🦋 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.
Geth rakudo/rakuast: a8de4bb672 | (Stefan Seifert)++ | src/Raku/ast/package.rakumod
Make $?ROLE, $?MODULE and $?CLASS available
 07:15
rakudo/rakuast: 24a66ea055 | (Stefan Seifert)++ | 2 files
Properly support access to role attributes from role methods

When a role's method tries to access the role's attribute, we need to generate a runtime lookup for $?CLASS, as attributes will be found on the concrete class instead of the role's type, i.e. nqp::getattr(self, $?CLASS, '$!attr') instead of nqp::getattr(self, MyRole, '$!attr')
rakudo/rakuast: 8ef48b0390 | (Stefan Seifert)++ | 3 files
Fix $?CLASS et al leaking outside the package's scope

In addition to adding generated lexicals for $?CLASS, ::?ROLE, etc. to a package's block, we need to declare those lexicals at BEGIN time, so they are available to traits, signatures, the parser etc. However doing so in PERFORM-BEGIN would get them added to the surrounding scope, thus making them still visible after the class body was finished. ... (8 more lines)
nine A good day begins with pushed fixes 07:23
That last one was hard earned though...took me almost a day :/ 07:24
Geth rakudo: 5a8aa66bc2 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | src/Perl6/Metamodel/CoercionHOW.nqp
Make coercions up to 1.6x as fast

  - calling method named after type, e.g. Int() -> .Int: 1.6x as fast
  - calling .COERCE: 4% faster
  - calling .new: 3% faster
This has been achieved by breaking up the large "coerce" method into ... (15 more lines)		 10:21
rakudo: d7d8e9d0f4 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | src/core.c/IO/Socket/INET.pm6
Make unsuccessful connect fail rather than die

Spotted in stackoverflow.com/questions/726398...socketinet
There is no documentation or tests covering the behaviour of a failed connect. It is actually the nqp::connect that throws (which turns out also to be not documented BTW).
Having it return a Failure feels more natural indeed.		 10:22
Xliff use NativeCall; my \mallocParams = :(size_t --> Pointer); mallocParams.say; sub a (&c:(|mallocParams), Int() $s) { my size_t $ss = $s; &c($ss).say; }; sub malloc (size_t --> Pointer) is native {*}; malloc(20).say; a(&malloc, 20);' 11:26
m: use NativeCall; my \mallocParams = :(size_t --> Pointer); mallocParams.say; sub a (&c:(|mallocParams), Int() $s) { my size_t $ss = $s; &c($ss).say; }; sub malloc (size_t --> Pointer) is native {*}; malloc(20).say; a(&malloc, 20);'
camelia ===SORRY!=== Error while compiling <tmp>
Unable to parse expression in single quotes; couldn't find final "'" (corresponding starter was at line 1)
at <tmp>:1
------> ve {*}; malloc(20).say; a(&malloc, 20);'⏏<EOL>
expecting …
Xliff m: use NativeCall; my \mallocParams = :(size_t --> Pointer); mallocParams.say; sub a (&c:(|mallocParams), Int() $s) { my size_t $ss = $s; &c($ss).say; }; sub malloc (size_t --> Pointer) is native {*}; malloc(20).say; a(&malloc, 20);
camelia (NativeCall::Types::size_t --> NativeCall::Types::Pointer)
Cannot resolve caller protect(Lock:U: Block:D); none of these signatures matches:
(Lock:D: &code, *%_)
in method setup at /home/camelia/rakudo-m-inst-2/share/perl6/core/sources/871…
Xliff Hmmm... I get "Signature constraint check failed in binding to parameter '&c'; expected :(|mallocParams) but got :(NativeCall::Types:..." 11:27
That almost looks like a way to attach function pointers to a symbol, which I desperately need. What kind of changes would it take to make it work? 11:28
^^ lizmat vrurg jnthn MasterDuke
lizmat Xliff: that's really outside of my mindset atm, MasterDuke is moving house and only online occasionally, and jnthn is not on this channel anymore 11:29
so maybe ask it on #moarvm
Xliff lizmat: Thanks. 11:54
Geth rakudo/lizmat-optimize-native-substr: 714eb6e832 | (Elizabeth Mattijsen)++ | src/core.c/Str.pm6
Add native candidates for sub substr

This makes:
   my str $a = "foo bar baz";
   my int $f = 4;
   my int $c = 2;
   say substr $a, $f; # bar baz
   say substr $a, $f, $c; # ba
about 2.8x as fast.		 21:13
rakudo: lizmat++ created pull request #4965:
Add native candidates for sub substr, making it 2.8x as fast

Powered by Cro and the Raku® Programming Language.

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

Thank you!