00:20 yewscion joined 00:23 yewscion left 00:37 floyza joined
floyza I've been defining the WHICH method to use custom classes as keys in sets, but I tried using one of those classes as a key for a hash, and it acted as though the default WHICH method was being used. What am I doing wrong? 00:41
For example: `class T { has Int $.x; method WHICH() { ValueObjAt.new("T|$!x") } }; my %a = T.new(:x(5)) => 10; say %a{T.new(:x(5))}` returns (Any)
01:22 MasterDuke joined 01:34 Manifest0 left 02:16 hulk joined, kylese left 02:27 zetaaaa left 02:29 nine left 02:30 nine joined
guifa floyza remember that a Hash is by default typed with Str keys and Any values 02:39
m: class T { has Int $.x; method WHICH() { ValueObjAt.new("T|$!x") } }; my %a{Mu} = T.new(:x(5)) => 10; say %a{T.new(:x(5))} 02:41
camelia 10
guifa make the key Mu (or probs Any too)
floyza guifa: Ah, ok. I guess that makes sense. I've been getting confused with typed keys when everything is converted into a string before being used as a key anyways. Thanks a bunch! 03:03
03:15 hulk left, kylese joined 03:16 MasterDuke left 03:26 yewscion joined 03:30 kylese left 03:35 kylese joined 03:37 yewscion left 04:11 vrurg left 04:23 vrurg joined 04:44 yewscion joined 04:49 yewscion left 05:02 floyza left 05:34 yewscion joined 05:38 yewscion left 06:00 yewscion joined 06:05 yewscion left 06:46 yewscion joined 06:50 yewscion left 07:21 yewscion joined 07:26 yewscion left 08:48 Manifest0 joined 09:41 zenmov_ left, zenmov joined 10:14 skyesoss joined 10:42 skyesoss left 10:43 skyesoss joined 10:44 sena_kun joined 10:45 skyesoss left, skyesoss joined 10:50 yewscion joined 10:52 skyesoss left 10:53 skyesoss joined 10:54 skyesoss left, skyesoss joined, yewscion left
Geth advent/main: 1505406670 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | raku-advent-2024/authors.md
Schedule 22
 10:55
10:55 skyesoss left
Geth ecosystem/main: 9a27fa50b9 | (Elizabeth Mattijsen)++ | META.list
Remove Number::Bytes::Human

It lives as a Raku Community module now		 10:56
11:05 Sgeo left 11:11 yewscion joined 11:16 yewscion left
Geth ecosystem/main: 23554bc2ee | (Elizabeth Mattijsen)++ | META.list
Remove Text::Caesar

It lives on as a Raku Community module		 11:36
11:46 yewscion joined 11:50 yewscion left 12:07 yewscion joined 12:11 yewscion left
tbrowder hi, i can use "subset X ..." and use test "is-ok $foo, X" to check it, no proplemo 12:17
* "isa-ok" 12:18
but i want to negate the test to show anothe $bar is not of type X. how? 12:22
use cmp-ok? trying that now offline... 12:23
12:41 yewscion joined
guifa you could use nok ($foo.isa: X) 12:43
tbrowder i just used "nok ($bar ~~ X)" which worked. 12:45
12:45 yewscion left
tbrowder guifa, thanks. i had tried a variant of yr method but couldn't get the syntax right 12:46
12:47 hellwolf left
tbrowder hm, did you mean ".is"? 12:48
guifa probably 12:50
normally ~~ works okay
but for testing I'd avoid it just in case the right hand has an ACCEPTS method that does osmething funky
tbrowder m: my $a = 2; nok ($a.isa: Int) 12:51
camelia ===SORRY!=== Error while compiling <tmp>
Undeclared routine:
nok used at line 1. Did you mean 'not'?
tbrowder m: use Test;
camelia ( no output )
tbrowder m: use Test; my $a=1; nok ($a.isa: Int) 12:52
camelia not ok 1 -
# Failed test at <tmp> line 1
tbrowder m: use Test; my $a=1; ok ($a.isa: Int) 12:54
camelia ok 1 -
Geth ecosystem/main: 4545ee0d38 | (Elizabeth Mattijsen)++ | META.list
Removed Lingua::En::Conjugate

It lives as a Raku Community module now
tbrowder cool!
haven't looked at type methods in a while, good stuff 12:55
thanks, guifa
13:11 Geth left, Geth joined
Geth ecosystem/main: 7dc247aa8b | (Elizabeth Mattijsen)++ | META.list
Remove Unicode::UTF8-parser

It lives on as a Raku Community module		 13:17
13:25 yewscion joined
antononcube @lizmat I my planned Raku Advent post "Chebyshev polynomials and fitting workflows" I have to refer to this issue: github.com/pierre-vigier/Perl6-Mat.../issues/70 13:28
Maybe, that is not a "good look" for an advent post? 13:29
lizmat there's quite a few open issues on that repo / module 13:30
13:30 yewscion left
antononcube Yeah... 13:30
lizmat without any response from the author it seems
seems author is no longer active at all
maybe it's time to fork 13:31
librasteve I observe that more and more modules are being adopted by the community - but my concern is that soon we will have unsupported / outdated community modules and will no longer be able to sustain a high quality
lizmat as a raku community module 13:32
librasteve to solve this, perhaps folks who are interested in the subjest matter should be encouraged to fork, improve and release
lizmat it seems that you and librasteve are users of it
so maybe the two of you should take it on? 13:33
I'll do the preliminary work of setting up the infra structure
antononcube Sounds good. (Although, I am not that willing to deal with "Math::Matrix".) 13:34
@lizmat Should I write, a Raku advent post on a different subject? 13:35
librasteve as devil's advocate, this sounds like "lets move this module to community and then steve and anton can be earmakred to do the maintenance"
however, I am proposing that "if you have a dependency on a module and find that it is still valuable to your work then you the individual should consider forking it and releasing a ne version under your authorship 13:36
"
@antononcube - your new Issue is a new feature request to improve performance - since you are asking for this, perhaps you should consider what it would take for you to make a PR - or make the change in your own fork? 13:38
antononcube Yes, I considering, cloning and doing PR.
librasteve btw - this is the model I followed with adding fragments to Cro - and now I am a proud member of the Cro maintenance crew
lizmat ok, I'll do the fork, and look at it to see if there's some easy stuff to be done
and then leave it to the community 13:39
antononcube Sounds good.
lizmat at least we'll be able to release new versions without having to take full responsibility for the module
antononcube Interesting point! 13:40
librasteve ok 13:41
lizmat should be there in 30 mins or so 13:44
meh, there's a fork by lichtkind, but that was never uploaded as such 14:05
going to ignore that for now
antononcube Hmm... I am not sure who is the primary developer -- both repositories refere to each other as "a new home." 14:14
lizmat yeah 14:23
and both worked on it further :-( 14:24
antononcube @lizmat Ok, I am bravely continuing writing that fitting/regression Raku Advent blog post. 14:31
lizmat cool! :-) 14:32
antononcube I find it to have too many "nice features." (To me at least.)
lizmat it definitely has a lot of pod :-) 14:33
antononcube Hmm... I meant the my post when finished is going to show some "very nice" features. As for "Math::Matrix", I debating how much should I accetuate on its use. 14:36
Is this the most "official" reference to the TPRC-2025, or there is a better one? www.perl.com/article/get-ready-for...onference/ 14:49
14:54 yewscion joined
lizmat it looks better than the official one :-) 14:54
antononcube Great! 14:55
lizmat antononcube librasteve Math::Matrix 0.4.1 just uploaded to zef
sorry it took more than twice as long :-) 14:56
issues can now be posted at github.com/raku-community-modules/...rix/issues 14:58
14:59 yewscion left
antononcube 👍 15:03
tbrowder lizmat: i'm still chugging on my too ambitious post. raku question on subsets: after a successful test of my very complex subsets, they really look and act as regexes. aren't they really named regexes behind the scenes? 15:09
lizmat not unless you use regexes in where clauses afaik
tbrowder ah, i am using where clauses 15:10
i always forget if they are really needed for subsets 15:11
but when trying to figure whether to use rx or not, naming the subset seems less confusing 15:12
lizmat not necessarily, but yeah: very often
tbrowder TIMTOWTDI and love it ! 15:13
15:23 El_Che left 15:26 El_Che joined 15:30 yewscion joined 15:35 yewscion left 15:51 yewscion joined 16:05 yewscion left
tbrowder *and we all love it i hope 16:39
antononcube: you here?
antononcube Yes! 😉 16:40
tbrowder putting link to my Christmas tree: <script src="gist.github.com/tbrowder/c781fdcc0...script> 16:41
antononcube Hmm... clicking on that link gives "Not found."
tbrowder ignore that!! 16:42
hm: gist.github.com/tbrowder/c781fdcc0...2dd5c50a7c
^^^
see in context at: raku-advent.blog/2020/12/11/day-11...h-a-class/ 16:43
as in TWEAK 16:44
see it now here? 16:46
antononcube Ok, yes. 16:47
tbrowder that is a valid Raku class
antononcube 💡 I think you posted that link (or a related) earlier this week, but I missed that it is a class. 16:48
tbrowder with embedded ornaments
i think i posted a bad link--got my years mixed up 16:49
with a subtle cross on top 16:50
and a tree stand 16:51
bye 16:52
16:55 yewscion joined
antononcube A, yeah, there is ✜ 16:55
17:00 yewscion left
SmokeMachine tbrowder: subsets can be used very differently from regex, eg: x.com/smokemachine/status/15132843...57127?s=46 17:19
17:19 Guest37 joined
Guest37 This comes from The Weekly Challenge #299, so spoilers ... 17:27
Gist code here: gist.github.com/nd3i/90789f85b4d53...2ab2859095
This has to do with building a regex in a string, then using the string as a regex matcher.
The base question is: why do #s 1 & 2 work but $3 does not?
17:51 sorenson left 18:19 abraxxa-home joined, abraxxa-home left 18:20 abraxxa-home joined 18:22 yewscion joined
Geth advent/main: 1cdf59eff3 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | raku-advent-2024/authors.md
19 -> 15, 15 -> 16, 16 -> 17, 17 -> 18, 18 -> 19
 19:46
lizmat Guest37: at first sight, that feels like a bug: it appears to be losing the positional capture there 19:55
Guest37 Thanks for having a look. 19:56
librasteve what is <|w> ? 20:07
tbrowder smokemachine: that looks pretty handy! i really need to revisit RedORM, maybe in the New Year... 20:08
Guest37 <|w> is a word boundary. It restricts the match to the beginning of a word. At least that's my understanding. 20:13
librasteve okaay - docs.raku.org/language/regexes#Word_boundary ... says <?wb> ... bit confused 20:16
timo i imagine the captured stuff would go in something other than $0
you may have to <blah=$re> and then $<blah> for that?
20:17 Guest37 left 20:23 Sgeo joined
timo to me, positional captures are "syntactical" at the level of the regex, and a <$foo> interpolation is more like an invocation than literally plopping the regex in your regex in the same position 20:25
21:05 abraxxa-home left 22:41 sena_kun left 23:02 skyesoss joined 23:19 sorenson joined 23:20 yewscion left 23:30 kst joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!