| disbot | <_.mu._> Scrolled back for the coding assistent issue but that wasn't really enlightening. Is there some ressource somewhere stating project policy or blog posts people have made to explain their stance? | 00:22 | |
| elcaro | Maybe it's already been mentioned, but I just realised there's no pre-compiled binaries for 2026.08 on rakudo.org/downloads | 00:32 | |
|
01:23
Voldenet_ joined,
Voldenet left
01:24
Voldenet_ is now known as Voldenet
|
|||
| Voldenet | > LLM is just an excellent tool to attract new users | 01:29 | |
| my upper management also believes this, but was this ever… confirmed anyhow? | |||
| It feels more like a meme | |||
| LLM is only useful if in you review its code | 01:34 | ||
| lucs: you're not matching full string, just 2ab | |||
| m: say '4a2ab' ~~ /:r (\d) 'a' 'b'/; say ~$0; | |||
| camelia | ｢2ab｣ 0 => ｢2｣ 2 |
||
| Voldenet | m: say '4a2ab' ~~ /:r ^ (\d) 'a' 'b' $/; # this doesn't match | 01:35 | |
| camelia | Nil | ||
| Voldenet | m: say '2ab' ~~ /:r ^ (\d) 'a' 'b' $/; # this does match | ||
| camelia | ｢2ab｣ 0 => ｢2｣ |
||
| lucs | Oh. Hmm... | 01:39 | |
| Voldenet | and no backtracking means that after matching part of the regex, engine will not backtrack | 01:48 | |
| m: say <aaab> ~~ /a+ b/ # engine will greedily match a, then b, this will work | 01:49 | ||
| camelia | ｢aaab｣ | ||
|
01:49
kylese left
|
|||
| Voldenet | m: say <aaab> ~~ /a+ ab/ # engine will greedily match a, then after exhausting all `a` letters, it'll never match a | 01:49 | |
| camelia | ｢aaab｣ | ||
| Voldenet | erm | ||
| m: say <aaab> ~~ /:r a+ ab/ # engine will greedily match a, then after exhausting all `a` letters, it'll never match a | |||
| camelia | Nil | ||
| Voldenet | m: say <aaab> ~~ /:r a+ b/ # this works with ratchet as well | 01:50 | |
| camelia | ｢aaab｣ | ||
|
01:50
kylese joined
|
|||
| [Coke] | el_che: did you do 2026.08 binary? | 01:58 | |
| patrickb: did you do 2026.08 binary? | |||
|
02:01
greppable6 left,
releasable6 left,
benchable6 left,
unicodable6 left,
nativecallable6 left,
committable6 left,
coverable6 left,
sourceable6 left,
quotable6 left,
linkable6 left,
shareable6__ left,
bloatable6 left,
bisectable6 left,
evalable6 left,
huggable6 left,
notable6 left,
tellable6 left
|
|||
| Voldenet | btw, in your example `4a` simply fails and then `2ab` gets matched, so there's no backtracking - you need example where one token is ambiguously matched by previous term and next term can't match it | 02:02 | |
| m: say <4a2ab> ~~ /:r \w+ ab/; # no matches | 02:03 | ||
| camelia | Nil | ||
| Voldenet | m: say <4a2ab> ~~ /\w+ ab/; # at first no matches, but after backtracking there's an obvious match | ||
| camelia | ｢4a2ab｣ | ||
|
02:04
releasable6 joined,
shareable6 joined,
evalable6 joined,
greppable6 joined,
huggable6 joined,
bisectable6 joined,
sourceable6 joined,
linkable6 joined
02:05
quotable6 joined,
coverable6 joined,
committable6 joined,
bloatable6 joined,
benchable6 joined
02:06
notable6 joined,
nativecallable6 joined,
tellable6 joined,
unicodable6 joined
|
|||
| lucs | Voldenet: I think I was confusing the backtracking as something that happens in the regex (in my example, can't go back to try the \d once it has been matched) rather than in the string (irrelevant in my example, just restart trying the match at 2, which succeeds). | 02:09 | |
| Thanks for the clear examples. | |||
|
02:15
kylese left,
kylese joined
02:18
bisectable6 left,
tellable6 left,
evalable6 left,
greppable6 left,
benchable6 left,
quotable6 left
02:21
bloatable6 left,
coverable6 left,
nativecallable6 left,
notable6 left,
committable6 left,
sourceable6 left,
huggable6 left,
shareable6 left,
releasable6 left
02:22
unicodable6 left,
linkable6 left
02:23
bloatable6 joined,
evalable6 joined,
greppable6 joined,
benchable6 joined,
quotable6 joined,
shareable6 joined
02:24
linkable6 joined,
nativecallable6 joined
02:25
coverable6 joined,
sourceable6 joined,
unicodable6 joined,
releasable6 joined,
bisectable6 joined,
committable6 joined,
huggable6 joined
02:26
tellable6 joined,
notable6 joined
|
|||
| disbot | <demanddystopia> i recall there was a site that had like a visualization of the grammar, incl like when it fails trying to find what, don't think that applied to single regexes though that'd be kinda useful too | 02:30 | |
| Voldenet | github.com/FCO/Grammar-Editor | 02:53 | |
| [Coke] | feel like the instructions should be for an installed version, not a git clone version. ^^ | 02:55 | |
| Voldenet | I remember seeing website version, can't find it though | 03:09 | |
| elcaro | fco.github.io/GrammarEditor/ | 03:27 | |
|
04:18
Pixi joined
04:20
Pixi` left
04:41
Pixi left
05:35
kjp joined
05:44
Sgeo left
05:48
abraxxa joined
05:50
gfldex left
05:51
gfldex joined
|
|||
| patrickb | [Coke]: I did not. 2026.08 fails to build because of a bug in zmij. Now that you remind me, I still need to bump the zmij version... | 06:57 | |
|
07:56
dakkar joined
09:34
oodani left,
oodani joined
|
|||