🦋 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.
00:08 reportable6 left 00:29 ggoebel joined 02:10 reportable6 joined 02:34 djinni` left 02:47 djinni` joined 03:57 linkable6 left, evalable6 left 03:59 linkable6 joined, evalable6 joined 04:04 Geth left, TempIRCLogger__ left, RakuIRCLogger left 04:06 lizmat left 04:07 lizmat joined 05:53 quotable6 left, statisfiable6 left, nativecallable6 left, benchable6 left, coverable6 left, evalable6 left, sourceable6 left, unicodable6 left, bisectable6 left, linkable6 left, reportable6 left, releasable6 left, notable6 left, committable6 left, bloatable6 left, greppable6 left, squashable6 left, shareable6 left, evalable6 joined 05:54 bisectable6 joined, releasable6 joined 05:55 shareable6 joined, greppable6 joined, nativecallable6 joined, committable6 joined 05:56 benchable6 joined, sourceable6 joined, quotable6 joined, coverable6 joined 06:10 reportable6 joined 06:54 unicodable6 joined, bloatable6 joined, squashable6 joined 06:56 notable6 joined 07:00 lizmat left, Geth joined 07:04 RakuIRCLogger left, RakuIRCLogger joined 07:08 RakuIRCLogger left, RakuIRCLogger joined
lizmat Files=1352, Tests=117129, 295 wallclock secs (35.10 usr 9.75 sys + 4101.30 cusr 332.28 csys = 4478.43 CPU) 07:12
07:12 TempIRCLogger joined 07:13 RakuIRCLogger left, RakuIRCLogger joined 07:17 TempIRCLogger left, TempIRCLogger joined 07:57 [Coke] left 08:46 [Coke] joined 08:56 statisfiable6 joined 09:53 linkable6 joined 11:06 TempIRCLogger left 11:40 hankache joined
Geth rakudo/add-Compiler.cver: 4bfe57b5dd | (Elizabeth Mattijsen)++ | 2 files
Add implementation-detail Compiler.cver method

The raw string version of the Compiler.new.version is needed in several places. This commit introduces a new Compiler.cver method that returns that raw string, and removes the private $!cver attribute from CURI, as it was basically just the raw compiler version.
No functional changes, just less copy-and-paste and a little less memory use for CURI instances (not that we have many of them).
11:48
rakudo: lizmat++ created pull request #4832:
Add implementation-detail Compiler.cver method
11:59 hankache left 12:00 hankache joined 12:08 reportable6 left 12:10 reportable6 joined 12:17 hankache left 12:30 hankache joined 12:32 hankache left
Geth rakudo: da54063c23 | (Elizabeth Mattijsen)++ | src/core.c/CompUnit/Repository/Installation.pm6
Streamline the checks for prefix writeability in CURI

They were incorrectly indented and smelled of Perl :-)
12:38
12:53 hankache joined
Geth rakudo: f4cba21c67 | (Elizabeth Mattijsen)++ | src/core.c/Rakudo/Internals.pm6
Make sure we serialize exports

This is a brute force attempt at #1920. It makes this script display "DONE" rather than segfault. The original script uses
  (require ::($module)), and that still usually segfaults.
   class TestMe {
... (14 more lines)
12:58
12:58 hankache left 12:59 linkable6 left 13:02 linkable6 joined 14:00 [Coke]_ joined 14:02 [Coke] left 14:03 ggoebel left 14:07 hankache joined 14:15 discord-raku-bot left
gfldex greppable6: ^compose 14:15
greppable6 gfldex, 9 lines, 2 modules: gist.github.com/cba81c9a7c237f128b...1ddf04ee81
14:15 discord-raku-bot joined
gfldex greppable6: \^compose 14:22
14:22 hankache left
greppable6 gfldex, 62 lines, 22 modules: gist.github.com/3702e423695ac9f0df...25ab11c9f4 14:22
gfldex Is there an alternative to phaser COMPOSE (that is NYI)? 14:45
lizmat .^compose I guess ? 14:46
gfldex I want to have a class method called when the role is composed. (As the phaser would do.)
Do I have to provide my own RoleHOW or is there a shortcut? 14:50
lizmat class Foo { }; BEGIN { Foo.method; Foo.^compose }
that is, if that method would change anything in the way the class would compose 14:51
if not, then BEGIN Foo.method would be enough ?
gfldex I may be asking for the wrong thing and shall rethink my question. :-) 14:52
greppable6: method \^ 14:55
greppable6 gfldex, 4 lines, 3 modules: gist.github.com/b4c84f7d22944b26cc...49ca6c7d06
15:37 discord-raku-bot left, discord-raku-bot joined 15:55 hankache joined 16:00 hankache left 16:11 AlexDanielIRC joined 16:15 ggoebel joined 16:30 ggoebel left, ggoebel joined 17:05 [Coke]_ is now known as [Coke] 17:15 sena_kun left 17:16 sena_kun joined
Geth whateverable: AlexDaniel++ created pull request #381:
Switch the IRC host to libera
17:19
rakudo: 1fd2d84359 | (Elizabeth Mattijsen)++ | src/core.c/CompUnit/Repository/Installation.pm6
Remove dead wood

Turns out the CURI $!cver attribute was basically unused
17:26
whateverable: AlexDaniel++ created pull request #382:
Fix FootgunDB to write into a temp file first
17:28
AlexDaniel Now, if you're wondering why am I submitting pull requests to whateverable, it's because I don't have access to the repo 17:29
so please review and merge :)
Geth whateverable: bb135ac850 | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | lib/Whateverable.pm6
Switch the IRC host to libera (#381)

Because freenode is no longer a thing (and everything moved to libera). Note that this tweak was hastily applied on the server and remained uncommitted, so actually this commit changes nothing, it's just about committing what is already there.
... (7 more lines)
17:31
whateverable: 14dfd36e49 | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | lib/Whateverable/FootgunDB.pm6
Fix FootgunDB to write into a temp file first (#382)

The comment in the code should explain everything. Without this the bots can trash their own files when the system runs out of space.
17:32
AlexDaniel lizmat: thanks! In the future please merge instead of squashing 17:34
lizmat I seem to never do the right thing in that respect :-( 17:35
why merge instead of squash?
AlexDaniel keeps gpg signatures, keeps commits, keeps commit descriptions, makes it possible to easily have a local version of master with some commits that are not merged yet 17:36
lizmat ok, next time I'll ask first :-)
AlexDaniel I guess that's the right way, yeah. Some people use git just to dump the changed files, others will craft by hand each commit and separate things if they belong to different commits… and then when it's squashed we're like, yeah… great, thank :D 17:38
thanks*
but all good no worries
OK now the good part of restoring tellable… finding the right backups :) 17:40
ha, `git: 'restore' is not a git command.`, really 😆
lizmat ++AlexDaniel
18:07 reportable6 left 18:45 sena_kun left 18:46 sena_kun joined
Geth rakudo: 412c4625d5 | (Elizabeth Mattijsen)++ | 2 files
Streamline R:I.EXPORT_SYMBOL a bit

  - abstract the actual binding in a sub
  - don't do method calls / hash lookups more than once
  - use AT-KEY / BIND-KEY for further speedup
  - remove one now unnecessary decont
  - improve readability by using better variable names
This code probably never get hot enough to be speshed or inlined. Any effects will only be seen at compile time for the various "is" traits.
18:57
whateverable/raiph-patch-1: fdb6574578 | raiph++ (committed using GitHub Web editor) | lib/Whateverable.pm6
Fix comment bitrot

  ❤️ Alex, Liz
19:03
whateverable: raiph++ created pull request #383:
Fix comment bitrot
19:04
AlexDaniel thanks :)
Geth whateverable: 1f7beb0eba | raiph++ (committed using GitHub Web editor) | lib/Whateverable.pm6
Fix comment bitrot (#383)

  ❤️ Alex, Liz
19:05
rakudo: 624b9271d9 | (Elizabeth Mattijsen)++ | src/Perl6/bootstrap.c/BOOTSTRAP.nqp
Use better variable name

This was probably copy-pasted from the Array version. But seeing
  $array.BIND-KEY() just feels wrong, and only leads to wasting time
  (it did for me). So make sure nobody will do that in the future.
19:09
whateverable: AlexDaniel++ created pull request #384:
Improve Tellable's .seen normalization on startup
19:16
19:24 lizmat_ joined 19:25 lizmat left 19:26 TempIRCLogger joined
Geth whateverable: 915225c4ad | (Aleks-Daniel Jakimenko-Aleksejev)++ | xbin/Tellable.p6
Improve Tellable's .seen normalization on startup

Tellable stores a dict of all latest messages by all users that it has seen. However, the keys in that dict are not exact nicknames. Instead, they are intentionally mangled simplified versions of nicknames. The goal of the normalization is to allow users to change their nicknames slightly and still receive messages, and have `.seen` that DWIMs. ... (12 more lines)
19:28
whateverable: 29bfa38a7f | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | xbin/Tellable.p6
Merge pull request #384 from AlexDaniel/improve-tellable-renormalization

Improve Tellable's .seen normalization on startup
19:41 lizmat_ left 19:42 lizmat joined 19:53 hankache joined, hankache left
AlexDaniel I see that there is some action in #raku-beginner. Should the bots join that channel too? 20:23
also, what's the name of the discord bot now? 20:25
my constant DISCORD-BRIDGE = any(‘raku-bridge’, ‘raku-bridge1’);
that's wrong, yes?
I think the last bot name is discord-raku-bot. Can anybody also tell me its fallback name (if discord-raku-bot is taken)? 20:27
reason why it matters: bots understand commands from the discord bot, m:, .tell, bisect: and all others
but for that they need to know the nickname
MasterDuke gfldex: ^^^ questions re discord-raku-bot 20:38
20:39 discord-raku-bot left 20:40 discord-raku-bot joined
gfldex AlexDaniel: the bot is called discord-raku-bot 20:40
I don't think we should move all the bots over there. It's a beginner channel after all. 20:41
AlexDanielIRC cool, no means no :) 20:42
gfldex: what's its backup name?
20:45 discord-raku-bot left, discord-raku-bot joined
MasterDuke tellable would probably be good there 20:46
AlexDanielIRC based on the logs, either it has never used it, or it is set to something that doesn't start with discord-raku-bot
MasterDuke: well, we should perhaps take some bots down anyway 20:47
squashable6: what is your purpose today? 20:48
squashable6 AlexDanielIRC, I cannot recognize this command. See wiki for some examples: github.com/Raku/whateverable/wiki/Squashable
MasterDuke linkable6 doesn't work right now, but i wish it did, that's a useful one
AlexDanielIRC linkable6: help
linkable6 AlexDanielIRC, Like this: R#1946 D#1234 MOAR#768 NQP#509 SPEC#242 RT#126800 S09:320 524f98cdc # See wiki for more examples: github.com/Raku/whateverable/wiki/Linkable
MasterDuke bloatable6 probably isn't needed everywhere
AlexDanielIRC R#1946
why doesn't it work though 🤔 20:49
20:49 linkable6 left
AlexDanielIRC well, that's very telling. 20:49
20:49 sena_kun left
AlexDanielIRC have no idea why it just left 20:50
20:51 sena_kun joined
AlexDanielIRC damn, most of these bots were so useful. Reportable I think wasn't touched in a while but it was a really cool one I think 20:51
MasterDuke i think it makes sense to keep them all in #whateverable, just maybe not all the other channels 20:52
AlexDanielIRC gfldex: so, what's the backup name? There has to be something that it uses if the main one is taken. Most bots in that situation will continue to use the backup name until restarted, don't know if that one is smarter 20:53
easiest way to test – kill it, take its name, start it, see what's the result
gfldex It uses the default alternate nick of IRC::Client, defined here: github.com/lizmat/IRC-Client/blob/...kumod#L350 21:18
Geth whateverable: AlexDaniel++ created pull request #385:
Fix the list of discord bridge bot names
21:37
AlexDanielIRC gfldex: thanks!
22:06 AlexDanielIRC left 22:52 AlexDanielIRC joined 23:26 TempIRCLogger left 23:37 AlexDanielIRC left 23:55 [Coke] left