🦋 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.
00:00 reportable6 left 00:01 reportable6 joined
ab5tract it took quite a lot of tweaking to get it all together 00:09
00:13 benchable6_ left, linkable6_ left, coverable6_ left, committable6_ left, bloatable6_ left, unicodable6_ left, quotable6_ left, shareable6_ left, bisectable6_ left, notable6_ left, greppable6_ left, statisfiable6 left, evalable6_ left, nativecallable6_ left, sourceable6_ left, releasable6_ left, tellable6_ left, squashable6 left, reportable6 left
ab5tract gah, just trying to generate a list of installed packages in Perl is unbelievably annoying 00:15
MasterDuke ab5tract: are you experimenting with my moarvm branch that turns cuuids into ints? 00:19
00:20 AlexDaniel joined
ab5tract MasterDuke: trying to pave the way for that by removing Inline::Perl5's dependence on random mega-ranged cuids 00:24
tellable6 2024-04-29T00:21:20Z #raku <AlexDaniel> ab5tract buildable is making builds for all branches
MasterDuke nice
AlexDaniel c: is-item-redux say 42
committable6 AlexDaniel, ¦is-item-redux: «42␤»
[Coke] I was never able to get it working on macOS!
AlexDaniel true, there needs to be a server that makes builds on macOS 00:25
MasterDuke 64-bits is obviously a *much* smaller state space for randomness, but maybe it'd be enough in practicality?
AlexDaniel if you want to run stuff natively that is
(if that note was about whateverable, haha)
[Coke] blin 00:26
ab5tract I think just exposing QAST::Block.next-cuid through an nqp op should be enough
[Coke] so yes?
AlexDaniel yeah
ab5tract but I'm trying to test that by importing every installed perl5 module through Inline::Perl5 and ... wow.. what a mess
AlexDaniel [Coke]: I mean, I have about 11629 builds for amd64, so if we want to run stuff on macos then somebody has to make some builds 00:27
[Coke]: I can store them, that's not a problem (will take less than 50GB which I now have)
ab5tract [Coke]: github.com/ab5tract/Blin and github.com/ab5tract/whateverable/ 00:28
_should_ get you there
AlexDaniel btw I joined here to announce that the old server will be shut down relatively soon. If there's anything relying on that server, then hmmmmm, brace yourself I guess?
ab5tract AlexDaniel: my use case for blin is much simpler -- testing single commits
AlexDaniel but the good news is that all whateverable bots were running on a new server for a few months, so nothing should be affected at all 00:29
but you never know :)
ab5tract AlexDaniel: the mothership setting in blin's config could use an update, assuming general access is intended 00:30
AlexDaniel and dockerizing ≈16 websites that I host was kinda fun and not fun at the same time :)
ab5tract I definitely hear that :) 00:31
AlexDaniel ab5tract: hmmm, what kind of update?
ab5tract I assume it points to the old server 00:33
but in general, Whateverable has seen major changes with no version bump
AlexDaniel It point to "whateverable.6lang.org", which points at the new server now :) 00:34
ab5tract I was stuck looking for Whateverable::Building for ages
AlexDaniel points*
ab5tract ok :)
AlexDaniel ab5tract: but feel free to bump versions and do anything else that you need? New changes to the `main` branch in whateverable repo are deployed automatically 00:36
also, I'm not monitoring this or other channels anymore, but I do follow the repo, so feel free to file tickets with any requests, I might even have time to do it O:3 00:37
pull requests are even more welcome, especially if you also take the time to press the merge button XD 00:38
ab5tract understoodl thanks
01:34 AlexDaniel left 07:51 sena_kun joined
nine We have EXPORT and EXPORTHOW. I think we really just need an EXPORTPHASERS, e.g. my package EXPORTPHASERS { our sub INIT::restore-inline-perl5 { self.restore_interpreter; self.restore_modules; } }; 08:04
ab5tract Hmmm 08:13
08:19 finanalyst joined
finanalyst .tell coleman please take a look at question on #356 of Raku/doc-website. how best to make ebook available in container? 08:22
tellable6 finanalyst, I'll pass your message to coleman
08:30 MasterDuke left
ab5tract nine: that would be nice to have but I don't feel like it should be a prerequisite for migrating cuid to ints 08:34
nine I disagree. There is no reason to rush this. Everything works just fine the way it is. There is just the hope of getting some minor performance improvement out of a switch. So why not solve a long standing problem first and therefore get rid of the blocker? Why invest time and effort into replacing one hack with another? 08:36
ab5tract idk, for me it just feels a bit like "my using a hack is okay, but _your_ using a hack is definitely not" 08:47
that said, I usually come around to your way of thinking on these things, so... 08:50
nine It's less "my vs. yours" and more "way back then vs. now" 08:51
I never wanted that hack. There was just no other way to do that and I didn't have the skills to provide a proper solution. 08:52
ab5tract fair enough
lizmat yeah, knowing now after spending several months in internals, also has changed my view on many hacks / solutions of my own :-) 08:53
ab5tract I guess my biggest concern with EXPORTPHASERS is whether that means two separate front end solutions or not 09:11
09:19 [Tux] left 09:24 [Tux] joined 09:25 sena_kun left
nine Now if you want to prepend RakuAST as a blocker and shave that yak first, I'm not at all disappointed :D 09:37
10:19 finanalyst left
ab5tract Would that be an entirely separate mechanism apart from EXPORTPHASERS? 10:38
[Tux] Rakudo v2024.04-10-ga4c7d5d19 (v6.d) on MoarVM 2024.04-2-g8ccb87755
csv-ip5xs0.260 - 0.260
csv-ip5xs-201.107 - 1.129
csv-parser1.548 - 1.553
csv-test-xs-200.142 - 0.142
test1.961 - 1.963
test-t0.422 - 0.424
test-t --race0.280 - 0.281
test-t-205.093 - 5.137
test-t-20 --race1.217 - 1.238
10:48
10:58 rba left 11:01 rba joined
Geth nqp: ab5tract++ created pull request #821:
Draft: Add nqp::getnextcuid
11:18
nine ab5tract: I don't understand your question 11:33
Would what be that separate mechanism? 11:34
nine might be more dense than usual since yesterday's hike was much more exhausting than planned due to half a meter of snow on the north flank 11:35
ab5tract would a RakuAST solution also use this new EXPORTPHASERS?
lizmat ab5tract: my understanding would be: yes
ab5tract because I could also imagine something where you just EVAL some ast
nine Sure. RakuAST is just one way to implement EXPORTPHASERS. Just like with EXPORTHOW 11:36
ab5tract so two frontends it is then :)
nine Yes. But neither should be particularly challenging or much work at all.
ab5tract bisectable6: subset S where { m/"aaa"/ }; class C { method Str { "aaa" } }; my S $s = C.new 14:04
bisectable6 ab5tract, Will bisect the whole range automagically because no endpoints were provided, hang tight
ab5tract, More than 3 changes to bisect, please try a narrower range like old=2020.06 new=HEAD 14:05
ab5tract, Output on all releases: gist.github.com/7f26ff8bb1cde01e32...1b89e494d1
ab5tract bisectable6: sub s($s where { m/"aaa"/ }) { dd :$s }; class C { method Str { "aaa" } }; s C.new 14:08
bisectable6 ab5tract, Will bisect the whole range automagically because no endpoints were provided, hang tight
ab5tract, Output on all releases: gist.github.com/f3b6a8c215e5e09bea...b04280c45f
ab5tract, More than 3 changes to bisect, please try a narrower range like old=2020.06 new=HEAD
ab5tract ... 14:13
how does this code make any sense? ` method match(Any:U: |) { self.Str; nqp::getlexcaller('$/') = Nil }` 14:14
m: Any.Str.say
camelia Use of uninitialized value of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something meaningful.

in block <unit> at <tmp> line 1
ab5tract okay, so I'm bumping up against a Bad Thing that I did.. apparently a4c7d5d193 was merged without proper testing (I swear I thought I ran spectest!) 15:07
linkable6 (2024-04-28) github.com/rakudo/rakudo/commit/a4c7d5d193 Provide conditional exit status for USAGE
ab5tract presuming that this is a nice thing to have and that we want to keep it, what is the actual approach to adjusting the test failures (which are all, shockingly, related to expecting 0 as a status code) 15:09
I feel like I was just reading a thread about this (how to assign tests to specific versions) but I can't seem to find it at the moment 15:12
bisectable6: ' $_ = "abc"; my $i = 1; s:i(1)/a/b/ ' 15:34
bisectable6 ab5tract, Will bisect the whole range automagically because no endpoints were provided, hang tight
ab5tract, ¦6c (79 commits): «WARNINGS for /tmp/qsGz3qFQjR:␤Useless use of constant string " $_ = \"abc\"; my $i = 1; s:i(1)/a/b/ " in sink context (line 1)␤» 15:35
ab5tract, Nothing to bisect!
ab5tract m: ' $_ = "abc"; my $i = 1; s:i(1)/a/b/ '
camelia WARNINGS for <tmp>:
Useless use of constant string " $_ = \"abc\"; my $i = 1; s:i(1)/a/b/ " in sink context (line 1)
ab5tract m: $_ = "abc"; my $i = 1; s:i(1)/a/b/ 15:36
camelia ===SORRY!=== Error while compiling <tmp>
Adverb i value must be known at compile time
at <tmp>:1
------> $_ = "abc"; my $i = 1; s:i(1)⏏/a/b/
ab5tract bisectable6: $_ = "abc"; my $i = 1; s:i(1)/a/b/
bisectable6 ab5tract, Will bisect the whole range automagically because no endpoints were provided, hang tight
ab5tract, ¦6c (79 commits): «4===SORRY!4=== Error while compiling /tmp/CbRPOxEd5k␤Adverb i value must be known at compile time␤at /tmp/CbRPOxEd5k:1␤------> 3$_ = "abc"; my $i = 1; s:i(1)8⏏4/a/b/ ␤ «exit code = 1»»
ab5tract, Nothing to bisect!
ab5tract ... then how come there is a failing test saying that this should work? :( 15:37
sigh.. today is one of those days that I think I should hand my brain back in for a warranty return 15:40
`#?rakudo todo 'be smarter about constant detection'` 15:41
ugexe dear sir/madam we've been trying to reach you about your brain's extended warranty 16:02
ab5tract lol! 16:05
[Coke] HA 16:21
lizmat notable6: weekly 16:33
notable6 lizmat, 2 notes: 2024-04-11T19:48:46Z <gfldex>: weekly www.youtube.com/watch?v=pFKLHOygtqQ&t=945s ; 2024-04-14T22:09:43Z <antononcube>: rakuforprediction.wordpress.com/20...-exercism/
lizmat notable6: weekly reset
notable6 lizmat, Moved existing notes to “weekly_2024-04-29T16:33:47Z”
lizmat oops, forgot to reset last week
[Coke] oops 16:34
17:53 sena_kun joined
lizmat And yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2024/04/29/2024-...nish-line/ 18:00
Geth rakudo/r5464: 14f1a866c7 | ab5tract++ | src/core.c/Any.rakumod
Supply an Any.match method candidate for Any:D

This addresses the use of non-`Str` candidates arriving in `where` constraints that attempt to match against a regex. (R#5464 / #5464).
  ```
... (26 more lines)
18:32
linkable6 R#5464 [open]: github.com/rakudo/rakudo/issues/5464 [dispatching] Invocant of method 'match' must be a type object of type 'Any'
rakudo: ab5tract++ created pull request #5559:
Supply an Any.match method candidate for Any:D
18:37
rakudo/r5464: 9522e3e3ee | ab5tract++ | src/core.c/Any.rakumod
Supply an Any.match method candidate for Any:D

This addresses the use of non-`Str` candidates arriving in `where` constraints that attempt to match against a regex. (R#5464 / #5464).
  ```
... (32 more lines)
18:41
linkable6 R#5464 [open]: github.com/rakudo/rakudo/issues/5464 [dispatching] Invocant of method 'match' must be a type object of type 'Any'
19:31 finanalyst joined
[Coke] m: Metamodel::InvocationProtocol.new 19:47
camelia Could not find symbol '&InvocationProtocol' in 'Metamodel'
in block <unit> at <tmp> line 1
19:48
[Coke] m: MetaModel::InvocationProtocol.new
camelia Could not find symbol '&InvocationProtocol' in 'GLOBAL::MetaModel'
in block <unit> at <tmp> line 1
[Coke] docs refer to Metamodel::InvocationProtocol but that doesn't compile
m: Perl6::MetaModel::InvocationProtocol.new 19:49
camelia Could not find symbol '&InvocationProtocol' in 'GLOBAL::Perl6::MetaModel'
in block <unit> at <tmp> line 1
[Coke] m: Perl6::Metamodel::InvocationProtocol.new
camelia Could not find symbol '&InvocationProtocol' in 'GLOBAL::Perl6::Metamodel'
in block <unit> at <tmp> line 1
[Coke] oh! it was removed and we didn't update the docs yet. 19:58
Geth roast: 918f55a151 | ab5tract++ | 3 files
Migrate USAGE -> GENERATE-USAGE

GENERATE-USAGE is a part of Raku for a while now and not just a Rakudo implementation detail.
This patch is a compromise for changing USAGE to conditionally set the exit status in the same way that GENERATE-USAGE does. ... (12 more lines)
20:22
roast: 6f7d6ae582 | ab5tract++ | 6.c/S06-other/main-refactored.t
Don't pollute test exit status

This is a leftover change missed in the previous commit.
21:14
coleman . 21:31
tellable6 2024-04-29T08:22:13Z #raku-dev <finanalyst> coleman please take a look at question on #356 of Raku/doc-website. how best to make ebook available in container?
21:50 finanalyst left 22:52 sena_kun left