🦋 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.
nine ab5tract: I could easily setup wireguard to get into my home network. I just have never really needed that. ssh has always been enough. But recently I upgraded my Wifi router to latest openWRT and I guess I forgot to enable even that. 06:40
m: say Q|for ^10 { }; 1|.AST.statements.head.sunk 06:41
camelia True
nine lizmat: ^^^
The last statement of an EVALed comp unit is not sunk. Otherwise you'd never get the produced value
Ah, now I read that you noticed that already :) 06:42
[Tux] Asked to up the version of Text::CSV. Inc in META6.json is the only thing to be done? And should that also require "6.d" (instead of 6.c)? 07:19
lizmat [Tux]: yes, META6.json is enough, unless you also specify versions in the "class" statement 07:24
6.d would be a safe bet by now
[Tux] done 07:28
lizmat did you do the release already? I'm hoping on updating Slang::Tuxic later today 07:29
making it ready for the RakuAST world
ah well, I'll be in the next release :-) 07:31
*it'll 07:43
nemokosch why does Text::CSV still live in p6c? 07:56
lizmat I guess someone needs to tell [Tux] how to migrate to the zef ecosystem :-) 07:58
nemokosch fez upload 🧠 08:02
anyway, I haven't said this in a while, maybe it's for the better... but I actually think the condition of CompUnit stuff AND also the ecosystem architecture is kind of a blocker for Raku adoption 08:03
and unlike improving the runtime, this probably doesn't require technical wonders, "only" some willingness to let go 08:04
Geth Slang-Tuxic/main: c585abc714 | (Elizabeth Mattijsen)++ | 13 files
Prepare for update to zef, test CI
08:06
nemokosch docs.raku.org/language/compilation#Why_change? there is this page and it's somewhat reassuring that by now I anticipated all of these points 08:08
It's just not enough justification for the price imposed on everybody, namely that the filesystem based resolution is useless for production purposes. 08:09
[Tux] lizmat, you *do* have commit. Just go ahead :)
lizmat True that... but I don't have Tux credentials on the zef ecosystem :-)
[Tux] Ah, so what does that Tux person need to do? 08:15
nemokosch if there is no zef:Tux then maybe get fez first and do fez register 08:22
lizmat deathbyperl6.com/faq-zef-ecosystem/
nemokosch zef install fez 08:23
Geth Slang-Tuxic/main: 4ae8eee5c0 | (Elizabeth Mattijsen)++ | META6.json
Oops, forgot to add dependency
08:25
nemokosch is this also a fresh release? 08:26
lizmat nemokosch you mean Slang::Tuxic ?
nemokosch yep 08:28
lizmat yeah, that will be a fresh release, with RakuAST support 08:33
[Tux] fez register finished 08:36
lizmat [Tux] ok, will do the rest to the Text::CSV module once I'm done with Slang::Tuxic, ok ? 08:37
your login name be "tux" I assume ? 08:38
so auth<zef:tux> ?
[Tux] or Tux. Don't know if it is case-sensitive. Thanks!
>>= Username: Tux
lizmat it is case sensitive afaik
ok, so Tux it is then 08:39
Geth Slang-Tuxic/main: 4a0935ac95 | (Elizabeth Mattijsen)++ | lib/Slang/Tuxic.rakumod
First stab at supporting RakuAST

Sadly, this still fails when run with RakuAST. Will need more investigation.
08:55
Slang-Tuxic/main: c84066d968 | (Elizabeth Mattijsen)++ | Changes
0.0.2
08:59
lizmat the RakuAST support is still incomplete, and may actually need some Raku grammar tweaks, otherwise it was ready for release 09:00
|Tux| regarding the META change: is there *anything* I need to do other than up the version in META on upcoming changes? 10:24
lizmat well, unless you introduce new dependencies ? 10:28
|Tux| so, change, inc version, push, done. ? 10:32
and do I switch from `zef` to `fez`? My time routine now invokes stuff like `zef update`, `zef install --exclude=perl Slang::Tuxic` and `zef upgrade --exclude=perl Slang::Tuxic` 10:35
lizmat that remains the same, fez is only for uploading dists 10:40
it's like the CLI for PAUSE
like zef is the cpan CLI 10:41
(at least, last time I looked at the cpan CLI about 10 years ago :-)
|Tux| never used the CPAN CLI 10:43
lizmat personally, I use App::Mi6, which wraps fez 10:44
notable6 10:50
notable6: weekly
ah, yeah AlexDaniel is working on that :-)
AlexDaniel lizmat: I tried to bring it up but I'm getting: 11:57
Invocant of method 'match' must be a type object of type 'Any', not an object instance of type 'IRC::Client::Message::Privmsg::Channel'. Did you forget a 'multi'?
lizmat interesting... hmmm
AlexDaniel also I don't think that there are any saved notes :) 11:58
for this week I mean
lizmat ok, good to know :-)
AlexDaniel the annoying part about all those errors is that I'm not getting any line numbers at all :( 11:59
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2023/09/25/2023-...eleaseses/ 12:01
Geth Inline-C/main: d40e7baa67 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | README.md
Announce archivation
13:21
Slang-Tuxic/main: 8fd132bebd | (Elizabeth Mattijsen)++ | 3 files
0.0.3
13:34
lizmat Slang::Tuxic now works with RakuAST
vrurg .ask Xliff In your GLib it looks like using something like `submethod BUILD (:$hash-table, :$!type) { $!h = $hash-table }` is a common pattern. Why not safer `submethod TWEAK(:hash-table($!h), :$!type) {}`? 13:50
Geth Slang-Tuxic/main: d5aa7a0229 | (Elizabeth Mattijsen)++ | 3 files
Check CI after having removed atkeyish hack
15:11
Slang-Tuxic/main: 6819c2afe0 | (Elizabeth Mattijsen)++ | 2 files
0.0.4
15:15
rakudo/main: fbacec3869 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Raku.rakumod
RakuAST: only produce sigil if there is one
17:46
rakudo: ab5tract++ created pull request #5394:
RakuAST: Several `*` tidbits
17:55
lizmat ab5tract: ready to merge ? 17:58
Geth rakudo/main: 54c5682f9d | (Elizabeth Mattijsen)++ | src/core.c/Rational.rakumod
Make sure Rational.raku is correct

The interface to the Rational role uses two positionals. Rat and FatRat have their own .raku logic, but the role didn't. This meant that it fell back on the generic .raku, which uses named arguments, which doesn't work. Now .raku is correct.
18:03
rakudo/main: a9d8b78382 | (Elizabeth Mattijsen)++ | src/Raku/Grammar.nqp
RakuAST: set $sigil safely even if there was no sigil match
18:21
rakudo/main: 7dc21fcff6 | (Elizabeth Mattijsen)++ | 2 files
Add a sigilless-variable token
19:14