🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | Logs available at irclogs.raku.org/raku-dev/live.html | For MoarVM see #moarvm
Set by lizmat on 8 June 2022.
00:21 guifa left 00:26 guifa joined
ab5tract last night I ran into some issues that appear to be related to having a comma in my path 09:06
raku was unable to find modules used by user (me) code 09:07
but it could find it through -M
Note that I don't think intellij itself was handling the situation very well either, nor the shell for that matter ;) 09:18
lizmat hehe.. delimiters... yeah 09:36
guifa Didn't we have some system that would let us check a change to Rakudo across the entire ecosystem? 16:59
ab5tract guifa: Blin is the one that is used for releases, melezhik has been working on a distributed solution called brownie 17:11
guifa ab5tract++ 17:12
ab5tract neither of them are quite "plug and play", unfortunately 17:13
guifa I was thinking about the reddit post on <alpha> and from my looking at Rakudo codebase, it shouldn't be a huge change to make, the question is what the damage would be on the ecosystem 17:14
ab5tract ah, nice! 17:16
it will get tested against the ecosystem before an actual release, so getting your MR merged is essentially signing up for an eventual ecosystem check 17:18
guifa Ah okay, I was thinking such a radical change would want to be ecosystem checked before being merged into core 17:34
'radical' (four lines of actual code I think would be touched in Rakudo, maybe one or two lines in nqp and moarvm) 17:37
whoa. it seems actually that our JS version of Rakudo might already have the change I want? lol 17:45
const ALPHANUMERIC = xregexp('\\pL|\\p{Nd}', 'yA');
ab5tract guifa: if it has un-manageable side effects in the ecosystem, it's straight forward to revert 17:49
guifa so I'm trying to track down where we actually introduce the _ into the definition
ab5tract which is to say, if it passes roast, it's mergeable. if it passes blin, it's releasable
guifa at least at the MoarVM level, so far I'm finding that we define MVM_CCLASS_ALPHABETIC as :L+:Nd 17:50
ugexe can you just look at git blame and follow the commits backwards? 17:51
you can probably even try asking co-pilot in the github interface itself on that line when _ was introduced 17:52
guifa ugexe: it's more that <alpha> gets handled in Raku-level regex to include _, but that's been in Raku since conception. But at the MoarVM level, the unicode handling bit keeps them separated -- so trying to see where they commingled. My guess is it's at the NQP level somewhere since I don't think MoarVM actually touches regex itself 17:56
found it! 18:00
github.com/Raku/nqp/blob/600929358....nqp#L1229
so actually, that's the only real substantive change. Well, alnum would probably be changed to CCLASS_ALPHANUMERIC if it exists, or ALPHABET or NUMERIC 18:02
18:20 Guest54 joined 18:28 Guest54 left
lizmat guifa: github.com/Raku/nqp/blob/main/docs...down#const 18:42
guifa ha yup it does :-) 18:43
easy change then
lizmat so isn't it a question of using CCLASS_WORD vs CCLASS_ALPHANUMERIC ?
guifa Basically yes. Change alpha to just CCLASS_ALPHABETIC (no more `|| * eq '_'`) and alnum to CCLASS_ALPHANUMERIC. Then in Rakudo, adjust the handful of references to alpha and alnum accordingly 18:46
disbot7 <librasteve> good to see that there is a way to check the fallout (if any) from this possible change - right now there are two opinions in the issue - github.com/Raku/problem-solving/issues/509 - so I think we need some more views on the pros and cons to develop a concensus 20:06
Geth rakudo/feedsy: dadfbf9fba | ab5tract++ | 2 files
Trying and failing
 20:08

Powered by Cro and the Raku® Programming Language.

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

Thank you!