releasable6 Next release in ≈2 days and ≈11 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 07:00
[Tux] Rakudo v2024.12-49-gb4d798329 (v6.d) on MoarVM 2024.12-22-gcaf97bf07
csv-ip5xs0.268 - 0.282
csv-ip5xs-201.110 - 1.138
csv-parser1.110 - 1.151
csv-test-xs-200.138 - 0.138
test1.794 - 1.889
test-t0.414 - 0.416
test-t --race0.273 - 0.276
test-t-205.045 - 5.402
test-t-20 --race1.265 - 1.284
 10:11
tux.nl/Talks/CSV6/speed4-20.html / tux.nl/Talks/CSV6/speed4.html tux.nl/Talks/CSV6/speed.log
lizmat [Coke]: the changelog draft still not up, right? 12:21
Geth rakudo: patrickbkr++ created pull request #5767:
Fix `make spectest` on Windows		 12:53
rakudo: patrickbkr++ created pull request #5768:
Hopefully fix MacOS CI for good		 12:56
Geth rakudo/main: 42ab44ea99 | (Patrick Böker)++ | tools/templates/Makefile-common-rules.in
Fix `make spectest` on Windows
 13:28
rakudo/main: 0ba4f0eedd | (Patrick Böker)++ (committed using GitHub Web editor) | tools/templates/Makefile-common-rules.in
Merge pull request #5767 from patrickbkr/fix-make-spec-on-win

Fix `make spectest` on Windows
rakudo/main: 253c812734 | (Patrick Böker)++ | azure-pipelines.yml
Hopefully fix MacOS CI for good

The preinstalled perl version keeps changing. Let's hope there is always only one version installed.		 13:29
rakudo/main: 545c9e4e09 | (Patrick Böker)++ (committed using GitHub Web editor) | azure-pipelines.yml
Merge pull request #5768 from patrickbkr/fix-mac-ci

Hopefully fix MacOS CI for good
[Coke] lizmat: correct, sorry. expected complete in 2.5 hours at the moment. 15:13
tellable6 2025-01-23T14:56:36Z #raku <guifa> [Coke]: they're going to open a second round for talks on Feb 1
[Coke] (on top of other things, had a prod issue for $dayjob and an internet outage at home.)
guifa, thanks.
ww.
ugexe github.com/rakudo/rakudo/blob/545c...od#L58-L63 17:02
does this seem wrong to anyone else?
it is using .chars on a path to try to remove a path prefix
i guess maybe its fine, i'm mostly just wondering if that would be fine on e.g. windows using wide characters in paths such as chinese 17:08
lizmat feels that should probably be using .relative somewhere? 17:10
ugexe yeah that would at least be easier to reason about 17:12
timo as long as it's using Str, it shouldn't accidentally break when using wide characters, since they should turn into a single grapheme cluster under unicode's grapheme cluster rules 17:21
[Coke] lizmat, jdv - should *all* "bump nqp" commits be under ignore and delete, or do they have to be considered per?
lizmat I think they're always listed under ignore? 17:22
[Coke] excellent. Just wanted to make sure I wasn't missing a nuance. 17:24
github.com/rakudo/rakudo/wiki/ChangeLog-Draft updated 17:33
let me know if anything looks wonky. checking moarvm one now.
Also may add a CI section. 17:35
github.com/moarvm/moarvm/wiki/changelog-draft 17:40
Done - please let me know any issues.
lizmat [Coke]: done my bits on github.com/rakudo/rakudo/wiki/ChangeLog-Draft 17:57
[Coke] quick turnaround, thanks! 17:58
jdv igmore 17:59
ha. ignore.
the associated commits are useful but not those, imho. idk:) up to you. 18:00
similar to a merge commit
[Coke] ? 18:10
jdv i mechanically put merge commits and bump commits down there 18:33
[Coke] ahhh, yes, that's what I did as well, I thought
excellent
jdv fabulouso 18:34
[Coke] Sí! 18:39
Geth rakudo/lizmat-post2025.01: 094e9c52fa | (Elizabeth Mattijsen)++ | src/core.c/Hash.rakumod
Make Hash.clone actually clone itself

Fixes #1238. The problem was caused by the fact that Hash.clone would create a fresh copy of its invcant, instead of cloning it. Creating a fresh copy is ok if the invocant is of type core Hash, but if anything is mixed in, then the fresh copy will lose any additional attributes (as #1238 showed.
Fixed by changing nqp::create(self) to nqp::clone(self)		 18:45
rakudo/lizmat-post2025.01: 8acfa06b8e | (Elizabeth Mattijsen)++ | 2 files
Make Hash/Map.clone multis

So that calling clone on a Hash/Map type object either works (if no twiddles were specified), or fails for the proper reason (if there were twiddles specified).
Found as part of research into #1238		 18:47
lizmat R#1238 18:48
linkable6 R#1238 [open]: github.com/rakudo/rakudo/issues/1238 [regression] but-ing a Hash only works the last time ( (%(:42foo) but A) but B )
Geth rakudo/lizmat-post2025.01: 1fda513a13 | (Elizabeth Mattijsen)++ | src/core.c/Mu.rakumod
Simplify Mu.clone

It either dies, or returns self. Found when looking at #1238		 19:08
rakudo/lizmat-post2025.01: 557e41dfd3 | (Elizabeth Mattijsen)++ | src/core.c/Array.rakumod
Rinse, repeat #1238 for arrays

Array.clone had the same issue as Hash.clone from the point of view of mixins.		 19:20
tbrowder ,tell [Coke]: I've added the 'modf' info to problem solving. 19:53
jdv .tell [Coke] I've added the 'modf' info to problem solving. (from tbrowder actually) 19:54
tellable6 jdv, I'll pass your message to [Coke]
tbrowder sorry bout ths ff 😅 20:02
tbrowder jdv thnx for the help 20:16
[Coke] thanks - I'll remove the wiki page if you haven't 20:35
tellable6 2025-01-23T19:54:14Z #raku-dev <jdv> [Coke] I've added the 'modf' info to problem solving. (from tbrowder actually)
tbrowder i'll remove it... 20:36
[Coke] sorry, just did! 20:42
it's in the history if you wanted to save a copy, I can send you the .md
tbrowder no, that's ok, i could never find it again anyway! 20:46
i basically copied over the original and tidied it a bit,, making sure the original (but closed) PR is referenced with links to tests, etc. 20:48
i'm sorry i can't make the Raku conf, do they ever allow paid ppl video access during the conference? 20:49
[Coke] like a hybrid in person/zoom thing?
I don't know if he's the right person to ask, but I'd ping Util to check 20:50
tbrowder yes!
will do
[Coke] I'm going, but only because the spouse likes going on long drives mainly for the drive. :)
tbrowder gotcha 20:51

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!