»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 Zoffix on 25 May 2018.
00:03 mcmillhj left 00:11 MasterDuke left 00:13 mcmillhj joined 00:18 rouking left, mcmillhj left 00:20 benjikun2 is now known as benjikun 00:27 mcmillhj joined 00:32 mcmillhj left 00:34 Bowlslaw left 00:50 fake_space_whale joined 00:51 MasterDuke joined 00:54 quester joined, quester left, quester joined 01:00 mcmillhj joined 01:02 zachk left 01:06 mcmillhj left 01:09 _Xliff_ left
quester p6: say $*PERL.compiler.version 01:10
camelia v2018.06.44.g.3.cbf.25.e.0.f
01:17 quester left 01:21 mcmillhj joined 01:23 markoong left 01:26 mcmillhj left 01:27 fake_space_whale left 01:34 MilkmanDan left
lucs m: say 49.sqrt ~~ Int # How do I get this to be True? 01:35
camelia False
01:35 MilkmanDan joined
AlexDaniel m: say say 49.sqrt.narrow 01:36
camelia 7
True
AlexDaniel m: say 49.sqrt.narrow
camelia 7
AlexDaniel m: dd 49.sqrt.narrow
camelia 7
AlexDaniel m: dd 50.sqrt.narrow
camelia 7.0710678118654755e0
AlexDaniel lucs: I don't know if that's a good solution
“Returns the number converted to the narrowest type that can hold it without loss of precision.” 01:37
maybe for some square roots it's not going to work
01:38 quester joined
lucs For my needs, it will do fine (smallish numbers). 01:38
Thanks.
01:41 epony left 01:42 silug joined 01:43 quester left 01:46 gabiruh left 01:49 quester joined 01:53 kybr left 01:55 mcmillhj joined 02:01 MasterDuke left, kybr joined, mcmillhj left 02:17 MasterDuke joined 02:28 turdmonkey joined, turdmonkey is now known as Bowlslaw
Bowlslaw Hi peoples 02:31
benjikun Bowlslaw: hi again 02:33
02:35 silug left 02:38 mcmillhj joined 02:42 quester left
Bowlslaw how's it goin benjikun? 02:42
02:42 quester joined
benjikun Bowlslaw: going alright 02:43
02:43 quester is now known as Guest49701, mcmillhj left, Guest49701 left
Bowlslaw how come I can't do a hash of arrays? 02:44
perlpilot Bowlslaw, I dunno ... why can't you do a hash of arrays? ;) 02:45
Bowlslaw my %hash = nums => [1..5], letters => ['a'..'e']
perlpilot I don't see anything wrong with that. 02:46
Bowlslaw ok well
when i try
02:46 silug joined
Bowlslaw say %hash{numbers} it gives me an "operator not found" error 02:46
but when I say %hash[0] it prints everything
perlpilot you want %hash<numbers> or %hash{'numbers'} 02:47
Bowlslaw o
OH YEAH
perl 6 uses <>...
everything works as expected now. i'm not retarded 02:48
(well maybe a little)
02:48 quester_ joined 02:51 perlpilot left 02:54 perlpilot joined, mcmillhj joined 02:58 quester_ left, mcmillhj left 02:59 quester_ joined
perlpilot just discovered CArray.allocate 03:02
(whoever implemented .allocate)++
03:04 quester_ left 03:05 quester joined
perlpilot though I'm currently getting a floating point exception when I try to use it (which is weird) 03:09
03:09 gabiruh joined 03:13 psychoslave joined 03:14 kybr left 03:17 psychoslave left
Bowlslaw is it beneficial to put a 'race for' inside of a react block? 03:18
03:19 mcmillhj joined 03:24 mcmillhj left
perlpilot Bowlslaw, I'm going to go with "no" because the question doesn't quite make enough sense to me. :-) 03:27
Bowlslaw lol 03:28
03:34 AlexDaniel left, AlexDaniel joined 03:36 mcmillhj joined 03:38 epony joined 03:41 mcmillhj left 03:44 vike1 left 03:51 psychoslave joined 03:56 psychoslave left 04:03 rindolf joined 04:05 wamba joined 04:06 kaare_ joined 04:09 rindolf left 04:10 mcmillhj joined 04:15 mcmillhj left 04:17 labster left, AlexDaniel left
Juerd lucs: Are you sure you want a type check there? How about "$foo %% 1"? 04:17
04:17 AlexDaniel joined 04:21 mcmillhj joined
lucs Juerd: Ha, nice. Yeah, I wouldn't have thought of that. 04:22
04:26 Bowlslaw left 04:30 mcmillhj left 04:33 psychoslave joined 04:51 mcmillhj joined 04:55 eliasr left 04:56 mcmillhj left 04:58 perlpilot left 05:11 wamba left 05:18 sauvin joined 05:19 psychoslave left 05:24 mcmillhj joined 05:29 mcmillhj left 05:32 psychoslave joined 05:37 kaare_ left 05:40 psychoslave left, kaare_ joined 05:52 psychoslave joined 05:56 psychoslave left 06:00 psychoslave joined 06:13 mcmillhj joined 06:18 mcmillhj left 06:23 Garland_g[m] left, CIAvash[m] left, lance_w[m] left, Matthew[m] left, mienaikage left, AlexDaniel`` left, mirlur[m] left, wictory[m] left, AlexDaniel` left, EuAndreh[m] left, unclechu left 06:24 ilmari[m] left 06:26 benchable6 left, greppable6 left, squashable6 left, bisectable6 left, releasable6 left, releasable6 joined, squashable6 joined, greppable6 joined, bisectable6 joined, benchable6 joined 06:31 lizmat left, mcmillhj joined 06:36 mcmillhj left 06:42 psychoslave left 06:54 mcmillhj joined, lizmat joined 06:59 mcmillhj left 07:05 MasterDuke left 07:32 mienaikage joined 07:34 mcmillhj joined 07:35 quester left, stmuk_ joined, psychoslave joined, jmerelo joined 07:39 mcmillhj left 07:54 domidumont joined
jmerelo Maybe someone is interested in contributing to this issue github.com/perl6/doc/issues/2133 07:55
07:55 psychoslave left 08:01 domidumont left 08:02 domidumont joined, |oLa| joined 08:03 |oLa| left 08:05 ChoHag left 08:09 mcmillhj joined 08:14 mcmillhj left 08:20 rindolf joined 08:22 AlexDaniel`` joined, AlexDaniel` joined, unclechu joined, mirlur[m] joined, wictory[m] joined, Matthew[m] joined, lance_w[m] joined, EuAndreh[m] joined, ilmari[m] joined, CIAvash[m] joined 08:23 Garland_g[m] joined
Geth doc: 23e56e3548 | (JJ Merelo)++ | doc/Language/syntax.pod6
Adds indirect invocation to syntax

And also adds preamble to example, instead of skipping test. Closes #2131
08:31
synopsebot Link: doc.perl6.org/language/syntax
Geth doc: aa35ae1453 | (JJ Merelo)++ | doc/Language/syntax.pod6
Fixes compilation error in preamble
doc: 45d68ff5d6 | (JJ Merelo)++ | doc/Type/Any.pod6
Revising Any.skip and Any.batch

In fact, they were done already. I have only added some clarification. This closes #1195
synopsebot Link: doc.perl6.org/type/Any
08:40 mikejw joined
mikejw how can I escape a dollar sign in a qqx command? 08:41
lizmat \$
mikejw hmm
lizmat do you need to interpolate other $foo in the string? 08:42
if not, you could do:
qqx:!s/ ... / 08:43
the :!s indicating that no scalar interpolation should take place
mikejw but I do want to use a variable as well
lizmat alternative: {'$'} 08:44
mikejw the command is -> aws s3 ls s3://bucketname/$name | awk '{ print \$4 }'
lizmat my $column = '$4'; -> aws s3 ls s3://bucketname/$name | awk '{ print $column }' 08:45
mikejw ah thanks :)
jmerelo lizmat: I have added stuff to R#1941. Maybe it's clearer now what I mean. Baseline is: some documentation is better than no documentation. 08:50
synopsebot R#1941 [open]: github.com/rakudo/rakudo/issues/1941 Document Rakudo-specific classes to support Perl 6 documentation
lizmat jmerelo: specifically about Rakudo::Iterator, what are you missing in its internal documentation ? 08:51
jmerelo lizmat: you mean, besides the comments? 08:52
lizmat yes
jmerelo lizmat: that one is better documented with comments than others, but a Pod would really help. I had to wade through the source to find its implementation, and then I saw it was huge. 08:53
lizmat: a simple list of methods would have been helpful, in a Pod, for instance. 08:54
lizmat: but that one is just an example. This all started with this github.com/rakudo/rakudo/blob/98e5...y/Spec.pm6 08:55
08:55 mcmillhj joined
mikejw still struggling here 08:55
lizmat FWIW, I'm slightly worried that providing a pod like that would be like a menu for people to choose from, which in the case of Perl 5 / perlguts
mikejw '$4' is stored in $column is evaluated to 'True'
jmerelo lizmat: you mean you are afraid that people would use in their own perl6 programs, creating a nightmare 08:56
lizmat led to the situation where nothing can be changed in the internals because
yes
mikejw: ??
jmerelo lizmat: but people are going to do that anyway. And a bit of documentation would help the documentors.
lizmat m: my $a = q/$4/; say qq/$a/ # mikejw 08:57
camelia $4
lizmat jmerelo: why would that help the documentors?
jmerelo lizmat: check out the code github.com/rakudo/rakudo/blob/98e5...y/Spec.pm6 08:58
lizmat yes, that could use quite some internal comments, I agree :-) 08:59
jmerelo lizmat: I mean, internal comments are documentation, some documentation is better than no documentation, and pod documentation is better than simple comments
08:59 mcmillhj left
lizmat well, this is more an issue for @LARRY I think 09:00
I'll gladly write pod for the stuff that I've mostly done if we agree on that being a good idea 09:01
jmerelo lizmat++
09:13 mcmillhj joined 09:18 mcmillhj left 09:20 ChoHag joined
mikejw I've just remembered you need to use the list 'modifier' when using qqx and capturing an array. I can't remember where I've used it before 09:23
qqx//.list ? 09:24
09:25 damnlie joined
lizmat I think qqx returns a Str with the complete output, so that would be .lines then # mikejw 09:28
mikejw ahh awesome that's it :)
lizmat or .words if you're interested in words (separated by whitespace) 09:29
or .comb( /\d+/ ) if you're interested in numbers :-)
El_Che lizmat: now you're showing off :) 09:30
lizmat or whatever regex or grammar you want to run on it :-)
guess I am.. :-)
mikejw thanks my dudes 09:33
09:34 mcmillhj joined
lizmat mikejw: and dudettes I presume? 09:36
mikejw of course
lizmat :-)
09:39 mcmillhj left 09:43 psychoslave joined
jmerelo notable6: Perl 6 survey results in a repo: github.com/perl6/p6survey 09:44
notable6 jmerelo, Noted!
lizmat jmerelo: I wish some normalization could have been done 09:45
e.g. with most wanted feauture
*feature 09:46
it is basically only 2: faster and JVM
ok, and Macros
09:46 rindolf left
lizmat so 3 09:46
ah, it scrolls: ok, sorry, more than 3, but still, some normalization would have been nice 09:47
perhaps for the next time :-)
09:48 psychoslave left, psychoslave joined 09:52 mcmillhj joined 09:56 mcmillhj left, rindolf joined
jmerelo lizmat: those are the raw questions; we can process them now as we want. That question was free form, so that's just expected. 10:01
10:12 psychoslave left 10:14 pmurias joined, mikejw left
pmurias lizmat: so it seems faster performance on truffle is the dream feature according to the survey ;) 10:14
lizmat pmurias: looking forward to a blog post I can mention in the P6W :-) 10:16
10:26 mcmillhj joined 10:31 mcmillhj left
El_Che pmurias: you are amalgaming responses into one. I am pretty sure that many people that answer "faster" don't want to run the jvm 10:34
10:38 mcmillhj joined 10:42 eliasr joined 10:43 mcmillhj left 10:53 MasterDuke joined
jmerelo lizmat: 56 out of the 220 answers mention speed in one form or other. I've created a script and uploaded it to the repo. 10:56
lizmat almost 25% wow 10:57
just over 25% even :-)
jmerelo lizmat: I might have missed some, or there could be false positives. Check out the script, just in case. 11:04
lizmat: maybe it would be better to create a tag could after normalizing to lowercase. I don't know. Any suggestion is welcome.
lizmat jmerelo: am right in the middle of something, and about to have visitors, so will look at it (much) later today / tomorrow
jmerelo lizmat: have fun.
lizmat jmerelo: thank you! 11:07
11:09 mcmillhj joined
pmurias El_Che: I assume most people don't care significanttttly what they are running on 11:09
11:13 psychoslave joined, markoong joined 11:14 mcmillhj left 11:27 jmerelo left 11:30 mcmillhj joined 11:34 mcmillhj left
Juerd Regarding speed, a simple benchmark shows that multi sub x(5) { ... } is consistently approx. 10% slower than multi sub x ($ where * == 5) { ... } both for positive and negative matches; this was unexpected. 11:45
11:50 domidumont left, vike1 joined
timotimo m: multi sub x(5) { say "yup" }; x("5") 11:51
camelia 5===SORRY!5=== Error while compiling <tmp>
Calling x(Str) will never work with any of these multi signatures:
(Int $ where { ... })
at <tmp>:1
------> 3multi sub x(5) { say "yup" }; 7⏏5x("5")
timotimo m: multi sub x($ where * == 5) { say "yup" }; x("5") 11:52
camelia yup
timotimo a slight semantic difference
Juerd timotimo: Oh, but it's still faster with Int $ where * == 5 :) 11:53
lizmat could it be that the (5) means where * ~~ 5, rather than where * == 5 ? 11:55
maybe Int could use its own .ACCEPTS ?
timotimo i'd assume that's where the difference comes from 11:56
Juerd lizmat: Perhaps, but where 5 and where * ~~ 5 are significantly slower.
lizmat anything visible with --profile ?
timotimo hm, didn't zoffix build an optimization recently?
Juerd I have to go, sorry 11:57
12:00 mcmillhj joined 12:04 mcmillhj left
lizmat has to go too 12:10
&
MasterDuke both of those cases spend a lot of time in bind_one_param, which isn't jitted 12:27
timotimo well, that's the slow path binder anyway, which is Just Bad
MasterDuke and the literal case spends a lot of time in ACCEPTS (src/core/Numeric.pm6:11), which the where case doesn't have at all 12:28
bind_one_param might just be because my benchmark code is bad 12:29
m: multi sub foo(5) { "five" }; multi sub foo($a) { "not five" }; my $a; for ^500_000 { $a = foo($_ % 6) }; say $a; say now - INIT now
camelia not five
6.19927022
12:30
MasterDuke m: multi sub foo(Int $ where * == 5) { "five" }; multi sub foo($a) { "not five" }; my $a; for ^500_000 { $a = foo($_ % 6) }; say $a; say now - INIT now
camelia not five
6.28544211
MasterDuke that's what i ran. got a bigger difference between them though
literal was 1s slower 12:31
12:47 benjikun2 joined 12:50 benjikun left, SCHAPiE joined, Zoffix joined
Zoffix Could it be you're just measuring a very fast operation, but you're still using `INIT` and so get a lot of noise in your measurements? 12:52
If I look at ASTs, the (5) case is just a type check and an `iseq_i` op, while the other two cases involve a whatevercode clone invocation and a check against its .ACCEPTS 12:53
and inside that whatevercode there's even a callstatic of == op 12:54
MasterDuke could be 12:55
the profiles do look pretty similar (when sorted by exclusive time). except for the addition of ACCEPTS in the literal case, which is its 5th most expensive routine 12:57
Zoffix In the literal case? 12:58
It shouldn't be there
MasterDuke called for every iteration of the for
Zoffix |1d check what's going on with `where 5` opt: colabti.org/irclogger/irclogger_log...06-30#l363 13:00
ZofBot Zoffix, Will remind you on 2018-07-01T09:00:40.807044-04:00 about check what's going on with `where 5` opt: colabti.org/irclogger/irclogger_log...06-30#l363
Zoffix gets ready to go to the non-lang specific coding meeting
13:01 llfourn joined
Zoffix Gonna hack a GTK app in Perl 6 :) 13:01
13:01 Zoffix left, mcmillhj joined
MasterDuke nice 13:02
13:08 mcmillhj left 13:13 Zoffix joined
Zoffix For some reason my 2-week old install of rakudo/zef can't find JSON::Tiny, while homebox finds it. 13:14
And laptop install says there are 0 modules in cpan 13:15
moritz zef update?
Zoffix moritz: yeah, from zef update 13:16
And there are zero in the file it fetches: raw.githubusercontent.com/ugexe/Pe...cpan1.json
and doing zef --/cached install JSON::Tiny on home box fails to find it too now 13:17
wr found the weakest link in cpan ecosystem? :)
we found the weakest link in cpan ecosystem? :)
does cpanm talk to local cpan mirror directly? 13:18
13:19 mcmillhj joined 13:20 smls joined
Zoffix eco: from:cpan 13:22
buggable Zoffix, Found 279 results: Bailador, zef, Inline::Perl5, JSON::Tiny, App::Mi6. See modules.perl6.org/s/from%3Acpan
13:23 mcmillhj left
smls When writing a Test::* module with additional test function, is it possible to implement them using `pass` and `flunk` from the Test module, but refer to the *caller*'s line numbers in the diagnostics output? 13:24
timotimo try "is hidden-from-backtrace" perhaps? 13:25
Zoffix installed JSON::Tiny via direct link to repo taken from modules.perl6.org and filed github.com/ugexe/zef/issues/258
crap GTK::Simple is failing install now -_-
"libgtk-3-0 is already the newest version" but GTK::Simple's tests say it can't find the lib 13:26
13:27 sena_kun joined
Zoffix it says "libgtk-3.so" tho, not "libgtk-3-0" 13:27
same with libglib-2.0. I got libglib2.0-0 13:28
MasterDuke doesn't nativecall need the name to be exactly the same or something like that?
maybe add a symlink as a workaround? 13:29
Zoffix From /lib/x86_64-linux-gnu/libglib-2.0.so.0 to /lib/x86_64-linux-gnu/libglib-2.0.so ? 13:30
smls timotimo: Doesn't work.backtrace
Zoffix smls: no, there's no way ATM
smls: is there some Test::Builder module in ecosystem? 13:31
eco: Test::Builder
buggable Zoffix, Test::Builder 'Flexible framework for building TAP test libraries': modules.perl6.org/dist/Test::Builde...%20Polulak
13:32 MilkmanDan left
Zoffix Never used it, but maybe that can help? 13:33
13:34 MilkmanDan joined 13:35 mcmillhj joined
Zoffix Here's how Test.pm6 figures out what line to report: github.com/rakudo/rakudo/blob/mast...#L715-L720 13:37
Maybe we should have something better and user-useable?
moritz also, "until !" better written as "while" 13:39
13:40 mcmillhj left
moritz will try to simplify that when my build finishes 13:40
Zoffix Filed R#1998 13:41
synopsebot R#1998 [open]: github.com/rakudo/rakudo/issues/1998 No way to use Test.pm6 routines in userland Test libraries
Zoffix MasterDuke++ `ln`ing the three libraries from lib.so.0 to lib.so worked 13:44
13:48 psychoslave left
MasterDuke doesn't Skarsnik have a rakudo PR open that attempts to do something with library version numbers? 13:49
13:50 mcmillhj joined 13:52 turdmonkey joined, turdmonkey is now known as Bowlslaw
Bowlslaw good morning 13:54
13:54 mcmillhj left
Bowlslaw I just discovered the ascii -> unicoe converter for perl6-vim 13:55
never used unicode in my code before..
my $λ = "this is lambda"; 14:11
wow, it even does the whole greek alpha with the range operator 14:15
14:17 mcmillhj joined 14:21 markoong left 14:22 mcmillhj left, markong joined
sena_kun junction is a unique thing to Perl 6, isn't it? At least under this name, as I cannot google similar thing in some other (not obscure one) language. 14:26
Bowlslaw sena_kun: I haven't seen anything like it elsewhere, either, but I am a newb 14:32
oh, cool! Reading "The Algorithm Design Manual" is coming in handy in my web crawler. I'm trying to implement a depth limit to my crawler, and it looks like I will need to apply "Iterative Deepening depth-first search" 14:33
timotimo it could be enough to just turn the URLs into a list or hash of "url" and "current depth" and when you're getting all the links from the page at one url, just push all links you've found with the "current depth + 1" 14:35
and only actually get the page if the current depth is below a certain threshold
Bowlslaw hmmm 14:36
as it is now, I collect all of the links from the page and push them into an array
so maybe... 14:37
%links = 'url1' => $depth, 'url2' => $depth
something like that?
timotimo that'll have the added effect that the same url won't be in there more than once 14:38
Bowlslaw I call .unique to ensure that 14:39
I think your idea is great 14:42
thanks
less code ;p
14:55 mcmillhj joined
Bowlslaw Are there any important differences between my-sub($var) and my-sub: $var ? 14:57
sena_kun absence of parens? :) 14:58
14:59 lizmat left
Bowlslaw parens lives matter 14:59
15:00 mcmillhj left
timotimo you can't my-sub: $var, though. only with methods 15:02
m: say: "hi"
camelia WARNINGS for <tmp>:
Useless use of constant string "hi" in sink context (line 1)
timotimo that's not a call, that's a label
Bowlslaw oo 15:04
El_Che it sounds like a foo fighters song
Bowlslaw so there isn't really a difference between parens and colon for method calls?
15:05 smls left
timotimo well, the precedence differs, of course 15:05
15:08 fascinum joined
Bowlslaw timotimo: thanks for your help, your hash suggestion was a great solution 15:08
also very perly ;p 15:09
autovivification ftw
What do you guys think about unicode and portability?
As in, what if you use a bunch of unicode symbols in your code, and then you or someone else, doesn't have proper uniode support? 15:10
15:10 psychoslave joined
El_Che Bowlslaw: the code will run, editing is something else 15:12
Geth gtk-simple: fe417e5b82 | (Zoffix Znet)++ (committed using GitHub Web editor) | README.md
Remove `use v6.c` from SYNOPSIS

When 6.d comes out this would be outdated and we might forget to remove it
15:13 mcmillhj joined
Bowlslaw El_Che: so the code will still run, but the rendering of the unicode symbols will be funky? 15:13
Zoffix Bowlslaw: it might, yeah 15:14
El_Che yes, I expect that
you're not force to use utf8 only charsij your output, though
d
Bowlslaw the code points would be read properly by Rakudo, is my guess 15:15
Zoffix TIL this works
m: class Foo { has $.a; has $.b; method new(|) {}; }; Foo.new: my $a = 42.abs, my $b = 50.abs; say [$a, $b] 15:16
camelia [42 50]
15:16 MasterDuke left
Zoffix I thought parens would be required around first `my` 15:16
Bowlslaw lol what in the world is that
Zoffix m: class Foo { has $.a; has $.b; method new(|c) { dd c }; }; Foo.new: my $a = 42.abs, my $b = 50.abs; say [$a, $b] 15:17
camelia \(42, 50)
[42 50]
Zoffix m: class Foo { has $.a; has $.b; method new(|c) { dd c }; }; Foo.new: (my $a = 42.abs), my $b = 50.abs; say [$a, $b]
camelia \(42, 50)
[42 50]
Zoffix hm...
m: class Foo { has $.a; has $.b; method new(|c) { dd c }; }; Foo.new: my $a := 42.abs, my $b := 50.abs; say [$a, $b]
camelia \((42, 50))
[(42 50) 50]
Zoffix m: class Foo { has $.a; has $.b; method new(|c) { dd c }; }; Foo.new: (my $a := 42.abs), my $b := 50.abs; say [$a, $b] 15:18
camelia \(42, 50)
[42 50]
15:18 benjikun joined
Zoffix Ahh... They are required, but only if you're binding :) 15:18
Cool.
15:18 mcmillhj left 15:20 benjikun2 left
Geth gtk-simple: a35fc1b870 | (Zoffix Znet)++ (committed using GitHub Web editor) | README.md
Use a clearer word
15:22
15:22 Zoffix left 15:24 mcmillhj joined
Geth gtk-simple: 197de0350e | (Zoffix Znet)++ (committed using GitHub Web editor) | examples/01-hello-world.pl6
Fix casing in method names in docs
15:25
15:27 Bowlslaw left
AlexDaniel squashable6: next 15:27
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 5 days and ≈18 hours (2018-07-07 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
15:29 mcmillhj left 15:32 psychoslave left 15:33 Zoffix joined 15:35 araraloren joined
Zoffix Any idea how to get a GTK app to work with react/whenever instead of .tap'ing all the buttons? I'm having issue with getting this to exit. The app just freezes when I press "Exit": gist.github.com/zoffixznet/78914f5...18074315bf 15:35
15:35 fascinum left
Zoffix Even tho the "hello" button works fine" 15:35
15:35 xtreak joined 15:36 stee joined
Zoffix Hm, this works: gist.github.com/zoffixznet/78914f5...18074315bf 15:37
tho a bit uggo, but meh
I guess I can just stick the react block into start 15:38
15:39 mcmillhj joined 15:43 mcmillhj left 15:46 domidumont joined 15:49 gabiruh left 15:54 domidumont left 15:55 mcmillhj joined 15:58 araraloren left 15:59 araraloren joined
Geth gtk-simple: 8036249751 | (Zoffix Znet)++ (committed using GitHub Web editor) | examples/02-toggles.pl6
We don't require unspace here anymore
15:59
15:59 mcmillhj left, gabiruh joined 16:02 psychoslave joined 16:05 jmerelo joined
Geth doc: 0752731914 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/traps.pod6
Another solution for `run «…»` trap

And some clarified comments.
16:06
synopsebot Link: doc.perl6.org/language/traps
16:06 turdmonkey joined
jmerelo Hi 16:07
sena_kun o/
araraloren night here o/
16:08 turdmonkey is now known as Bowlslaw
Zoffix \o 16:08
jmerelo Afternoon here in hot Granada, Spain
araraloren Seems like everywhere is so hot :) 16:09
Zoffix Same here in hot Canada :)
+46C
"feels like" +46C
jmerelo Zoffix: :-)
AlexDaniel 😮
araraloren can not imagine 16:10
Zoffix Yeah, my igloo is melting :P
AlexDaniel araraloren: 14°C here, not hot at all
araraloren: Estonia
jmerelo AlexDaniel: Wow. 16:11
araraloren here night is not hot, 23°C
jmerelo You know one question missing in the survey? Country 16:12
16:12 mcmillhj joined
jmerelo I have the impression that we folks in Europe are not too abundant. Mornings in the chat channel are really quiet. 16:12
huf Zoffix: is that the real temperature or is it really humid? 16:13
Zoffix Says "+32C 'feels like +44C', humidity 63%" 16:14
huf ugh
though i dunno what they base the feels like on
araraloren here is Tsingtao/QingDao
huf 63 isnt *that* high. but 32C is hot enough
16:15 Bowlslaw left, wamba joined 16:16 psychoslave left 16:17 mcmillhj left
sena_kun jmerelo, in my little experience, many coders just tend to wake up relatively "late". :) Personally, I don't know any folk who wakes up before 10. 16:17
Zoffix benjikun: is github.com/perl6/p6survey/ the final results or is there a better-viewable version coming along? That CSV file is hard to read. 16:18
jmerelo sena_kun: that could be, but still. It's always more lively in the afternoon than in the morning. And there's basically noone early in the morning, European time. El_Che, timotimo, moritz, lizmat, that's it. 16:19
Zoffix: check this out docs.google.com/forms/d/19qSBpGWWc...wanalytics 16:20
Zoffix jmerelo: is that the final URL for result?
jmerelo Zoffix: I would say so, yes.
Zoffix jmerelo: as in, if I make that URL into perl6.org/survey/2018.06-results is that fine?
jmerelo Zoffix: right. 16:21
Zoffix jmerelo: oh, and is the survey done now or are we still waiting for results.
jmerelo Zoffix: it's not going to change; and the answers are not going to change either
Zoffix: no, it's closed.
Zoffix cool
Geth perl6.org: dbd5604bea | (Zoffix Znet)++ (committed using GitHub Web editor) | source/.htaccess
Add short URL for 2018.06 p6 survet results
16:22
Zoffix Pretty cool that a quarter of Perl 6 users haven't used Perl 5... "extensively" 16:25
16:26 psychoslave joined
jmerelo Zoffix: Yep, but that's still too little, I guess. 16:26
I think that the bulk of the community should come from outside the Perl community. 16:27
Zoffix: this is the first survey, right? 16:28
araraloren do we have next season, I am miss that survey :)
Zoffix jmerelo: dunno
jmerelo araraloren: I think the intention of benjikun is to run it once every quarter or so.
Zoffix 10% have used P6 in the past but not any more. Interesting that they still were aware of the existence of this survey 16:29
jmerelo Zoffix: maybe they just hang out because of the community :-)
araraloren oh
16:29 psychoslave left
AlexDaniel age distribution seems to be ok :) 16:30
Zoffix 23% haven't used zef? shocking
16:30 psychoslave joined
jmerelo Zoffix: maybe they used panda :-) 16:30
AlexDaniel a bit too little on <18 side maybe
Zoffix AlexDaniel: pretty surprised to see so many people over 70
AlexDaniel over 60 you mean 16:31
Zoffix I mean over 60
Makes me feel very young :P
jmerelo Zoffix: not really. If you think about the middle age of the Perl community from where most come from...
Zoffix So about 35% already use Perl 6 for work, a lot higher than I expected 16:32
jmerelo Zoffix: I would have to do some cross-question analysis.
jmerelo thinks about creating a Sankey diagram.
Zoffix And only a quarter use Rakudo Star
20% say they "cloned the repo"... Wonder if that means they always use whatever happend to be HEAD... 16:33
AlexDaniel yes but also 9.2% + 13.3% for third-party and distro packages, that's very nice I think 16:34
16:35 fake_space_whale joined
AlexDaniel so a lot of people using proper releases, even if not rakudo star 16:35
16:37 psychoslave left
AlexDaniel so people want better speed and performance 16:37
Zoffix dunno if stmuk still plans to quit doing Rakudo Stars at the end of the year, but if that's still the plan, maybe we should kill Star entirely, if it's just 25%.
Yeah, the amount of speed/perf responses is a lot larger than I anticipated. 16:38
Would be nice to know speed of *what* part of the language
AlexDaniel “Grammar Optimization” “Faster Grammars” :)
most people probably want general speedups
16:39 fascinum joined
AlexDaniel “Stability and performance.” “performance/stability” – at least some people mentioned stability 16:40
“Rename Perl 6 to Rakudo!!” “Keep the Perl 6 name.” :D
Zoffix Also listed as favourite feature, so yeah, looks like optimizing grammars should be high on our priority list. Especially since it'll help that awful 80s compile time after each change :)
AlexDaniel Zoffix: R#1999 maybe? 16:47
synopsebot R#1999 [open]: github.com/rakudo/rakudo/issues/1999 [meta][regex][severe] Roadmap for Regex/Grammar optimizations
AlexDaniel damn 1 away from 2000
weekly: 2000 tickets/PRs on github
notable6 AlexDaniel, Noted!
AlexDaniel weekly: (in rakudo repo)
notable6 AlexDaniel, Noted!
Zoffix :) 16:48
jmerelo AlexDaniel: hey, you got the D#2000 ticket and didn't "note" it! And this is not even 2000, it's 1999!
synopsebot D#2000 [closed]: github.com/perl6/doc/issues/2000 [site] Content of pages without ToC appears on the left side
jmerelo jealous 16:49
16:49 fascinum left
AlexDaniel weekly: Also ≈one month ago we hit 2000 tickets/PRs on github.com/perl6/doc 16:49
notable6 AlexDaniel, Noted!
AlexDaniel jmerelo: didn't notice :)
jmerelo AlexDaniel: yay!!! 16:50
El_Che lots of rakudobrew
Zoffix yeah, too much :)
AlexDaniel is RT down?… 16:52
Zoffix I think perl.org is down. Like our CPAN dist fetcher is down 16:53
stmuk_ well more people are using star at 25.7% than the monthly debs etc at 9.2% 16:54
so I'd hope someone picks up R* after 2018.10 16:55
the worrying stat was 13.3% distro packages :)
Zoffix :D
AlexDaniel why worrying?
Zoffix They're likely outdated 16:56
s/likely/might be/;
AlexDaniel not really, actually
stmuk_ yeah you will get people asking for support for 3 year old versions
El_Che stmuk_: I am not surprised. Most distros have old packages and rakudo star is still the official way to run rakudo 16:57
AlexDaniel so debian stretch users have 2016.12, testing is 2018.05 and unstable is 2018.06
El_Che everything else is many clicks away and explicitly non-official
AlexDaniel and I think there's nothing bad about that at all
Zoffix I got 2013.12.1 rakudo in the repos on my current laptop :P
2016.12 is ancient IMO 16:58
araraloren I got old rakudo too
El_Che distro packages will go way up as time passes and rakud stabilizes
stmuk_ I was surprised by the number using windows 22.4% compared with Mac 25.6%
araraloren good night :) 16:59
El_Che stmuk_: why are you thinking of passing R* to someone elses, it looks like you're doing a good job
AlexDaniel fedora is even fresher: rpms.remirepo.net/rpmphp/zoom.php?rpm=rakudo
araraloren go to sleep now
AlexDaniel (if that's the right link)
16:59 araraloren left, Zoffix left
AlexDaniel I think the situation with distro packages is very good, and definitely not worrying 16:59
El_Che AlexDaniel: I run the most recent ubuntu on my laptop, but servers are work run the latest centos and that's means OOOOOLD 17:00
jmerelo I don't have the privs to publish this in the web, but here's the sankey diagram for agegroup vs perl use
github.com/perl6/p6survey
AlexDaniel El_Che: yeah but what's the rakudo version in the last stable centos? 17:01
El_Che none
AlexDaniel ok, so no problem
:)
El_Che AlexDaniel: it is, most people won't find the repos and there is no way in hell admin will compile rakudo on production
AlexDaniel: it's a huge problem
admins 17:02
AlexDaniel El_Che: not in this context. We were talking about how using distro packages is “worrying”, based on survey results that a lot of people *do* use distro packages
El_Che and by the time rakudo makes it to the repos it will be an old release that probably won't even work with zef
stmuk_ El_Che: I've been doing it for ages and it probably fresh blood
^ needs 17:03
AlexDaniel El_Che: yes, I'm not saying that it is not an issue, just not in this context :)
jmerelo Here you go the Sankey diagram for age group vs. perl use github.com/perl6/p6survey/blob/mas...e-2018.png
AlexDaniel jmerelo: some labels would make a huge difference :)
jmerelo Gist of it is all users over 40 are prior Perl users. 17:04
stmuk_ I'll probably asking for a R* guy after Oct (or earlier if anyone reading this is interested)
jmerelo AlexDaniel: right is age group, right is prior perl user.
I'll see if I can do better. 17:05
AlexDaniel jmerelo: I understand, but people looking at the graph on twitter or wherever it gets reposted won't
El_Che stmuk_: good luck. It's difficult finding replacements in FOSS
AlexDaniel jmerelo: it's a really nice graph by the way :)
jmerelo AlexDaniel: man, it's R. It can probably be done, but it's never easy...
AlexDaniel stmuk_: thank you for your work ♥
jmerelo AlexDaniel: It's nicer and interactive if published, but I don't have admin privs in that repo. 17:06
AlexDaniel jmerelo: you sure? ;) ;) ;)
jmerelo AlexDaniel: Ah! Some magic has happened! I can do it now! It's the Camelia, the fairy godmother, listening to my wishes :-) 17:07
El_Che stmuk_: whatever happens we need to make sure the transistion happens smoothly becasue of the number of people using star
AlexDaniel camelia++
stmuk_ AlexDaniel: I've not gone yet :) 17:08
AlexDaniel stmuk_: I know, it doesn't stop me from thanking you :)
stmuk_ El_Che: sure
:)
17:08 markong is now known as markoong
El_Che stmuk_: after the holidays you'll have full batteries again and continue the stewardship :) 17:08
jmerelo AlexDaniel: and here you go perl6.github.io/p6survey/html/sank...rluse.html
17:09 jast joined
AlexDaniel jmerelo: STILL NO LABELS!!!!! 17:09
jmerelo AlexDaniel: still trying to find out how to label the two parts...
AlexDaniel :)
jmerelo: wait you can publish html just like that??
I didn't know that
jmerelo I mean, it's not trivial. Columns are generated automatically. I'll have to hack something... 17:10
AlexDaniel never wondered how github.io stuff worked
jmerelo AlexDaniel: right. Just tell it to publish the master branch; you can also do it in a directory or a gh-pages branch.
El_Che I use a /docs directory
a github pages branches is just to weird 17:11
jmerelo El_Che: that's what was used for a long time...
AlexDaniel: check now perl6.github.io/p6survey/html/sank...rluse.html 17:17
It's lost the colors, I don't know why...
AlexDaniel jmerelo: you mean “Perl 5 user”?
jmerelo Er
AlexDaniel jmerelo: right side is no longer sorted 17:18
17:18 domidumont joined
jmerelo AlexDaniel: I think it was just by chance... I've generated it again 17:19
The script can generate crossing between any two columns you want. I'm taking requests. 17:20
AlexDaniel jmerelo: it's hard to read if not sorted
jmerelo AlexDaniel: you won't happen to work in academia? 17:21
AlexDaniel jmerelo: 😈
17:24 ChoHag left
jmerelo AlexDaniel: luckily, it could be solved on the Perl 6 side. 17:27
17:33 ChoHag joined
AlexDaniel jmerelo: what about matching“How long have you been using Perl 6?” and “Have you ever experienced issues with using your Perl 6 programs with newer Rakudo versions?” in a similar way? 17:34
jmerelo AlexDaniel: let's see 17:38
AlexDaniel: here you go perl6.github.io/p6survey/html/sank...ssues.html 17:44
AlexDaniel jmerelo: left side is not sorted!!! 17:45
jmerelo AlexDaniel: it's randomly sorted :-)
AlexDaniel: check out the code. If there's a number, I sort it numerically. If there's not, well, tough luck. 17:46
AlexDaniel m: say ‘jmerelo’.comb.pick(*).join
camelia erljome
AlexDaniel erljome: alright!
jmerelo Point is, you have to do it by hand. Not trivial. 17:47
AlexDaniel: still, there's some information you can obtain, right? For instance, most people using for less than a year have had no issue. 17:48
AlexDaniel: no one less than a month has had an issue.
Most people not using it anymore is _not_ because they had an issue.
17:51 wamba left, wamba joined
AlexDaniel yeah 17:55
jmerelo AlexDaniel: this one is interesting perl6.github.io/p6survey/html/sank...score.html 18:03
AlexDaniel: most people who give a low score have stopped using it. 18:04
thowe how many respondents? 18:13
18:13 xtreak left, Possum joined
thowe I think it's interesting that almost half have been using less than a year. 18:14
jmerelo tobs: 220
18:17 domidumont left
sena_kun Can anyone please re-phrase "It (constructor) can be used to create objects by providing it with arguments" line from introduction? I am not sure what the intended meaning is. 18:18
18:21 fake_space_whale left
sena_kun Seems like a much more simple line is just "It takes arguments and creates an object". 18:22
AlexDaniel yes, that sounds clearer to me, but maybe change “it” to “constructor” 18:24
sena_kun Thanks for confirmation. Well, "it" or "constructor" does not really matter, as I am not messing with English version. :)
jmerelo sena_kun: can you please open an issue? 18:27
sena_kun jmerelo, for perl6intro repo? Not sure if needed(though likely is). The bad thing is that things get old quickly if you have a bunch of translations around. But I am considering it now, I also see minor style issues. 18:29
jmerelo sena_kun: I thought that was for perl6/doc. But yes, you can create issues for that too.
sena_kun jmerelo, too much docs detected! :) 18:31
jmerelo sena_kun: perl6/doc is rather a reference. I mean, wherever you found the error, it will help if you create an issue to fix it :-) 18:32
18:32 psychoslave joined
sena_kun jmerelo, agreed. I used to work with doc repo some years ago(huh, checked now and I am even ranked 16th as contributor), yet have other Perl6 things to work with now. :S But, of course, I won't refuse to open an issue when needed. 18:35
18:36 Possum left, Possum joined 18:37 llfourn left 18:41 llfourn joined
jmerelo sena_kun: thanks! 18:54
sena_kun: and come back to perl6/doc. We need you!
sena_kun: you are also one of the authors with the most files modified github.com/JJ/TPF-Grant/blob/7e971...-files.csv 18:58
sena_kun: and the one with the highest entropy github.com/JJ/TPF-Grant/blob/7e971...ntropy.csv
18:59 psychoslave left 19:00 HaraldJoerg joined, stee left
sena_kun jmerelo, Comma needs me too. :) 19:05
19:06 Possum left
jmerelo sena_kun: of course :-). And we all need Comma. 19:06
sena_kun as for files: the thing is that I cannot decently contribute as my English is not native enough plus I don't know much about internals to write docs. Yes-yes, I know about "Not the one who knows better, but the one who is willing to do", yet still. So I concentrated mainly on issues like "grep/sed all http links to https and check the result" or "unify a big bunch of files", basically a large amount of style edit-centered work. Hence the 19:08
files and entropy.
afk. o/
jmerelo sena_kun: where did you see that sentence? I'm using it in my presentation 19:09
sena_kun jmerelo, what sentence? About constructor?
jmerelo sena_kun: I mean "not the one who knows better, but the one who's willing"
sena_kun Ah. It was from your comment, so obviously from you. :)
jmerelo sena_kun: he 19:10
I thought you had picked it up from somewhere else, and I inadvertently did it too. i wanted to quote it properly.
sena_kun: but your explanation of entropy makes sense. 19:11
Going AFK now. Dinner!
See you
19:12 jmerelo left 19:13 Zoffix joined 19:14 Possum joined
Zoffix jmerelo, but… all but one people who no longer use it have rated the docs as above average. Where are you seeing any correlation between docs quality and people stopping using it? 19:15
I don't see any correlation at all in that pairing. 19:16
19:17 MasterDuke joined
Zoffix *all but 3 19:18
19:19 wamba left
Zoffix And IMO the Issues with Perl 6 question is tainted and can't be used. It wasn't clear enough. I imagine a whole bunch of pre-Christmas adopters filled one of the "yes" options even if breakage was pre-stable-release. 19:19
19:28 stee joined 19:29 zakharyas joined 19:30 reisinge joined 19:44 zachk joined, zachk left, zachk joined, reisinge left 20:21 Xliff joined
Xliff \w 20:21
Err... 20:22
\i
Whatevs.
20:22 Zoffix left
Xliff Has anyone gotten Inline::Perl5 to install undfer Rakudo Star? 20:22
20:24 fake_space_whale joined 20:25 louisono joined 20:46 Xliff_ joined, louisono left 20:49 Xliff_ left 20:50 Xliff left 21:02 pmurias left 21:07 MasterDuke left 21:10 lizmat joined 21:30 kaare_ left 21:39 gabiruh left
zostay Yes, I was using it for the first time just this week and it just worked for what I needed. 21:43
22:10 rindolf left 22:15 gabiruh joined 22:29 zakharyas left 22:48 sena_kun left 22:53 gabiruh left
tbrowder_ i enjoyed taking the survey, but it didn’t allow going back so i fumbled some answers. 23:04
also, imo, some answer choices begged for an “other” or additional detail. 23:05
El_Che tbrowder_: it was a good kickstart 23:07
tbrowder_ anyone using 2018.06 having any pod problems? p6 seems to be evaluating code inside a pod block which is a bit disconcerting.
El_Che: you are correct! i’m not knocking it, just saying... 23:08
El_Che tbrowder_: I know, I am just stating the obvious :)
AlexDaniel yeah, next one should be better :)
tbrowder_ ref pod: i regularly use =begin comment/=end comment blocks to cut out code and haven’t noticed errors from them with p6 before. 23:10
AlexDaniel m: gist.github.com/AlexDaniel/fb0880a...bd0f4ee919 23:11
camelia ( no output )
AlexDaniel 6c: gist.github.com/AlexDaniel/fb0880a...bd0f4ee919
committable6 AlexDaniel, ¦6c (31 commits): «»
AlexDaniel tbrowder_: ↑ so what's the issue exactly? 23:12
benjikun .tell Zoffix Sorry, was asleep. Yeah this round's survey is complete, I was going to create charts for each question, but I think the google form results will do fine for now. 23:16
yoleaux benjikun: I'll pass your message to Zoffix.
El_Che benjikun: coordinate with jmerelo, he seemed to be working an graphs as well 23:17
benjikun I saw the one he was making for GitHub contributions, we could certainly make some custom ones together for the survey results as well 23:18
AlexDaniel weekly: 300 people in perl6 org on github: github.com/orgs/perl6/people 23:19
notable6 AlexDaniel, Noted!
AlexDaniel wekkly: (hm, maybe not all of them are visible publicly, so possibly it's not noteworthy) 23:20
hello? 23:21
ha
weekly: (hm, maybe not all of them are visible publicly, so possibly it's not noteworthy)
notable6 AlexDaniel, Noted!
El_Che lol
tbrowder_ AlexDaniel: i need to golf the problem and do a little more checking, but i’m getting errors like undeclared vars from code inside =begin/=end comment blocks. 23:22
AlexDaniel tbrowder_: sounds like you have an unclosed quote " somewhere before =begin 23:23
or maybe /
tbrowder_ could be, but, if that’s the case, maybe LTA error msg 23:24
AlexDaniel oh yeah: RT#75592 [LTA] Long running unclosed quotes trigger unhelpful message
tbrowder_ and granted, it was pretty rough code ahead of the pod 23:25
23:26 HaraldJoerg left 23:27 synopsebot left 23:35 raschipi joined 23:42 benjikun2 joined 23:46 benjikun left