»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.
squashable6 🍕 JJ++ wrote a comment on “Vow not documented”: github.com/perl6/doc/issues/588#is...-370189825 00:01
00:21 Xliff joined
Xliff o 00:21
\o, even
00:27 comborico1611 left 00:29 greppable6 joined
timotimo º 00:36
00:58 mniip left 01:00 mingdao joined 01:15 jimav left 01:20 pierre_ joined 01:25 konsolebox left, konsolebox joined 01:30 konsolebox left 01:32 konsolebox joined 01:43 pierre_ left 01:56 konsolebox left 01:57 konsolebox joined 02:06 quester joined, quester left, quester joined 02:20 Rawriful left
cfa ah, looks like zoffix fixed the .contains issue 02:23
02:26 konsolebox left 02:29 konsolebox joined 02:34 rindolf joined 02:35 Xliff left 02:36 konsolebox left 02:37 konsolebox joined 02:43 regina joined, kaare_ left 02:44 ilbot3 left 02:56 ilbot3 joined, ChanServ sets mode: +v ilbot3 03:02 konsolebox left 03:03 konsolebox joined 03:08 kaare_ joined 03:14 kurahaupo left 03:15 konsolebox left 03:22 konsolebox joined 03:26 konsolebox left
rouking Zoffix: grant proposal looks great. I'm excited to see FatRats for high-precision literals. What are some examples of constants' behaviour being poorly defined? 03:26
03:28 konsolebox joined
rouking Nevermind, I see. 03:30
03:36 pharv joined 03:47 ufobat___ joined 03:50 ufobat_ left 04:06 wamba joined 04:10 konsolebox left 04:11 konsolebox joined 04:16 kurahaupo joined 04:18 konsolebox left 04:19 konsolebox joined
ZzZombo Aw dammit, grammar blocks my progress again! github.com/rakudo/rakudo/blob/5217....nqp#L2367 explicitly forbids anything other than roles to be parameterizable. 04:19
Why oh why??
and since there is also no `COMPOSE` for `EXPORTHOW`, I'm not gonna fiddle with the stock roles. 04:20
A separate declarator was my safe bet until this moment. 04:21
04:22 martinp2 joined, Actualeyes joined
ZzZombo m: dd $*W 04:30
camelia &CORE::infix:<orelse>(Failure.new(exception => X::Dynamic::NotFound.new(name => "\$*W"), backtrace => Backtrace.new), *.self)
ZzZombo m: say $*W
camelia Dynamic variable $*W not found
in block <unit> at <tmp> line 1
ZzZombo yeah, why the &CORE::infix:<orelse>?
04:31 konsolebox left 04:33 konsolebox joined 04:35 martinp2 left 04:42 konsolebox left 04:43 konsolebox joined 04:48 konsolebox left 04:50 konsolebox joined 04:58 kurahaupo_ joined, regina left 05:00 kurahaupo left 05:05 khw left 05:09 kurahaupo_ left 05:10 kurahaupo joined 05:11 pharv left 05:12 pharv joined 05:17 pharv left 05:23 shinobi-cl left 05:42 jeromelanteri joined 05:46 tt left 05:47 Xliff joined, jeromelanteri left 05:49 tt joined 05:52 quester left 05:55 pierre_ joined 06:22 perlbot left, simcop2387 left 06:26 simcop2387 joined, simcop2387 left, simcop2387 joined 06:27 perlbot joined 06:28 perlbot left, simcop2387 left 06:37 simcop2387 joined 06:40 perlbot joined 06:42 entonian joined 06:48 entonian left 07:07 darutoko joined 07:08 BenGoldberg left, leah2 left, ufobat___ left 07:15 ufobat___ joined 07:25 leah2 joined 07:29 wamba left
Geth doc: finanalyst++ created pull request #1811:
c function tutorial added
07:37
squashable6 🍕 finanalyst++ opened pull request “c function tutorial added”: github.com/perl6/doc/pull/1811
🍕🍕🍕 First contribution by finanalyst++! ♥
masak finanalyst++ # nice to see you again! 07:56
08:07 wamba joined 08:30 AlexDaniel joined
AlexDaniel squashable6: status 08:34
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈3 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/ff8860ce8033abcbd1...6fb34de1e4
AlexDaniel huggable: doc build 08:36
huggable AlexDaniel, nothing found
AlexDaniel what was the link for the build log 08:37
huggable: doc build :is: docs.perl6.org/build-log/
huggable AlexDaniel, Added doc build as docs.perl6.org/build-log/
AlexDaniel This type cannot unbox to a native string: P6opaque, Failure
docs.perl6.org/build-log/build-201...+00:00.log
uh?
it's using old rakudo version so that is not an issue 08:38
like, it's not a rakudo bug
can it be because of github.com/perl6/doc/commit/d468b6...0d8346541e 08:43
squashable6 🍕 AlexDaniel++ wrote a comment on “c function tutorial added”: github.com/perl6/doc/pull/1811#iss...-370212347 08:48
🍕 AlexDaniel++ wrote a comment on “c function tutorial added”: github.com/perl6/doc/pull/1811#iss...-370212425 08:49
Geth doc: 83ddea53ec | (Richard Hainsworth)++ (committed using GitHub Web editor) | doc/Language/nativecall.pod6
c function tutorial added
08:50
synopsebot Link: doc.perl6.org/language/nativecall
doc: 4c543c4fa1 | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | doc/Language/nativecall.pod6
Merge pull request #1811 from finanalyst/master

c function tutorial added
squashable6 🍕 AlexDaniel++ merged pull request “c function tutorial added”: github.com/perl6/doc/pull/1811
squashable6 🍕 AlexDaniel++ wrote a comment on “RAKUDO_ specific environ…”: github.com/perl6/doc/issues/990#is...-370212954 08:59
🍕 AlexDaniel++ closed issue “RAKUDO_ specific environ…”: github.com/perl6/doc/issues/990
09:01 sena_kun joined
squashable6 🍕 AlexDaniel++ opened issue “Doc build is broken”: github.com/perl6/doc/issues/1812 09:02
🍕 AlexDaniel++ labeled issue “Doc build is broken”: github.com/perl6/doc/issues/1812
🍕 AlexDaniel++ edited issue “Doc build is broken”: github.com/perl6/doc/issues/1812
🍕 AlexDaniel++ edited issue “Doc build is broken”: github.com/perl6/doc/issues/1812 09:03
AlexDaniel [Coke]: help? 09:04
I can probably revert the change and it'll fix itself, maybe
but why is it happening at all?
09:05 Xliff left
squashable6 🍕 AlexDaniel++ wrote a comment on “say vs put vs dd”: github.com/perl6/doc/issues/1802#i...-370213389 09:06
🍕 AlexDaniel++ closed issue “say vs put vs dd”: github.com/perl6/doc/issues/1802
09:08 trnh joined 09:10 Xliff joined 09:28 noganex left
Geth doc: 9b60680b1c | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/faq.pod6
Circular dependencies (FAQ)

Resolves #1419.
I sorta stiched together various thoughts from the IRC discussion. Not sure if any of this is making sense, please review and change accordingly.
09:32
synopsebot Link: doc.perl6.org/language/faq
squashable6 🍕 AlexDaniel++ closed issue “Circular dependencies (FAQ?)”: github.com/perl6/doc/issues/1419
09:34 noganex joined 09:40 noganex left 09:48 trnh left 09:53 trnh joined 09:55 espadrine joined 09:59 trnh left 10:04 noganex joined 10:07 trnh joined
Geth doc: 562ae50469 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Type/Cool.pod6
Explain what “grapheme” is, give more examples

Resolves #1403, or so I think. Another page where this can be explained is language/unicode, but I think most users will wonder what graphemes are when stumbling upon the description of `chars`. Maybe it can be simply copied over to language/unicode.
10:08
synopsebot Link: doc.perl6.org/type/Cool
squashable6 🍕 AlexDaniel++ closed issue “grapheme is not searchable”: github.com/perl6/doc/issues/1403
AlexDaniel .tell samcv plz review github.com/perl6/doc/commit/562ae50469
yoleaux AlexDaniel: I'll pass your message to samcv.
samcv AlexDaniel: looking 10:09
yoleaux 10:08Z <AlexDaniel> samcv: plz review github.com/perl6/doc/commit/562ae50469
samcv AlexDaniel: I think you make it too complex in the first paragraph. It would make more sense if you explain how graphemes are user visible characters 10:10
because going into Prepend Extend etc most people are going to be like: "what am i reading. all i wanted to know is what graphemes are"
so it should be written (at least the first paragraph), assuming the reader has no clue what graphemes are 10:11
or even just add one or two sentences before the more technical unicode stuff. but I think most people will understand user visible characters rather than codepoints can be different but may not understand extend/prepend and what that means 10:12
TEttinger what's a character, like pikachu?
samcv TEttinger: not yet anyway. or ever i guess. since Unicode doesn't allow trademarked things 10:13
TEttinger trademarks expire eventually 10:14
in the postapocalyptic unicode emoji wasteland... there are only branded emoji
10:15 pmurias joined
Geth doc: c0159cac65 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Type/Cool.pod6
Extend the description of graphemes

Adds a ramp to make the techincal details more approachable. samcv++
Discussion: irclog.perlgeek.de/perl6/2018-03-04#i_15882036
10:16
AlexDaniel TEttinger: trademarks potentially last forever :P
pmurias the new lobster emoji seems like a fine substitute for the pikachu character: www.xkcd.com/1953/
AlexDaniel unlike patents and copyright that have somewhat… “reasonable”… time limits 10:17
samcv: I don't know if that helps ↑
samcv AlexDaniel: much better. maybe not perfect but I feel much better now :) 10:22
AlexDaniel squashable6: status 10:32
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈1 hour. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/54b7cb7b685e033641...68f39eb67f
10:33 trnh left 10:42 pharv joined
squashable6 🍕 AlexDaniel++ wrote a comment on “Reduce travis test time”: github.com/perl6/doc/issues/1765#i...-370218906 10:42
🍕 AlexDaniel++ wrote a comment on “Next perl6/doc squashath…”: github.com/perl6/doc/issues/1778#i...-370218959 10:43
🍕 AlexDaniel++ closed issue “Next perl6/doc squashath…”: github.com/perl6/doc/issues/1778
🍕 AlexDaniel++ labeled issue “How to document rakudo bugs?”: github.com/perl6/doc/issues/1486 10:44
🍕 AlexDaniel++ wrote a comment on “How to document rakudo bugs?”: github.com/perl6/doc/issues/1486#i...-370219078 10:45
10:46 JJ joined 10:47 pharv left
AlexDaniel squashable6: status 10:53
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈1 hour. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/30a9bb8976769236d3...8e3734b81c
10:53 pharv joined 10:56 quotable6 left
JJ I'm going to keep working on broken links. There are literally thousands of them. Any help will be appreciated 10:56
10:56 quotable6 joined
moritz if there are thousands of broken links, then it stands to reason that some link generation in the HTML generator is broken 10:57
JJ I'm working from the top of the file down, if someone wants to pitch in, please do it from the bottom up. Or just warn. Anyone wants some clues on how to to proceed, ping me. I'm @jjmerelo in Telegram
AlexDaniel there are not many 404-s though
and these are the actual broken links, everything else is poor html generation 10:58
which **really** needs some fixing
like, some hero to fix it :)
JJ @moritz @alexdaniel I'm working on the 404s. However, some 200 are also kinda broken since they point to fragments and might have some underlying problem.
Like they point to a section that has disappeared and so on. Anyway, 404s are priorities.
AlexDaniel squashable6: status 10:59
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈1 hour. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/5620b5044f158c436d...ba84e44d46
JJ @AlexDaniel a whole posse of heroes, probably :-)
AlexDaniel moritz: by any chance do you have any clue on how we can unbreak the doc build on docs.perl6.org ? 11:00
moritz: github.com/perl6/doc/issues/1812
CI is very green, locally it's fine 11:01
samcv: someone assigned you on github.com/perl6/doc/issues/1517 11:02
samcv: can you please either fix it or unassign yourself?
moritz docs.perl6.org/build-log/build-201...+00:00.log
AlexDaniel: should I try installing a newer rakudo? 11:03
AlexDaniel moritz: it shouldn't matter, but that won't hurt
moritz This is Rakudo version 2017.11-78-g1dbf5f589 built on MoarVM version 2017.11-27-gfac29990
AlexDaniel it feels like it needs a make clean or something… not sure
moritz I'll try that first
samcv AlexDaniel: thanks i'll look. there must be a page github has to show you things you are assigned to? i should probably find it
AlexDaniel samcv: github.com/issues/assigned 11:04
squashable6 🍕 samcv++ wrote a comment on “Document unicode normalization”: github.com/perl6/doc/issues/1517#i...-370220151 11:05
🍕🍕🍕 First contribution by samcv++! ♥
samcv i have to go to bed now, but I left a comment, if you can reply. thanks AlexDaniel :)
i'll work on it tomorrow when i wake up
JJ I don't get what's wrong with this L<LEAVE|/language/phasers#LEAVE>? Why does it get turned into docs.perl6.org/type/$SOLIDUSphasers? 11:06
squashable6 🍕 AlexDaniel++ wrote a comment on “Document unicode normalization”: github.com/perl6/doc/issues/1517#i...-370220408 11:09
11:09 kaare__ joined, kaare_ left
Geth doc: 79866964d8 | (JJ Merelo)++ | doc/Type/Any.pod6
More fixes refs #561
11:11
synopsebot Link: doc.perl6.org/type/Any
JJ Since some paragraphs are actually included in many different pages, there are actually not so many 404s. I think this should be done at the repo document level, not at the website level. Maybe add some test? 11:16
11:17 kaare_ joined, kaare__ left
AlexDaniel JJ: like github.com/perl6/doc/issues/1433 ? 11:18
although I'm not sure how it works 11:19
JJ: fwiw there's also a “debug” mode on the website, but I don't know if it works
JJ: if you scroll to the bottom of any page it says "[Debug: off]" 11:20
JJ @alexdaniel right. And I commented on it. Sorry, I've been all night (or a short TZ-jumping version of it) on a flight, I don't have all my wits about me
@AlexDaniel every 404 is a different issue... Plus I just realized the link to the doc at the top might be outdated. 11:21
AlexDaniel click on that and it'll turn it on, IIRC it checks for broken links
JJ @AlexDaniel gotta do that for every different page, right? 11:22
AlexDaniel JJ: no, just turn it on once and it'll stay on
for example, if you go to docs.perl6.org/language/5to6-perlfunc
it says:
Broken link: 404 for docs.perl6.org/type/printf found. Please report at irc.freenode.net#perl6
hm… it no longer says that for me 11:24
anyway, it kinda works! :)
Geth doc: b81387c281 | (JJ Merelo)++ | doc/Language/operators.pod6
StrDistance is not actually documented refs #691

But eliminates broken link #561
synopsebot Link: doc.perl6.org/language/operators
JJ @AlexDaniel to find out the broken links I'm using this checklink -b -D 25 -q doc.perl6.org | tee "$(date '+%F')" That's not so big a deal... Problem is broken links keep being added and every one poses a different problem, like last one. 11:26
AlexDaniel yea
squashable6 🍕 JJ++ opened issue “X::Attribute::Required is not documented”: github.com/perl6/doc/issues/1813 11:30
AlexDaniel moritz: o! It worked! 11:34
moritz: thank you very much!
Geth doc: baf0c72dc0 | (JJ Merelo)++ | doc/Type/Signature.pod6
Adds link to ACCEPTS.

Refs #561
synopsebot Link: doc.perl6.org/type/Signature
squashable6 🍕 AlexDaniel++ wrote a comment on “Doc build is broken”: github.com/perl6/doc/issues/1812#i...-370221965 11:35
🍕 AlexDaniel++ closed issue “Doc build is broken”: github.com/perl6/doc/issues/1812
AlexDaniel JJ: fwiw broken link to TWEAK here: docs.perl6.org/language/traps#BUIL..._arguments 11:36
11:37 pierre_ left
JJ @AlexDaniel thanks, I'll try that next... 11:38
AlexDaniel hm I don't know if 👩‍👧‍👦🏿 is a reasonable example 11:39
11:39 kaare_ left
AlexDaniel huggable: clog 11:39
huggable AlexDaniel, nothing found
AlexDaniel huggable: log
huggable AlexDaniel, Comes out 2.41x faster than core .log10 and can handle Ints larger than 1.8e308: gist.github.com/zoffixznet/2539e8b...43e173fe8f
AlexDaniel huggable: irc log
huggable AlexDaniel, nothing found
AlexDaniel huggable: irclog
huggable AlexDaniel, nothing found
11:40 kaare_ joined, pmurias left
AlexDaniel chromium renders all characters separately… which is totally broken 11:41
and firefox renders white family with a skin tone modifier separately
I guess there's no point in having an example that can't be rendered by anything? 11:42
it's still 1 character according to perl 6 though… 11:43
(re docs.perl6.org/type/Cool#index-entry-Grapheme )
11:44 pharv left
AlexDaniel I think <pre> tag breaks proper rendering of characters 11:44
Geth doc: 1d3a0f3840 | (JJ Merelo)++ | doc/Language/concurrency.pod6
That Vow is defined in the example

Refs #561
synopsebot Link: doc.perl6.org/language/concurrency
doc: b385e85a23 | (JJ Merelo)++ | doc/Language/traps.pod6
Adds link to TWEAK

Refs #561 Reported by @AlexDaniel
synopsebot Link: doc.perl6.org/language/traps
11:45 pharv joined
JJ squashable6: status 11:46
squashable6 JJ, 🍕🍕 SQUASHathon is in progress! The end of the event is in 13 minutes. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
JJ, Log and stats: gist.github.com/b085dfb99c0e4c0832...ff6c195b9a
JJ OK; one more link to go
11:46 konsolebox left 11:47 pmurias joined
squashable6 🍕 AlexDaniel++ labeled issue “Convert static site to dynamic site”: github.com/perl6/doc/issues/1246 11:48
11:49 pharv left
squashable6 🍕 JJ++ opened issue “Thread::Instrument::Usag…”: github.com/perl6/doc/issues/1814 11:51
11:52 konsolebox joined
moritz AlexDaniel: doc build seems to be fixed 11:52
AlexDaniel I know! already closed the issue! :) thank you 11:53
moritz yay, old-fashioned me is still good for something :)
you're welcome
Geth doc: 5289381371 | (JJ Merelo)++ | doc/Language/glossary.pod6
Routine not syntax refs #561
11:57
synopsebot Link: doc.perl6.org/language/glossary
doc: 15f03be788 | (JJ Merelo)++ | doc/Type/Lock.pod6
URL had moved refs #561
synopsebot Link: doc.perl6.org/type/Lock
doc: 8f782921dc | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Type/IO/Handle.pod6
Correct note about file handles

Resolves #1561.
synopsebot Link: doc.perl6.org/type/IO::Handle
squashable6 🍕 AlexDaniel++ closed issue “Unclosed handles on exit”: github.com/perl6/doc/issues/1561
JJ squashable6: status 11:58
squashable6 JJ, 🍕🍕 SQUASHathon is in progress! The end of the event is in 1 minute. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
JJ, Log and stats: gist.github.com/028a3c756d157090ba...69bba00194
Geth doc: c1865ccea3 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Type/IO/Handle.pod6
Make the note clearer

Otherwise people might think that closing anything is not needed at all. Issue #1561.
12:01
AlexDaniel squashable6: status
squashable6 AlexDaniel, Next SQUASHathon in 32 days and ≈21 hours (2018-04-07 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel squashable6: log 2018-03-03
squashable6 AlexDaniel, Log and stats: gist.github.com/7fe03d8bcf694750ea...30afa7650c
JJ OK, people, I'm off now. Will continue working on broken links in the next few days... First on the test, probably. 12:02
AlexDaniel JJ: thanks! See you ;) 12:03
JJ Thanks to you all! It's been a pleasure (in a weird kind of way). Cheers! 12:04
AlexDaniel I find it amazing that 17 people touched the doc during the squashathon 12:08
12:08 trnh joined
AlexDaniel yes obviously only a few of them were actually closing tickets and committing, but 17 people is a lot IMO 12:08
12:08 JJ left
AlexDaniel lol and I need some English lessons to work on the doc repo properly :P 12:10
Geth doc: 1ffc90ee81 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/faq.pod6
Another solution for circular dependencies

Discussion: irclog.perlgeek.de/perl6-dev/2018-...i_15882396
12:21
synopsebot Link: doc.perl6.org/language/faq
DrForr Yow. Like the new moules.perl6.org dist/ directory look 12:29
*modules 12:30
12:35 Actualeyes left 12:40 eliasr joined 12:46 natrys joined
DrForr There's a slight difference between cpan-upload's .pause and App::Mi6's .pause format - 'username' vs 'user' in App::Mi6. 12:50
tbrowder .tell AlexDaniel your English is as good as most native-English speakers whose articles I read online these days, and better than many! 12:52
yoleaux tbrowder: I'll pass your message to AlexDaniel.
DrForr Or maybe something else is broken, as using neither 'user' nor 'username' for the entry seem to work. I'll poke around. 12:53
12:54 ufobat_ joined 12:57 ufobat___ left 13:21 pierre_ joined 13:22 Kaiepi left 13:35 shinobi-cl joined 13:42 aborazmeh joined, aborazmeh left, aborazmeh joined
Ulti m: say +lines().grep(/the/) 13:44
yoleaux 22 Feb 2018 23:47Z <Zoffix> Ulti: you have a PR unbreaking module listing: github.com/MattOates/Stats/pull/6
camelia 0
Ulti m: say +lines().race.grep(/the/) 13:45
camelia A worker in a parallel iteration (hyper or race) initiated here:
in block <unit> at <tmp> line 1

Died at:
No such method '!cursor_start' for invocant of type 'Str'
Ulti ^ whats this about
moritz looks like left-over debugging code, plus a bug 14:03
14:05 aborazmeh left 14:10 konsolebox left 14:11 konsolebox joined
timotimo i don't think that's left-over, that looks like a deliberate helpful thing 14:19
m: say +lines().race.grep(/the/); CATCH { .perl.say } 14:20
camelia X::Method::NotFound+{X::HyperRace::Died}.new(start-backtrace => Backtrace.new, invocant => "»Wann treffen wir drei wieder zusamm?«", method => "!cursor_start", typename => "Str", private => Bool::False)
A worker in a parallel iteration (hyper or r…
timotimo m: say +lines().race.grep(/the/); CATCH { default { .perl.say } }
camelia X::Method::NotFound+{X::HyperRace::Died}.new(start-backtrace => Backtrace.new, invocant => "»Wann treffen wir drei wieder zusamm?«", method => "!cursor_start", typename => "Str", private => Bool::False)
timotimo as you can see it's a proper exception
14:21 konsolebox left 14:23 konsolebox joined
timotimo or do you mean why does that explode? 14:29
cursor_start is supposed to be called on the regex, not on the string, right?
MasterDuke it dies in this loop github.com/rakudo/rakudo/blob/mast...m6#L41-L44 at line 43 14:31
timotimo m: /foo/("hello") 14:32
camelia No such method '!cursor_start' for invocant of type 'Str'
in block <unit> at <tmp> line 1
timotimo right, it's not supposed to call the matcher if it's a regex
who wants to fix it? :) 14:34
14:45 Actualeyes joined 14:57 trnh left
rindolf timotimo: hi, sup? 15:02
MasterDuke: hi
MasterDuke hey 15:03
15:03 trnh joined
rindolf MasterDuke: sup? 15:04
timotimo yo rindolf
rindolf did you get to the bottom of the wrong result problem? 15:05
timotimo i didn't investigate further, sorry :(
MasterDuke rindolf: i'm trying to figure out how make native array accesses use native candidates 15:07
15:08 zakharyas joined, pierre_ left
rindolf timotimo: ah 15:09
15:09 kaare__ joined
rindolf timotimo: where is the offending code? 15:09
15:09 pierre_ joined, kaare_ left
MasterDuke that would (almost) mean you could just stick an 'int' in front of your variables and reduce the runtime to a quarter 15:09
15:10 zakharyas left
rindolf MasterDuke: gl, and thanks 15:10
15:11 zakharyas joined
MasterDuke np, hopefully it'll help other code also 15:11
15:14 pierre_ left 15:18 Kaiepi joined, kaare__ left 15:22 trnh left 15:24 trnh joined
buggable New CPAN upload: CommonMark-0.0.5.tar.gz by JGOFF cpan.metacpan.org/authors/id/J/JG/...0.5.tar.gz 15:24
15:25 kaare__ joined, xtreak joined 15:34 trnh left 15:37 trnh joined 15:40 sharksoft joined
rindolf timotimo: where is the offending code? 15:41
15:41 skids joined
rindolf timotimo: i'd like to debug it 15:41
timotimo gotta find it in the irclog
15:42 trnh left
timotimo ok, gotta edit it again i think? 15:42
rindolf timotimo: ah 15:44
15:44 sharksoft left 15:45 sharksoft joined
timotimo oh i need to get primesieve again 15:46
rindolf timotimo: i wrote a version wo needing primesieve 15:47
15:47 jimav_ left
timotimo i don't have that any more :D 15:47
rindolf timotimo: ah 15:48
timotimo: see paste.debian.net/1013068/ 15:49
timotimo um, huh 15:50
i don't see the improper output any more
4181 0 15:51
ret = 5468808766 ; 468808766 15:52
is this correct for the last two lines of output?
for 2..19
rindolf timotimo: let me see 15:53
timotimo gist.github.com/timo/c6f33191957d5...978b50de7e 15:54
rindolf ret = 5468808766 ; 468808766 15:55
15:58 khw joined 15:59 trnh joined, sharksoft left 16:03 trnh left 16:05 zakharyas left
rindolf timotimo: you used ./primesieve which tripped me 16:07
timotimo oh, right
i just extracted it next to the script in /tmp
16:10 zakharyas joined
cfa morning all 16:12
so, silly question: i gather type smiley :D stands for definite (as in .DEFINITE); what does :U stand for? 16:13
moritz undefined
cfa is there a corresponding method?
or is it just !foo.DEFINITE?
(also, why definite/undefined rather than definite/indefinite or some such?) 16:14
16:16 Kaiepi left
jdv79 anyone have a good snippet to wait for all new entries on a channel or timeout? 16:18
timotimo cfa: you have a point, especially since you can give your types a custom "defined" method, but that's not what gets checked for :D or :U 16:20
cfa also, while i'm asking dumb questions, here's a small documentation proposal: i think it could be made clearer that slicing adverbs on hashes etc. are defined on the operators themselves---i remember reading through the docs a while back and wondering where this magical :exists came from. 16:21
16:21 regina joined
cfa (knowing it's just an argument to the postcircumfix:<{ }> multi makes things way clearer) 16:22
timotimo: yeah. tbh i would've expected Str:D and Str:I (for indefinite)
timotimo yeah, adverbs are kind of a funny syntactical feature
the way it attaches to operators in expressions 16:23
cfa yeah
but understanding that's what's happening clears some confusion up
like, the hash docs say: "Beware of the relatively loose precedence of operator adverbs, which may require you to add parentheses in compound expressions:"
but the idea that :exists is actually part of %hash<key> isn't necessarily clear
jdv79 mine is something like await Promise.anyof(Promise.in(10),start {...await $c...}) but that inner await seems to allow the outer to unblock...or someting 16:24
cfa timotimo: but ofc, maybe it _is_ clear and i'm just slow :)
timotimo the information is split very far apart i'd say 16:25
cfa yeah
er, assuming that's re: docs
timotimo yes 16:26
the part about the individual slicing adverbs probably want to mention the connection to the operators themselves 16:27
cfa that'd get my vote, yes
like, i answered this qn by grepping/acking src/core, which isn't ideal 16:28
timotimo true
cfa i'm speaking as a perennial p6 beginner here fwiw
timotimo your perspective is very valuable
and not in the "you are a valued customer" sense 16:29
cfa i have another (p5) thing that i think could possible be made clearer too if that's of interest
though this one may have been discussed to death already
timotimo i can't speak to p5 matters, i've written like 20 lines in total in my life :) 16:30
cfa :) 16:31
well, i might as well mention here 16:33
docs.perl6.org/language/regexes
i think it'd be worth cautioning that a naive `my ($foo, $bar) = 'abcd' ~~ /(a) .* (b)/` won't do what p5 folks expect
since that pattern is super common in perl 5
and then advising on whether to use .list or switch to named captures 16:34
i get the impression that the latter are the p6 preference
timotimo you can also write @("abcd" ~~ /(a) .* (b)/) 16:36
cfa didn't consider that
but great; all this would be useful to clarify
(the @( ... ) example isn't on that page either)
tbh one of my big surprises coming from p5 was the massive change to matching 16:37
timotimo yup, it was drastically redesigned
cfa and that p5 example i gave above is pervasive (match, assign groups to declared scalars in list context)
tbh i still don't know whether i should be using .list / your @() example or $<foo>, $<bar> 16:38
timotimo it's basically equivalent
cfa the latter seems a bit more idiomatic though?
i find ().list pretty ugly 16:39
because it seems to say: we've given you a match object but now you need to do something else to get your groups bound
vs. just having them bound for immediate use
timotimo doesn't destructuring also work here?
cfa you tell me! i'm the ignoramus 16:40
jdv79 nevermind. silent exception.
timotimo my ($foo, $bar) := 'abcd' ~~ /(a) .* (b)/; say "foo is $foo"; say "bar is $bar";
m: my ($foo, $bar) := 'abcd' ~~ /(a) .* (b)/; say "foo is $foo"; say "bar is $bar";
camelia Unexpected named argument '' passed
in block <unit> at <tmp> line 1
timotimo oops :)
cfa np
timotimo well, that's not how
cfa but yeah, all these questions could be answered on that regex doc i think?
timotimo m: ('abcd' ~~ /(a) .* (b)/).Capture.say
camelia 「ab」
0 => 「a」
1 => 「b」
timotimo m: ('abcd' ~~ /(a) .* (b)/).Capture.perl.say 16:41
camelia Match.new(list => (Match.new(list => (), made => Any, pos => 1, hash => Map.new(()), orig => "abcd", from => 0), Match.new(list => (), made => Any, pos => 2, hash => Map.new(()), orig => "abcd", from => 1)), made => Any, pos => 2, hash => Map.new(("" …
cfa like, i'd imagine you don't want every page of docs to be about changes from p5
but still
timotimo well, we do have learn-p6-from-p5 documents
cfa yeah
so maybe the regex page could give further advice on how to do capturing idiomatically
and the from p5 page could clarify some of this stuff 16:42
i also ran into an issue with an early p6 test where i used a Str param in the signature and tried to pass a match group in
(though ~ resolved it, it was still a surprise) 16:43
sorry all, i'm just dumping stuff i've run into
timotimo you can use coercive parameters for this purpose :)
sena_kun m: gist.github.com/Altai-man/b5b71a74...979d93f171
camelia 0
1
Promise.new(scheduler => ThreadPoolScheduler.new(initial_threads => 0, max_threads => 64, uncaught_handler => Callable), status => PromiseStatus::Planned)
too bad
16:44
cfa timotimo: eg?
sena_kun Looks like a bug to me, isn't it? The promise should be Kept there.
16:44 SyrupThinker joined
timotimo m: sub test(Str() $thingie) { say "here's the thing: $thingie; $thingie.perl()" }; test("hello" ~~ /.../) 16:44
camelia here's the thing: hel; "hel"
cfa timotimo: hunh; guess i need to look that up 16:45
timotimo sena_kun: it could be the closed thing only sets itself to true once you've received the last piece? 16:47
m: my $c = Channel.new; $c.send(1); $c.send(2); $c.close; say $c.closed.status; .say for $c.List; say $c.closed.status
camelia Planned
1
2
Kept
timotimo sena_kun: ^- this will be the reason 16:48
sena_kun timotimo, closed returns a promise, but it isn't kept on `close` call for some reason(if I understood you correctly).
ah.
timotimo correct, it gets kept once the last item has been received on the other end
sena_kun timotimo, thanks. 16:50
jdv79, I think you may want something like gist.github.com/Altai-man/b5b71a74...979d93f171 16:51
dunno if it's any better though 16:52
timotimo i do believe we have something in the ecosystem for cancellable promises or something
buggable: eco Cancel
buggable timotimo, Nothing found
rindolf timotimo: it seems to yield the correct result now (even for 2..24) but it is still very slow 17:14
17:23 noreem joined
cfa another documentation issue: nothing for 'unit' in the search dropdown 17:27
(this should probably lead to language/modules.) i'm also not sure whether 'unit sub MAIN' is documented anywhere 17:28
17:31 cognominal joined 17:37 xtreak left 17:46 cdg joined
titsuki m: class A {}; grammar B is A { }; say B ~~ Grammar 17:50
camelia False
titsuki is this a bug? 17:51
moritz I think I submitted it as a bug years ago :-) 17:55
titsuki moritz: Thanks! I see.
18:04 zakharyas left 18:06 zakharyas joined 18:07 cdg left, cdg joined
Ulti can someone explain to me why the ecosystem is moving to CPAN and I assume requires PAUSE? 18:08
given how awful my experience has been with that system historically and right this second trying to recover my ancient account 18:09
I have worked with Perl 5 for 15 years and not once put something on CPAN, that Perl 6 made it incredibly nice to author modules was a reason I produced any 18:10
moritz the reason is mostly that github isn't a CDN, so we shouldn't try to scale while relying on github 18:11
and given that Perl already has a module distribution system, why not use it?
and maybe we didn't know about your awful experience...
18:11 cdg left
Ulti dunno just feels odd I can entirely agree the way its done at the moment with the github registry is really bad 18:13
18:24 zakharyas left 18:25 zakharyas joined 18:33 domidumont joined 18:38 natrys left, lizmat joined 18:48 domidumont left
Geth doc: b92cf18ab0 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/HomePage.pod6
Replaced dated reference to speculations with roast

Closes github.com/perl6/doc/issues/1817
18:49
18:49 Zoffix joined
Zoffix cfa: it'd be more helpful if you reported problems with documentation in the repository instead: github.com/perl6/doc 18:49
cfa: if you're interested, we can get you a commit bit, so you could fix the problems too. 18:50
Ulti: I wouldn't call it "moving". Rather we have two ecosystems: CPAN (paused based) and p6c (gitlab and github based) 18:51
18:51 natrys joined 18:52 zakharyas left
Zoffix Ulti: and I suggest using some command line utility to upload to PAUSE (App::Mi6 can do it), which significantly ameliorates the awfulness of dealing with PAUSE. 18:52
Zoffix authors 200+ CPAN modules and all that was involved is running `dzil release` command once 18:53
18:54 zakharyas joined
Zoffix cfa: I'd say the idiomatic equivalent of P5's `my ($foo, $bar) = 'abcd' ~~ /(a) .* (b)/` would often be using named captures instead: `'abcd' ~~ /$<foo>=a .* $<bar>=d/;` 18:57
m: 'abcd' ~~ /$<foo>=a .* $<bar>=d/; say $<foo bar>
camelia (「a」 「d」)
Zoffix m: 'abcd' ~~ /$<foo>=a .* $<bar>=d/; say ~«$<foo bar>
camelia (a d)
Zoffix cfa: or .comb 18:58
m: say "as2134dsad r3w45w3rwrfe 3245".comb: /\d+/
camelia (2134 3 45 3 3245)
Ulti isnt actually creating a PAUSE account a human limited process though?
or has that changed
Zoffix Yeah, I think that's still handled by a human. 18:59
19:02 darutoko left
Zoffix ZzZombo: "yeah, why the &CORE::infix:<orelse>?". Because that's what .perl of a handled Failure object is. The `handled` attribute is not publicly twirlable, so the orelse call (that calls .defined) is used to set .handled to True 19:05
19:05 sharksoft joined
Zoffix s: &dd 19:08
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/fb88...y.pm6#L586
cfa Zoffix: thanks 19:09
Zoffix: i guess the difference there is that $/ will get clobbered but the subsequent match, so you need to assign $<foo> and $<bar> to (say) $foo and $bar 19:10
but the => by the
or rather, by a
Zoffix ZzZombo: looks like that `try` in &dd causes all Failures to get handled. Gonna commit a fix shortly 19:11
cfa and noted re: suggestions for doc fixes; i guess i'm just discussing stuff here in case i'm stupidly missing stuff already documented rather than spamming the doc repo with issues
Zoffix cfa: yeah, that's one of the issues.
m: my ($foo, $bar) := ~«'abcd'.match: /(.)..(.)/; dd [$foo, $bar] 19:14
camelia ["a", "d"]
cfa i'll admit that's a bit cryptic
Zoffix :D 19:15
« is a hyper operator and it's hypering prefix ~, which stringifies stuff.
m: my ($foo, $bar) = 'abcd'.match: /(.)..(.)/; say [$foo, $bar]
camelia [「abcd」
0 => 「a」
1 => 「d」 (Any)]
Zoffix m: my ($foo, $bar) = .list with 'abcd'.match: /(.)..(.)/; say [$foo, $bar] 19:16
camelia [「a」 「d」]
cfa i guess this is okay too 19:19
m: ('abcd' ~~ /(a) .* (d)/).map: ~*
camelia ( no output )
cfa hunh
Zoffix m: say ('abcd' ~~ /(a) .* (d)/).map: ~*
camelia (a d)
cfa oh, camelia won't display the return value? 19:20
m: True
camelia WARNINGS for <tmp>:
Useless use of constant value True in sink context (line 1)
Zoffix Right
cfa derp
other than conciseness, is ~<< preferable to .map: ~* here?
(to be honest i find both obfuscate the match a little) 19:21
Zoffix doesn't see reasons other that conciseness
19:22 noreem left
Zoffix In this particular case, I mean. The » and .map aren't equivalent: docs.perl6.org/language/traps#Usin...changeably 19:22
cfa nods 19:23
Zoffix m: my ($foo, $bar) = m/(.)..(.)/».Str with 'abcd'; say [$foo, $bar]
camelia [a d]
cfa cute 19:24
Zoffix m: -> $ (Str() $foo, Str() $bar) { say [$foo, $bar] }('abcd' ~~ /(.)..(.)/)
camelia Unexpected named argument '' passed in sub-signature
in block <unit> at <tmp> line 1
cfa it'd be nice to have an adverb that handles this i guess
Zoffix gah
c: 2017.02 -> $ (Str() $foo, Str() $bar) { say [$foo, $bar] }('abcd' ~~ /(.)..(.)/) 19:25
committable6 Zoffix, ¦2017.02: «Unexpected named argument '' passed in sub-signature␤ in block <unit> at /tmp/jAEASjSpD4 line 1␤␤ «exit code = 1»»
cfa gah to my crappy suggestion or the output?
Zoffix The output. I didn't expect a nameless named arg :)
cfa ha ha
Zoffix But I'm not a fan of adding an adverb either :}
cfa ah, i figured
Zoffix: filed two doc tickets 19:44
Zoffix cfa: sent you an invite to perl6 GitHub org. You can accept it on github.com/perl6/ That gives you commit bit to most of our repos, include docs repo, so you can commit whatever fixes you can directly 19:46
cfa wow, thanks
19:52 comborico1611 joined 19:53 buggable left, buggable joined, ChanServ sets mode: +v buggable 19:54 mniip joined 19:58 sharksoft left, buggable left 19:59 buggable joined, ChanServ sets mode: +v buggable
Zoffix e: dd $*W 20:11
evalable6 Failure.new(exception => X::Dynamic::NotFound.new(name => "\$*W"), backtrace => Backtrace.new)
Zoffix ZzZombo: fixed ^. Also, the $*W is available only in compile time (and it's not a user-supported object, so if you use it, stuff may break without notice)
e: BEGIN dd $*W
evalable6 (Perl6::World without .perl method)
20:12 Zoffix left 20:19 zakharyas left 20:25 regina left 20:26 Kaiepi joined 20:31 cognominal left 20:35 pmurias_ joined, pmurias left, pmurias_ is now known as pmurias, freeberz_ left
pmurias how much does Rakudo support deploying apps/scripts on windows? 20:37
20:37 freeberz joined 20:38 cognominal joined 20:40 vike left
Geth whateverable: cee3f5cfee | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Whateverable.pm6
Print $*PID when testing

So that you can attach gdb and other stuff to it.
20:40
20:42 cog_ joined 20:43 cognominal left
MasterDuke buggable: eco wix 20:44
buggable MasterDuke, App::InstallerMaker::WiX 'Tool to make basic WiX installers for Perl 6 apps, bundling MoarVM/Rakudo.': github.com/jnthn/p6-app-installermaker-wix
20:45 rindolf left
MasterDuke pmurias: ^^^ 20:45
20:49 cdg joined 20:50 pmurias left 20:53 cdg left 20:56 rindolf joined 21:00 stmuk joined, vike joined 21:01 stmuk_ left 21:04 buggable left 21:12 wamba left 21:21 pmurias joined
Geth whateverable: 6d369380f3 | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Whateverable.pm6
Time out later if we're downloading builds

Most tests have a timeout, and some of these tests need tens of rakudo builds (which we have to download if we're running locally). With this change it will ask the test suite to wait some extra time which is roughly equal to the amount of time we spend on fetching the builds.
21:23
21:29 Kaiepi left 21:30 Kaiepi joined
pmurias MasterDuke: thanks 21:36
22:04 SyrupThinker left
rindolf timotimo: here? 22:08
timeless: your new program generates the right results
timeless: sorry
22:13 robertle left 22:18 khw left 22:20 natrys left 22:21 AlexDaniel left
timotimo i'm here 22:22
22:26 lookatme left, lookatme joined
MasterDuke rindolf: fwiw, this gist.github.com/MasterDuke17/cb74d...8e136a91f7 is the version i was working on (i cut out the primes just to make pasting it easier) 22:29
if RT #128655 is fixed it should take around 275s 22:30
synopsebot RT#128655 [open]: rt.perl.org/Ticket/Display.html?id=128655 [OPTIMIZER] Mixup in candidates through optimizer
rindolf MasterDuke: ah 22:32
MasterDuke and if somebody figures out that native array access optimization, you won't need to do the `my int $a = @d[$m]; my int $b = @d[$m-$p]` business
22:39 sharksoft joined 22:42 comborico1611 left 22:52 sharksoft left
timotimo eh. i'd like to see how many visitors i get for my blog, but it doesn't seem like ghost has something for that built-in 22:52
and i'm not sure if i should put google analytics in
thowe what's your blog URL? 22:53
timotimo wakelift.de
piwik aka matomo might be good? 22:54
22:58 buggable joined, ChanServ sets mode: +v buggable 23:01 Zoffix joined
Zoffix rindolf: so what's this mystery program of yours that everyone seems to be optimizing? What's the full source code? 23:01
timotimo i think it's a project euler solution? 23:02
rindolf Zoffix: wait a sec 23:03
23:03 sena_kun left
rindolf Zoffix: see github.com/shlomif/project-euler/t...-euler/618 and projecteuler.net/problem=618 23:04
Zoffix rindolf: I only see P5 and Python versions... 23:05
rindolf Zoffix: i have a p6 version too
Zoffix rindolf: well, that's what I was asking about.
rindolf Zoffix: see gist.github.com/timo/c6f33191957d5...978b50de7e 23:06
23:07 cdg joined
Zoffix m: use QAST:from<NQP>; BEGIN $*W.compile_time_evaluate: $, QAST::Op.new: :op<callmethod>, :name<new>, QAST::WVal.new(:value(Buf)), QAST::Op.new: :op<bitand_i>, QAST::IVal.new(:value(10)), QAST::IVal.new: :value(255) } 23:08
camelia 5===SORRY!5=== Error while compiling <tmp>
An exception occurred while evaluating a BEGIN
at <tmp>:1
Exception details:
Type check failed in initializing element #0 to Buf; expected uint8 but got Int (10)
in code at <tmp> line 1…
Zoffix Doesn't look like RT#128655 got much to do with the actual ops candidates
synopsebot RT#128655 [open]: rt.perl.org/Ticket/Display.html?id=128655 [OPTIMIZER] Mixup in candidates through optimizer
Zoffix gonna play around with it a bit 23:09
23:09 Zoffix left
Geth perl6-examples: 6e7a4b07ed | (Shlomi Fish)++ | categories/euler/prob618-shlomif-timotimo.p6
Add functional, but slow solution to E618.

Based on
  github.com/shlomif/project-euler/t...-euler/618 .
23:10
MasterDuke Zoffix: lines 41 and 48 23:11
23:12 cdg left 23:13 Zoffix joined
Zoffix MasterDuke: what's that? The ops in the last gist? 23:13
MasterDuke when i first made all the variables ints it didn't get any faster, because there is only an Int,Int candidate for the bit-ops (and the array accesses in the math expressions also make things Int) 23:14
yeah
i uncommented the int,int versions of the bit-ops in Int.pm6, rebuilt rakudo, pulled those array accesses out into their own int variables and it took one fourth the time 23:15
Zoffix neat 23:16
23:18 konsolebox left 23:19 konsolebox joined
MasterDuke i got the correct results, but then i did try running a spectest and there were failures, so there's definitely still something up with those ops 23:22
timotimo but of course ublock origin blocks piwik/matomo by default %) 23:23
wakelift.de/2018/03/05/delays-and-delights/ - my first TPF grant report post is up now 23:24
Zoffix \o/ 23:25
timotimo notable6: help
notable6 timotimo, Like this: notable6: weekly rakudo is now 10x faster # See wiki for more examples: github.com/perl6/whateverable/wiki/Notable
timotimo notable6: weekly timo's first tpf grant report blog post wakelift.de/2018/03/05/delays-and-delights/
notable6 timotimo, Noted!
Zoffix timotimo++ 23:26
m: use nqp; Buf.new: nqp::bitand_i(10, 10) 23:34
camelia Type check failed in initializing element #0 to Buf; expected uint8 but got Int (10)
in block <unit> at <tmp> line 1
Zoffix m: use nqp; Buf.new: nqp::hllize(nqp::bitand_i(10, 10))
camelia ( no output )
Zoffix m: use nqp; dd Buf.new: nqp::hllize(nqp::bitand_i(10, 10))
camelia Buf.new(10)
23:36 pmurias left
Zoffix Wonder if I should stick that into Buf or as a more general solution 23:36
guess Buf, for now. It reaches out into List's guts to get this explosion, so I guess there's less chance a lost of stuff is similarly affected 23:38
.oO( how the hell did @Zoffix[2017] think the bug was in the dispatcher :S )
23:42
23:42 |oLa|2 left
Zoffix MasterDuke: were the test failures you got in t/spec/S03-operators/bit. ? 23:59