🦋 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.
raydiak m: sub foo($a) { $a.VAR.^lookup("WHICH").raku }; say foo(1); 02:37
camelia proto method WHICH (Mu: |) {*}
raydiak shouldn't that be showing the Scalar method instead of the Mu method? 02:38
oh, I see it's a multi. guess that's just how the dispatch works. don't mind me :) 02:46
Kaiepi data time 10:51
greppable6, role
greppable6 Kaiepi, 3295 lines, 361 modules: gist.github.com/529b72e273d560fa21...db080caee6
Kaiepi wondering if, despite the additional complexity in its typecheck, it's faster to typecheck a role's doees first by virtue of them being more common 10:52
Kaiepi even if it is, it results in a slower build, so i'm not sure it favours the types in the setting, let alone the ecosystem 11:07
Kaiepi so it follows making the parametric role and its group consider parents first should be helpful, which reduces build times and looks to be efficient enough for Buf[uint8] ~~ Blob[uint8] not to take much of a hit 11:27
Geth nqp/new-disp: 6c28ff34b2 | Altai-man++ (committed using GitHub Web editor) | tools/templates/MOAR_REVISION
Bump new-disp MoarVM
 12:31
rakudo/new-disp: 9c83541e2c | Altai-man++ (committed using GitHub Web editor) | tools/templates/NQP_REVISION
Bump new-disp nqp
sena_kun was there a new-disp rebase recently? I see a lot of commits are now split to be authored/committed and I also don't see my bump commits in the history of new-disp at all? 12:33
nine sena_kun: yes, I rebased the branch on Sunday 12:40
I may have lost those bumps when solving merge conflicts. Sorry about that 12:41
sena_kun nine, no problem, just confirmed, thanks. 12:42
[Coke] ... are those bumps correct for the new rebase? 14:33
I just updated moarvm, built, installed, moved to nqp, updated, built... moar too old.
(found 395, needed 397) 14:34
sena_kun [Coke], that's weird. whateverable built an image just fine and I am finishing a blin run with it?
[Coke], can you try just rakudo with `--gen-moar --gen-nqp`?
[Coke] I don't want the defaults, I want latest. 14:36
... oh. the version number is pulled during config, isn't it. Ugh. 14:38
Ok, will start from scratch.
sena_kun [Coke], in case of MoarVM my new-disp branch has the latest commit 5f563ba1a0a21becb41a9dc1534ac95fb64148d2. git describe returns me `2021.07-397-g5f563ba1a`, is it different for you?
github.com/MoarVM/MoarVM/commits/new-disp agrees with me, no? 14:39
[Coke] it's not different. the issue is that when I updated, I just did an 'nmake install', which does NOT rerun the git describe check. 14:53
sena_kun [Coke], so is it fixed now or? 14:54
[Coke] I need to do a full reconfig and test. 14:55
Will take me a little bit, but I'm sure it'll be fine.
sena_kun great then
vrurg Actually, there is something wrong with versions. Seemingly, new-disp moar and nqp has lesser versions, than the one from master. I had to supply Configure.pl with --force-rebuild, but still ended up with removing nqp/ and install/ and clean rebuild. 15:07
[Coke] ISTR force-rebuild didn't do what I wanted. 15:12
so yes, when I want to really force a rebuild, I end up doing rm -rf install/ nqp/ and re-configuring. 15:13
vrurg For now I consider it as glitches of switching between different branches. But when new-disp is ready for becoming the mainline, there are likely rough edges to be resolved. 15:14
[Coke] huh. build seems to hang on windows on master. 15:21
(inside of nqp)
sena_kun: yup, with a full reconfig, it's fine 16:02
sena_kun yay
at least it wasn't wrong, phew. :)
sorry to take your time with this
[Coke] no worries! 17:15
It's not you, it's weird 3-tier build system. :)
gfldex Should Rakudo::Internals::LoweredAwayLexical leak in v6.e? (I'm fooling around with CLIENT::LEXICAL.) 20:10
vrurg gfldex: It may. Why not? The symbol exists, despite of the downgrade. 20:25
gfldex I would like to get the value begind that symbol instead of Rakudo::Internals::LoweredAwayLexical . 20:31
vrurg gfldex: unlikely possible. The symbol has been optimized away. 22:11

Powered by Cro and the Raku® Programming Language.

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

Thank you!