Geth Blin: hythm7++ created pull request #25:
Add --pm option to switch between package managers		 01:05
¦ roast: vrurg assigned to Kaiepi Issue S32-io/IO-Socket-Async.t incorrectly fails when IPv6 is disabled on a host github.com/Raku/roast/issues/684 02:28
bartolin reports that nothing (new) is broken on JVM ;) 05:32
codesections how long would it take to cut a new release that adds a welcome message to Rakudo if we need to do so rather urgently? I'm on TPF's legal committee, and we're having some difficulty getting the Raku trademark; we're still taking to the lawyer, but it's possible that our best option will be to add a welcome message to Rakudo similar to what we have in the --version output 11:17
sena_kun codesections, adds nothing, but a welcome message? Tonight. 11:18
codesections (e.g., something like "Welcome to Rakudo v2020.10 implementing Raku v6.d")
sena_kun codesections, point?
lizmat when should this welcome message be shown ? 11:19
sena_kun Ah, of course, including _only_ releasing, not someone writing the code.
lizmat I guess the install script that says "Rakudo installed succesfully!" now, could be adapted for that? 11:20
but should this also run with a package installer ? 11:21
sena_kun lizmat, I would expect to see this message when repl is started. 11:22
lizmat that should be easy to fix :-) 11:23
sena_kun lizmat, let's wait for more details from codesections 11:29
codesections (sorry, my connection dropped for a bit, but I didn't miss any messages) 11:30
sena_kun ok
codesections, so how urgent is "urgent"?
codesections the legal requirement is for "launch screens that show the mark in an introductory message box that appears after opening the program". As I said, we're still talking to the lawyer, but I _think_ a into message when starting the REPL would fit thatbox 11:31
lizmat Files=1336, Tests=113604, 218 wallclock secs (29.53 usr 8.46 sys + 3047.00 cusr 283.43 csys = 3368.42 CPU) 11:32
Geth rakudo: e435a08798 | (Elizabeth Mattijsen)++ | src/core.c/hash_slice.pm6
Further simplify %h{a;b;c} candidate

The Str handling is not different from non-Str handling, so remove that special case. Is the same number of conditionals ran anyway.		 11:42
lizmat the rest of that handling is pretty underdeveloped and may cause breakage when developed properly 11:43
so that will probably wind up in 6.e
afk for a few hours& 11:52
[Coke] OOC, what needs to be in the message? Raku™ ? 13:32
moritz Your Friendly Potters. 13:33
[Coke] :) 13:34
Also, is there a ticket somewhere in Raku-land about this?
(making sure the community is on board with what YAS is doing) 13:35
codesections [Coke]: the relevant issue is at github.com/Raku/problem-solving/issues/193 I haven't updated it with this latest idea but will do so if it goes anywhere. 13:40
codesections the language we asked the attorney about is "Welcome to Rakudo v2020.10 implementing the Raku Programming Language (v6.d)" 13:42
[Coke] codesections: thanks
[Tux] Rakudo version 2020.09-23-ge435a0879 - MoarVM version 2020.09-8-g60070970c
csv-ip5xs0.836 - 0.882
csv-ip5xs-208.130 - 8.152
csv-parser26.121 - 26.179
csv-test-xs-200.390 - 0.408
test7.728 - 7.823
test-t1.859 - 1.961
test-t --race0.861 - 0.862
test-t-2032.150 - 33.540
test-t-20 --race9.498 - 9.881
 13:54
ugexe buffer 1 17:14
Geth rakudo: 551b577d86 | (Elizabeth Mattijsen)++ | 2 files
Re-implement %h{a;b} from scratch for 6.e

Differences with 6.c implementation:
  - if a single key is requested, return a scalar instead of a list
  - correctly work with :exists and Whatever
  - add support for :delete
  - make :!delete and :!exists do the right thing
As part of handling github.com/rakudo/rakudo/issues/3937 .		 17:31
Geth rakudo: f106c23107 | (Elizabeth Mattijsen)++ | t/02-rakudo/03-corekeys.t
Oops, forgot to add &postcircumfix:<{; }> to core.e keys
 17:54
lizmat that should make CI happy again 17:57
gfldex lizmat: did you conside to implement design.raku.org/S09.html#line_419 ? 18:01
I did that in a module and found it very useful.
lizmat gfldex: I'm focusing on github.com/rakudo/rakudo/issues/3937 atm 18:02
gfldex: which module are we talking about? is it in the ecosystem ?
gfldex github.com/gfldex/perl6-rakudo-slippy-semilists
lizmat the shaped array business has it roots deep in NQP, that would need some serious looking into 18:04
still, it feels like something very useful to have... 18:05
gfldex I tried to implement it in CORE and failed for that reason.
lizmat yeah... that's what I mean... I think I looked at implementing it at one time a few years ago
travis-ci Rakudo build failed. Elizabeth Mattijsen 'Re-implement %h{a;b} from scratch for 6.e 18:33
travis-ci.org/rakudo/rakudo/builds/734367307 github.com/rakudo/rakudo/compare/e...1b577d86d0
lizmat travis definitely lags CircleCI 18:34
by about an hour
f106c23107 fixed this
linkable6 (2020-10-09) github.com/rakudo/rakudo/commit/f106c23107 Oops, forgot to add &postcircumfix:<{; }> to core.e keys
Geth rakudo: bd029ae4ce | (Patrick Böker)++ | tools/build/binary-release/build-windows.ps1
Fix Windows binary release script

The `raku` path passed to `prove` was wrong.		 20:39
rakudo: 99acfa9cac | (Patrick Böker)++ | 3 files
Don't run spec tests during binary release build

They run > 2:00 or hang indefinitely *after* running all tests in
  `S03-sequence/misc.t`. Not sure what's going on there.
rakudo: 6ff7c5be23 | (Patrick Böker)++ (committed using GitHub Web editor) | 3 files
Merge pull request #3938 from patrickbkr/binary-release-fixes

Binary release pipeline fixes
rakudo: 4dbdcb8d07 | (Patrick Böker)++ | tools/lib/NQP/Config/Rakudo.pm
Build: A wrong NQP version should abort the build

Irrespective of using the `--with-nqp` configure flag a wrong NQP version is always fatal.		 20:53
rakudo: f3a70a5867 | (Patrick Böker)++ | tools/lib/NQP/Config/Rakudo.pm
Switch to 'Raku' github project in Configure.pl

This mirrors the move that happened a while ago.
rakudo: b8191e8160 | (Patrick Böker)++ (committed using GitHub Web editor) | tools/lib/NQP/Config/Rakudo.pm
Merge pull request #3939 from patrickbkr/wrong-nqp-abort

Build: A wrong NQP version should abort the build

Powered by Cro and the Raku® Programming Language.

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

Thank you!