Documentation Channel for #raku | This channel is logged | Roadmap: github.com/raku/doc/wiki
Set by [Coke] on 23 May 2022.
05:06 rf left 08:20 sena_kun joined 14:12 rf joined 16:23 jjatria left 16:24 jjatria joined
Geth doc/main: f92dca051c | (Elizabeth Mattijsen)++ | doc/Language/tables.rakudoc
Remove trailing whitespace from code blocks
16:30
doc/main: 621b2c9f85 | (Elizabeth Mattijsen)++ | doc/Language/pod.rakudoc
C<M<>> is *not* the way to show verbatim M<>

V<M<>> is the way and currently the only way that is consistent in rendering between Pod::To::Text and RakuDoc::To::Text. Soe change all the appropriate instances
lizmat only 1 left
and that's caused by improper flattening of L<> and X<> inside V<>
and with github.com/rakudo/rakudo/commit/eb169c3fc6 now all raku documentation files render the same under RakuAST as under the legacy grammar 17:15
except almost 5x as fast 17:17
[Coke] ^^ 17:20
[Coke] lizmat++ 17:21
is it worth turning on that comparison test going forward until the old way goes away? 17:22
lizmat I think so 17:26
[Coke] Roger. 17:28
Geth doc/main: 303aa45080 | (Will Coleda)++ | xt/rakuast-compare.rakutest
Always run this test under 'make xtest'

  lizmat++ got all tests passing, keep testing until the new
AST model is the only method (then we can remove the test file)
17:37
17:48 rf left
[Coke] Failed some example compilations in Language/tables.rakudoc 17:54
(running full make xtest)
Geth doc/main: 9758544590 | (Will Coleda)++ | 2 files
Fix error when var not set.

add test plan back.
use RAKUAST when populating cache
17:57
[Coke] ... I think I may leave that one failing for a bit until I redo the examples compilation to use AST parsing instead of old school parsing to get the code blocks. 18:10
lizmat $ast.grep({ $_ ~~ RakuAST::Doc::Block && .type eq 'code' | 'implicit-code' && .resolved-config<lang> eq 'raku' }) 18:35
[Coke] ^^ something like that
19:32 rf joined
[Coke] dd 'doc/Language/faq.rakudoc'.IO.slurp.AST.grep(* ~~ RakuAST::Doc::Block); # Cannot find method 'from-paragraphs' on object of type NQPMu 19:57
... my fault 19:58
RAKUAST makes t/03-tests-valid.rakutest very unhappy 20:14
wonder if I need to add a shim in here to turn RAKUAST on and off per test file before invocation 20:16
20:53 sena_kun left
lizmat you need to add use v6.e.PREVIEW ? 22:16