🦋 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: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
Geth roast: vrurg++ created pull request #807:
Add tests for Lock::Soft
02:25
Xliff lizmat: The gist of it? Take a sub or method callable and do "my $tree = &callable.ast" -- Then perform analysis, introspection and modifications as necessary. If modifications are done, create a new callable from the changed AST. 07:07
japhb: It wouald be great for Red. It would be huge for my projects as well. However, I would settle for workable macros.
Also a way to make this possible in nativecall: "int (*function_ptr) (char *param1, Struct *param2)" 07:08
Rather: "typedef int (*function_ptr) (char *param1, Struct *param2) MyParamTypeForFunctionPointer" 07:09
And then use MyParamTypeForFunctionPointer in signatures.
lizmat Files=1353, Tests=117197, 293 wallclock secs (36.33 usr 9.81 sys + 3995.49 cusr 332.50 csys = 4374.13 CPU) 08:28
Kaiepi o/ 14:32
touching generics led to me falling down the role rabbit hole again, but in reverse this time around 14:33
which makes things a little easier
i know that, generally, lists in the MOP need to be HLLized in order to be iterated. i'm wondering if there are certain lists that would be better off staying a deconted VMArray, e.g. mro and role_typecheck_list 14:34
the RTL's handling would need changes in order to be like that though 14:35
vrurg .tell ugexe I somehow overlooked that github.com/rakudo/rakudo/pull/4917 is worth your review too. 19:20
tellable6 vrurg, I'll pass your message to ugexe
[Coke] . 19:22
(spent like 30m recovering from shitty home wifi router, yikes) 19:23