🦋 Welcome to the MAIN() IRC channel of the Raku Programming Language (raku.org). Log available at irclogs.raku.org/raku/live.html . If you're a beginner, you can also check out the #raku-beginner channel!
Set by lizmat on 6 September 2022.
Xliff How can I get documentation for an installed module? 00:01
00:07 jpn joined
ab5tract Xliff: unfortunately I don't think it's easy atm 00:07
00:12 jpn left
ab5tract That is to say, I'm not sure of an equivalent to, say, `perldoc` 00:15
though perhaps, weirdly enough, Podlite might work for this use case.. 00:16
Xliff $ zef install rakudoc; rakudoc <module> 00:17
ab5stract: I was surprised as well.
ab5tract did `rakudoc` do what you needed? 00:19
if so, double surprise :)
Xliff Actually, yes! 00:28
How R U btw?
I appreciate your willingness to try my software, but for now it works only for Ubuntu 00:29
I don't have library data on other distributions
If you are NOT on ubuntu. Would appreciate the names for the following libs: glib, gio, gobject, pango, gdk and gtk+3 00:30
00:34 sftp left, sftp joined 01:08 jpn joined 01:13 jpn left
ab5tract Xliff: sure, I'll give it a try 01:21
I'm a long-time Linux user that de-camped to macOS last year
01:21 hulk joined 01:23 kylese left
ab5tract gtk is notoriosly unsupported over here, but that's as good a reason to give it a try as any 01:30
01:49 jpn joined 01:55 jpn left
jdv www.reddit.com/r/programming/comme..._creators/ 02:03
perl 6 and raku were created by people of different ages?:) 02:04
02:15 hulk left, kylese joined
antononcube @jdv Yeah, but just different middle ages. 02:16
02:16 sftp left
Hmm... I can make a better version of that graph in Raku. 02:20
And an even better one in Mathematica.
jdv what do you mean about middle ages? 02:23
antononcube Maturity of the creator(s), I guess... And that the average age Raku users is 56. 02:26
jdv the graph is about creator age. not user age.
the age of the creator or perl 6 and raku are both the same cause they are the same thing and it was larry 02:27
*of
antononcube I know it is about the creator age. I am basically saying "made by old people for old people." 02:33
02:44 jpn joined 02:48 DarthGandalf left 02:50 jpn left
ugexe Larry wasn’t the same age when he created each 03:07
Oh wait I thought it was Perl and Raku, not Perl 6 and Raku 😬 03:08
03:41 jpn joined
antononcube This is what I got (so far) using "JavaScripdt::D3": 03:42
cdn.discordapp.com/attachments/633...e8bac&
03:46 jpn left
I have to mark bubble charts more “tunable” when it comes to labels colors and font sizes . 03:51
Probably the correlation of age-at-creation vs appeared vs rank is more interesting. 03:57
cdn.discordapp.com/attachments/633...032a0&
cdn.discordapp.com/attachments/633...8f2fc& 04:07
Xliff antoncube: Can you send me that data set? 04:51
Ah! datawrapper.dwcdn.net/rT0yG/1/ -- There is a link at the bottom to get the .csv 04:57
05:05 jpn joined 05:10 jpn left 05:21 jpn joined 05:26 jpn left 05:30 Sgeo left
librasteve according to the data, Larry was 33 when he released perl - so smack in the middle of the Gaussian 05:30
05:40 kylese left, kylese joined 05:45 ranie joined 05:46 ranie left
if raku is to continue to thrive, we need to appeal to a new generation of coders … there’s a renaissance of languages for them to try Nim or Elm or Zig or Roc instead 05:51
06:13 jpn joined 06:18 jpn left 06:40 Chanakan left
holmdunc Elm is a bit moribund already and fading away. Doubt many teenagers use it 06:53
Though that's due to the idiosyncratic stonewall way the project is managed, not a reflection of the language features 07:02
07:14 jpn joined 07:26 jpn left 07:54 DarthGandalf joined 08:16 dakkar joined
patrickb m: "a".lines 08:21
camelia ( no output )
patrickb m: dd "a".lines
camelia ("a",).Seq
patrickb m: dd "".lines 08:22
camelia ().Seq
patrickb Is this a bug or a feature?
(The empty string returning no lines instead of an empty line.)
roast has no test covering this. 08:40
lizmat m: dd "".split("\n") 08:42
camelia ("",).Seq
lizmat good question 08:43
it's at least consistent with "foo".IO.lines where "foo" is an empty file 08:44
08:44 sena_kun joined 08:54 sftp joined
antononcube @Xliff The provided TSV file is with incomplete lines. So, I cheated — I imported it in R first and exported it as “full lined” CSV. (It was too late at night for me to make an ad hoc proper importer.) That R importing introduced zeroes for missing data — so, the plots I pasted above can be misleading. 09:09
patrickb Datapoint the current behavior of Str.lines makes the following necessary: github.com/patrickbkr/Terminal-Wid...026efd1d40 09:13
lizmat you can also write that as if $.text and the code blocks reversed 09:15
patrickb i.e. mapping lines of text will always return a list, except for the empty string where it returns nil
lizmat fwiw, I'm afraid changing it to produce an empty string, might break quite a bit of code
patrickb: alternately, you could use .split("\n") 09:17
patrickb In any case this should be roasted and doced. 09:20
09:21 jpn joined
lizmat and offered split(/ <?after \n> /) for the !chomp alternative 09:38
10:14 dpk left 10:28 jpn left 10:29 jpn joined, dpk joined 10:53 jpn left 11:19 jpn joined 11:23 jpn left 11:46 vrurg left 12:14 jpn joined 12:19 jpn left
tbrowder Xliff: a 12:32
lizmat clickbaits rakudoweekly.blog/2024/05/20/2024-...ry-primed/ 12:33
tbrowder for a quick look you can look up the module in Raku.land
for most modules that should give you source repo where you can read its README 12:34
but it all depends on the author--you may have to read the code. in any event, file an issue for doc suggestions 12:35
13:00 jpn joined 13:16 vrurg joined
patrickb I just found github.com/Raku/old-design-docs/bl...tent-lines That states `Str.lines` should behave similar to `Str.comb(/ \S+ /, $limit)`. 13:34
m: dd "".comb(/ \S+ /)
camelia ().Seq
patrickb That was a wrong paste. Should be similar to `Str.comb(/ ^^ \N* /, $limit)` 13:36
m: dd "".comb(/ ^^ \N* /)
camelia ("",).Seq
patrickb So that'd be a minor datapoint to have .lines return a non empty list. I wonder if a Blin run could suffice to find out about the extent of fallout if we'd change this. 13:38
bisectable: dd "".lines 13:43
bisectable6 patrickb, Will bisect the whole range automagically because no endpoints were provided, hang tight
patrickb, ¦6c (79 commits): «().Seq␤»
patrickb, Nothing to bisect!
antononcube Except patronizing patricks. 13:54
lizmat bisectable6: old=2018.01 dd "".lines 13:58
bisectable6 lizmat, On both starting points (old=2018.01 new=cf87ccf) the exit code is 0 and the output is identical as well
lizmat, Output on both points: «().Seq␤»
lizmat bisectable6: old=2017.01 dd "".lines
bisectable6 lizmat, On both starting points (old=2017.01 new=cf87ccf) the exit code is 0 and the output is identical as well
lizmat, Output on both points: «().Seq␤»
lizmat bisectable6: old=2016.01 dd "".lines
bisectable6 lizmat, On both starting points (old=2016.01 new=cf87ccf) the exit code is 0 and the output is identical as well
lizmat, Output on both points: «().Seq␤»
lizmat bisectable6: old=2015.01 dd "".lines
bisectable6 lizmat, Bisecting by exit code (old=2015.01 new=cf87ccf). Old exit code: 1
lizmat, bisect log: gist.github.com/1029a14eba401e41e8...a2940b3476 13:59
lizmat, (2015-04-16) github.com/rakudo/rakudo/commit/e1...2d4c7ee415
lizmat hehe... that's the introduction of "dd" :-) 14:00
bisectable6: old=2015.01 new=2016.01 say "".lines
bisectable6 lizmat, Bisecting by output (old=2015.01 new=2016.01) because on both starting points the exit code is 0
lizmat, bisect log: gist.github.com/7c5d746465483b703d...b995a131fb
lizmat, (2015-08-29) github.com/rakudo/rakudo/commit/ea...7d1c7e724d
lizmat bisectable6: old=2015.01 new=2016.01 say "".lines.elems 14:01
bisectable6 lizmat, On both starting points (old=2015.01 new=2016.01) the exit code is 0 and the output is identical as well
lizmat, Output on both points: «0␤»
lizmat bisectable6: old=2014.01 new=2015.01 say "".lines.elems
bisectable6 lizmat, On both starting points (old=2014.01 new=2015.01) the exit code is 0 and the output is identical as well
lizmat, Output on both points: «0␤»
lizmat bisectable6: old=2013.01 new=2014.01 say "".lines.elems
bisectable6 lizmat, No build for revision “2013.01”
lizmat patrickb: looks like it has had this behaviour since like forever 14:02
design docs or not
patrickb I've looked up the original implementation of lines. It behaves the same. 14:03
antononcube @jdv Thanks for bringing up the "age of creation" post! It was motivating to finish up some plotting routines features. 14:31
@Xliff Thanks for the link the plot and data ! Also, I was wrong about the TSV file being with incomplete lines. (This is the TSV file I used : pldb.io/posts/age.tsv .) 14:34
ab5tract patrickb: if you push a branch which fixes it, I'll blin it for you 14:53
patrickb ab5tract: Whoop! Will do!
antononcube Here is my latest Raku-made age-of-creation plot. Raku's / Larry's bubble is the top one in the top-rightmost rectangle. 15:02
cdn.discordapp.com/attachments/633...0bfc8&
15:48 soverysour joined 16:07 jpn left
librasteve my opinion of Elm and Roc success comes from Richard Feldman “Unscripted” podcast, so not entirely objective 16:24
good podcast tho 16:27
i was wondering if raku would benefit from the equivalent of the elm html package approach to building more maintainable websites package.elm-lang.org/packages/elm/html/latest/ … forgive me maybe we have it already? what do you’ll think? 16:30
you’all 16:31
16:34 jgaz joined 16:38 dakkar left
holmdunc Agreed. I like this one too which has a similar format signalsandthreads.com 17:04
ab5tract librasteve: reminds me of SmokeMachine's presentation last year, though that was a bit more focused on components 17:10
it would be very cool to have something like what elm offers 17:11
librasteve pretty much raku can do elm - (except for purity and javascript target … which may dilute the benefits a bit) 17:23
17:24 jpn joined
SmokeMachine Why not js target? 17:28
fco.github.io/MemoizedDOM/todo6.html 17:29
17:29 soverysour left
ab5tract SmokeMachine: slick! 17:31
17:32 jpn left
SmokeMachine (It takes long to compile… but please, take a look att the code on the browser… you will see it’s Raku…) 17:32
(I just used other people stuff..) 17:37
17:38 soverysour joined, soverysour left, soverysour joined
lizmat how long is long? 17:40
SmokeMachine Less than a minute on my arc…. 17:46
usercontent.irccloud-cdn.com/file/...313613.JPG
lizmat I guess Safari doesn't like it 17:47
SmokeMachine it took 5s on my safari... 17:49
lizmat I waited several minutes :-(
SmokeMachine usercontent.irccloud-cdn.com/file/.../image.png
on a Mac? or on iphone? it didnt open on my iphone... 17:50
17:52 jgaz left
SmokeMachine it's very old... I wonder how better it would work with current code 17:52
(it used to work on iphone... :( ) 17:53
lizmat MBP in Intel 17:54
SmokeMachine mine is arm... but it used to work on my old intel as well...
lizmat on an M1: same< just the text "loading..." :-( 17:55
SmokeMachine usercontent.irccloud-cdn.com/file/....57.18.gif 17:58
is there any error message on console? 17:59
lizmat TypeError: stack[stackIndex].getFunctionName is not a function. (In 'stack[stackIndex].getFunctionName()', 'stack[stackIndex].getFunctionName' is undefined) 18:07
TypeError: Raku.eval is not a function. (In 'Raku.eval(code)', 'Raku.eval' is undefined) 18:08
18:37 soverysour_ joined 18:49 soverysour left 18:51 xinming left, PolarBearXL joined 18:52 soverysour joined 18:53 xinming joined 18:55 jgaz joined 19:15 soverysour left
[Coke] is it me or is Z<broken> in pod? 19:27
This is Z<not> OK. -> "This is not OK" when run through raku --doc 19:28
lizmat contents => ["This is ", Pod::FormattingCode.new(type => "X", meta => [], config => {}, contents => ["not"]), " OK"])])] 19:31
is what $=pod gives me
[Coke] "X" ?? 19:32
lizmat yeah, oddly, RakuAST does the same:
RakuAST::Doc::Paragraph.new(
"This is ",
RakuAST::Doc::Markup.new(
letter => "X",
opener => "<",
closer => ">",
atoms => (
"not",
)
),
" OK\n\n"
)
[Coke] ... Should be Z, no? 19:33
19:35 xinming left
lizmat feels like, yes 19:36
19:37 xinming joined
lizmat [Coke]: fwiw it deparses incorrectly, so please make an issue for it 19:42
I will fix it in RakuAST :-) and maybe in the legacy grammar, unless someone beats me to it
[Coke] $=pod is giving me Z here.
lizmat hmmmm 19:44
indeed... my source was wronng :-( 19:45
the deparse in RakuAST is also ok *phew* 19:46
% RAKUDO_RAKUAST=1 raku --rakudoc 1 19:47
This is OK
so "fixed in RakuAST" :-)
[Coke] does rakuast use Pod::To::Text? 19:53
19:58 jpn joined, PolarBearXL left
lizmat no, RakuDoc::To::Text (from memory) 19:58
lib/RakuDoc/To/Text.rakumod 19:59
[Coke] I think it's just NYI ater 20 years and no one noticed. 20:01
working on a patch. 20:02
20:14 jpn left
[Coke] github.com/rakudo/rakudo/pull/5575 20:15
gfldex Is there a way to add a multi candidate to the front of the list, without a lexical proto and forwarding to the sub in core? 20:18
lizmat not that I know of 20:33
afk& 20:35
21:00 soverysour_ left 21:03 soverysour joined, soverysour left, soverysour joined 21:07 soverysour left 21:47 jpn joined 21:52 jpn left
patrickb ab5tract: I've created a patch. It breaks loads of spectest and some STDOUT output of compile. That settles the matter. No need for a Blin. Thanks anyways! 22:02
22:02 Sgeo joined 22:18 sena_kun left
[Coke] I have a patch. is there a way to do the equivalent of git commit -p where I pick and choose what gets applied? 23:44
(Trying to get some of kaipei's old doc patches finally applied) 23:47
er, kaiepi!