🦋 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. |
|||
ab5tract | github.com/rakudo/rakudo/issues/5589 | 08:51 | |
This one is so interesting.. haven’t been able to poke at it (left the laptop at home) | 08:52 | ||
But the | operation on the array results in nothing, but earlier | is used against the method chain to obtain the parameters.. and this seems to work without issue | 08:53 | ||
09:16
finanalyst joined
09:21
sena_kun joined
|
|||
finanalyst | .tell lizmat just sent you an email about C<> & V<> markup that been config'ed | 10:55 | |
tellable6 | finanalyst, I'll pass your message to lizmat | ||
lizmat_ | finanalyst: so if I understand correctly, you'd want all markup to be always parsed and disregard any :allow() during parsing ? | 10:57 | |
tellable6 | 2024-06-01T19:10:20Z #raku-dev <patrickb> lizmat: github.com/rakudo/rakudo/commit/77...r142633905 | ||
2024-06-02T10:55:20Z #raku-dev <finanalyst> lizmat just sent you an email about C<> & V<> markup that been config'ed | |||
finanalyst | problem with allow is that it can now be configured with a config for a markup code and for a block, | 10:59 | |
lizmat_ | ok, so you're saying it's become the renderers responsibility | ||
10:59
lizmat_ left
|
|||
finanalyst | Also, we had a discussion that meta data should handled by Renderer | 10:59 | |
11:00
lizmat joined
|
|||
lizmat | ok, gotcha... am in the middle of patrickb's bug report, will look at this after that | 11:01 | |
finanalyst | understood | ||
11:25
finanalyst left
|
|||
Geth | rakudo/main: 9502bcbbb2 | (Elizabeth Mattijsen)++ | 3 files Hopefully final fix for CPP name mangling - copied the original implementation to a new test file - made new cpp_param_letter implementation "our" for testability - added tests to run all possible types - compare results of original with new implementation - found and fixed some discrepancies |
11:46 | |
lizmat | patrickb ^^ | ||
patrickb | lizmat: Thanks! Sadly CI still fails with the same error. Maybe I can have a look in the late evening. | 13:33 | |
14:01
finanalyst joined
|
|||
lizmat | patrickb: then I guess my assumption abour $P / $R / $K being always empty, is incorrect | 15:56 | |
Geth | rakudo/main: cc52ff1251 | (Elizabeth Mattijsen)++ | 2 files Restrore NativeCall::Compiler::MSVC to its original state And adapt tests for it accordingly. Mostly to see how CI will react to this. |
16:24 | |
lizmat | patrickb ^^ | ||
Geth | rakudo/main: 946ab9024a | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Fixups.rakumod RakuAST: don't flatten markup inside C<> Let a renderer make that decision |
16:40 | |
patrickb | dev.azure.com/Rakudo/rakudo/_build...ew=results | 17:23 | |
lizmat: Now it succeeded | |||
lizmat | ok, then I'm gonna leave it at that | 17:42 | |
Geth | rakudo/main: c99d4369a4 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Fixups.rakumod RakuAST: don't flatten markup inside V<> either Let a renderer make that decision, as per the 2.0 rakudoc specification |
17:44 | |
18:22
finanalyst left
|
|||
patrickb | lizmat: Thanks for wading through this. | 18:59 | |
lizmat | well, I caused it, so it's only fair | 19:00 | |
patrickb | Do you think this is worth a point release? | ||
Still annoying;-) | |||
lizmat | not sure | 19:01 | |
it wasn't borked in 2024.04 right ? | 19:02 | ||
patrickb | Nope. Currently prevents me doing the binaries. I could either fake something up to make the build go through, or a point release. | 19:03 | |
lizmat | then I guess a point release is in order :-( | ||
patrickb | jdv: Can you do a point release containing lizmats MSVC native call fixes? | 19:06 | |
lizmat: Could you create a branch containing only the relevant changes? i.e. 2024.05 + cherry-picking the respective commits? | 19:07 | ||
lizmat | yeah, can do | 19:08 | |
patrickb | Thanks! | 19:09 | |
Geth | rakudo/main: 1a61ed025e | (Elizabeth Mattijsen)++ | 2 files RakuAST: allow for =formula, as per rakudoc 2.0 The reason this didn't work is because =formula shares the prefix with =for. Fixed for now by adding a specific token for =formula. Turns out the legacy grammar **also** has an issue with any abbreviated rakudoc items that share the prefix with a directive, such as =beginna and =enda. I wonder why nobody found this to be a problem before in the past 10+ years: I guess a general fix is just not that easy, *or* would slow down parsing in general too much. |
19:52 | |
rakudo/2024.05.1: 4 commits pushed by (Elizabeth Mattijsen)++ | 20:05 | ||
lizmat | patrickb jdv ^^ | ||
afk& | 20:16 | ||
Geth | nqp/main: e7d1bbe3df | (Daniel Green)++ | src/vm/js/Operations.nqp One step toward maybe unbreaking the JS build |
21:53 |