🦋 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.
SmokeMachine m: say "boo".comb.permutations.unique 06:53
camelia ((b o o) (b o o) (o b o) (o o b) (o b o) (o o b))
SmokeMachine should unique work for a list of lists? 👆
m: say <b o o>.WHICH; say <b o o>.WHICH 06:58
camelia List|4052417101152
List|4052417102688
SmokeMachine Isn't List a value type? 06:59
quib m: say "boo".comb.permutations.unique(with => &[eqv])
camelia ((b o o) (o b o) (o o b))
SmokeMachine m: say "boo".comb.permutations.unique(:with(&join)) 07:00
camelia ((b o o))
SmokeMachine m: role ValueList { method WHICH { ValueObjAt.new(self.map(*.WHICH).join: "|") } }; say "boo".comb.permutations.map({$_ but ValueList}).unique 07:14
camelia ((b o o) (o b o) (o o b))
SmokeMachine Should List be value objects? 08:21
Voldenet I'm suspecting that value semantics for things will need to get implemented at some points 08:49
because NativeCall doesn't handle structs now, only pointers to structs
which probably would add value semantics to regular subroutines 08:50
which would probably lead to implementing value semantics for lists
Voldenet it's a lot of "probably" in that, changing those semantics for lists could be breaking change somewhere 08:54
Btw, that approach using `WHICH` is really slow 09:00
eqv can compare lists more cheaply
m: role ValueList { method WHICH { ValueObjAt.new(self.map(*.WHICH).join: "|") } }; my @a = "booooooo".comb.permutations.map({$_ but ValueList}).unique;say now - BEGIN now 09:01
camelia 0.654847952
Voldenet m: my @a = "booooooo".comb.permutations.unique(:with(&[eqv])); say now - BEGIN now
camelia 0.168533007
antononcube I cannot upload with fez this package: github.com/antononcube/Raku-WWW-Gemini . I get the message "Error reading META from your last upload". Can someone spot an error in the META file: github.com/antononcube/Raku-WWW-Ge...META6.json 11:20
Well, it looks like fez does not like the dependencies specifications in that MEAT6 file -- it is not clear to me why. 11:41
Xliff \o 11:44
replit.com/@Xliff/Weekly260#main.raku - Weekly 260, second problem
For the life of me, I can't understand the first.... I might need coffee.
antononcube Damn! The upload with fez is still problematic. (For “WWW::Gemini” linked above.) 12:25
ab5tract hmm, `META6.json` looks fine to me 12:30
antononcube Thanks! 13:19
tbrowder__ antononcube: i can't see my msg here, but i'm forking yr prob repo. you would be better off if you would use App::Mi6 to help manage yr repos. 13:29
have you also tried to zef install . --debug ?
tbrowder__ it installed locally just fine so far 13:33
tbrowder__ using mi6 to upload is much better. it does some xtra checking 13:36
using Mi6::Helper "helps" start a new repo project. converting an old one is a small but worthwhile effort. i would be happy to submit PR tondo that if you wish. 13:38
afk& 13:39
tbrowder__ Also make sure zef is updated 13:44
antononcube @tbrowder Thanks — I will try mi6. 13:45
@tbrowder I think you wrote an Advent article discussing "Mi6::Helper" (or something similar.) 14:03
Somehow it is not easy to find... 14:04
tbrowder__ use Raku.land 14:21
mi6 is App::Mi6 14:22
antononcube Ok.
tbrowder__ mi6-helper is in Mi6::Helper
i probably need to update mi6-helper repo tags in the meta file 14:24
antononcube @tbrowder "[...] i would be happy to submit PR tondo that if you wish." -- please do. 14:56
[Coke] docs team could definitely use some editors or people working the bug queue. Anyone interested in contributing to the docs site, it's very easy to suggest small edits, and I'm definitely open to folks who want to add or edit something more substantial. 16:09
please join us in raku-doc (or ping me in this channel if you prefer)
lizmat someone should make lobste.rs have a "raku" category: lobste.rs/search?q=raku&what=s...der=newest 16:32
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2024/03/11/2024-11-membering/ 18:12
jdv what does it take to be on this grants commitee? 18:17
codesections: ^ 18:20
aruniecrisps I saw this from a little while ago, does anyone know the status of this? gist.github.com/codesections/841c9...d0e06dacfb 18:25
Skyb0rg I'm installing the Rakudo Star Bundle, however I am confused on the installation location settings. I would like to set the installed location to /usr/local, but install the files into /usr/local/stow/rakudo-star-2024.02 to work with GNU stow. Is this possible? 18:35
lizmat lobste.rs/s/p8ptwq/can_we_have_raku_tag_please 18:49
jdv: grant committee chores are basically: reading proposals and voting by email 18:50
with sometimes email discussions inbetween
weekly: lobste.rs/s/p8ptwq/can_we_have_raku_tag_please 18:51
notable6 lizmat, Noted! (weekly)
Geth Pod-To-HTML: e8e8390347 | (JJ Merelo)++ | .travis.yml
Closes #99 :recycle: to Github Actions
18:55
Pod-To-HTML: e003354cf3 | (JJ Merelo)++ | .github/workflows/test.yaml
#99 forgot this
18:57
Pod-To-HTML: 575469090e | (JJ Merelo)++ | README.md
_Really_ closes #99
19:01
Geth Pod-To-HTML: 105b72880f | (JJ Merelo)++ | 2 files
Add tags closes #97
19:03
antononcube @tbrowder Thanks for the advice to use "App::Mi6" -- it helped identify the problem. 19:36
[Coke] you can ask codesections or someone on the board to introduce you to the GC chair. 19:49
several of us used to be involved ore directly but it's... been a while
antononcube @lizmat "Anton Antonov continues to dissect speeches and articles using LLM functions. This week even twice [...]" -- given some of the screaming I get one might think I am doing vivisection. 19:50
lizmat heh... modern politics appears to be about screaming a lot :-( 19:52
antononcube I try to just present the tools, so people can consume politics writings better. (I hope.) But I think my expositions are too succint to be considered unbiased. 19:54
For example, I applied the prompt "FindPropagandaMessage" to Stephen Wolfram's latest article ("Can AI solve science?") and people from the Wolfram community is ask me / told me why did I do that. According to them the article is non-political. 19:56
Well, perhaps, but applying the prompt and reading the outcomes we can state that for sure. 19:57
Anyway, I removed that propaganda section for Wolram's site. Raku community, of course, got the full version. 19:58
@vrurg Because I got distracted working on Stephen Wolfram's article and "WWW::Gemini" I did not make the LLaMA video I promised. I hope I will do it today or tomorrow.
lizmat antononcube: having more tools will help in these days of PR overflowing 20:00
antononcube: but trust in these tools is even more important
otherwise they become a PR tool themselves :-(
antononcube This is seems relevant: 20:05
cdn.discordapp.com/attachments/633...017a9&
lizmat :-) 20:07
SmokeMachine Voldenet, aruniecrisps: If interested, I've "started playing" with the idea of migrations: github.com/FCO/Red/issues/15#issue...1989495613 21:42
librasteve arun: as an observer it looks to me like code sections immutable data structures never got off the ground 22:49
imo raku can get close to this model with the relatively recent ValueList, ValueObj modules since these pin the contents of immutable structures (so they cannot contain containers and can have WHICH semantics eg for set ops) 22:51
deep immutables are not free in that your code can lose data locality / cache performance and so while i think having deep immutables is a worthy long term goal, there is not much you cannot already do with the Value modules 22:54