatever you have in your “provides” section 16:54
guifa: so a module Foo::Bar that provides Baz::Baz should be installable with `zef install Baz::Baz` 16:55
maybe you can use that to provide some sort of backward compatibility, but maybe not and it'll confuse everything :)
guifa ooh, didn’t realize that
That might not be bad. I don’t think *that* many people are using language tags
squashable6 🍕 llelf++ wrote a comment on “Implement ¿? and ¡! as alternative to ""”: github.com/rakudo/rakudo/pull/3218...-544172196 16:56
guifa I originally named the module Intl::BCP47 after the document that defined them, but I’ve implemented a few other related standards so the name is kinda misleading. Besides it makes more sense to just call the module Intl::LanguageTag with all the rest of the stuff I’ve been doing in Intl:: . 16:57
The exports are all the same though, so I think I can just provide Intl::BCP47 for the time being and it shouldn’t break any code 16:58
17:01 pmurias left 17:13 sena_kun left 17:14 sena_kun joined
timotimo finally set up proper travis-ci for moarperf; now people will be able to download releases from github that contain the finished javascript files, so they don't have to run "npm install" and get millions of javascript packages onto their machines 17:14
there'll probably also be releases on cpan that will be similar in that way, hopefully 17:15
17:32 wildtrees joined 17:33 wildtrees left
squashable6 🍕 Kaiepi++ synchronize pull request “Add tests for R#3035”: github.com/perl6/roast/pull/584 17:35
guifa WTF 18:09
reading through the T extension options for the “transform destination” data
one of them is “remove every character in the string"
so ‘en-t-d0-remove’ is “text in English that had ever single character removed” lolol 18:10
18:14 go|dfish joined
cpan-raku New module released to CPAN! XML::Actions (0.4.3) by 03MARTIMM 18:38
18:49 go|dfish left 18:59 Tirifto left 19:04 pmurias joined
El_Che if someone has a clue why this rakudo build fails on travis (with a fedora 30 container) but runs find on the same container on my machine: travis-ci.org/nxadm/rakudo-pkg/builds/599998931 19:11
20:09 pmurias left 20:11 pmurias joined
squashable6 🍕 lizmat++ wrote a comment on “Empty #| cause compilation error”: github.com/rakudo/rakudo/issues/32...-544194008 20:17
🍕 lizmat++ closed issue “Empty #| cause compilation error”: github.com/rakudo/rakudo/issues/3242
20:19 finanalyst_ joined 20:32 chloekek joined 21:00 finanalyst_ left 21:01 cinch_ is now known as cinch, cinch left, cinch joined
squashable6 🍕 niner++ wrote a comment on “Weird issue in epoll module”: github.com/rakudo/rakudo/issues/32...-544197846 21:14
Doc_Holliwood how do i require a minimal version of a module in a `use` statement? 21:23
ah, it dwims 21:24
who'd have thought
21:34 ggoebel joined
Doc_Holliwood is it too late to propose builtin methods? 21:37
i'd really like to have a .last and .nth method for lists
sena_kun Doc_Holliwood, it likely is in a sense that we can't just add stuff in CORE, you'd need to somehow prove that it is useable over what we have now, then consider it for next language spec version, then wait for its release, then for a compiler implementing that, and you'll only be able to use that from fresh enough compilers that support the spec... 21:42
tellable6 2019-10-19T15:31:38Z #raku <AlexDaniel> sena_kun: well, I don't think we thought through the idea of different compilers, partial compliance and stuff like this
2019-10-19T15:33:23Z #raku <AlexDaniel> sena_kun: fudging is ok but very LTA in my opinion
gist.github.com/13c179d51fae450460...fd6433445a
sena_kun Doc_Holliwood, what's wrong with [*-1] and [n]? 21:43
m: my $a = (1,2,3,4,5); say $a[*-1]; say $a[2];
camelia 5
3
AlexDaniel goddamn tellable6 stop posting gists :) 21:44
thank you for your service, of course, tellable6, but you can do better
tellable6 AlexDaniel, \o/
Doc_Holliwood sena_kun, it forces you to break the flow in a chain of list functions. `@foo.grep().nth(3).split().pick` is prettier than `(@foo.grep())[2].split().pick` 21:45
also there is already a `first` method
sena_kun m: (1,2,3,4,5).grep(*)[3].pick;
camelia ( no output )
sena_kun m: (1,2,3,4,5).grep(*)[3].pick.say;
camelia 4
sena_kun the flow seems ok to me. :)
Doc_Holliwood, also, if you REALLY want that, you can use AT-POS directly, I think 21:46
21:46 delimited left
sena_kun m: (1,2,3,4,5).grep(*).AT-POS(3).say; 21:46
camelia 4
tobs m: (1,2,3,4,5).grep(*).[3].pick.say; # indexing can also be a method
camelia 4
sena_kun heh
Doc_Holliwood yeah. 21:47
heh
tobs++
yeah, no. when i try any of those i either get `Missing infix inside []` or `Malformed postfix call (only alphabetic methods may be detached)` 21:53
squashable6 🍕 niner++ wrote a comment on “Weird issue in epoll module”: github.com/rakudo/rakudo/issues/32...-544200300
Doc_Holliwood the latter is the method lookup
squashable6 🍕 Altai-man++ synchronize pull request “Test hyperop results in a slip”: github.com/perl6/roast/pull/587 21:56
Doc_Holliwood nevermind
squashable6 🍕 Altai-man++ wrote a comment on “Test hyperop results in a slip”: github.com/perl6/roast/pull/587#is...-544200488
22:07 eseyman left 22:14 delimited joined, delimited is now known as Guest31645, eseyman joined, Guest31645 left
sena_kun 6c: use NativeCall; my @t2682 := CArray[int32].new(1,2,3); say @t2682[*-1]; 22:15
committable6 sena_kun, gist.github.com/4e4eb9af3f30d6a522...7bf0d984dc
sena_kun 6c: use NativeCall; my @t2682 := CArray[int32].new(1,2,3); say @t2682[0..*-2]; 22:21
committable6 sena_kun, gist.github.com/d7a10b93fd2f62dd01...68c0c3724a
sena_kun committable6, help 22:23
committable6 sena_kun, Like this: committable6: f583f22,HEAD say ‘hello’; say ‘world’ # See wiki for more examples: github.com/perl6/whateverable/wiki/Committable
AlexDaniel just swap 6c: with bisect: if you want to bisect it 22:24
sena_kun AlexDaniel, I think the thing I am writing test for was broken in between of two releases, so want to check that my golf indeed can trigger the correct error on old version... 22:25
AlexDaniel sena_kun: between which releases?
sena_kun but yes, bisecting is another option, but I'd like to check the fixing commit and the one before that
AlexDaniel, github.com/rakudo/rakudo/commit/0d...7578df7d89 <- the commit in question 22:26
AlexDaniel ahh right the parent of the fixing commit will be the one
c: 0d2658b806^,0d2658b806 use NativeCall; my @t2682 := CArray[int32].new(1,2,3); say @t2682[0..*-2];
committable6 AlexDaniel, gist.github.com/b53a76047ad2a110a1...5855df3f5d
AlexDaniel let's see
that? 22:27
sena_kun AlexDaniel, yup! Thanks, pushing it...
by the way, rakudo tests are not accepted due to a release being close or?
AlexDaniel if they pass they can be merged right away 22:28
sena_kun AlexDaniel, they don't (as in full spec), but the failing ones are pretty unrelated... and different on different travis configs
AlexDaniel you mean this? github.com/rakudo/rakudo/pull/3249
sena_kun AlexDaniel, yes, for example this one, also github.com/rakudo/rakudo/pull/3245 22:29
AlexDaniel ahh I see, there's something failing in CI 22:30
anyway the PR is good to go
squashable6 🍕 Altai-man++ wrote a comment on “Some issue exposed by Math::FFT::Libfftw3”: github.com/rakudo/rakudo/issues/26...-544202590
🍕 Altai-man++ opened pull request “Test indexing with whatever for CArray”: github.com/rakudo/rakudo/pull/3250 22:31
sena_kun a-and another one...
AlexDaniel sena_kun: ok, looking at your rakudo pr, why not a roast test? 22:32
sena_kun AlexDaniel, which one? :)
AlexDaniel sena_kun: I understand that your test is looking at the error message, which is a no-no for roast, but the functionality seems to be something that is defined by the language, no? 22:33
sena_kun one is nativecall, which is rakudo-specific, AFAIK
AlexDaniel sena_kun: this one: github.com/rakudo/rakudo/pull/3249
sena_kun error messages are no-no for roast
22:33 lucasb left
AlexDaniel sena_kun: right, but it's error vs no error, so shouldn't we be checking that in roast? 22:34
like a basic lives-ok or something
sena_kun hmmm
squashable6 🍕 AlexDaniel++ wrote a comment on “Test indexing with whatever for CArray”: github.com/rakudo/rakudo/pull/3250...-544202859 22:35
🍕 AlexDaniel++ closed issue “Some issue exposed by Math::FFT::Libfftw3”: github.com/rakudo/rakudo/issues/2682
🍕 AlexDaniel++ merged pull request “Test indexing with whatever for CArray”: github.com/rakudo/rakudo/pull/3250
sena_kun AlexDaniel, I think it doesn't. The rule being tester here is that return constraints are considered when looking at "If this method implemented or not". If it is, we are ok, but if it isn't, it is a compiler level error, it does not compile at all, so we can't dies-ok it. 22:40
s/tester/tested/
AlexDaniel sena_kun: eval-dies-ok ? 22:43
sena_kun AlexDaniel, well... I wonder what's the semantic here would be. We test things that work and some things that must not work. But we don't test any possible input outside of allowed one as "doesn't work". The line between "If it is untested, we consider it to not work" and "We test it does not work" is a bit blurry, imho. 22:44
m: (:(Str(Int) $a)).gist.say 22:50
camelia (Str(Int) $a)
sena_kun 6c: (:(Str(Int) $a)).gist.say
committable6 sena_kun, gist.github.com/2c35eb0ea7f7f59ff7...e12ab1e10a 22:51
22:52 lelf left, lelf joined
AlexDaniel sena_kun: I'm not sure what you mean, I see it as simply “role multi methods must be implemented with the right type” 22:52
sena_kun AlexDaniel, maybe I am overthinking it... 22:53
AlexDaniel for example, don't we already tests that check this (without return types)?
sena_kun we do. I need to check though if we are checking it with a correct return type or not... 22:55
AlexDaniel sena_kun: something like this: github.com/perl6/roast/blob/93cc2e...#L179-L190 22:56
your test is pretty much the same, just using the return type
right?
ok I highlighted the wrong lines but it's around there :) 22:57
22:57 lelf left
sena_kun hm, yes, I guess it is 22:57
22:57 pmurias left
sena_kun anyway, I am finishing a new one and going to bed, I am afraid... 22:57
are results of .gist specced? 22:59
squashable6 🍕 AlexDaniel++ wrote a comment on “Test return constraints for role implementors”: github.com/rakudo/rakudo/pull/3249...-544204099
sena_kun I am looking at github.com/rakudo/rakudo/issues/2416 the test is very easy, but I am really not sure where should it go.
AlexDaniel sena_kun: no, .gists are not specced and shouldn't be I think
arguably even .perl has a range of OK results, I don't know if we test it but we shouldn't 23:00
so putting them in rakudo is definitely safe
sena_kun AlexDaniel, so we can just close it? ;)
AlexDaniel sena_kun++ # shredding it :)
sena_kun: what's the connection to .gist? I'm not following 23:01
sena_kun AlexDaniel, see gist.github.com/2c35eb0ea7f7f59ff7...e12ab1e10a 23:02
the test should be essentially "Gist on a signature with a coercion shows the coercion type"
AlexDaniel ohh I see
just add a test in rakudo, why not 23:03
sena_kun oki
AlexDaniel my point is that they shouldn't go in roast for sure, but adding a test in rakudo to make sure there's no regression is always a good idea :)
sena_kun 02-rakudo/99-misc.t, I guess... 23:04
woah
what I am seeing...
AlexDaniel a lot of gist tests in roast? I see that too :S 23:06
sena_kun no, not these
sigh
AlexDaniel some tests are abusing gists to test the functionality… 23:07
we definitely need a great roast refactor :(
sena_kun github.com/rakudo/rakudo/blob/mast...#L153-L156 and github.com/perl6/roast/blob/master...#L180-L186 23:08
AlexDaniel vrurg: a bunch of roast tests that check the output of .gist? That doesn't make any sense, right??
sena_kun rakudo-one was originally added by jnthn, and I really wonder if we want to remove it now once it's in roast... 23:09
ok, sleep time for me.
o/
AlexDaniel sena_kun: sleep well!
your PRs are great :)
sena_kun AlexDaniel, so do you! Take care, thanks for all your help today.
23:11 sena_kun left 23:14 chloekek left, ggoebel left 23:28 ggoebel joined