00:05 david7832 left 00:24 oodani left 00:27 oodani joined 00:42 tejr left 01:08 hulk joined 01:09 kylese left 01:10 tejr joined 02:15 hulk left, kylese joined 03:06 soverysour joined 03:11 soverysour left 03:37 annamalai left 03:38 annamalai joined, annamalai left, annamalai joined
[Coke] m: say shell("echo $SHELL"); 04:09
evalable6__ (exit code 1) ===SORRY!=== Error while compiling /tmp/LMEdYd_zpr
Variable '$SHELL' is not declared. Did you mean '&shell'?
at /tmp/LMEdYd_zpr:1
------> say shell("echo <HERE>$SHELL");
[Coke] m: say shell('echo $SHELL'); 04:10
evalable6__
Proc.new(in => IO::Pipe, out => IO::Pipe, err => IO::Pipe, os-error => Str, exitcode => 0, signal => 0, pid => 15078, command => ("echo \$SHELL",))
[Coke] m: shell('echo $SHELL', :out).out.say;
evalable6__ (exit code 1) This pipe was opened for reading, not writing
in block <unit> at /tmp/fnOTh3b7rv line 1
[Coke] ... I get /bin/zsh when running that in the REPL. Should we override it to /bin/sh? 04:11
any ideas why github.com/Raku/doc/actions/runs/2...9919740054 doesn't have a rakudo that knows about exits-ok? 04:14
04:54 Aedil joined 06:46 soverysour joined 06:50 soverysour left 07:13 ShimmerFairy left, ShimmerFairy joined 07:19 soverysour joined 07:24 soverysour left 07:40 sibl joined 07:42 Aedil left, soverysour joined, soverysour left, soverysour joined 08:25 sibl left 08:33 soverysour left 08:58 soverysour joined, soverysour left, soverysour joined 09:01 Sgeo left 09:08 soverysour left 09:24 soverysour joined, soverysour left, soverysour joined 09:37 soverysour left 10:16 david7832 joined
david7832 ab5tract: Thanks! 10:17
tellable6 2026-04-03T11:09:36Z #raku <ab5tract> david7832: you might be interested in the ‘is item’ trait
david7832 While I couldn't get is-item to show an infuence on precedence, it's definitely interesting. 10:18
10:22 soverysour joined 10:32 soverysour left 10:51 annamalai left 10:58 soverysour joined, soverysour left, soverysour joined 11:36 david7832 left 11:57 annamalai joined 11:59 ds7832 joined, ds7832 left, ds7832 joined 12:01 soverysour left, soverysour joined, soverysour left, soverysour joined 12:06 soverysour left
ab5tract Yeah it is designed for tweaking dispatch, rather than precedence 12:18
ds7832 Ah, by precedence I meant to say "precedence with regard to multiple dispatch", not operator precedence (I guess that's what you understood?). So what I was trying to say was, I couldn't see any effect on the priority of multis in multi dispatch, or at least not the desired effect that e.g. f(@a is item) should take priority over f(@a) for 12:36
f($[1,2,3])
Coke: I think you have to use shell 'echo $0'.  For echo $SHELL, I get /bin/bash even when running it directly from sh, fish or zsh (outside from any raku). So I think $SHELL is just the environment variable saying what your system's default shell is, so it's the same everywhere 12:39
lizmat % echo $0 12:40
-zsh
% echo $SHELL
/bin/zsh
on my MBP
ds7832 What happens when you run that from another shell than zsh? 12:42
(presuming it's zsh from MBP and the % ) 12:43
lizmat bash-3.2$ echo $0
bash
bash-3.2$ echo $SHELL
/bin/zsh 12:44
ds7832 Thanks, I think that fits the picture: $0 gives the executable (if used inside a shell that understands $0 as such – zsh, bash and sh do, while fish does not), $SHELL your system default 12:45
(I use OpenSUSE with bash as default shell, which explains what I got) 12:46
I actually just yesterday looked into shell-independent ways to print the current shell's name. These seem to be the best: 12:48
sh -c 'ps -p $$ -o ppid=' | xargs ps -o comm= -p
sh -c 'ps -p $$ -o ppid=' | xargs -I'{}' readlink -f '/proc/{}/exe'
from askubuntu.com/a/1022440
12:50 soverysour joined 12:54 soverysour left
ds7832 ... although these two differ when run from inside raku: the first gives me sh, the second /usr/bin/bash.   (But no difference between REPL and a raku program called via raku program.raku) 12:55
13:27 ds7832 left 13:38 soverysour joined 13:57 soverysour left 13:59 soverysour joined, soverysour left, soverysour joined 14:16 annamalai left 14:17 annamalai joined 14:31 hvxgr__ left 14:46 annamalai left 15:25 annamalai joined 15:41 annamalai left, annamalai joined 16:03 Aedil joined 16:28 Pixi` joined 16:29 Pixi left 16:48 wayland joined 16:49 wayland76 left 16:58 simcop2387 left 17:02 simcop2387 joined 17:52 wayland76 joined 17:53 wayland left, soverysour left 17:54 Sgeo joined 18:15 sftp joined, sftp left, sftp joined 18:19 sftp left 18:39 soverysour joined, soverysour left, soverysour joined 18:46 merp left 18:48 merp joined 19:25 Pixi` is now known as Pixi, Pixi left 19:26 Pixi joined 19:59 Aedil left 20:38 soverysour left 20:46 soverysour joined
ugexe RAKUDO_RAKUAST=1 raku -e 'say Q:cc/\g/' 20:51
===SORRY!=== 20:52
Cannot find method 'typed_panic' on object of type Raku::QGrammar+{cc}+{stop}
github.com/ugexe/rakudo/blob/4fb9c....nqp#L6445
discovered with claude code
(should be typed-panic)
disbot2 <antononcube> 😱
21:11 soverysour left
ugexe RAKUDO_RAKUAST=1 raku -e 'if 1 { } elif 0 { }' 21:49
===SORRY!===
Cannot stringify object of type NQPArray
TIL elif would compile at all 21:51
s/would/should/
22:22 wayland76 left
ugexe m: say StrDistance.new(before => "", after => "a").Int 22:56
evalable6__ 0
ugexe github.com/ugexe/rakudo/blob/4fb9c...od#L27-L28 22:57
s/\^\@/0..@/
also discovered with claude code
[Coke] Anyone want a camelia laptop sticker (assuming this print works) 23:15
23:19 sftp joined, sftp left, sftp joined
SmokeMachine I've been playing with AI agents and SDD and decided to start playing with that on my rewrite of Sourcing... Can I get anyone's opinion on that? fco.github.io/Sourcing-2/ 23:28
ugexe m: raku -e 'my $x = "abcabc"; say m:٢nd/abc/ given $x' 23:29
evalable6__ (exit code 1) ===SORRY!=== Error while compilin…
ugexe, Full output: gist.github.com/b0b643f6767ee1d2aa...0b0467caa4
ugexe quite a few issues around using non ASCII digits, rakudo using \d somewhere, and then later rakudo using + on that value which is silently turned to 0 23:31
oh nice, that one is fixed in rakuast already 23:32
m: say ١.Int; say "ABC" ~~ /:1i abc/; say "ABC" ~~ /:١i abc/; 23:38
evalable6__ 1
｢ABC｣
Nil
ugexe m: my $p = "/usr/bin".IO; $p.is-absolute; say $p.absolute("/bar") 23:55
evalable6__ (exit code 1) Type check failed for return value; expected Str:D but got Mu (Mu)
in block <unit> at /tmp/5NBY79J8yg line 1
ugexe IO/Path.rakumod:246-247 — .absolute($CWD) assumes that if $!is-absolute is set, then $!os-path is also populated. But .is-absolute (line 80-84) sets $!is-absolute without touching $!os-path 23:56

Powered by Cro and the Raku® Programming Language.

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

Thank you!