🦋 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.
uzl[m] <leont> "On stackoverflow privileges..." <- SO royalty 😄 00:04
leont Hey I worked hard for it! ;-) 00:05
uzl[m] <leont> "Hey I worked hard for it! ;-)" <- 💪 I hear people complain about SO (apparently it's not beginner-friendly according to some) but SO and StackExchange overall is a great website 00:12
leont It's the worst site, except for all the others ;-) 00:18
uzl[m] Perfect, perspective is the key! At times it feels like the internet is shrinking haha 00:32
nemokosch are you using the "IRC outlook", without the avatars and all? 00:52
guifa www.youtube.com/watch?v=G7Sa_Q0XG1A 02:45
Video link will probaly die in a bit, as they accidentally attributed it to sawyerx and not me
nemokosch You make a good point about the use case ("will this arbitrary regex work with my code") 09:04
But I think the much bigger deal would be the opposite: write Raku, run everywhere 09:04
Especially since I think it would be quite uncontroversial that Raku regex syntax is superior 09:07
guifa I think someone had asked about a module to get Raku syntax in another language. I'm definitely open to figuring that out. 10:42
but I absolutely agree, Raku regex > PCRE 10:45
tonyo . 11:17
i actually prefer PCRE (haven't watched the video yet) 11:18
nemokosch I had to work with python regex a little bit and it's not just the syntax (even though I do think the Raku syntax is lot clearer)... 12:19
No nesting of captures is kind of a show stopper
That's a significant reason why you often don't even need to write a fully-blown grammar in Raku 12:20
tonyo yea, python leaves a lot to be desired in regex 12:42
and a lot of other areas
tonyo i only prefer PCRE because of the brevity..having to literal-ize non specials is difficult to read 12:45
nemokosch Interesting, I think that's actually way more consistent and easier to read 14:19
I would say this not-trying-to-optimize-for-some-special-case is actually a rare and very positive occurrance for regex 14:20
tonyo generally yea but regex is optimizing for a special case, otherwise index-of would probably suffice 15:48
[Coke] new feature for the zef deps app, want to show all the licenses needed for a particular (set of) dependencies. 15:57
lizmat clickbaits rakudoweekly.blog/2023/07/17/2023-...-and-asts/ 16:44
nemokosch I don't really like the overall idea of treating regex as something subject to this categorization 17:50
but even then, index-of would rather be the special case of a regex, if anything
jdv anyone know who owns the discord bridge? 17:52
nemokosch anyway, having to remember which characters must be escaped in which dialect, which mustn't, how they interact with other features coming from vim/shell/quoting/whatever, is really not fun
sometimes (in reality: often) one will need to escape the escapings and count how many times something has to be escaped or in what context it is appropriate to evaluate it 17:53
in Raku, you don't just get a Swiss army knife that tears these escape problems down (quoting the literal content) but it's actually a pattern that I think should be overall more encouraged. Don't make me guess which character has a special meaning in your intended dialect and environment - if it's a literal, mark it, and if you don't, it does have a special meaning. 17:56
roguerakudev jdv: zephyr and I are handling it 19:06
tonyo it's not fun but it's also not that difficult to remember, you're also not penalized for just escaping most things. there's way more syntax to raku regex. that said the raku regex is way more powerful and a much cooler way to handle tokenization/language extension 19:57
nemokosch considering it's value, I'd say it's way more difficult to remember than it adds value 20:07
also, if one ends up "escaping most things", that's basically an acknowledgement that Raku got this right 😛 20:09
also, I wanted to confirm my memories 20:34
in vi, + and * behave exactly the opposite way
+ needs to be escaped in the regex to have the regex meaning and not escaped to have the literal meaning 20:35
* needs to be escaped to be literal and not escaped to be quantifier
tonyo most people don't need to escape most things..it's designed for pattern matching and it does it succinctly and nicely. needing to escape plain text matching seems like a regression 21:01
nemokosch another way to look at it would be the elimination of "bare words" which never appeared in a context that is worth remembering 21:08
I think it's actually cleaner to write 'foobar', or 'dead horse', or '[general]'; all of them on their own as well but especially the consistency this pattern achieves is something to be valued... especially when it's about strings or non-alphanumeric characters, I just can't understand what's so precious about saving two keystrokes 21:12