»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:02 buggable left 00:03 buggable joined, ChanServ sets mode: +v buggable
labster m: my @a = 1,"foo"; my @b = @a.deepmap(*); 00:05
camelia rakudo-moar 0db081: OUTPUT«Type check failed in binding to &block; expected Callable but got Whatever (*)␤ in block <unit> at <tmp> line 1␤␤»
notviki AlexDaniel: maybe you should schedule it? The only way I see of reading is in the edit mode and IIRC that messes up the HTML code when it autosaves 00:06
m: DateTime.now.utc.say
labster m: my @a = 1,"foo"; my @b = @a.deepmap();
camelia rakudo-moar 0db081: OUTPUT«2016-12-23T00:06:39.101908Z␤»
rakudo-moar 0db081: OUTPUT«Too few positionals passed; expected 2 arguments but got 1␤ in block <unit> at <tmp> line 1␤␤»
notviki AlexDaniel: seems like it could even be published :)
AlexDaniel notviki: let's publish then? :) 00:07
notviki Yeah, publish :)
labster: you need to give it a callable to deepmap with
m: my @a = 1,"foo"; my @b = @a.deepmap(*.uc);
camelia ( no output )
notviki m: my @a = 1,"foo"; my @b = @a.deepmap(*.uc); dd @b
camelia rakudo-moar 0db081: OUTPUT«Array @b = ["1", "FOO"]␤»
labster But I just want to clone an array?
notviki hehe 00:08
AlexDaniel notviki: done
notviki labster: well, that'd be *.clone, I'd guess but I bet that's pretty broken
labster It's only a shallow clone. Or so it says in the docs.
notviki labster: I mean .deepmap with *.clone
labster Oh interesting. I mean this is all a lot of work for what will be a somewhat common use case. 00:09
00:09 pmurias left
notviki hehe 00:10
Perl 6 users are so spoiled :)
labster who are we supposed to be tormenting? :P 00:11
AlexDaniel read the latest advent post for a clue, perhaps? ;)
labster design.perl6.org/S12.html#Cloning Little explain, less spec, wow.
Tormenting our unit tests, apparently. 00:12
00:13 bjz left, raiph joined 00:15 Herby_ joined
Herby_ \o 00:15
labster o/ 00:18
00:19 wamba left 00:22 rindolf left
notviki AlexDaniel++ good post 00:26
AlexDaniel \o/ thanks
00:31 MasterDuke joined
[Coke] anyone have any pointers about getting utf8 to work in a tmux connection? 00:34
before I start tmux on irc.p6c.org, I can use utf8 - after tmux is started, nothing. 00:35
sjn [Coke]: I'm using tmux 2.1 and have no problems with utf8 00:37
(tmux -V shows the version) 00:38
sjn uses nb_NO.utf8 as locale, and othetwise a recent linux 00:39
[Coke]: what does it look like when it's not working? 00:40
[Coke] ah. irc.p6c.org is only on 1.9
sjn;it just doesn't send the unicode through. Here's an ellipsis: 00:41
here:
sjn here's mine: … 00:48
geekosaur is mildly confused, since utf8 support was one of the things tmux had over screen for a long time 00:49
00:49 SCHAAP137 joined
sjn [Coke]: can I have a login on p6c to see if I see someting weird too? <_< :) 00:51
00:51 japhb_ left
Herby_ I keep fumbling with p6 regexes. If I have a string of names, "NAME=Bob\nNAME=Carl\nNAME=Sally", how do I go about getting an array of (Bob, Carl, Sally)? 00:53
TEttinger I'm curious too, Herby_ 00:56
I know there's a fair amount of difference between standard PCRE-style regexes and Perl6's regexes/parsers/grammars 00:57
00:58 cibs left
Herby_ TEttinger, yeah my regex knowledge is shaky at best and I stumble even more when trying to use p6 regexes 00:58
TEttinger my regex knowledge is not shaky but my perl6 knowledge is OOH LOOK A KITTY 00:59
01:00 cibs joined
timotimo m: "NAME=Bob\nNAME=Carl\nNAME=Sally".comb(/ '=' <( .*? )> \n /).perl.say 01:01
camelia rakudo-moar 0db081: OUTPUT«("Bob", "Carl")␤»
timotimo oh
m: "NAME=Bob\nNAME=Carl\nNAME=Sally".comb(/ '=' <( .*? )> $$ /).perl.say
camelia rakudo-moar 0db081: OUTPUT«("Bob", "Carl", "Sally")␤»
Herby_ \o/
two questions: why the '< >' around the capture group, and how would I push that into an array for later use 01:02
timotimo it's actually the <( thing and the )> thing
it has essentially the same effect as if you put the stuff before it into a lookbehind and the stuff after it into a lookahead
though i could imagine it's faster
Herby_ m: my @names = "NAME=Bob\nNAME=Carl\nNAME=Sally".comb(/ '=' <( .*? )> $$ /); say @names; 01:04
camelia rakudo-moar 0db081: OUTPUT«[Bob Carl Sally]␤»
01:04 leah2 left
TEttinger with more traditional regexes in Clojure I'd do: (re-seq #"(?<=NAME=)\S+" "NAME=Bob\nNAME=Carl\nNAME=Sally") 01:04
Herby_ i was trying to use m :g//
TEttinger not sure how perl6 does lookbehind
timotimo it's <?after foo>
TEttinger thanks
Herby_ thanks, timotimo 01:05
TEttinger is .comb short for something or is it like "combing the desert for clues"
timotimo my regex doesn't account for spaces, though
it's the latter, TEttinger
Herby_ timotimo: one more question for you if you're feeling adventurous 01:08
timotimo sure
Herby_ if I have the string ""NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47", how would I return tuples of (NAME,AGE)
timotimo adventure is my second middle name
Herby_ so [(Bob,23),(Carl,31),(Sally,47)] 01:09
or something similar
timotimo just a sec
Herby_ kk
timotimo are name and age always in the right order?
Herby_ yep. always NAME, and their AGE on the next line 01:10
timotimo well, that's easy then
m: "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".comb(/ '=' <( .*? )> $$ /).rotor(2 => 2).perl.say
camelia rakudo-moar 0db081: OUTPUT«(("Bob", "23"), ("Sally", "47")).Seq␤»
timotimo oops?
m: "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".comb(/ '=' <( .*? )> $$ /).rotor(2 => 0).perl.say 01:11
camelia rakudo-moar 0db081: OUTPUT«(("Bob", "23"), ("Carl", "31"), ("Sally", "47")).Seq␤»
timotimo there we go
Herby_ you're money
01:12 japhb left 01:13 espadrine left
b2gills m: #~(you may want them in pairs though) say "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".comb(/'=' <(.*?)> $$ /).map(*=>*).perl 01:13
camelia ( no output )
notviki m: "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".comb(/"="<(\N+/).rotor(2 => 0).perl.say
camelia rakudo-moar 0db081: OUTPUT«(("Bob", "23"), ("Carl", "31"), ("Sally", "47")).Seq␤»
b2gills m: #`(you may want them in pairs though) say "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".comb(/'=' <(.*?)> $$ /).map(*=>*).perl 01:14
camelia rakudo-moar 0db081: OUTPUT«(:Bob("23"), :Carl("31"), :Sally("47")).Seq␤»
01:14 japhb joined
b2gills .rotor(2=>0) can be spelled as .rotor(2) 01:14
notviki m: "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".comb(/"="<(\N+/).Hash.say 01:15
camelia rakudo-moar 0db081: OUTPUT«{Bob => 23, Carl => 31, Sally => 47}␤»
Herby_ how would you tackle that with traditional regex, not utilizing comb and rotor?
well, traditional p6 regex
01:16 SCHAAP137 left, newbie1 left
notviki .ask samcv seems there's now merge conflict on the Linguist thing: github.com/github/linguist/pull/33...-267160040 01:16
yoleaux notviki: I'll pass your message to samcv.
Herby_ i have a small python script that parses text for me and utilizes re.findall(), i'd like to swap it out for a p6 script 01:17
timotimo well, findall is basically comb, so ... :P
notviki Herby_: so swap it out? Why do you need it to be regex only?
timotimo you can also give :match to comb to get match objects instead of only the strings
then it'll behave a lot more like m:g
m: ("NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47" ~~ m:g/ '=' <( .*? )> $$ /)>>.Str.rotor(2 => 0).perl.say 01:18
camelia rakudo-moar 0db081: OUTPUT«(("Bob", "23"), ("Carl", "31"), ("Sally", "47")).Seq␤»
timotimo ^- without comb
01:19 BenGoldberg joined
notviki m: "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".match(:g, /<after "="> (\N+)/)».caps».flat.say 01:19
camelia rakudo-moar 0db081: OUTPUT«(((0 => ｢Bob｣) (after => #<failed match>)) ((0 => ｢23｣) (after => #<failed match>)) ((0 => ｢Carl｣) (after => #<failed match>)) ((after => ｢=EGA␤lraC=｣) (0 => ｢31｣)) ((after => ｢=EMAN␤13=EGA␤lraC=EMAN␤32=｣) (0 => ｢Sally｣)…»
notviki This is really weird output... wtf is it backwards!
Herby_ notviki: i need to brush up on comb and rotor. my realworld data is not as cleanly structured 01:20
user inability, not language :)
notviki m: m: "NAME=Bob\nAGE=23\nNAME=Carl\nAGE=31\nNAME=Sally\nAGE=47".split(/<[\n=]>/).say 01:21
camelia rakudo-moar 0db081: OUTPUT«(NAME Bob AGE 23 NAME Carl AGE 31 NAME Sally AGE 47)␤»
notviki doh
Herby_ but thanks timotimo and notviki for showing me some examples, I can work backwards from there
01:22 raiph left
timotimo good good 01:22
01:23 aborazmeh joined, aborazmeh left, aborazmeh joined 01:24 SCHAAP137 joined 01:25 dataf3l joined
MasterDuke notviki: btw, perl6 -ne 'say "{$_}"' foo, #130383, is uneffected by my recent IO::ArgFiles PRs 01:27
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130383
01:27 troys is now known as troys_
MasterDuke but i'm still trying to get phasers working, #129093, so maybe a fix for one will also fix the other 01:28
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=129093
01:30 Actualeyes joined 01:33 skids joined 01:40 kurahaupo left
samcv .tell notviki thanks for pointing that out. Wish github had notified me. Will fix when I get home 01:50
yoleaux samcv: I'll pass your message to notviki.
01:16Z <notviki> samcv: seems there's now merge conflict on the Linguist thing: github.com/github/linguist/pull/33...-267160040
02:02 dataf3l left 02:03 pyrimidi_ joined, pyrimidine left 02:06 cpage_ joined 02:18 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 02:25 pyrimidine joined, pyrimidi_ left 02:30 Actualeyes left 02:32 Actualeyes joined 02:33 leah2 joined 02:45 ilbot3 left 02:47 ilbot3 joined 02:50 kalkin- left, kalkin-_ joined 02:55 troys_ is now known as troys 03:03 BenGoldberg left 03:06 BenGoldberg joined 03:13 xtreak joined 03:15 pyrimidi_ joined, pyrimidine left 03:25 troys is now known as troys_ 03:27 colomon left 03:33 pyrimidi_ left 03:35 labster left 03:36 Actualeyes left 03:38 aindilis joined 03:44 BenGoldberg left 03:46 BenGoldberg joined 03:51 AlexDaniel left, Actualeyes joined 03:57 noganex_ joined 04:00 noganex left 04:03 cpage__ joined 04:04 cpage_ left, cpage__ is now known as cpage_ 04:13 Actualeyes left 04:28 cpage_ left 04:31 pyrimidine joined 04:32 dugword joined
dalek c: 3bdc823 | (Armand Halbert)++ | doc/Type/Str.pod6:
Wrote documentation for Str method match
 04:35
c: a0cb931 | (Armand Halbert)++ | doc/Type/Str.pod6:
Wrote documentation for Str method match
synopsebot6 Link: doc.perl6.org/type/Str
c: b0c9044 | (Armand Halbert)++ | doc/Type/Str.pod6:
Added say to the code examples
synopsebot6 Link: doc.perl6.org/type/Str
c: aa85248 | (Armand Halbert)++ | doc/Type/Str.pod6:
Merge branch 'master' of github.com:ahalbert/doc
synopsebot6 Link: doc.perl6.org/type/Str
c: 4766ed3 | (Armand Halbert)++ | doc/Type/Str.pod6:
Fixed typos
synopsebot6 Link: doc.perl6.org/type/Str
c: 202fa2c | titsuki++ | doc/Type/Str.pod6:
Merge pull request #1074 from ahalbert/master

Wrote documentation for Str method match
synopsebot6 Link: doc.perl6.org/type/Str
04:41 khw left 05:09 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 05:14 raiph joined 05:18 cpage_ joined 05:19 Cabanossi left 05:20 Cabanoss- joined, Cabanoss- is now known as Cabanossi 05:22 aborazmeh left 05:29 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg, rpburkholder left 05:33 zacts left 05:34 troys_ is now known as troys
samcv ok. back home now. time to fix those highlighting things 05:35
05:39 aborazmeh joined, aborazmeh left, aborazmeh joined, cognominal left 05:45 BenGoldberg left, Ben_Goldberg joined 05:46 Ben_Goldberg is now known as BenGoldberg, skids left 05:50 cognominal joined 05:51 mr-foobar joined 05:54 dugword left 05:56 aborazmeh left 05:57 mr-foobar left 06:00 xtreak left 06:04 xtreak joined
dalek c: c0a734b | samcv++ | doc/Language/regexes.pod6:
Trigger doc rebuild to pull in highlighter updates

Also add in a missing semicolon.		 06:09
synopsebot6 Link: doc.perl6.org/language/regexes
06:13 sivoais_ is now known as sivoais, sivoais left, sivoais joined 06:19 bjz joined 06:25 labster joined 06:29 darutoko joined 06:38 troys left
samcv m: say 0xD800.chr 06:39
camelia rakudo-moar 0db081: OUTPUT«Error encoding UTF-8 string: could not encode codepoint 55296␤ in block <unit> at <tmp> line 1␤␤»
06:39 pyrimidine left, pyrimidi_ joined
samcv bisect: all say 0xD800.chr 06:39
bisectable6 samcv, On both starting points (old=2015.12 new=0db0810) the exit code is 1 and the output is identical as well
samcv, Output on both points: WARNINGS for /tmp/GmkKOTzqTc:␤Useless use of "all say 0xD800.chr" in expression "all say 0xD800.chr" in sink context (line 1)␤Error encoding UTF-8 string: could not encode codepoint 55296␤ in block <unit> at /tmp/GmkKOTzqTc line 1␤
samcv T_T
m: say 0xD801.chr 06:43
camelia rakudo-moar 0db081: OUTPUT«Error encoding UTF-8 string: could not encode codepoint 55297␤ in block <unit> at <tmp> line 1␤␤»
samcv m: say 0xD804.chr
camelia rakudo-moar 0db081: OUTPUT«Error encoding UTF-8 string: could not encode codepoint 55300␤ in block <unit> at <tmp> line 1␤␤»
samcv m: say 0xE804.chr
camelia rakudo-moar 0db081: OUTPUT«␤»
06:44 thayne joined
samcv how can i write a script to test against unicode's grapheme test if it can't even encode a character they use on like 1/7 of the tests :\ 06:44
06:46 RabidGravy joined 06:55 riatre left
dalek Iish: 1c47b5f | titsuki++ | examples/pg.p6:
pg: Use .dispose instead of .disconnect
 06:57
Iish: a9ba5ef | titsuki++ | examples/pg_arrays.p6:
pg_arrays: Use .dispose instead of .disconnect
Iish: 9707c6f | titsuki++ | lib/DBIish.pm6:
POD: Use .dispose instead of .disconnect
Iish: d806786 | titsuki++ | / (3 files):
Merge pull request #81 from titsuki/use-dispose

Use dispose
06:57 riatre joined
holli prefers ASCII anyway 06:59
07:05 bwisti left 07:08 BenGoldberg left 07:13 xtreak left
samcv heh 07:14
07:14 xtreak joined 07:15 AlexDaniel joined 07:18 bjz left
samcv so of the tests that break because of that character we mysteriously can't put in a string, we fail 54 of the 744 tests that i am able to run 07:21
07:24 AlexDaniel left
samcv basically testing how we break of characters if you are intimate with what graphemes are 07:25
i've been learning way too much about unicode the last two weeks… 07:27
07:31 Actualeyes joined
samcv m: 0x200D.uniprop('Line_Break').say 07:33
camelia rakudo-moar 0db081: OUTPUT«BK␤»
samcv wat
t-that's not right
holli #NotMyOutput
;-)
samcv should be ZWJ, no wonder some of these tests are failing 07:34
samcv opens yet another MoarVM ticket for unicode :P
m: 0x103D.uniprop('Line_Break').say 07:35
camelia rakudo-moar 0db081: OUTPUT«BK␤»
samcv t-that's not right either
they're all probably wrong i might think
u: { .uniprop('Line_Break') ne all('BK', '') } 07:36
unicodable6 samcv, U+0001 START OF HEADING [Cc] (control character)
samcv, U+0000 NULL [Cc] (control character)
samcv, U+0002 START OF TEXT [Cc] (control character)
samcv ok. so all of them are broken except for those three... pluss all the ones that would be BK anyway 07:37
thank you unicodable
07:38 holli left 07:40 xtreak left, xtreak joined 07:42 xtreak left 07:43 xtreak joined
samcv u: { .uniprop('Grapheme_Extend') == True } 07:46
unicodable6 samcv, U+0300 COMBINING GRAVE ACCENT [Mn] (◌̀)
samcv, U+0301 COMBINING ACUTE ACCENT [Mn] (◌́)
samcv, U+0302 COMBINING CIRCUMFLEX ACCENT [Mn] (◌̂)
samcv, gist.github.com/3d784ac319ba82ac23...c38148ca1b
07:46 labster left 07:50 pierre__ joined
samcv actually i think i did that wrong 07:51
u: { .uniprop('Line_Break') ne any('BK', '') }
07:51 Actualeyes left
unicodable6 samcv, U+0000 NULL [Cc] (control character) 07:51
samcv, U+0001 START OF HEADING [Cc] (control character)
samcv, U+0002 START OF TEXT [Cc] (control character)
samcv, gist.github.com/c254846001d1b03583...74416bedf4 07:52
07:55 labster joined, xtreak left 08:02 pyrimidine joined, pyrimidi_ left 08:10 wamba joined, xtreak joined 08:11 djbkd joined 08:14 Tonik joined 08:24 andrzejku joined
samcv m: Uni.new(0x1D1C0,0x1D1BA, 0x1D165, 0x1D16F).say 08:24
camelia rakudo-moar 011df4: OUTPUT«Uni:0x<1d1c0 1d1ba 1d165 1d16f>␤»
samcv m: Uni.new(0x1D1C0,0x1D1BA, 0x1D165, 0x1D16F).Str.say
camelia rakudo-moar 011df4: OUTPUT«𝆺𝅥𝅯𝆺𝅥𝅯␤»
08:25 rpburkholder joined 08:34 domidumont joined 08:38 domidumont left 08:39 domidumont joined, djbkd left 08:41 domidumont left 08:43 AlexDaniel joined 08:45 Tonik left 08:50 xtreak left
DrForr Morning. 08:50
samcv good morning 08:51
DrForr o/
09:00 cibs left 09:02 cibs joined 09:12 xtreak joined 09:14 domidumont joined 09:31 pierre__ left 09:36 pierre__ joined 09:47 pierre__ left 09:52 rindolf joined 10:00 AlexDaniel left, AlexDaniel joined 10:03 dj_goku left 10:07 AlexDaniel left 10:29 rpburkholder left 10:32 xtreak left, Guest33965 is now known as ponbiki 10:33 eyck joined 10:44 labster left 10:52 grondilu joined 10:54 grondilu_ left 10:57 xtreak joined 11:04 Actualeyes joined 11:11 thayne left 11:12 bjz joined 11:16 xtreak left 11:17 gregf_ joined
arnsholt mst++ # MSTPAN 11:32
11:35 TEttinger left 12:02 grondilu_ joined, pyrimidine left, pyrimidine joined 12:04 grondilu left 12:19 barkode joined 12:20 bjz left 12:26 bjz joined
dalek rl6-most-wanted: 0e79516 | (Tom Browder)++ | most-wanted/modules.md:
add a calendar module
 12:30
rl6-most-wanted: 2c0b3a3 | (Tom Browder)++ | most-wanted/modules.md:
change category name
 12:37
12:44 barkode left 12:45 grondilu joined 12:48 grondilu_ left
dalek rl6-most-wanted: 5f71e99 | (Tom Browder)++ | most-wanted/modules.md:
add WIP module
 12:57
notviki well... crap. 13:14
yoleaux 01:50Z <samcv> notviki: thanks for pointing that out. Wish github had notified me. Will fix when I get home
notviki 2016.11 R* is broken on Windows 7 :/
notviki has just confirmed RT#130391 13:15
rt.perl.org/Ticket/Display.html?id=130391
13:17 bjz left, bjz_ joined
lizmat is it fixable ? 13:17
easily?
notviki Well, the error looks like the bug in 2016.10 R* that we thought we fixed.
it was something about mkdir or something
notviki rakes MoarVM commits
Oh. Someone smart started adding commit shas to changelog :) github.com/rakudo/rakudo/blob/nom/...geLog#L125 13:18
github.com/rakudo/rakudo/commit/44a4c75 13:19
13:22 teksteiner joined
notviki :o zef works fine :/ 13:27
oh, I thought someone already suggested `zef` to that dude on Facebook 13:28
oh they did... 13:29
"Failed to rename 'C:\Users\winklest\.zef\store\projects.json' to 'C:\Users\winklest\.zef\store\p6c\packages.json': Failed to rename file: no such file
or directory
in block at C:\rakudo\share\perl6\site\sources\F9AFEBDFA35D22BCCF53CDB4667B5C8E843F7754 (Zef::ContentStorage::Ecosystems) line 65
can't reproduce that...
Though I don't have .zef at his location.... It's stuffing everything into one of my network shares -_- 13:31
lizmat ah, maybe the network shares have a different renaming policy 13:38
notviki vOv 13:41
13:44 pmurias joined
pmurias Is class file to large expected when building rakudo-j? 13:44
dalek Iish: eda3fa7 | adaptiveoptics++ | lib/DBDish/Pg/Native.pm6:
Added map for bpchar type (char blank-padded)
Iish: 478fdb2 | RabidGravy++ | lib/DBDish/Pg/Native.pm6:
Merge pull request #82 from adaptiveoptics/master

Added map for bpchar type (char blank-padded)
notviki So how do I get "npm" command that's now needed for docs? 13:57
.ask samcv since you know it best, would you mind updating github.com/perl6/doc/blob/master/C...f-contents with the new deps for new highlighting mode? I'm guessing we no longer need pygmentize and Inline::Python, but need `npm`? I'm having trouble getting `npm` on debian wheezy 14:01
yoleaux notviki: I'll pass your message to samcv.
MasterDuke pmurias: there was some talk about that in #perl6-dev earlier
14:02 stanley joined
dogbert17 notviki: doesn't npm come as part of the node js installation? 14:06
notviki dogbert17: not on wheezy 14:07
14:07 Actualeyes left
notviki is following antler.co.za/2014/04/install-node-j...-wheezy-7/ now 14:07
but I'm seeing a bunch of errors -_-
dogbert17 testing the new highlighter for the first time 14:08
notviki dogbert17: BTW, I'm making sass mandatory now 14:09
dogbert17 uh oh :-)
notviki hm...
That site tells to run curl www.npmjs.com/install.sh | sh as root, but when I try to download that script, I get "The certificate's owner does not match hostname `www.npmjs.com'"
tsk tsk
14:10 Actualeyes joined
dogbert17 I think I got that working as a matter of fact, don't remember how though 14:10
sass that is
notviki ls
grrrr 14:12
./install.sh -> "npm ERR! Please try running this command again as root/Administrator." 14:13
sudo ./install.sh -> "You need Node.js to run this program."
notviki gives up
dogbert17 attempt 1 = FAIL: duplicated path /tmp/dogbert-97571-pod_to_pyg.pod 14:14
MoarVM panic: Memory allocation failed; could not allocate 1799720 bytes
how much memory is needed to build the docs with highlighting? (have 2 gigs in my vm) 14:16
notviki fails on debian jessie too 14:18
"npm WARN This failure might be due to the use of legacy binary "node"" 14:19
gyp ERR! node -v v0.10.29
dogbert17 that is old
notviki v0.12.16 is on hack and it builds fine 14:20
damn hipsters with their shiny tools :} 14:22
dogbert17 notviki: the doc build seems to need 3 gigs+ of memory !! 14:23
notviki heh
dogbert17 grr: MoarVM panic: Memory allocation failed; could not allocate 1900488 bytes
shortly before fail: ogbert@dogbert-VirtualBox ~ $ free 14:24
total used free shared buffers cached
Mem: 2063544 1937816 125728 1252 328 22008
-/+ buffers/cache: 1915480 148064
Swap: 1046524 1046076 448
notviki Well... Plan C for my small change then. 14:25
Which is: edit file in a checkout on a local box I'm sshed into, then scp it to one of my servers that has its key on hack, then scp to hack, then build to ensure the change doesn't break anything -_- 14:26
dogbert17: as for memory, dunno. Ask samcv if there's a way to reduce usage
dogbert17 will do 14:27
notviki .tell samcv FWIW, building on Debian Jessie failed too. It was complainging about failed node-gyp install :/ 14:29
yoleaux notviki: I'll pass your message to samcv.
notviki :| 14:30
dogbert17 m: multi sub cross() { }
camelia ( no output )
notviki now pod2bigpage is failing "Required named parameter 'name' not passed"
Today is Failure day!
timotimo dogbert17: would be nice if you ran the doc update with /usr/bin/time, as it'll tell us the peak memory usage
dogbert17 timotimo: on it
notviki: would you like to help a lazy git? 14:31
notviki with what?
dogbert17 closing an RT
RT #126508
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=126508
notviki Don't tell me you still don't have access to close them yourself... 14:32
14:32 bjz_ left
dogbert17 that's the lazy git part :-) 14:32
notviki dogbert17: so why is it being closed?
dogbert17 look a few lines above
notviki So? 14:33
It's not proof that bug is fixed.
Tests are proof.
:)
dogbert17 where would you want them, i.e. which file
notviki tree -f | grep cross 14:34
S32-list/cross.t looks like a good candidate
dogbert17 ok, will use that 14:35
notviki goes for Plan D 14:40
rindolf notviki: hi! What did you try to build? 14:41
dalek c: fe7cc28 | (Zoffix Znet)++ | / (2 files):
Consolidate new highlights build under `make html`
 14:42
notviki rindolf: docs.perl6.org
14:43 ab6tract joined
rindolf notviki: ah 14:43
ab6tract how hard is the deadline of getting an advent post in before midnight of the publishing day? 14:45
because i'm struggling :(
notviki m: say (DateTime.new(:2016year, :24day, :12month).Instant - DateTime.utc.now.Instant)/3600 14:47
camelia rakudo-moar f0398f: OUTPUT«Cannot look up attributes in a DateTime type object␤ in block <unit> at <tmp> line 1␤␤»
notviki waat
m: say (DateTime.new(:2016year, :24day, :12month).Instant - DateTime.now.utc.Instant)/3600
camelia rakudo-moar f0398f: OUTPUT«9.20535812672176␤»
notviki ab6tract: well, I think if you post it in the next 9-33 hours, it'd be fine :) 14:48
MasterDuke bisect: multi sub cross() { }
bisectable6 MasterDuke, Bisecting by exit signal (old=2015.12 new=f0398fb). Old exit signal: 11 (SIGSEGV)
MasterDuke, bisect log: gist.github.com/19a3304266e93daa7f...d777cd1b82 14:49
MasterDuke, (2016-07-28) github.com/rakudo/rakudo/commit/99...ce01f6eeb0
notviki ummm... running `make help` in docs now tries to run `init-highlights`.... 14:50
Any idea what's wrong?
14:51 travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'Consolidate new highlights build under `make html`' 14:51
travis-ci.org/perl6/doc/builds/186354627 github.com/perl6/doc/compare/c0a73...7cc28dbdc7
14:51 travis-ci left
ab6tract notviki: ooof that makes me feel much better 14:51
i am on call tomorrow and will be stuck at the computer, should be able to get it done by lunch AMS time
notviki ohhh... apparently I can't use `` in @echo 14:52
dalek c: 8381367 | (Zoffix Znet)++ | Makefile:
list init-highlights in make help
dogbert17 notviki: will the do? 14:55
# RT 126508
ok EVAL('multi sub cross() { }'), "multi sub cross shouldn't SEGV"
s/the/this/
14:56 Praise left
notviki committable6: 2016.04 use Test; lives-ok 'multi sub cross() { }' 14:56
committable6 notviki, gist.github.com/b69fbf6d848d67bade...848f4ce71a
14:56 Praise joined, Praise left, Praise joined
notviki committable6: 2016.04 use Test; eval-lives-ok 'multi sub cross() { }' 14:56
committable6 notviki, ¦«2016.04»: «exit signal = SIGSEGV (11)»
14:56 bwisti joined
notviki m: use Test; eval-lives-ok 'multi sub cross() { }' 14:56
camelia rakudo-moar f0398f: OUTPUT«ok 1 - ␤»
notviki dogbert17: eval-lives-ok 'multi sub cross() { }', "multi sub cross doesn't SEGV" 14:57
dogbert17 thx
pmurias notviki: the current node is 7.something so 0.12.16 doesn't count as shiny ;) 15:00
notviki: the node-gyp issues I have seen where cause by the "debian people" renaming the node binary 15:01
to avoid conflicts with some used by 3 people program 15:02
15:02 travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'list init-highlights in make help' 15:02
travis-ci.org/perl6/doc/builds/186356808 github.com/perl6/doc/compare/fe7cc...8136725e0c
15:02 travis-ci left
dogbert17 notviki: PR sent 15:03
notviki dogbert17: looks good. merge it. 15:04
dogbert17 done
notviki \o/ dogbert17++
dogbert17 will you close the RT, promise to hound [Coke] later 15:05
notviki Done 15:06
dogbert17 notviki++ 15:09
dalek c: 5ec5bec | (Zoffix Znet)++ | .travis.yml:
Make travis use `make html`
timotimo it wouldn't be bad if we put "/usr/bin/time" into travis' commands, too 15:10
that way we'd also be able to track memory usage over time
15:11 holli joined 15:12 skids joined
dalek c: e4cff5c | (Zoffix Znet)++ | .travis.yml:
use /usr/bin/time in travis for moar info
 15:12
jeek Make me use what now? 15:16
notviki ? 15:17
jeek <- Travis 15:18
15:19 travis-ci joined
travis-ci Doc build errored. Zoffix Znet 'use /usr/bin/time in travis for moar info' 15:19
travis-ci.org/perl6/doc/builds/186361583 github.com/perl6/doc/compare/5ec5b...cff5ccc26e
15:19 travis-ci left
timotimo cool 15:19
jeek: thanks for running all that stuff for us and all the other FOSS people
notviki timotimo: it ain't got/usr/bin/time
jeek: you should shave your mustache... 15:20
timotimo wow, damn it
dalek c: 158db97 | (Zoffix Znet)++ | .travis.yml:
Revert "use /usr/bin/time in travis for moar info"

There ain't no pancake mix in there.
timotimo pancake mix? o_O 15:21
rindolf notviki: hi! I am getting this w perl6/doc - paste.debian.net/904241/
dogbert17 timotimo: gist.github.com/dogbert17/1181b513...a394bd16a0
notviki timotimo: www.youtube.com/watch?v=qwyZ0ji1GRU
timotimo that's just 1.8 gigs, though? 15:22
dogbert17 yeah, but it crashed, lemme restart my vm with three gigs instead
timotimo notviki: wow, that's fantastic
15:22 dogbert17 left
notviki rindolf: well, you didn't follow the full build steps so you're missing prereqs: github.com/perl6/doc/blob/master/C...#podtohtml 15:23
15:24 brrt joined 15:25 dogbert17 joined
rindolf notviki: «panda install Pod::To::HTML Pod::To::BigPage» fails here. 15:27
notviki rindolf: how?
15:27 Praise left, Praise joined, Praise left, Praise joined
rindolf notviki: see paste.debian.net/904242/ 15:28
timotimo rindolf: you need to run "rakudobrew rehash" 15:30
to make the rakudobrew give you a binary for pod2onepage
moritz Pod::To::HTML:ver<0.3.6>:auth<>:api<> already installed 15:31
sounds like the installation has worked before
notviki
.oO( rakudobrew probably should be nixed from doc/CONTRIBUTING.md )
moritz notviki: +1 15:32
rindolf timotimo: I ran it and I get the same problem
timotimo oh?
notviki rindolf: same problem is what?
rindolf: the paste you showed tells you the module is already installed.
timotimo ah, i expect you already had Pod::To::HTML, but not Pod::To::Bigpage
but it bailed out early because Pod::To::HTML is already installed
rindolf notviki: moritz : timotimo : rakudobrew is also in .travis.yml
timotimo so just do panda --force install The::Two::Thigns
rindolf timotimo: ok/ 15:33
notviki rindolf: yes, but we know what we're doing. rakudobrew is not for end users.
rindolf: and then run rakudobrew rehash to get the pod2bigpage script from Pod::To::BigPage properly installed 15:34
buggable: eco bigpage
buggable notviki, Pod::To::BigPage 'Render many pod6-files into one (big) html-file.': github.com/gfldex/perl6-pod-to-bigpage
notviki :}
dogbert17 timotimo: failed on 3 gig vm as well: gist.github.com/dogbert17/1181b513...a394bd16a0
timotimo i thought you were going to run it on hack (which should have enough ram) 15:35
dogbert17 I'm wondering if it could be a 32 bit issue
15:44 gregf_ left
notviki Anyone knows bash and could tell me why I'm getting "compile-sass.sh: line 12: syntax error near unexpected token `elif'" ? gist.github.com/zoffixznet/f6ce0a5...abbeea68e8 15:46
moritz docs.python.org/3.6/whatsnew/3.6.html 15:47
timotimo oh, they decided against skipping version number 6
moritz I can't help but think "format string? Perl (and PHP, shell, ruby, ..) have had string interpolation for ages" 15:48
timotimo oooh, they stole our underscores for numeric values
moritz and underscores in number literals? old hat too
15:50 brrt left
rindolf notviki: OK, seems to run fine now, but it does seem kinda time consuming 15:50
notviki Well, yeah
timotimo they have async and yield combined, we don't have that i don't think 15:51
moritz don't use rakudobrew if it hinders more than it benefits you
15:52 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Make travis use `make html`' 15:52
travis-ci.org/perl6/doc/builds/186360952 github.com/perl6/doc/compare/83813...c5bec0b0a8
15:52 travis-ci left
timotimo rindolf: yes, it takes a long time and also a lot of memory 15:53
rindolf timotimo: ah 15:54
dogbert17 have never used Python I know that people at work who swears by it use a much older version for some reason 15:55
rindolf timotimo: wow! 41% of my 3 GB of RAM 15:56
Putting firefox to shame. ;-)
timotimo it might end up taking even more, though
dogbert17 probably a bit more
notviki solved my thing; missing semicolons after exit 1; 15:58
15:58 cyphase left 16:00 newbie1 joined
dalek c: 8fff433 | (Zoffix Znet)++ | / (2 files):
Add SASS compiler script and make target

  - Tries to use `sass` first
  - If no `sass` is present, tries to compile with CSS::Sass module		 16:01
16:02 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Revert "use /usr/bin/time in travis for moar info" 16:02
travis-ci.org/perl6/doc/builds/186364068 github.com/perl6/doc/compare/e4cff...8db975a4ea
16:02 travis-ci left 16:03 cyphase joined
notviki So I need to bring in either apt-get install ruby-sass or CSS::Sass Perl 5 module... do we have some sort of list where this stuff is added? 16:04
like we have /home/rakudobrew/MODULES for P6 modules
notviki doesn't see anything in github.com/perl6/infrastructure-do...6c.org.pod 16:05
timotimo what, on hack?
hm. i think we just ask our administrators to install stuff?
notviki Yeah, on hack
OK then, I'll just install it
16:07 mr_ron joined
mr_ron m: say ("ab", "cd").flatmap( *.comb ) # (a b c d) rather than ((a b) (c d)) 16:08
camelia rakudo-moar f0398f: OUTPUT«(a b c d)␤»
notviki m: say ("ab", "cd").flatmap( *.comb.item ) 16:09
camelia rakudo-moar f0398f: OUTPUT«((a b) (c d))␤»
mr_ron Looking at the earlier discussion with Herby and docs issue 851 and wondering if the idea is a useful improvement github.com/perl6/doc/issues/851
16:13 zacts joined 16:14 cyphase left
dalek c: 0bc9048 | (Zoffix Znet)++ | / (3 files):
Enable SASS Compiler and TOSS style.css

html/css/style.css is now a generated file and no longer needs to be in the repo.
  `make sass` target now compiles SASS
  `make html` now also calls `make sass`		 16:15
16:16 zakharyas joined 16:18 Khisanth left 16:19 cyphase joined
dalek c: a07e3b4 | (Zoffix Znet)++ | Makefile:
list all phony targets
 16:19
c: a0cd377 | (Zoffix Znet)++ | util/compile-sass.sh:
tabs » spaces
 16:21
c: f799c3a | (Zoffix Znet)++ | doc/Type/Str.pod6:
Remove trailing whitespace
 16:22
synopsebot6 Link: doc.perl6.org/type/Str
moritz why do we have the Supply/Supplier distinction? 16:23
16:24 travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'Enable SASS Compiler and TOSS style.css 16:24
travis-ci.org/perl6/doc/builds/186376244 github.com/perl6/doc/compare/8fff4...c904809b9a
16:24 travis-ci left 16:25 vendethiel- joined
notviki moritz: Supplier sends stuff and Supplies provide it? 16:26
I recall jnthn added that right close to Christmas and was much happier with the new UI
s/UI/API/;
16:27 vendethiel left, travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'list all phony targets' 16:27
travis-ci.org/perl6/doc/builds/186377179 github.com/perl6/doc/compare/0bc90...7e3b40b817
16:27 travis-ci left
notviki jeek: Travis! Stop erroring out! :P 16:28
moritz notviki: that's a kinda weak explanation; before the distinction you could send values directly to the Supply; looked easier to me
jeek :( 16:29
notviki moritz: I'm sure jnthn can explain much better
16:30 travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'tabs » spaces' 16:30
travis-ci.org/perl6/doc/builds/186377621 github.com/perl6/doc/compare/a07e3...cd377d6115
16:30 travis-ci left, travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'Remove trailing whitespace' 16:30
travis-ci.org/perl6/doc/builds/186377661 github.com/perl6/doc/compare/a0cd3...99c3a3f063
16:30 travis-ci left
dalek c: 65e9332 | (Zoffix Znet)++ | .travis.yml:
Make Travis happy
 16:30
mr_ron notviki: is .item documented user facing? ... noticed $(), @() and %() don't seem to be 16:31
16:31 Khisanth joined
timotimo it's supposed to be user facing 16:32
mr_ron Was looking at opening a doc issue on contextualizers later ... 16:33
dalek c: 3b1674f | (Zoffix Znet)++ | assets/sass/style.scss:
Toss generated warning from style.scss

We no longer keep any generated files in the repo		 16:37
16:39 vendethiel- left
mr_ron sorry - issue #626 on item 16:40
16:40 travis-ci joined
travis-ci Doc build failed. Zoffix Znet 'Add SASS compiler script and make target 16:40
travis-ci.org/perl6/doc/builds/186372968 github.com/perl6/doc/compare/158db...ff43397a01
16:40 travis-ci left
notviki Well, since the show/hide on TOC is broken—and I assume has been for a while—I'm gonna nix that feature altogether 16:42
on doc site I mean 16:43
16:46 dugword joined
notviki huh, last travis is weird. it's a sucessful exit at the end. But... there were errors for tabs. I thought those tests were not run on travis? 16:47
dalek c: 54fb4b3 | (Zoffix Znet)++ | / (6 files):
SASSify all CSS

Make the site use a single sassy CSS file.		 16:50
perlpilot moritz, notviki: didn't the Supply/Supplier distinction have something to do with live vs. on-demand supplies? 16:57
dalek c: 06930ae | (Zoffix Znet)++ | / (3 files):
Remove TOC hide/show feature

It's only used on narrow screens, since we now show TOC on the side on wide screens. It's also is half-broken and has been for at least several months and no one noticed, so I doubt the feature is in high demand.
By removing it we can get rid of Cookies jquery plugin and the associated notice about our site using cookies.
  .oO( mmm.... coookies )		 17:00
moritz idlewords.com/talks/superintelligence.htm "If AdSense became sentient, it would upload itself into a self-driving car and go drive off a cliff." 17:04
notviki :/ 17:05
samcv hi all 17:06
yoleaux 14:01Z <notviki> samcv: since you know it best, would you mind updating github.com/perl6/doc/blob/master/C...f-contents with the new deps for new highlighting mode? I'm guessing we no longer need pygmentize and Inline::Python, but need `npm`? I'm having trouble getting `npm` on debian wheezy
14:29Z <notviki> samcv: FWIW, building on Debian Jessie failed too. It was complainging about failed node-gyp install :/
samcv no problem notviki, i'll do that
updating the PR for github now so they can pull in the changes too 17:07
notviki .ask [Coke] are you sure your new fuzzy search is working? I tried searching "trait-mod" and trait_mod does not show up. This is BEFORE I made the changes you'll see I have made by the time you read this, so I doubt it's me breaking anything...
yoleaux notviki: I'll pass your message to [Coke].
samcv have never resolved conflicts where upstream had changed references to other git repos inside the same repo before but i think i did it right
dalek c: 79f541f | (Zoffix Znet)++ | / (4 files):
Merge sift4 script into search.js script

To save us from making an extra HTTP request		 17:08
notviki samcv: OK. I'm gonna make a couple of other modificaftions to CONTRIBUTING.md first then 17:09
17:10 dugword left, travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Make Travis happy' 17:10
travis-ci.org/perl6/doc/builds/186379124 github.com/perl6/doc/compare/f799c...e9332ccdb9
17:10 travis-ci left, khw joined 17:12 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Toss generated warning from style.scss 17:12
travis-ci.org/perl6/doc/builds/186380292 github.com/perl6/doc/compare/65e93...1674f998f2
17:12 travis-ci left 17:13 Herby_ left
dalek c: 0f5a4a8 | (Zoffix Znet)++ | CONTRIBUTING.md:
Remove rakudobrew from isntructions

  - Tell to use Rakudo Star instead
  - switch panda to zef (will be part of next R*)		 17:13
17:14 ufobat joined
dalek c: 15b46af | (Zoffix Znet)++ | CONTRIBUTING.md:
Document how to obtain a SASS compiler
 17:17
notviki samcv: I'm done.
And I'm done with all the changes to doc.perl6.org I promised to do a few months back \o/ 17:18
Now onto some PHP hacking for the rakudo.org download page \o/
samcv which changes? 17:21
nice notviki
notviki samcv: make the build chain automatically compile SASS instead of us having to do it manually 17:23
samcv nice
dogbert17 samcv: memory usage when running htmlify with highlighting seems at bit high, do you know if there's anything which can be done about that? 17:26
samcv how high is it?
and what's using the ram? moarvm or node?
dogbert17 2gig+, in fact it crashes on my system
samcv hmm i will look at that then 17:27
dogbert17 samcv: example here: gist.github.com/dogbert17/1181b513...a394bd16a0
samcv do you see what is using it?
RabidGravy Oooh when did "Unsupported use of \b; in Perl 6 please use <|w> for word boundary" come in?
yoleaux 22 Dec 2016 23:57Z <notviki> RabidGravy: buggable: eco pastebin::gist
RabidGravy notviki++ nice 17:28
moritz RabidGravy: it was a recent NQP patch, I think 17:29
commit c5f38888ecf1053eab6d9aa29086e23e0d3f0e83
Author: Zoffix Znet [email@hidden.address]
Date: Sun Dec 4 02:46:19 2016 -0500
Forbid bare \b and \B in regexes
17:29 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'SASSify all CSS 17:29
travis-ci.org/perl6/doc/builds/186382979 github.com/perl6/doc/compare/3b167...fb4b3bf98e
17:29 travis-ci left
RabidGravy I'm cool with it but I'd basically copied some P5 regex for something and munged tell they worked, and that worked last time I tested it 17:30
notviki RabidGravy: it didn't work tho. In P5 it's a word boundary but in P6 it's a backspace char
RabidGravy: we found a bug you missed.
You're welcome.
17:31 grondilu left
RabidGravy the strange thing is that it actually worked, I guess my coverage wasn't quite as good as I thought it was 17:33
notviki :)
17:33 grondilu joined
notviki .tell [Coke] FWIW "var OK_distance = Math.min(max_distance, len -1);" is useless since the conditional above it guarantees len is > 2 so it won't be less than max_distance there. Dunno if this helps you debug whether the fuzzy search works 17:34
yoleaux notviki: I'll pass your message to [Coke].
samcv dogbert17, running now. looks like i'm up to 1GB for moarvm
node is using 41MB
moarvm memory keeps going up but node is staying constant 17:35
idk what the issue is. could be proc::async leaking? idk
17:38 grondilu_ joined, grondilu left
notviki .tell [Coke] oh, I guess len can be == 2 and then len -1 is 1.... never mind :) 17:39
yoleaux notviki: I'll pass your message to [Coke].
17:41 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Remove TOC hide/show feature 17:41
travis-ci.org/perl6/doc/builds/186385140 github.com/perl6/doc/compare/54fb4...930aed0d4c
17:41 travis-ci left 17:45 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Merge sift4 script into search.js script 17:45
travis-ci.org/perl6/doc/builds/186387010 github.com/perl6/doc/compare/06930...f541fb0885
17:45 travis-ci left
notviki *sigh* travis should really improve their bot 17:53
17:54 FROGGS joined
notviki build passed, so stfu now. 17:54
especially with the joins and parts 17:55
notviki shakes fist at cloud
RabidGravy RAAAAR!
17:58 Ven joined, cyphase left
notviki Commencing holday celebrations in 3... 2... 1... 18:00
\o/
notviki relocates
dogbert17 samcv: there could definitely be a memory leak in there somewhere 18:01
18:04 ufobat left, cyphase joined 18:11 zakharyas left 18:20 bpmedley joined 18:42 cyphase left 18:45 wamba left 18:47 cyphase joined 18:51 pierre_ joined, raiph left, pierre_ left 18:52 pierre_ joined, dugword joined 18:57 cdg joined 18:59 madjestic left 19:02 telex left 19:03 cdg left 19:04 telex joined 19:07 skids left 19:12 baest_ joined 19:13 dataf3l joined, baest left 19:17 domidumont left
pmurias is someone working on rust interop? 19:23
19:23 madjestic joined
arnsholt Not to my knowledge 19:24
notviki I was gonna... but got sidetracked as always. 19:25
arnsholt Might be possible to jury-rig something from NativeCall though. How different is the code generated from C conventions?
notviki Well, you can tell rust to compile a C lib and use that from NativeCall 19:26
Though the approach looked somewhat fragile last time someone tried that in the channel.... we concluded rust's ownership was causing issues
Like, if you touch owned things from Perl 6 it segfaults instead of politely refusing to work. 19:27
mst pmurias: somebody's done rust-for-XS
notviki :o
mst pmurias: so you can always go via Inline::Perl5
(what could possibly go wrong ...)
19:29 xinming left 19:30 pierre_ left 19:31 xinming joined 19:38 darutoko left 19:39 TEttinger joined
notviki TIL 1, 2, { $^a + $^b } … receives 1 in $^a and 2 in $^b as opposed to in reverse 19:44
19:44 labster joined 19:45 lichtkind joined, lichtkind__ left, lichtkind left 19:46 lichtkind joined 19:49 djbkd joined 19:50 ufobat joined
RabidGravy I hate it when things don't segfault when you're trying to debug why they segfault 19:55
20:03 cyphase left
notviki m: my @G = 0; @G = 0, {++$ - @G[@G[$++]]} … ∞; say @G[^30] 20:05
camelia rakudo-moar f0398f: OUTPUT«(0 1 1 2 3 3 4 4 5 6 6 7 8 8 9 9 10 11 11 12 12 13 14 14 15 16 16 17 17 18)␤»
notviki <3 Perl 6
en.wikipedia.org/wiki/Hofstadter_s...G_sequence 20:06
20:06 FROGGS left
notviki m: my @G = 0, {++$ - @G[@G[$++]]} … ∞; say @G[^30] 20:07
camelia rakudo-moar f0398f: OUTPUT«(0 1 1 2 3 3 4 4 5 6 6 7 8 8 9 9 10 11 11 12 12 13 14 14 15 16 16 17 17 18)␤»
notviki :o this works
20:08 cyphase joined 20:29 Buliarous joined, lichtkind_ joined 20:30 Buliarous left
pmurias mst: Inline::Perl5 is a solution to almost everything, we should have a bundled Star+Perl5 offering or something of that sort ;) 20:31
20:33 lichtkind left 20:34 bjz joined
labster m: say *.WHAT; say *.^methods; 20:46
camelia rakudo-moar f0398f: OUTPUT«(Whatever)␤{ ... }␤»
rindolf RabidGravy: a Heisenbug! 20:48
notviki m: say *.WHAT; say WHAT *.^methods; 20:49
camelia rakudo-moar f0398f: OUTPUT«(Whatever)␤(WhateverCode)␤»
notviki (WHAT is not a true method, hence the difference between the two)
labster I thought that the .^ operator acted differently here, guess this is the way to do it: 20:50
m: say *.HOW.methods(*) 20:51
camelia rakudo-moar f0398f: OUTPUT«(ACCEPTS perl Str)␤»
moritz my $w = *; say $w.^methods 20:52
m: my $w = *; say $w.^methods
camelia rakudo-moar f0398f: OUTPUT«(ACCEPTS perl Str)␤»
labster m: *.VAR.^methods
camelia ( no output )
labster m: *.VAR.^methods.say 20:53
camelia rakudo-moar f0398f: OUTPUT«(ACCEPTS perl Str)␤»
labster I don't know why that works, but it does.
20:55 pierre__ joined 21:01 rindolf left 21:05 pierre__ left 21:08 labster left 21:09 rindolf joined, labster joined 21:11 andrzejku left 21:14 teksteiner left
mr_ron m: use Test; my $x = 1; is(do {my $x = 3; $x}, 3, "my"); is(do {temp $x = 4; $x}, 4, "temp"); is(do {let $x = 5; $x}, 5, "let") 21:15
camelia rakudo-moar f0398f: OUTPUT«ok 1 - my␤not ok 2 - temp␤␤# Failed test 'temp'␤# at <tmp> line 1␤# expected: '4'␤# got: '1'␤ok 3 - let␤»
moritz m: use Test; my $x = 1; is(do {temp $x = 4; +$x}, 4, "temp") 21:17
camelia rakudo-moar f0398f: OUTPUT«ok 1 - temp␤»
moritz mr_ron: the do block doesn't decontainerize its return value 21:18
so it doesn't return 4, but the variable $x, which is (correctly) reset to its original value immediately after block exit
mr_ron I sort of understand but wonder if there might not be some ambiguity. Is the evaluated $x "my $x" or "temp $x" and why? 21:21
Wait ... starting to understand ... 21:24
21:24 Axord joined
moritz there is only one $x in that example 21:24
temp just temporary replaces the value
mr_ron m: use Test; my $x = 1; is(do {temp $x = 4; $x++}, 4, "temp"); say $x 21:28
camelia rakudo-moar f0398f: OUTPUT«ok 1 - temp␤1␤»
mr_ron m: use Test; my $x = 1; is(do {temp $x = 4; $}, 4, "temp"); say $x 21:31
camelia rakudo-moar f0398f: OUTPUT«not ok 1 - temp␤␤# Failed test 'temp'␤# at <tmp> line 1␤# expected: '4'␤# got: (Any)␤1␤»
mr_ron Not quite understanding why the two cases return different $x 21:32
m: use Test; my $x = 1; is(do {temp $x = 4; $x}, 4, "temp"); say $x
camelia rakudo-moar f0398f: OUTPUT«not ok 1 - temp␤␤# Failed test 'temp'␤# at <tmp> line 1␤# expected: '4'␤# got: '1'␤1␤»
21:32 bjz left
moritz ++ creates a copy 21:32
so $x++ doesn't return $x
it returns a copy of the value in $x
21:32 djbkd left
mr_ron A little unintuitive to me but understandably correct with explanation ... thank you 21:34
21:37 hankache joined
moritz it's unintuitive because you write to and read from the same variable within one expression 21:38
even if no temp is involved, that can be very surprising
in C that is even undefined behavior
21:38 labster left 21:41 domidumont joined 21:42 djbkd joined, djbkd left 21:44 Ven left 21:48 pyrimidine left 21:51 hankache left 21:54 dugword left 21:56 domidumont left 21:57 perlpilot left 21:59 cyphase left 22:01 Axord left 22:03 cyphase joined
mr_ron m: my @a = <1 3>; my @b = do {temp @a; @a.push(5)}; @b.say; @b = do { @a.push(5) }; @b.say 22:04
camelia rakudo-moar fc599d: OUTPUT«[1 3]␤[1 3 5]␤»
22:05 rindolf left 22:06 djbkd joined 22:10 Tonik joined 22:16 lukaramu joined 22:23 labster joined 22:25 bjz joined 22:26 Rawriful joined 22:33 BenGoldberg joined 22:35 bwisti left 22:36 Ben_Goldberg joined, BenGoldberg left 22:37 Ben_Goldberg is now known as BenGoldberg
pmurias mr_ron: $x++ must create a copy because expresssion first do the side effects and then return a value 22:45
22:45 pyrimidine joined 22:50 bjz left 22:55 TEttinger left 22:58 geekosaur left 23:00 geekosaur joined 23:01 TEttinger joined
samcv scroll down to the end docs.perl6.org/language/quoting#Heredocs:_:to 23:01
how am i supposed to highlight two heredocs starting on the same line?
er and what does that even do. do they overlap?
oh i guess not overlapping. aaaah this is my worst nightmare for how to highlight XD 23:03
notviki m: gist.github.com/zoffixznet/ec3755e...6af8621e19
camelia rakudo-moar b306de: OUTPUT«["FIRST\nMULTILINE\nSTRING\n", "SECOND\nMULTILINE\nSTRING\n"]␤»
notviki
.oO( why do we even have that atrocity in the docs... )
 23:04
23:05 kurahaupo joined
notviki And I bet that's not spectested. 23:05
samcv :D
it's kind of ugly tbh. just
i first thought maybe they overlapped or something one ending one place the other another place
but them not overlapping is ever weirder 23:06
23:06 killbill joined 23:07 AlexDaniel joined
samcv notviki, well design.perl6.org/S02.html#Heredocs the S02 has it… not sure how i'm going to highlight this. maybe it's not possible. unless i like… idk it will be hard 23:14
notviki samcv: I'd ignore that case. Anyone stupid enough to write such code deserves LTA highlighting :)
samcv :)
notviki Is there a reverse of `is pure`? 23:19
m: multi infix:<-> (Str $, Str $) { say "hi" }; say "start"; quietly "x" - "x" 23:20
camelia rakudo-moar b306de: OUTPUT«hi␤start␤»
timotimo if you don't put "is pure" on it, it won't be pure
but i'd expect code that checks for purity to see if there's a method for pureness and whether that method returns 1 or not
notviki Well, what if there an is pure on the proto? Can I mark one multi as not pure?
Basically, our proto for infix:<-> is pure, and that causes any of the custom user's infix:<->'s to be constant folded as well. 23:21
And (as above) produce undesirable results).
m: proto infix:<->(|) {*}; multi infix:<-> (Str $, Str $) { say "hi" }; say "start"; quietly "x" - "x"; say 42 - 42 23:23
camelia rakudo-moar b306de: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Calling infix:<->(Int, Int) will never work with any of these multi signatures:␤ (Str, Str)␤at <tmp>:1␤------> 3 say "start"; quietly "x" - "x"; say 42 7⏏5- 42␤»
dalek c: 1ba632e | samcv++ | doc/Language/regexes.pod6:
Trigger doc rebuild to pull in highlighter fixes

Pulling in fixes for the highlighting of the ==> operator
synopsebot6 Link: doc.perl6.org/language/regexes
notviki m: proto infix:<->(|) {*}; multi infix:<-> (|c) {CORE::("&infix:<->")(|c)} ;multi infix:<-> (Str $, Str $) { say "hi" }; say "start"; quietly "x" - "x"; say 42 - 42 23:27
camelia rakudo-moar b306de: OUTPUT«start␤hi␤0␤»
notviki At least there's a way to work aroun dit.
Well, if the docs have a section bragging about how you can expand core ops... it should mention the constant-foldiness 23:29
timotimo we can have a second look before doing the constant folding to see if the proto is in the core setting but the candidate we're trying isn't 23:43
notviki oh, cool 23:44
23:45 quietfanatic joined 23:50 skids joined 23:58 grondilu joined
samcv \O/ they just merged the updated perl 6 highlighter for github's linguist :D 23:59
notviki Wooooo \o/
samcv++

Powered by Cro and the Raku® Programming Language.

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

Thank you!