🦋 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.
00:00 reportable6 left 00:03 reportable6 joined 00:19 discord-raku-bot left 00:20 discord-raku-bot joined 00:45 vrurg left, vrurg joined 01:56 epony left 02:40 Kaiepi left 03:20 epony joined 03:23 ab5tract joined 04:04 ab5tract left, ab5tract joined 04:39 epony left 04:47 ab5tract left 04:56 ab5tract joined, epony joined 04:57 ab5tract left, ab5tract joined 05:21 ab5tract left 05:22 ab5tract joined 05:45 ab5tract left 05:46 ab5tract joined 05:53 ab5tract left 06:00 reportable6 left 06:01 reportable6 joined 07:01 evalable6 left 07:02 evalable6 joined
nine And that one's parent commit "347d4eb425d95ef76e293a20367a40b0bda5f6f1" is still ok 08:32
08:49 Kaiepi joined
nine nine@sphinx:~/rakudo ((7d4c4ae07...))> RAKUDO_RAKUAST=1 ./rakudo-m -Ilib -e 'use experimental :rakuast' 08:54
nyi cclass compilation
This is it: RAKUDO_RAKUAST=1 ./rakudo-m -Ilib --ll-exception -e '/<[a..zA..Z]>/' 08:57
nyi cclass compilation
09:03 sena_kun joined
Geth rakudo/main: 2814e016a0 | (Stefan Seifert)++ | src/Raku/ast/regex.rakumod
RakuAST: implement support for regex char classes with multiple ranges
nine And we're up at 87 passing tests and 601 passing spectests
Nemokosch I got sick towards the end of holidays which mostly killed my motivation and now it's back to work and procrastination... 09:41
but there are things I did check because of how things work (badly) currently
smartmatch of s///, m//, tr// kind of stuff, for example 09:42
or typed declarations of &variables
nine vrurg: in case you're looking for a new task, t/02-rakudo/11-deprecated.t would pass if 'use v6.d;' actually did something in RakuAST 09:52
12:00 reportable6 left 12:02 reportable6 joined
Geth rakudo/main: dcfb306ba7 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Deparse.pm6
Handle deparsing of ternaries with heredocs properly

For whatever number or depth of embedding, courtesy of the power of dynamic variables!
lizmat e.g.$a ?? qq:to/FOO/ !! $b ?? qq:to/BAR/ !! qq:to/BAZ/ 12:29
13:41 Kaiepi left
nine lizmat: out of curiousity: why use so many nqp ops in deparsing code? 13:51
Nemokosch back to m// - this rather shouldn't be a QuotedRegex node, should it? 14:02
> # A subsbitution, such as `s/abc/def/`, `S/not_in/place/`, or `s/abc/ = 'def'`. 14:07
Oh there is quite some peculiar stuff in here
for example, I don't think `s/abc/ = 'def'` worked previously 14:08
lizmat nine: it's possibly a premature optimization 14:13
I was wondering that myself while working on this again, actually, after almost 2 years :-)
14:14 Kaiepi joined
lizmat I guess my thinking was that we'd want deparsing to be as fast as possible for debugging purposes 14:16
also: that work predates the newdisp work, so things have changed since then 14:17
Geth rakudo/main: b8f56ee6d1 | (Elizabeth Mattijsen)++ | 20 files
Only deparse for each test once

Almost all tests were basically deparsing twice: once for the diag, and once to do actual tests on. Since deparsing could get pretty expensive, doing it only once should help with resource usage during development and CI testing.
Oddly enough, a few tests failed when using a "canned" deparse. Not sure yet what caused that, marked those tests as such for later investigations.
rakudo/main: 2006fcff28 | (Stefan Seifert)++ | t/12-rakuast/heredocs.rakutest
Add missing comma in test
vrurg nine: you mean we lack support for `use v6.<rev>;`? I thought it is not a priority. 14:42
Nemokosch by the way: don't yall think now that we have a rakuast irc channel, we could try to concentrate these discussions there? I would have guessed that was the idea 14:44
vrurg Wrong assumption it seems. :)
nine I don't see the point of having multiple channels with the same people when only one of them will be active at a time anyway. Current topic of raku development is rakuast 14:45
vrurg: we're gonna have to have use v6.<rev> support at some point anyway. I don't see any priorities at all. The whole thing will only be done once all the bits are implemented. 14:46
Nemokosch nine: on one level, I agree with you; on another level, it's kinda disturbing that any "Raku development" notification can pop up any time, just like any other topic could, but it happens often even 14:48
vrurg "Priority" is too strong word perhaps, but non-passing test is a good reason to do something. I just need to finish up some $dayjob tasks this week in first place.
nine vrurg: don't we all :) 14:50
vrurg I'm more lucky with this because there are good periods when they totally forget about my existence; and bad periods when they have a couple of new document formats to parse and all has to be done yesterday. :) 14:52
lizmat I guess in the future, #raku-rakuast would be more for users of RakUAST, rather than developers 14:53
I'm going to keep it around until then 14:54
vrurg BTW, some status, just out of curiosity: my work project lib/ is already 820K, all .rakumod/.rakutest files are ~13800 lines in total. No auto-generated code. Dist 'provides' has 112 modules. 14:57
lizmat pretty impressive! 14:58
vrurg The only problem in production is memory corruption causing sporadic but regular errors like `no symbol found` in modules using NativeCall. Luckily, it's a cron-launched application, so I just re-process things on the next run. 14:59
15:08 ab5tract joined 15:12 ab5tract left 15:13 ab5tract joined 15:34 ab5tract left 15:35 ab5tract joined
Geth nqp: patrickbkr++ created pull request #791:
Don't error without git but all needed files are present
rakudo/main: 30cf5c7914 | (Elizabeth Mattijsen)++ | 2 files
Remove need for RakuAST::Deparse to be instantiated

  - replace all public attributes by methods
  - don't instantiate in RakuAST::Node.DEPARSE
  - remove params to RakuAST::Node.DEPARSE
  - should allow a class to be specified in .DEPARSE, but I haven't
   figured yet how to take an optional argument in this Raku/ast
... (6 more lines)
rakudo: patrickbkr++ created pull request #5149:
Don't errors without git but all needed files are present
17:11 linkable6 left, evalable6 left 17:12 linkable6 joined 17:13 evalable6 joined 17:27 sena_kun left 17:28 sena_kun joined 18:00 reportable6 left 18:02 reportable6 joined 18:20 [Coke] joined
[Coke] . 18:21
18:51 discord-raku-bot left 18:52 discord-raku-bot joined 19:24 opoku joined 19:41 opoku left 20:24 Kaipei joined 20:28 Kaiepi left 20:37 hythm joined 20:49 NemokoschKiwi joined 21:06 NemokoschKiwi left
Geth rakudo/main: 0562c89494 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Deparse.pm6
Part 1 of HLLizing RakuAST::Deparse

About 2/3rds through, but getting too tired now, more tomorrow
[Coke] ^^ is that "just" de-nqp-izing? 21:37
I think I'm just re-phrasing the commit comment
23:26 sena_kun left
lizmat yeah, it's de-nqpizing, mostly, except for a few cases 23:53