🦋 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.
00:02 reportable6 left 00:06 reportable6 joined 02:11 japhb joined 02:23 frost joined 02:28 frost left 02:39 frost joined 02:40 frost left, frost joined, frost left 02:43 frost joined 03:25 Xliff joined
Xliff \o\ 03:25
m: use NativeCall; class A is repr<CStruct> { HAS $.array[30] is CArray }; my $a = A.new; $a.elems.say; 03:26
camelia 5===SORRY!5=== Error while compiling <tmp>
The [] shape syntax with the $ sigil is reserved
at <tmp>:1
------> 3lass A is repr<CStruct> { HAS $.array[307⏏5] is CArray }; my $a = A.new; $a.elems.s
expecting any of:
state…
Xliff m: use NativeCall; class A is repr<CStruct> { HAS @.array[30] is CArray }; my $a = A.new; $a.elems.say;
camelia 1
Xliff ^^ Isn't this a bug? 03:27
^^ lizmat timotimo ugexe codesections vrurg
moon-child m: use NativeCall; class A is repr<CStruct> { HAS @.array[30] is CArray }; my $a = A.new; $a.array.elems.say; #it doesn't seem to just be deferring to its element's method, either
camelia 0
Xliff A nd if I cast from a pointer, I get "Don't know how many elements a C array returned from a library" 03:29
Something seems wonky.
At any rate, wanted to get that out before I snozz for the night. 03:30
Geth roast: McSinyx++ created pull request #736:
Test cases where mkdir should fail
03:55
04:32 vrurg joined 04:34 vrurg_ left 04:55 Xliff left 05:24 nebuchadnezzar joined 06:02 reportable6 left 06:05 reportable6 joined
nine m: my @a = 0..100; loop { @a.hyper.map({$_}).sum; } 08:32
camelia A worker in a parallel iteration (hyper or race) initiated here:
in block <unit> at <tmp> line 1

Died at:
emit without supply or react
in block <unit> at <tmp> line 1
nine How on earth?
moon-child doesn't seem to need the map to trigger 08:35
nine indeed
moon-child nor the sum, works with just @a.hyper[0]
nine Actually this is enough: 08:36
m: my @a; loop { @a.hyper[0]; }
camelia (signal XCPU) 08:37
nine m: my @a; loop { @a.hyper[0]; }
camelia A worker in a parallel iteration (hyper or race) initiated here:
in block <unit> at <tmp> line 1
in block <unit> at <tmp> line 1

Died at:
emit without supply or react
in block <unit> at <tmp> line 1
in block <unit> a…
nine I fear, it's a spesh bug 08:38
Initially I suspected commit 6f42d19eb0 and reverting that made my program massively more stable. But in the end I still got that error after enough tries. 08:39
08:40 linkable6 left 08:42 linkable6 joined 09:15 dogbert17 joined 09:20 squashable6 left 09:21 squashable6 joined
dogbert17 nine: it can fail even when spesh is disabled 09:27
09:31 RakuIRCLogger left, AlexDaniel left, RakuIRCLogger joined 09:34 AlexDaniel joined
dogbert17 tsan shows some quite suspicious output though 09:34
gist.github.com/dogbert17/dea0a2c2...80be3b1710 09:37
nine dogbert17: ah, thanks! I got it to fail even with MVM_SPESH_LIMIT=29 which is hit early in the compilation, so I figured the same. But haven't been able to reproduce with spesh completely disabled so far 09:42
dogbert17 the comments about the destroyed mutexes I haven't seen before 09:44
nine At least it also appears in a debug build 10:14
Geth rakudo: db074a8bc4 | (Elizabeth Mattijsen)++ | 2 files
Abstract single elem path of [] into R:I

This allows the more common path of having multiple values in a
  [] to be inlined, causing a small but noticeable improvement.
10:53
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
11:20 squashable6 left 11:21 squashable6 joined 12:02 reportable6 left 12:03 reportable6 joined 13:33 lizmat left 13:34 lizmat joined 14:34 Xliff joined 14:56 morayj joined 15:15 b2gills left 15:50 Xliff left 17:05 morayj left 17:06 morayj joined 17:48 frost left 18:02 reportable6 left 18:04 reportable6 joined
Geth nqp: 1385091dc7 | (Daniel Green)++ | tools/templates/moar/Makefile.in
Simplify MoarVM backend runner's Makefile template

Remove one redundant entry ('moar') and two unneeded entries ('sha1',
  'tinymt').
18:22
nqp: c0d5add92f | MasterDuke17++ (committed using GitHub Web editor) | tools/templates/moar/Makefile.in
Merge pull request #728 from MasterDuke17/simplify_moar_runner_makefile_template

Simplify MoarVM backend runner's Makefile template
timo .tell Xliff yeah that does look like a bug 19:21
tellable6 timo, I'll pass your message to Xliff
Geth rakudo/circumfix-array-typed: 323427ffa6 | (Elizabeth Mattijsen)++ | 4 files
Introduce [1,2,3]:Int syntax

Arrays created with [] do not have typing. It is one of Raku's traps that people do: sub foo(Int @a) { }, and then get a typecheck error when they pass [1,2,3] as an argument. This commit would allow specification of [1,2,3]:Int as a valid argument.
... (15 more lines)
19:50
rakudo: lizmat++ created pull request #4406:
Introduce [1,2,3]:Int syntax
timo interesting 20:07
ugexe i was thinking `foo(Int @ = [1,2,3])` would work, but nope 20:12
lizmat m: sub foo(Int @a) { dd }; foo my Int @ = 1,2,3 # closest thing 20:13
camelia sub foo(Int @a)
tonyo there is the beautiful: foo(Array[Int].new(1,2,3)) 20:50
lizmat or foo Array[Int].new: 1,2,3 20:56
less lispy :-) 20:57
vrurg @[Int][1, 2, 3] 21:01
sena_kun can anyone please check `zef install Tika` on 2020.05? 21:03
lizmat checking 21:05
[Tika] Could not connect socket: Connection refused 21:06
21:06 morayj left
Geth rakudo: cd4a72b6a7 | (Elizabeth Mattijsen)++ | 6 files
Implement Foo.pick(**)

This will produce all values of Foo.pick(*), and then produce all values of Foo.pick(*) again, and again, ad infinitum.
This removes some unnecessary protos. And adds a Rakudo::Iterator.Reiterate method to produce values from a Callable that produces an iterator.
The Any.pick(HyperWhatever) candidate needs to be marked as "default" because of ambiguity with List.pick(Any).
21:06
tonyo live author tests shouldn't go in your t/
sena_kun if this is on 2020.05, then fine. I wonder just how on Earth blin sometimes make the modules to pass.
which means we have no blockers so far
lizmat maybe because sometimes the server it tries to reach, is reachable ?
sena_kun releasable6, status
releasable6 sena_kun, Next release in ≈2 days and ≈21 hours. 1 blocker. 0 out of 18 commits logged
sena_kun, Details: gist.github.com/bbcbafcd593ab21c5e...adff6abd7e
sena_kun blin ones
Geth roast: 9439f47653 | (Elizabeth Mattijsen)++ | S32-list/pick.t
Add tests for .pick(**)
21:15
lizmat enough core hacking for yours truly today& 21:18
sena_kun lizmat++
22:38 tyilanmenyn joined 22:39 tyil left 22:41 tyilanmenyn is now known as tyil 23:55 Util_ left