00:10 stanrifkin left
[Coke] tonyo: github.com/tony-o/raku-fez/issues/135 01:18
new error today about not being able to find my orgs. AND it still can't find my module 01:19
(and all uploaded versions including 0.0.1 are visible to zef, just not fez. :| ) 01:21
I don't care so much that it's there now, but delete seems broken. 01:22
02:10 annamalai joined 02:14 kylese left 02:15 kylese joined 02:32 sibl joined 02:39 sibl left 03:15 kylese left, kylese joined 03:20 sibl joined 03:27 committable6 left, quotable6 left, notable6 left, evalable6 left, sourceable6 left, bisectable6 left, coverable6 left, bloatable6 left, greppable6 left, tellable6 left, shareable6 left, benchable6 left, unicodable6 left 03:32 committable6 joined, greppable6 joined 03:33 bisectable6 joined, evalable6 joined, linkable6 joined, tellable6 joined, releasable6 joined, nativecallable6 joined 03:34 quotable6 joined, sourceable6 joined, benchable6 joined 03:35 notable6 joined, bloatable6 joined, coverable6 joined, unicodable6 joined, shareable6 joined 04:13 itaipu left 04:31 itaipu joined 04:34 lichtkind_ joined 04:37 lichtkind__ left 05:23 Aedil joined 05:25 Aedil left 05:28 Aedil joined
tonyo alright, thanks i'll try to track it down. currently travelling for work and in and out 06:28
06:34 sibl left 06:58 Sgeo left 07:50 abraxxa joined 08:15 Aedil left 08:24 Aedil joined 08:50 mahlay left 08:55 dakkar joined 09:05 mahlay joined 09:14 tejr left 09:34 wayland76 joined 09:38 tejr joined
wayland76 Hi all! My understanding is that the standard way of creating a Slang is to make a Role, and then effectively mix it into the main grammar with Slangify. I've tried making a grammar that inherits from Raku::Grammar, but it fails ( gist.github.com/wayland/242006cc53...066cc7b5cc ). Is there a way of making a grammar that inherits rather than mixes in? 09:54
I'm asking because my current grammar says: This element has not been resolved. Type: RakuAST::Term::Self
...and I haven't quite been able to figure out why yet.
...and I was hoping that inheriting from Raku::Grammar might solve that problem. 09:55
(the problem is likely coming from the part that's basically a duplicate of "method-def" from the grammar, but with a few tweaks). 09:56
disbot11 <antononcube> @melezhik. Sorry, I am too preoccupied with problematic "Jupyter::Chatbook" executables. 10:28
<antononcube> The new "WWW::Ollama" package allows local running on for LLM models. Initially, I made that execution to be "seamless", i.e. the executable ollama is found/located and started automatically and the LLM model specifeid in the request is downloaded. 10:30
<antononcube> This turns out is not a good idea from a certain sysadmin point of view. So, now by default the automatic running is disabled. 10:32
<antononcube> Also, I had the idea to have the Ollama executables for different platforms to be part of the package (resources.) But that would make it too big, ≈95MB. 10:34
10:58 librasteve_ joined
librasteve_ .seen SmokeMachine 11:17
tellable6 librasteve_, I saw SmokeMachine 2026-01-03T21:33:57Z in #raku: <SmokeMachine> That’s why I haven’t commented anything on that repo yet… I’m waiting to know if I want to be part of that or not
lizmat wayland76: would "use v6.e. PREVIEW" or "use experimental :rakuast" be of help?
SmokeMachine librasteve_: hi!
librasteve_ hi!
lizmat wayland76: ah, no, it seems you already have that, sorry for the noise :-)
librasteve_ SmokeMachine: I just made a PR on Cromponent (putting some more deps in the META6.json) to resolve Issue #15 … hoping you can review soon since I am now working on testing with Air::Component 11:19
SmokeMachine I saw it, thanks… I just didn’t have time to test it yet. But I’ll need to review my code because based on your MR I’m using 2 different UUIDs… 11:20
And when I looked at it, it seems the tests were failing 11:22
disbot11 <librasteve> yeah - the tests are good now (I realsed that there were still some deps on my machone - so I did rakubrew nuke to make sure I caught them all) 11:27
<librasteve> would be best to all stick to UUID::V4 if that works for you (Air & raku.org are currently pinned to UUID::V4:ver<1.0.0>:auth<zef:masukomi>:api<1>) 11:29
<librasteve> SmokeMachine: I have made a couple more Issues - it would be great to get Cromponent Auth working with Air ... getting Air to CRUD level is one of my goals for the new year 12:17
tbrowder hi. pls tell me again how to test that a script works. in a test file i'm using: lives-ok { run "pro", "arg"; }, "runs ok"; done-testing; 12:27
i can run it fine, but when i do "mi6 test" it complains about test plans and parse errors. i think there is an undocumented trick to do that 12:30
ah, jj's test helper? 12:31
Test::Script::Output should work, but there is another trick ugexe or other knows... 12:36
jj's needs work, bummer 12:37
ah, my bad: all i had to do was add the missing", :out;" and voila! 12:40
13:34 melezhik4 joined
melezhik4 . 13:34
started a discussion on reddit www.reddit.com/r/devops/comments/1...is_is_yet/ - if I get enough positive feedbacks, maybe I'll start implement this ( under the hood it's going to be a Raku ) 13:35
disbot11 <antononcube> Anyone here named “Gary”? 13:37
<antononcube> cdn.discordapp.com/attachments/633...3f906&
13:40 melezhik4 left 14:09 wayland joined, wayland76 left
tbrowder hi, ref Spreadsheet::XSLX, i'm looking for advice. i have added a simple Raku script to a new /examples directory which shows how to convert a CSV file to an XLSX file (Micosoft Excel 2007+). it's in my tb-examples branch of my fork of the raku-community-modules repo on github. i would appreciate any comments on it before i submit a PR. 14:17
typo again: XLSX is the correct spelling (very hard for anyone prone to dyslexia) 14:19
15:56 abraxxa left
librasteve_ notable6: weekly 16:11
notable6 librasteve_, 6 notes: gist.github.com/771c77b76d741b9f85...bb1ef316c5
disbot11 <antononcube> @Coke You might be interested in this post: community.wolfram.com/groups/-/m/t/3604657 16:16
16:48 xinming_ left 16:49 xinming joined
librasteve_ rakudoweekly.blog/2026/01/12/2026-...solutions/ 17:18
notable6: weekly reset
notable6 librasteve_, Moved existing notes to “weekly_2026-01-12T17:18:59Z”
17:34 dakkar left
[Coke] antononcube - thanks! 17:40
glad I wasn't the only one who had the same idea. :) 17:41
lucs I've got a function that builds a regex string. 18:03
For example, build-regex(42) returns "^^ 42".
How do I "plug" its result in a regex match?
Something like this failing example: 18:04
$str ~~ / <build-regex(42)> /
I'd also be happy if it can be done with match() instead of //. 18:06
[Coke] docs.raku.org/language/regexes#Reg...erpolation 18:07
<{code}>Runs Raku code inside the regex, and interpolates the stringified return value as a regex. 18:08
lucs Aha, thanks, will read...
[Coke] if you already had a variable, you could use <$variable> - need the braces since you're calling some code. 18:09
lucs Dang, I have to reread all that stuff once in a while.
Yeah, I knew about the <$variable> thing, but introducing a variable felt just a bit LTA in my case. 18:11
Worked fine (of course :-) ) 18:12
Thanks.
18:14 swaggboi left
disbot11 <antononcube> @Coke I am aking the author of that WL post/notebook to make this kind of movie(s). 18:16
<antononcube> cdn.discordapp.com/attachments/633...eb51b&
<antononcube> The circle chords in the movie correspond to pairs of integers that have Fibonacci representations with distance 1. The chords in red correspond to pairs differences that are prime numbers. 18:17
<antononcube> Of course, the next logical task is to make that movie in Raku. 18:18
18:23 swaggboi joined
[Coke] If you want to use my module, please let me know if there's anything I can do to make it useful and not just a toy! 18:24
disbot11 <antononcube> @Coke Ok. Please implement the Dual Zeckendorf representation: resources.wolframcloud.com/Functio...sentation/ 18:28
[Coke] I... have to figure out what the even is, adding to my homework. :) 19:16
lucs Problem having a subroutine attribute in a class: 20:50
gist.github.com/lucs/251df56a1c890...bd17d83208
ugexe call with $baz.func.(42) 20:58
lucs !
Thanks!
Um, anywhere I should read up in the docs about this? 20:59
Seriously, that was the shortest modification I've ever had to make to code I couldn't make work. 21:03
Period.
tbrowder why not use a method? 21:05
ugexe i dunno where you'd read about. maybe `($baz.func)(42)` makes it more clear
lucs tbrowder: Um, the way I was coding led up to having it like that, maybe it could have been a method. 21:08
Looking...
tbrowder: Hmm... Not sure how I could make it into a method. 21:10
Each Baz instance could have a different &func.
ugexe public attributes create public methods 21:11
tbrowder actually the docs i think addresses something like that but it’s hard to find in current version
over my head but interesting 21:13
lucs ugexe: Aha, yeah, clearer when seen as `($baz.func)(42)`.
ugexe m: class Foo { has $.bar = 42; }; my $foo = Foo.new; my $method = $foo.^find_method("bar"); say $method($foo) 21:19
camelia 42
tbrowder yes that is in docs hard to find 21:20
in one of the sections showing the ^ operator on an object 21:21
ugexe if you want to avoid the indirect syntax you can use a private attribute and a separate method to avoid the ambiguity of $baz.func(42) as well 21:25
m: class Foo { has &!cb; submethod BUILD(:&!cb) {}; method cb(|c) { &!cb.(|c) } }; my $foo = Foo.new: cb => sub ($n) { say "n: $n" }; $foo.cb(42)
camelia n: 42
ugexe it uses a private attribute to store the callback so there is no public method created for it. instead we create our own public method and dispatch to the attribute using the e.g. (self.func)(42) syntax. but consumers of the class can use e.g. $foo.func(42) 21:26
lucs I'm noting down all this, but it's a bit too advanced for my understanding and for my current need. 21:30
[Coke] .^ is mentioned near the top of docs.raku.org/language/mop 21:41
ugexe m: class Foo { has $.abc = 42 }; say Foo.new.abc(); say Foo.new.abc(123); 21:42
camelia 42
Too many positionals passed; expected 1 argument but got 2
in method abc at <tmp> line 1
in block <unit> at <tmp> line 1
ugexe as you can see there is no `method abc` yet we can call .abc() (including the parens) 21:43
the .abc(123) has the same error you were running into
because the public method that is automatically create for public attributes does not take any arguments
you can either disambiguate the call via ($foo.func)(42) / $foo.func.(42), or you can store your func in a private attribute instead (no public methods are created for public attributes) and then create your own public method 21:45
the later means callers of your code don't have to do that disambiguation
lucs Interesting. I'm starting to see how it all hangs together. 21:50
21:50 wayland left 23:09 Sgeo joined 23:39 Aedil left
[Coke] m: my @asdf=<1 2 3>; for @asdf.kv -> ($key, $value) { say $value } 23:42
camelia Cannot unpack or Capture `0`.
To create a Capture, add parentheses: \(...)
If unpacking in a signature, perhaps you needlessly used parentheses? -> ($x) {} vs. -> $x {}
or missed `:` in signature unpacking? -> &c:(Int) {}
in block <unit> at …
[Coke] Would be nice if the error there suggested removing the breaking parens. 23:43
... DAMMIT
he said, reading the FULL error.
[Coke] goes to crawl back in his hole.
ugh. there is trailing whitespace on the default LICENSE file. :| 23:47
antononcube - done 23:51
github.com/coke/math-zeckendorf/issues/1 23:52
Will wait a bit before cutting a release, since the last time I did it immediately, I missed something obvious. :)