🦋 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: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
Geth rakudo: vrurg++ created pull request #4580:
Make enums and Bool throw on .new		 02:30
roast: vrurg++ created pull request #762:
Add tests for enums throwing on .new		 02:31
japhb Does anyone have preferences for tab completion behavior? I'm working on github.com/japhb/Terminal-LineEditor (to be a pure Raku alternative to Readline and Linenoise, perhaps useable in the Rakudo REPL), and I'm thinking about how to approach tab completion. 06:28
The behavior of Linenoise and the bash prompt for tab completion are very different: Linenoise rotates through available possible completions every time you hit <tab>, but only ever shows you that one.
Bash instead tries to advance as far as it can with a common prefix, but if there's no common prefix and you <tab> again, it lists all the available completions 06:29
(and redisplays the prompt and the input-in-progress) 06:30
Neither one gives you any information about the options other than the pure text, and I have a feeling more advanced editors do more than this (in fact I think Comma's been demonstrated to have a quite advanced version) 06:31
Anyone with feelings on this, please leave them here (or in an issue on the repo) so I can review and cogitate 06:32
nine I like Bash's behaviour. Or maybe I'm just more used to it. 06:36
Mondenkind japhb: I like zsh's behaviour 07:06
it shows you a list of completions on the _following_ line when you press tab; and then pressing tab repeatedly cycles through them, while continuing to show them 07:07
my gut is that if I wanted something much fancier than that I would no longer want a line editor. But I am not sure what you mean by 'information about the options'. Like annotating return values of methods? Or using colour for [something]? 07:08
(err, return _types_) 07:14
MasterDuke i think the windows terminal and vim both do it the linenoise way, but i do prefer the bash behavior 07:26
Mondenkind finds bash kind of useless. Why scroll the screen? Why do I want to see the results of my previous completions? 07:27
Mondenkind here's a solution: make the tab completion approach configurable! 07:30
releasable6 Next release in ≈3 days and ≈7 hours. 1 blocker. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 11:00
Geth nqp: 19bac5b297 | (Jonathan Worthington)++ | src/core/dispatchers.nqp
Add megamorphic handler for nqp-istype

Also tweak to look at the object's mode flags rather than the type's ones for deciding if to call type_check, since it exists on the LHS meta-object.		 11:07
rakudo/megamorphic-handlers: f9c60866cb | (Jonathan Worthington)++ | src/core.c/ObjAt.pm6
Ensure ValueObjAt is composed

Otherwise we end up doing late-bound type checks for it.		 11:08
rakudo/megamorphic-handlers: a51423d5a2 | (Jonathan Worthington)++ | src/Perl6/Metamodel/ParametricRoleGroupHOW.nqp
Publish a type check cache for role groups

This can at least contain the types a role "pretends" to be (Any, Mu) and the role itself, which means we avoid having to make calls to the type_check method in a number of situations.
Kaiepi releasable6, status 12:22
releasable6 Kaiepi, Next release in ≈3 days and ≈6 hours. 1 blocker. Changelog for this release was not started yet
Kaiepi, Details: gist.github.com/da5c80edea53592cde...f4b386d627
[Tux] Rakudo v2021.09-279-g1d8bf66a5 (v6.d) on MoarVM 2021.09-667-gc43ee9dff
csv-ip5xs1.369 - 1.434
csv-ip5xs-2015.284 - 15.323
csv-parser4.589 - 4.701
csv-test-xs-200.371 - 0.371
test7.232 - 7.251
test-t1.669 - 1.699
test-t --race1.016 - 1.022
test-t-2024.182 - 24.860
test-t-20 --race8.034 - 8.096
 13:03
tonyo [Tux]: wow .. i remember when csv-parser was in the 20-30s range in that test 15:12
why is text::csv missing from that?
is it because there are two text::csv's now? 15:14
[Tux] test-t is raku Text::CSV, Text::CSV from perl5 is not tested as I already test Text::CSV_XS and Text::CSV_PP : tux.nl/Talks/CSV6/speed5.html 15:27
tonyo i mean there's a second text::csv in the raku ecosystem 15:48
oh, i see, they're both yours
Geth rakudo/bugfix-4488: d0a739f1ea | tony-o++ | src/core.c/operators.pm6
bug fix for #4488
 17:42
rakudo: tony-o++ created pull request #4582:
Fixes `require STUB:file<>` bug #4488		 17:43
Geth roast: vrurg++ created pull request #763:
Rename is_wrapped -> is-wrapped		 18:24
rakudo/bugfix-4488: eb1c1f4f13 | tony-o++ | src/core.c/operators.pm6
Revert "bug fix for #4488"

This reverts commit d0a739f1ea9f9c87d2188503b53567a9e795f40a.		 18:37
rakudo/bugfix-4488: d5cb092d85 | tony-o++ | src/core.c/operators.pm6
bug fix for #4488
tbrowder why does one msg say log yr chngs for release and the other says ChangeLog hasn’t started? 21:47
which is correct?
the wording for the heredoc issue is not quite correct on one version, so i want to make sure it gets fixed 21:53
sena_kun tbrowder, the truth is that it is not started yet, but one can start and log. 22:07
releasable6, status
releasable6 sena_kun, Next release in ≈2 days and ≈20 hours. 1 blocker. 0 out of 279 commits logged
sena_kun, Details: gist.github.com/f0af768d38bfc5f819...3d92383c88
sena_kun tbrowder, ^
tbrowder so what can i do to fix the wording on the heredoc entry? 22:09
sena_kun tbrowder, were your patches merged? 22:10
tbrowder yes 22:12
i just changed the PR title to read a bit better (PR #4567) 22:13
sena_kun tbrowder, for the best results you can log them yourself. The draft is at github.com/rakudo/rakudo/wiki/ChangeLog-Draft and you click "Edit Page" and write a description. Your fix goes under `Fixes` category, I think, and then you add an entry like `+ $text `. I'll format commit hashes myself, just leave a link to PR there. 22:15
tbrowder i see yr new draft, but what happened to the other draft? 22:16
sena_kun what's "the other draft"?
tbrowder in the query by Kaiepi above 22:17
sena_kun that's the only one 22:18
tbrowder the response has a link to a draft page
sena_kun the releasable6 bot checks this wiki page. it still contained the old changelog, so it said "has not started yet", I updated it and it says another thing now. 22:19
If you mean gist.github.com/Whateverable/da5c8...f4b386d627 then it's not a changelog, but rather something to write a changelog from.
tbrowder ok, i’ll fix it… 22:20
sena_kun thanks! 22:21

Powered by Cro and the Raku® Programming Language.

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

Thank you!