🦋 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 can be inspected at colabti.org/irclogger/irclogger_log/raku-dev | For MoarVM see #moarvm
Set by lizmat on 26 April 2021.
Geth rakudo: 5331a1d801 | (Stoned Elipot)++ | src/core.c/Exception.pm6
Do not include falsy attribute in exception's message
 08:17
rakudo: c7a078dec7 | niner++ (committed using GitHub Web editor) | src/core.c/Exception.pm6
Merge pull request #4365 from stoned/xnc-no-reason

Do not include falsy attribute in exception's message
lizmat am I the only one who would like to have .words to be more like .comb(/\w+/) instead of .split(\s+) ? 09:05
gfldex lizmat: does \w+ actually work for all natural languages? 10:12
lizmat it's using the Unicode property afaik
so I assume the answer is yes? 10:13
gfldex m: dd 'Herr Müller-Lüdenscheit'.comb(/\w+/);
camelia ("Herr", "Müller", "Lüdenscheit").Seq
gfldex nope
lizmat ? 10:14
m: dd 'Herr Müller-Lüdenscheit'.words;
camelia ("Herr", "Müller-Lüdenscheit").Seq
lizmat gfldex: what am I missing?
gfldex In german double names are considered one word. Conceptually, .words sounds like it gives me the acutal words. 10:15
lizmat m: dd 'Herr Müller_Lüdenscheit'.comb(/\w+/);
camelia ("Herr", "Müller_Lüdenscheit").Seq
lizmat German is using the wrong hyphen :-) 10:16
gfldex m: dd 'Herr Müller‐Lüdenscheit'.comb(/\w+/);
camelia ("Herr", "Müller", "Lüdenscheit").Seq
gfldex that might actually be a bug 10:17
m: dd 'Herr Müller‑Lüdenscheit'.comb(/\w+/); 10:18
camelia ("Herr", "Müller", "Lüdenscheit").Seq
gfldex ‑ <-- that is the "don't line break here‼" variant 10:19
lizmat as I said, \w is defined in some unicode property 10:20
if unicode doesn't consider - or ‑ to have that property, well, end of story then I'd say ?
gfldex The meaning of a hyphen (and other stuff) depends on the natural language. That sounds like ecosystem territory to me. So changing the meaning of .words in CORE seams to carry little utility. 10:23
lizmat well, there's always .comb( / << \w+ >> / ) but that's soooo much slower :( 10:24
gfldex Does it have to be?
lizmat good question: after trying to make Match faster, I figured the only way to do that, would be to refactor a lot of regex engine internals 10:25
I'm not ready to do that just yet..
stoned75 hi. 10:32
Should I be surprised by the following ?
> 1.^mro(:roles)[1].^name 10:33
Real
> 1.^mro(:roles)[1].HOW
Perl6::Metamodel::ConcreteRoleHOW.new
> 1.^mro(:roles)[1].WHAT
(low-level object `Real`)
> 1.^mro(:roles)[1].gist
Cannot find method 'gist' on 'Real': no method cache and no .^find_method in block <unit> at <unknown file> line 1
>
and idem for 1.^mro(:roles)[2] 10:34
lizmat m: Real ~~ Any 10:35
camelia ( no output )
lizmat m: dd Real ~~ Any
camelia Bool::True
lizmat m: dd 1.^mro(:roles)[1].WHAT
camelia (Real without .raku or .perl method)
lizmat stoned75: looks like that's the nqp role "Real" leaking out 10:36
stoned75 so I suppose it is not expected ? 10:37
lizmat no, expected 10:56
more a case of DIHWIDT
? 10:57
thundergnat Don't really care for the idea of switching .words to mean .com(/\w+/). 14:31
m: say .comb(/\w+/), .split(/\s+/) given <don't won't can't Epsteian-Barr>
camelia (don t won t can t Epsteian Barr)(don't won't can't Epsteian-Barr)
Geth nqp: usev6++ created pull request #722:
[JVM] Keep correct REPR.name for native arrays		 20:35

Powered by Cro and the Raku® Programming Language.

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

Thank you!