🦋 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.
patrickb I've Bern thinking about having SSL in core a lot some time ago. My conclusion was that there is not much point as long as installing it via zef is simple and doesn't require a compiler on Windows. Adding some SSL module to our precompiled packages is worth a thought, but I'd really like to keep those packages slim. People should be encouraged to use zef. 12:01
*been
ugexe it would help zef though in that it could use pure raku code for fetching modules 12:25
ab5tract_ interesting point 12:53
ugexe the more practical application is for users to adjust socket examples 12:56
ugexe like an example might be careful to show a site that can serve http 12:57
but if a new users just adjusts the url to point at a site that only does https, then it obviously isn't going to work and they've have to figure out why 12:58
[Coke] I'm +1 for that being out of the box or having an amazing error message. 12:59
(like we do with REPL where we recommend specific modules that you probably want0
nemokosch what is the current situation with SSL? All I can think of is the old, barely maintained modules from sergot 13:00
nine m: my $a; sub foo { say $a }; BEGIN $a = 1; $a = 2; foo; 14:49
camelia 2
nine That's as expected. However:
m: my $a; sub foo { say $a }; BEGIN $a = 1; BEGIN foo; $a = 2; foo;
camelia 1
1
nine Funny how easy it is to find problems with the old compiler when pondering the replacement.
Continuing on irclogs.raku.org/raku-dev/2023-10-22.html#11:42 as demonstrated above, even if we post-process the QAST for BEGIN time execution, this QAST must only be used for actual BEGIN time execution. Problem is that we set the result of the compilation of this QAST as $!do for the code object, so the same QAST will be used even for non-BEGIN time execution. 14:51
Geth rakudo/main: 3fea473880 | (Elizabeth Mattijsen)++ | 4 files
RakuAST: generalize slang handling

  - Cool.AST now takes a positional for slang, which can either be
   a string (e.g. "NL") or a specific slang role to mixin. Automatically
   loads the necessary modules.
  - RakuAST::DEPARSE now loads the deparsing module of a slang if the
   slang is specified as a string
... (8 more lines)
17:26
AlexDaniel buildable6: info 19:10
buildable6 AlexDaniel, (⏳ Building…) 34341 rakudo-moar builds in 12315 archives (86.1 GB); 0 moarvm builds in 0 archives (0 GB)
AlexDaniel okay, this is wrong…
buildable6 AlexDaniel, Done building! 19:12
Geth whateverable/main: dd2990dee3 | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Whateverable/Building.pm6
Consider long-term storage when building

Otherwise Buildable just keeps building commits that are already built and archived.
19:12
AlexDaniel this should make it right
AlexDaniel buildable6: info 19:20
buildable6 AlexDaniel, (⏳ Building…) 32513 rakudo-moar builds in 10487 archives (72.3 GB); 0 moarvm builds in 0 archives (0 GB)
AlexDaniel 6c: say 42 19:28
committable6 AlexDaniel, ¦6c (73 commits): «42␤» 19:29
AlexDaniel all: say 42
committable6 AlexDaniel, gist.github.com/7f64f40f1f48f7ba2d...289729117c
nemokosch 6c?
six cent 19:30
AlexDaniel well yes, it's all the commits that support `use v6.c`
um, all releases I mean
nemokosch so... all releases? 19:45
AlexDaniel nemokosch: no, `all` means all rakudo releases, starting from 2014 :) 20:21
buildable6: info
buildable6 AlexDaniel, (⏳ Building…) 32522 rakudo-moar builds in 10496 archives (72.4 GB); 0 moarvm builds in 0 archives (0 GB)
nemokosch I would really like to think there is no reason to run on something from 2014... 20:25
AlexDaniel there's also no reason not to be able to do it :) 20:38
it's just a bit of tech excellence in whateverable itself. Storing all commits up to 2014 in such a limited amount of space. 20:39
AlexDaniel c: lizmat-vent-0 say 42 21:39
committable6 AlexDaniel, ¦lizmat-vent-0: «42␤»
AlexDaniel just a quick tip that this is still available and has always been :) 21:40
so you can test and show stuff on branches that are not even merged into main
bisecting works too, just like everything
nemokosch lizmat: stmt-prefix-do mmache 23:11
was this intentional?
AlexDaniel: yeah I'd say it's kind of a weird flex 23:12
I didn't know you can fall for such vanities 😛