🦋 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.
00:09 reportable6 left 01:12 reportable6 joined 03:17 evalable6 left
Geth rakudo: vrurg++ created pull request #4777:
Defuse a Failure
03:41
04:38 linkable6 joined 04:46 japhb left 04:57 japhb joined 05:16 frost joined 05:19 evalable6 joined 06:09 reportable6 left 07:02 frost left 07:11 reportable6 joined
Geth rakudo: 65738a2195 | (Vadim Belman)++ | src/core.c/Exception.pm6
Defuse a Failure

Minor fix to the fix in #4774. It's not sufficient to react to a failure. The failure has to be defused too.
07:44
rakudo: 74e1dcc65f | niner++ (committed using GitHub Web editor) | src/core.c/Exception.pm6
Merge pull request #4777 from vrurg/method-not-found-on-my-class

Defuse a Failure
07:45 linkable6 left, linkable6 joined 08:33 frost joined
lizmat Files=1351, Tests=117097, 288 wallclock secs (34.57 usr 9.76 sys + 4015.16 cusr 332.25 csys = 4391.74 CPU) 08:42
Geth rakudo: 05447137f0 | (Elizabeth Mattijsen)++ | src/core.c/Exception.pm6
Don't take the long road to defusing a Failure

  .so is not a Failure method, .Bool is.
08:55
09:47 frost left 09:48 frost joined 10:24 frost left 10:25 cognominal joined 10:26 cognominal left
jdv vrurg: next time please hold off until the release is over 11:29
11:58 sena_kun joined 12:07 reportable6 left 12:08 frost joined 12:50 frost left 12:51 squashable6 left 13:01 frost joined 13:05 Voldenet left
[Coke] we can revert if needed 13:10
lizmat jdv: actually vrurg only did a PR, it was nine who merged
13:11 Voldenet joined
jdv lizmat: is that what github.com/rakudo/rakudo/pull/4774 says? 13:12
lizmat heh, I was going by what Geth said: 13:13
jdv [Coke]: that's but i'm already past that point:)
lizmat [08:44:29] <Geth>¦ rakudo: 74e1dcc65f | niner++ (committed using GitHub Web editor) | src/core.c/Exception.pm6
Ah, I was thinking you talked about #4777 13:14
jdv i believe they are related and kinda proves my point that a freeze would be nicer 13:15
hopefully it doesn't break anything:) 13:17
lizmat too 13:19
gfldex m: sub wut($c is rw where { $c := 'oi‽']) {}; wut(42); 13:31
camelia ===SORRY!=== Error while compiling <tmp>
Cannot bind to '$c' because it was bound in a signature and variables
bound in signatures cannot be rebound unless they were declared with the
'is rw' or 'is copy' traits
at <tmp>:1
------> su…
gfldex I can see the `is rw` quite clearly…
m: sub wut($c is rw where { $c := 'oi‽'}) {}; wut(42); 13:32
camelia ===SORRY!=== Error while compiling <tmp>
Cannot bind to '$c' because it was bound in a signature and variables
bound in signatures cannot be rebound unless they were declared with the
'is rw' or 'is copy' traits
at <tmp>:1
------> su…
gfldex m: sub wut($c is copy where { $c := 'oi‽' }) { say $c }; wut(42); 13:33
camelia ===SORRY!=== Error while compiling <tmp>
Cannot bind to '$c' because it was bound in a signature and variables
bound in signatures cannot be rebound unless they were declared with the
'is rw' or 'is copy' traits
at <tmp>:1
------> su…
gfldex m: sub wut($c is copy where { $c = 'oi‽' }) { say $c }; wut(42);
camelia oi‽
jdv that looks like a change in the last release 13:34
part of codesections work on immutables or something 13:35
[Coke] jdv: (past) oops! 13:36
at one point I pushed for having a release branch so we didn't have to worry about master, but folks at the time were not interested.
if it's something you want to consider, you're free to, of course. Pros: you decide what's going in the release and anyone committing to master can't break it. Cons: there's cleanup merge work after the release. 13:37
(I am sure there are other cons.)
jdv yeah... 13:39
13:52 frost left
gfldex bisectable6: sub wut($c is copy where { $c := 'oi‽' }) { say $c }; wut(42); 13:54
bisectable6 gfldex, Will bisect the whole range automagically because no endpoints were provided, hang tight
gfldex, Output on all releases: gist.github.com/2bee9308a4f475e2cd...d62994f9bb
jdv anyone else see IO::Socket::Async::SSL's tests fail?
bisectable6 gfldex, Bisecting by output (old=2021.10 new=2021.12) because on both starting points the exit code is 1
gfldex, bisect log: gist.github.com/4f163d2592dc303d2a...dadd94bd55
gfldex, (2021-11-07) github.com/rakudo/rakudo/commit/32...175ae9dabc
gfldex, Output on all releases and bisected commits: gist.github.com/af3a7b02c7f03d7c2c...303e4f2bcd 13:55
jdv and after forcing that one then Cro::TLS test's fail 13:58
lizmat :-( 13:59
it was mention yesterday on #cro
*mentioned
jdv does anyone know if those modules issues or rakudo issues? 14:00
s/those/those are/
14:01 discord-raku-bot left
japhb jdv: Jonathan was of the opinion it was more likely a Rakudo regression 14:01
(The Cro::TLS one, at least) 14:02
14:02 discord-raku-bot joined
jdv lovely 14:03
japhb logs.liz.nl/cro/2022-02-10.html 14:04
A problem with Blin's categorization, perhaps
(Meaning, that's why Blin runs didn't point it out sooner.) 14:05
lizmat japhb: irclogs.raku.org nowadays :-)
jdv i imagine a rakudo release that can't run Cro is a blocker 14:06
japhb lizmat: Oh sorry, I've had those tabs in my saved set for so long ....
jdv: Yeah, I'd agree
jdv which is great to find out mid-releasing 14:07
japhb Yeah, my apologies for not having brought the red flag over from #cro
jdv ideally it would be noted in a blocker issue on rakudo. hmm... 14:08
japhb (Though given the importance of it, I'd recommend release managers join #cro; it's relatively low traffic.)
14:08 reportable6 joined
jdv all the blin runs since 2021.12 have Cro::$something as zeffail 14:21
sena_kun it's not a problem while the rakudo itself is not released, we can just create a blocker ticker, if it's resolved quickly on the rakudo side - release with the fix, if it's a VM one than we can release a moar point, which is not that bad, and proceed with rakudo on the point. 14:26
jdv yeah... meh. 14:27
sena_kun s/ticker/ticket/ 14:32
jdv can anyone in here help figure out this cro rakudo regression? 14:39
my best guess is it might have someting to do with IO::Socket::Async::SSL... 14:48
lizmat comfirm that I cannot install IO::Socket::Async::SSL on MacOS on head 14:49
*confirm
gist.github.com/lizmat/5a976b30c82...062cff5036 14:50
jdv nine: around? there's a guess the uint work may be involved in this regression 15:06
15:22 MasterDuke left
jdv i guess the release is on hold for now 15:25
15:32 frost joined
gfldex m: say [].sum; 15:51
camelia 0
gfldex m: say Int.sum; 15:53
camelia Cannot resolve caller sum(Int:U: ); none of these signatures matches:
(Any:D: *%_)
in block <unit> at <tmp> line 1
16:15 frost left 16:33 Xliff left 16:37 Kaipi left 16:38 Kaipi joined 16:52 squashable6 joined
gfldex I can't find any test for sum of the empty List or Array in roast. Is this just emergent behaviour? 17:05
m: say [].abs; 17:06
camelia 0
lizmat gfldex: that last one is just number of elements in the empty array 17:07
same for .sum: nothing to add, so the result is 0 17:08
Int.sum is illegal, so I think the error is correct fwiw
gfldex That's more of a docs issue. `.abs` and friends state that the invocant is coerst to Numeric. 17:09
I'm asking because it bit me with `sub MAIN(*@a) {…}`. 17:10
lizmat coercing a [] to numeric yields the number of elements 17:13
so [].abs is the same as 0.abs
[Coke] m: say Int.Numeric 17:15
camelia Use of uninitialized value of type Int in numeric context
0
in block <unit> at <tmp> line 1
17:48 sena_kun left
jdv lizmat: can you edit en.wikipedia.org/wiki/MoarVM for the new version? 18:04
lizmat "You are currently unable to edit Wikipedia." :-( 18:05
jdv aparrently i can't cause my ips are blocked... both my phone and my other inet.
wut...
lizmat some nonsense about an open Proxy / webhost 18:06
jdv can anyone do that?
[Coke]: can you?
18:08 reportable6 left 18:10 reportable6 joined
Geth Benchmark/main: f651c26268 | (Elizabeth Mattijsen)++ | 14 files
1.0
18:12
lizmat that's one less community module without version + migrated to zef 18:13
jdv [Coke]: nevermind - i got a rando friend to do it 18:16
Geth Benchmark/main: 81d4701606 | (Elizabeth Mattijsen)++ | .github/workflows/test.yml
Fix Github Actions CI
18:18
rakudo/release-2022.02: e04b0b2cb5 | (Justin DeVuyst)++ | README.md
Minimal Copyright update
18:26
rakudo/release-2022.02: 9b4f4d1381 | (Justin DeVuyst)++ | 3 files
Update changelog + announcement

Deliberately not logged:
  [36716b18][0935f5b2][ef35db06][c9b2e0b4][3aa23bf9][ba723099][fb909efc]
  [67906e48][25562dd8][28f8dddc][7227512c][acc093c1][de06617c][2ee3a46e]
  [36af7d0e][ac009507][fca7a6f9][0cd05ca0][b9ffbb97][8bba2aec][d743ebfd]
... (6 more lines)
lizmat jdv: isn't there an nqp release as well ?
nine jdv: sorry, just came home 18:27
jdv nine: jnthn got it
lizmat: yeah, getting to it. i think this is just the order the "Ake" thing wants 18:28
lizmat jdv ok, just checking :-)
jdv let me know if somehting blows up though:) 18:29
lizmat will do, but first dinner&
jdv nine: thanks though
nine Ah, so it was a module error after all? So the release is back on track? 18:33
jdv yes, the fix was tiny. it is...underway again 18:34
nine Cool.... jnthn++, jdv++ 18:35
jdv seems there's a bunch of roast errata fails - some todos, some not. 19:28
lizmat could you gist them ? 19:31
jdv gist.github.com/jdv/5d662beb26ad52...0388879f31 19:36
the todos don't have to block release, right? 19:39
otherwise its just 7 or 8 depending on the branch 19:40
lizmat todo's shouldn't block, I don't thnk 19:41
so that leaves t/spec/S12-meta/primitives.t t/spec/S15-nfg/grapheme-break.t
t/spec/S26-documentation/block-leading.t t/spec/S26-documentation/block-trailing.t
t/spec/S26-documentation/why-both.t t/spec/S26-documentation/why-leading.t t/spec/S26-documentation/why-trailing.t 19:42
jdv right 19:44
and t/spec/S02-types/pair.rakudo.moar, no? 19:47
lizmat oops, yes 19:50
jdv all except grapheme-break.t have recent changes on master 19:51
Geth Test-Class/main: c094abc080 | (Elizabeth Mattijsen)++ | 19 files
Initial migration work
20:00
Test-Class/main: e6573cd1a4 | (Elizabeth Mattijsen)++ | Changes
1.0
lizmat jdv: hmmm 20:01
20:02 discord-raku-bot left, discord-raku-bot joined
lizmat afk& 20:04
jdv i think i have fixes for all but grapheme-break.t on 6.c-errata 20:37
any ideas?
Invalid surrogate codepoint 55296 in block <unit> at t/spec/S15-nfg/grapheme-break.t line 61 20:42
Geth roast/6.c-errata-fixups-2022.02: 1509be59e2 | (Vadim Belman)++ (committed by Justin DeVuyst) | 5 files
Make sure role group delegates .WHY to its default candidate

This required to remove tests which were testing for `Role.WHY` returning `Nil`. But as was suggested in this comment:
  github.com/rakudo/rakudo/pull/3549...-687587580
... (10 more lines)
20:45
roast/6.c-errata-fixups-2022.02: 1e444f8129 | (Vadim Belman)++ (committed by Justin DeVuyst) | S12-meta/primitives.t
Fix incorrect tests

Interaction of a few errors made the test erroneously passing:
  - typechecking cache was initialized incorrectly
  - omitted typing with Mu of some meta-class method parameters resulted
   in hidden calls to type_check method
... (8 more lines)
roast/6.d-errata-fixups-2022.02: 1a2fc4a055 | (Vadim Belman)++ (committed by Justin DeVuyst) | 5 files
Make sure role group delegates .WHY to its default candidate

This required to remove tests which were testing for `Role.WHY` returning `Nil`. But as was suggested in this comment:
  github.com/rakudo/rakudo/pull/3549...-687587580
... (10 more lines)
roast/6.d-errata-fixups-2022.02: cce470d0e6 | (Elizabeth Mattijsen)++ (committed by Justin DeVuyst) | S02-types/pair.t
Adapt to correct behaviour of Pair.ACCEPTS(Pair)
20:47
roast/6.d-errata-fixups-2022.02: e525fc141a | (Vadim Belman)++ (committed by Justin DeVuyst) | S12-meta/primitives.t
Fix incorrect tests

Interaction of a few errors made the test erroneously passing:
  - typechecking cache was initialized incorrectly
  - omitted typing with Mu of some meta-class method parameters resulted
   in hidden calls to type_check method
... (8 more lines)
roast: jdv++ created pull request #796:
6.c errata fixups 2022.02
roast: jdv++ created pull request #797:
6.d errata fixups 2022.02
20:48
jdv does anyone have ability to merge at roast? it appears i do not. 20:50
nine me neither :/ 20:52
jdv welp, i guess the release is blocking on those being merged and someone figuring out what is up with 6.c-errata t/spec/S15-nfg/grapheme-break.t 20:55
or should i just move on without that stuff?... 20:56
nine I think you can continue anyway
jdv ok 20:57
nine I think those release steps are there to ensure that we don't accidentally break older language versions. We do it on purpose in this case and once those PRs are merged, the tests will be happy again
jdv except that grapheme one. but yeah cool. 20:58
nine Invalid surrogate codepoint 55296 21:04
Is what that ends in
jdv saw that. still no ideas:) 21:05
nine Test file has been removed in master in a commit named: Remove duplicate test and testgen. GraphemeBreakTest.t provides the same and has been updated for Unicode 9.0 21:09
jdv oh, looks purely vestigal cause master and 6.d don't have it
ah, right. i'll try that. thanks. 21:10
nine It's surrogate-D800 it doesn't like. Every test that includes this code point fails. And D800 is 55296 in decimal 21:15
U+D800 is not a valid Unicode code point / character. 21:16
It belongs to the "High Surrogates" block which goes from 0xD800 to 0xDB7F but is not defined in the Unicode 9.0 database of code points. 21:17
www.unicode-symbol.com/u/D800.html
This commit probably caused the test failures: github.com/rakudo/rakudo/commit/c9...98d339ec58 21:21
Commit message says: "Turns out, the test was actually wrong, after consulting with
our resident Unicode guru Samantha McVey."
I think one could infer from this, that U+D800 is indeed invalid and we're correct in throwing that error and that it's a simple oversight that the 6.c-errata branch wasn't updated accordingly. 21:22
surrogate-D800 was still present in the Unicode 8 test file the test was originally based on: www.unicode.org/Public/8.0.0/ucd/a...akTest.txt but was removed in Unicode 12 21:26
The reason they disappeared was: "Surrogate code points were moved from Control to XX for the Grapheme_Cluster_Break property, to eliminate the need to have isolated surrogate code points in the test cases." www.unicode.org/versions/Unicode12...se_Changes 21:28
21:49 moon-child is now known as bowl-of-petunias, bowl-of-petunias is now known as moon-child 21:54 linkable6 left 21:55 linkable6 joined 21:59 linkable6 left
[Coke] jdv++ sorry my IRC attention is very spotty. 22:07
Geth roast/6.c-errata-fixups-2022.02: 61ef05b92c | (Justin DeVuyst)++ | 6 files
Fix "grapheme break" test.

See github.com/Raku/roast/commit/db27cf25.
22:12
jdv nine: ^ that might be sufficient 22:13
Geth nqp: 59df321e04 | (Justin DeVuyst)++ | tools/templates/MOAR_REVISION
[release] Bump MoarVM revision to 2022.02
22:14
nqp: 13861d6670 | (Justin DeVuyst)++ | VERSION
[release] Bump VERSION to 2022.02
rakudo/release-2022.02: 7f21bbd8ff | (Justin DeVuyst)++ | tools/templates/NQP_REVISION
[release] Bump NQP revision to 2022.02
rakudo/release-2022.02: 7c08c5fedb | (Justin DeVuyst)++ | VERSION
[release] Bump VERSION to 2022.02
rakudo: jdv++ created pull request #4781:
Release 2022.02
22:31
rakudo/master: 5 commits pushed by (Justin DeVuyst)++ 22:34
jdv okie dokie then. I believe the release is over. 22:46
if someone could merge this - github.com/rakudo/rakudo.org/pull/77 - that'd be great 22:47
patrickb: release 22:48
vrurg jdv: merged 22:49
jdv oh, one more thing... 22:50
vrurg: can you try to edit en.wikipedia.org/wiki/Rakudo to update it?
i'm blocked. as is liz. ...
vrurg jdv: nope. I don't have an account with them. 22:51
lizmat jdv++ 22:54
jdv who is "A"?... github.com/rakudo/rakudo/releases/tag/2022.02 22:57
i have no idea what that is all about
vrurg: you might be able to anon style
but thanks for the merge 22:58
in any case, nite. i'm out. 23:02