🦋 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.
vrurg greppable6: ParametricRoleHOW 01:24
greppable6 vrurg, 8 lines, 6 modules: gist.github.com/9f273db44a3e9926cf...da36424ccb
lizmat when everybody is back from holiday, I'd like to convene a discussion / jitsi meet on how to progress on RakuAST in 2023 12:30
gfldex fast progress would be nice :) 13:53
Geth rakudo/main: 5 commits pushed by (Stefan Seifert)++ 13:56
nine That's 600 passing spectests!
lizmat: unfortunately make tests seems to have regressed quite a bit. git bisect points at the "Implement use experimental :rakuast" commit 13:57
s/make test/make test/ 14:02
Nemokosch to be honest, the RakuAST frontend doesn't seem very hard to grasp. It might even be easier than the previous way of doing things 14:04
the thing is - alas, as usual - it's *heavily underdocumented*
nine It is definitely much easier than the previous iteration. The hardest part of developing RakuAST is really finding out what the heck the old frontend does. 14:10
guifa is committed to documenting RakuAST nodes when he finishes the new version of Intl::Format::Number 16:27
I've been using already a number of them that don't have tests — so I'll also add those
Nemokosch an example is: say, you know what piece of code you would like to cover, behavior-wise. It's not at all obvious what AST object structure belongs to it 16:49
from names like Declaration::Simple
guifa Nemokosch: that might require an longer tutorial. I think the best approach is to write something starting from literals, basic ops, and going up from there in small building blocks. At least, that was the approach I took for the advent post 17:28
tellable6 guifa, I'll pass your message to Nemokosch
lizmat guifa: I'm working on RakuAST::Heredoc testing 19:55
nine: 0f86d6f72a7d05fbdfda32a8193f59c5d34f7278 basically got reverted, so I don't understand how that could matter? 19:56
Geth rakudo/main: 4535adc087 | (Elizabeth Mattijsen)++ | t/12-rakuast/TODO
Mark RakuAST::Categorical for tests
20:25
lizmat m: dd val "42\n" # /me wonders whether the "\n" should be in the string representation 20:47
camelia IntStr.new(42, "42\n")
nine lizmat: maybe my bisection result was bogus. Had to skip a few commits because they were on the old master branch where I obviously couldn't run those rakuast tests. What I can definitely say is that between my last commit in December and these new ones from today several tests broke. 22:09