🦋 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 01:03 linkable6 left, evalable6 left, evalable6 joined 01:05 linkable6 joined 02:00 ab5tract left 03:12 guifa joined 03:15 guifa__ left 03:24 MasterDuke left
Geth ¦ problem-solving: 0rir assigned to coke Issue Ranges' purposes github.com/Raku/problem-solving/issues/382 04:31
05:39 samcv_ left 05:40 samcv joined 06:00 reportable6 left 06:02 reportable6 joined 11:47 ab5tract joined 12:00 reportable6 left 12:02 reportable6 joined 12:13 ab5tract left 12:19 ab5tract joined 12:22 squashable6 left 12:24 squashable6 joined 12:25 ab5tract left
Geth rakudo/main: fc11963d37 | (Elizabeth Mattijsen)++ | 6 files
Add X::NoZeroArgMeaning exception class

And use it to create failures at the appropriate places. Also introduce a Str.no-zero-arg implementation-detail method to reduce the overall bytecode size.
[Coke] looks like jvm is broken, trying to bisect (SO SLOW), went back 2 release, and that had a completely different failure. 14:25
we should do a bare minimum check on JVM&JS backends for each release.
2023.06 has the same failure as 2023.05 (not current failure) 14:33
nemokosch absolutely unsurprising. Is there anyone who is maintaining even JVM (let alone JS which is probably broken for a long time) at the moment? 14:35
ugexe i think when it started failing for the JVM it ran out of memory. i think i noticed this some months after it started, but when i raised the memory limit there was a new failure (the BOOTInt one) that had to have occured sometime after the JVM stoppped working initially
avoiding these situations is a boon of a workflow that only merges green CI runs 14:36
[Coke] oh, so this was not a new failure but a revealed one? then this bisect is useless, I bet. 14:37
(merge only green) yes, please. 14:38
ugexe yeah, you'd have to somehow get github.com/rakudo/rakudo/pull/5309 in for commits where the failing started
nemokosch that would probably open Pandora's box, whether that's a good thing or not... 14:43
ultimately, even MoarVM is in a sorry state. There are like 3 people working on it and nobody to consolidate issues or PR's. It's all ad-hoc. 14:44
lizmat re the BOOTint failure: do we have a full stacktrace of that? if so, please make an issue with it and I'll look at it later today 14:47
[Coke] lizmat: there's an issue. don't have a stacktrace. 15:06
will see if I can make one. 15:07
ugexe dev.azure.com/Rakudo/rakudo/_build...f&l=72 15:08
[Coke] oh, that is in the ticket, yes. 15:14
github.com/rakudo/rakudo/issues/5337 15:16
ugexe seems like maybe something to do with the value used for the indent level
[Coke] I don't know where that Metamodel file is. 15:17
ugexe it gets generated, so i suspect if you tried to build locally that it would show up 15:19
[Coke] ah, I didn't finish a rebuild at head. 15:27
make -j might also be broken for jvm 15:50
(though I did just have to reboot halfway through, who knows)
16:05 codesections joined
nemokosch raku.org/compilers/features.html this page is also generated from an archived repo 16:17
[Coke] the line it's complaining about bootint on is nqp::die($die_message) 16:32
nemo: "why not pugs" link goes nowhere and can be removed at some point. 16:33
ugexe what is calling the nqp::die though 16:34
the nqp::die is probably trying to generate a different error message with indentation, but it also fails to parse whatever value indent is 16:35
nemokosch yeah well, my point is that this whole thing may be best to be removed... 16:40
[Coke] (remove) that's fine with me. was most helpful when the .net version was a going concern. 16:42
nemokosch also, there are "snippets". Here is an example: raku.org/snippets/99-bottles.html 16:59
it's very apparent that this isn't meant to show up in itself; it's probably a template 17:00
yet to find where it appears
the reason this matters is that only these "snippets" require a highlight feature 17:01
I think it's unused... the index page, for example, has the highlighting hard-coded 17:17
17:18 linkable6 left, evalable6 left
[Coke] might be from the examples site? 17:20
but yes, we should update those to use the same highlighting as we use in docs
17:21 linkable6 joined, evalable6 joined
nemokosch for the time being, I'd try to minimize the costs. We don't have a great long-term solution for highlighting 17:29
18:00 reportable6 left 18:02 reportable6 joined 18:14 squashable6 left 18:15 squashable6 joined 18:19 sena_kun joined 18:22 sena_kun left 18:23 sena_kun joined
github.com/rakudo/rakudo/issues/5313 free "story points" 19:15
[Coke] wonders if jmespath.org/ would be useful in raku, or if the fact that we can easily convert json to raku objects means we don't need a syntax to search inside of the object. 21:01
Would this be of use to anyone?
(asking in -dev to avoid some noise)
nemokosch the cost-benefit doesn't look great on that one /2c 21:16
22:45 sena_kun left
vendethiel And this exists already: github.com/jnthn/json-path 23:04
Geth Template6/main: a77e6aa724 | (Polgár Márton)++ | 5 files
23:56 evalable6 left, linkable6 left 23:58 evalable6 joined 23:59 linkable6 joined