[Coke] welcome aboard 00:00
00:53 arkiuat left 00:58 deoac joined 01:09 arkiuat joined
korvo Solved it! Humming-Bird::Core::listen() requires a :no-block argument in order to actually `start` concurrently. 01:13
disbot7 <aruniecrisps> @librasteve sure I can do that, what exactly are you looking for? 02:28
<aruniecrisps> Or can it just be any snippet? 02:30
02:36 kylese left 02:37 kylese joined 02:49 deoac left
disbot7 <jubilatious1_98524> @aruniecrisps have you tried NFKD decomposition? docs.raku.org/routine/NFKD 03:04
03:07 deoac joined 03:15 kylese left, kylese joined 03:28 itaipu left 03:44 itaipu joined 04:11 Aedil joined 04:13 cm left 04:15 deoac left 04:20 lichtkind_ joined 04:21 cm joined 04:23 lichtkind left 04:38 itaipu left
Voldenet jubilatious1_98524: it's not a contradiction – to group things you need a container for the grouping (a scope) and another value that escapes that scope is the only one preserved 05:16
disbot7 <jubilatious1_98524> @Voldenet so what's the 'escape key'? 05:17
Voldenet 'escape key' hm? 05:18
what do you mean
m: my $result = do { my $a = 1; my $b = 2; $a + $b }; say $result # this is the pattern - $result gets assigned, but $a and $b are not preserved 05:19
camelia 3
Voldenet it can get a lot more complicated with FINALIZER module for instance 05:20
disbot7 <jubilatious1_98524> Have you posted this do { ... } idiom previously? I might have missed it. 05:21
Voldenet yes a while earlier, but without 'do'
disbot7 <jubilatious1_98524> It doesn't seem exactly like the Rust code you posted, which talked about immutability of variables manipulated in that blog code-example. 05:23
Voldenet in C# there's `using(var x = new SomeIDisposableImpl()) { … }` for tracking IDisposable values
it's a bit similar to using do-block
where you can `my $item = do { using FINALIZER; my $x = FoosRepository.new(:conn(await $connections.connect()); $x.items.fetch(:id(42)); }; ` 05:25
disbot7 <jubilatious1_98524> In my Raku one-liners, I use BEGIN blocks all the time: unix.stackexchange.com/search?q=us...7738+BEGIN
Voldenet obviously rust is a different kind of language, but you can make things immutable by turning it into List on the end 05:26
for example
disbot7 <jubilatious1_98524> Here are some (other) examples with state variables: unix.stackexchange.com/search?q=us...7738+state
<jubilatious1_98524> So I thought a hypothetical {STATE .... } keyword would replicate what you wanted to bring over from Rust. 05:28
Voldenet I agree that BEGIN blocks are very useful, I still do time things with `now - BEGIN now` :>
though I guess do-block is just something that's totally unnecessary
it hardly does anything
so using it in this context is just not very obvious 05:29
disbot7 <jubilatious1_98524> (Never heard of FINALIZER before...).
Voldenet very cool module, it's something that should be in stdlib, basically it implements the IDisposable (C#) AutoCloseable (java) or Drop (rust) patterns 05:31
so instead of directly using DBIish, you'd use method attaching .dispose to current scope 05:33
disbot7 <jubilatious1_98524> do blocks make for loops imperative, I'll use it instead of (complicated)map calls: unix.stackexchange.com/a/734258/227738 05:37
Voldenet I use for loops if I plan for them to have side effects 05:38
.map is for things without side effects 05:39
05:39 [Coke] left
disbot7 <jubilatious1_98524> As long as you assign outside the do for loop, there doesn't seem to be any difference from map. 05:41
05:42 [Coke] joined
Voldenet m: say do for ^10 { $_ * 10 } # this feels a bit weird to write compared to `(^10).map({ $_ * 10 })` 05:42
camelia (0 10 20 30 40 50 60 70 80 90)
Voldenet of course, in the end they're both equivalent in what they do, so it's only a matter of preference 05:45
disbot7 <jubilatious1_98524> m: my @a = 0..9; @a = do for @a { <a b c d e f g h i j>[ $_ ] }; .say for @a.pairs; 05:48
<jubilatious1_98524> raku -e 'my @a = 0..9; @a = do for @a { <a b c d e f g h i j>[ $_ ] }; .say for @a.pairs;' 0 => a 1 => b 2 => c 3 => d 4 => e 5 => f 6 => g 7 => h 8 => i 9 => j
05:48 arkiuat left
SmokeMachine For making stuff immutable, one could also use something like this: raku.land/zef:FCO/ValueClass 05:48
Or maybe something like these: raku.land/zef:FCO/Functional::LinkedList raku.land/zef:FCO/Functional::Queue or raku.land/zef:FCO/Functional::Stack 05:50
Voldenet (it's not the same as in rust - in rust mutability will stop some optimizations and immutability could make them possible) 05:51
rust is just very different
SmokeMachine Sorry, I’m not comparing with rust… (I don’t know rust) 05:52
Voldenet I mean, language knows the concept of mutability
in raku, language will asume that ValueClass is mutable 05:53
the only thing preventing mutation is Proxy STORE dying 05:54
06:18 arkiuat joined 06:23 arkiuat left
Voldenet the above example doesn't need for btw 06:27
m: my @a = 0..9; @a = do for @a { <a b c d e f g h i j>[ $_ ] }; .say for @a.pairs; # this
camelia 0 => a
1 => b
2 => c
3 => d
4 => e
5 => f
6 => g
7 => h
8 => i
9 => j
Voldenet m: my @a = <a b c d e f g h i j>[0..9]; .say for @a.pairs; 06:28
camelia 0 => a
1 => b
2 => c
3 => d
4 => e
5 => f
6 => g
7 => h
8 => i
9 => j
korvo LangJam GameJam is done. I made a little S-expression language declaring an idle game and a friend wrote a little scenario. I'll have a blog post ready in a day or two.
Voldenet m: my $a = 0; my $b = 1; say do for ^10 { my $n = $b + $a; $b = $a; $a = $n; $n } # this is something that totally wouldn't work with .map 06:29
camelia (1 1 2 3 5 8 13 21 34 55)
06:52 arkiuat joined 06:57 arkiuat left 07:26 arkiuat joined 07:31 arkiuat left
disbot7 <jubilatious1_98524> m: my $a = 0; my $b = 1; for ^10 { my $n = $b + $a; $b = $a; $a = $n; print "$n " }; put(); 07:44
<jubilatious1_98524> raku -e 'my $a = 0; my $b = 1; for ^10 { my $n = $b + $a; $b = $a; $a = $n; print "$n " }; put(); ' 1 1 2 3 5 8 13 21 34 55 07:45
<jubilatious1_98524> @Voldenet did you not just show that the code worked without map? 07:46
08:00 arkiuat joined 08:05 arkiuat left 08:18 Aedil left 08:28 arkiuat joined 08:33 arkiuat left 08:41 Sgeo left
Voldenet yes, maybe I mistyped, but from my understanding it might not work with map 08:54
m: my $a = 0; my $b = 1; my $x = (^10).map({ my $n = $b + $a; $b = $a; $a = $n; $n }); say $x, $x
camelia (1 1 2 3 5 8 13 21 34 55)(1 1 2 3 5 8 13 21 34 55)
Voldenet Yes, it happens to work as expected, maybe it's even documented
08:55 jjidozzz joined
Voldenet but side effects feel wrong in such snippet – it might just work because initial sequence is ^10 or it might work for lists etc. 08:56
It doesn't conceptually map values at all, so it might not work 08:58
09:02 arkiuat joined
Voldenet in all other languages such things may or might not work: dotnetfiddle.net/Widget/zNieq1 09:03
09:07 arkiuat left
Voldenet m: my $a = 0; my $b = 1; my $x = (1..*).race(:1batch).map({ my $n = $b + $a; sleep rand; $b = $a; sleep rand; $a = $n; $n }); say $x[^10] # just throwing in "race + sleep" makes it not work 09:22
camelia (1 1 1 2 1 1 3 3 3 4)
09:29 arkiuat joined 09:34 arkiuat left 10:03 arkiuat joined 10:08 arkiuat left 10:23 arkiuat joined 10:30 Aedil joined 10:36 arkiuat left 10:48 arkiuat joined 10:54 arkiuat left 11:08 arkiuat joined 11:12 arkiuat left 11:26 arkiuat joined 11:35 arkiuat left 12:05 arkiuat joined
tbrowder if you are interested in tonyo's help on DateTime formatters, check my code at github.com/tbrowder/LocalTime. Besides the published code you can see code based on tonyo's suggestion in the ./tools directory. if you can still search #raku look for tonyo / tbrowder/ formatter in 2022 novemberish 12:07
12:10 arkiuat left 12:12 Ekho left
tbrowder also see tonyo's name in the file names and code in the ./dev directory 12:13
12:21 itaipu joined 12:25 jjidozzz left 12:31 arkiuat joined 12:34 melezhik joined 12:36 arkiuat left
lizmat Q: what do you think was a noticeable new module in the Raku Ecosystem? (yes, you can mention your own :-) 12:38
12:46 sibl joined 12:47 arkiuat joined 12:50 sibl left 12:52 arkiuat left 13:11 arkiuat joined 13:16 arkiuat left
disbot7 <antononcube> @lizmat How about 5 answers? 13:26
lizmat sure!
disbot7 <antononcube> Agh, you are too gracious. I thought you might prompt me to reduce to 1, at most two.
lizmat there's no guarantee I'll include all 5 :-) 13:28
disbot7 <antononcube> Hmm... this can be a motivation to provide ML supplied answers, across the whole ecosystem. 13:29
<antononcube> Or, just using "plain" statistics. 13:30
lizmat using "plain" statistics is not really an issue 13:31
I'm more interested with what resonated in the community (rather than just with me)
disbot7 <antononcube> First of all, I think simple metrics, like, frequencies of updates, number of lines, discussions on IRC, do not tell the whole picture. 13:32
lizmat indeed, they do not
13:32 arkiuat joined
disbot7 <antononcube> But, that can be used to support certain candidates -- post-selection. 13:32
lizmat take e.g. raku.land/zef:patrickb/Anolis 13:33
just a simple module, 1 release so far, but I know how much work went in there to get to that point
disbot7 <antononcube> Yes, I looked at/browsed the page but did not install it yet. 13:34
lizmat see also raku-advent.blog/2025/12/21/a-terminals-tale/ :-)
disbot7 <antononcube> Just about to read it -- it was published "later" than I thought.
<antononcube> (Meaning, not at midnight UTC-0.)
lizmat yeah, technically it was on time, just not easily found :-)
disbot7 <antononcube> Agh, yeah -- some of my posts were like that too. 13:35
<antononcube> Back to my statistics angle -- I was looking for an excuse to try and use "Draku" by Brian Duggan for those. 13:36
<antononcube> *those statistics. 13:37
13:37 arkiuat left 13:42 itaipu left 13:45 arkiuat joined 13:52 arkiuat left 13:59 itaipu joined
disbot7 <librasteve> lizmat: my top modules are Jupyter::Chatbook and FatRatStr (wot i rote) 14:15
<librasteve> first because I use it every day and it has “taught me” LLM best practice prompt engineering, second because it demonstrates the usefulness Allomorphs and makes FatRat more practical 14:17
14:18 arkiuat joined
disbot7 <antononcube> @librasteve I think @lizmat said "new modules" which I interpret as "2025 modules." 14:22
<antononcube> "Jupyter::Chatbook" is, of course, great, but it is also an "old" module. 14:23
<antononcube> And thanks for proposing it!
14:53 melezhik left
tbrowder i second the FatRatStr package--very allomorph helpful 15:40
arkiuat tbrowder: thanks for the info on DateTime :formatter. I still wish I could find those ruminations about the original design that I remember reading somewhere. 15:45
15:54 itaipu left 16:10 itaipu joined 16:24 melezhik joined
lizmat librasteve you don't need sets to be able to use set operators! 17:00
m: my @cities = <London Oslo Paris Amsterdam Berlin>; my @visited = <Berlin Oslo>; say "I still need to visit the following cities: " ~ (@cities (-) @visited);
camelia I still need to visit the following cities: Paris London Amsterdam
lizmat in response to news.ycombinator.com/item?id=46346132 17:02
17:06 silug left
Geth advent/main: d7804d25d0 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | raku-advent-2025/authors.md
Update an move review to the 25th
 17:08
lizmat One slot in the Advent Calender opened up! 17:09
17:14 silug joined 17:42 [Coke] left, [Coke] joined 17:48 arkiuat left 17:51 arkiuat joined
disbot7 <jubilatious1_98524> m: my @cities = <London London Oslo Paris Amsterdam Berlin>; my @visited = <Berlin Oslo Oslo>; say "I still need to visit the following cities: " ~ (@cities (-) @visited); 18:00
<jubilatious1_98524> raku -e 'my @cities = <London London Oslo Paris Amsterdam Berlin>; my @visited = <Berlin Oslo Oslo>; say "I still need to visit the following cities: " ~ (@cities (-) @visited);' I still need to visit the following cities: London Paris Amsterdam
18:13 arkiuat left 18:28 arkiuat joined
grondilu got: '===> Failed to find dependencies: pq:ver<5>:from<native>' when trying to install Red (raku.land/zef:FCO/Red) 19:05
oh nevermind that's documented in the readme 19:08
19:13 melezhik left 19:21 abraxxa-home joined 19:23 arkiuat left 19:28 arkiuat joined 19:33 arkiuat left 19:38 abraxxa-home left 19:42 arkiuat joined 19:44 Sgeo joined 19:54 Ekho joined
disbot7 <librasteve> lizmat: ah - should have stringified - thanks! 20:02
lizmat librasteve also lose the .Set :-) 20:03
disbot7 <librasteve> holy moly - that's cool
lizmat m: dd <a b c> (-) <a>
camelia Bool = Set.new("c","b")
lizmat m: dd ~<a b c> (-) <a> 20:04
camelia Bool = Set.new("a b c")
lizmat m: dd ~(<a b c> (-) <a>)
camelia "c b"
lizmat set operators either produce a QuantHash, or a Bool
but the input doesn't have to be a QuantHash 20:05
20:05 abraxxa-home joined
lizmat m: dd <a b c> (elem) <a b c d e f> 20:05
camelia Bool::False
lizmat m: dd "a" (elem) <a b c d e f>
camelia Bool::True
lizmat m: dd <a b c> (<) <a b c d e f>
camelia Bool::True
lizmat m: dd <a b g> (<) <a b c d e f> 20:06
camelia Bool::False
disbot7 <librasteve> wonders how to retain the list order
lizmat well... it doesn't as a Set is a QuantHash
disbot7 <librasteve> just an idle wonder in case that's a requirement of "array math" ... didnt expect to get it from Set 20:07
lizmat m: my @c = <a b c d e f g>; my @s = <a e>; dd (@c (-) @s){@c}:k 20:08
camelia ("b", "c", "d", "f", "g")
lizmat that would keep the order 20:09
disbot7 <librasteve> .oO - very glad I asked that is awesome 20:11
<librasteve> betcha can't do that in Ruby
lizmat :-) 20:12
disbot7 <librasteve> finally groks index adverbs (about time) 20:16
<antononcube> @lizmat "One slot in the Advent Calender opened up!" -- fairly late notice... 20:37
21:38 Aedil left 21:39 abraxxa-home left
tonyo . 21:55
whew, glad it was a mention from tbrowder and not something with the eco ;) 21:56

Powered by Cro and the Raku® Programming Language.

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

Thank you!