00:15 patrickbkr[m] left, patrickbkr[m] joined, AlexDaniel` left 00:20 AlexDaniel` joined 00:34 MasterDuke left 00:50 dogbert17 joined 00:54 dogbert11 left 01:28 dogbert17 left 01:29 dogbert17 joined 01:47 JRaspass left 01:48 AlexDaniel` left, unclechu left 02:18 AlexDaniel` joined 02:22 unclechu joined 02:32 kvw_5 joined 02:35 kvw_5_ left 02:45 leont left 06:05 Geth left, Geth joined 07:14 domidumont joined 08:14 sena_kun left 08:35 squashable6 left, squashable6 joined 08:37 patrickb joined 09:39 domidumont left 09:58 JRaspass joined
Geth ¦ problem-solving: patrickbkr assigned to codesections Issue The PS-process' acceptance policy doesn't match reality github.com/Raku/problem-solving/issues/272 10:34
10:55 sena_kun joined 11:45 [Tux] left
Geth rakudo/rakuast: 85484c0ea8 | (Elizabeth Mattijsen)++ | 2 files
Follow b2gills++ suggestion about deparsing

See github.com/rakudo/rakudo/commit/06...t-47855336
11:49 [Tux] joined 12:12 leont joined
Geth rakudo/enums-as-Map: 615030d279 | (Elizabeth Mattijsen)++ | 2 files
Same effect by dispatch

As pointed out by vrurg++ TIL I learned about the Enumeration role.
12:25 squashable6 left, squashable6 joined 12:51 domidumont joined 13:08 lizmat_ joined 13:10 lizmat left 13:28 JRaspass left 13:45 uzl[m] joined 14:45 leont left 14:46 leont joined 15:08 frost-lab left 15:34 JRaspass joined 15:35 lizmat_ is now known as lizmat 15:36 maggotbrain left 15:47 maggotbrain joined, maggotbrain left 16:14 JRaspass left
Geth rakudo: affdd05831 | (Elizabeth Mattijsen)++ | src/core.c/precedence.pm6
Set up .prec info for &[<=>]

It also appeared to not exist
rakudo/rakuast: ef4fb699fe | (Jonathan Worthington)++ | 5 files
Basic RakuAST implementation of `for` modifier

The `for` statement AST node is refactored a little to allow for some code sharing (which will be helpful when we start producing the various optimized forms). It always assumes sink context for now, which isn't quite correct, and will need tweaking. Otherwise, it works.
rakudo/rakuast: 5fbe66bf74 | (Elizabeth Mattijsen)++ | 2 files
Revert "Follow b2gills++ suggestion about deparsing"

This reverts commit 85484c0ea8c1b22b777c886bef80aec2e4abec58.
This breaks simple cases such as %h<s> :-( I guess the deparser needs more context.
rakudo/rakuast: 8538709dbf | (Elizabeth Mattijsen)++ | 2 files
Add StatementModifier::For deparsing and test
18:23 domidumont left 18:57 MasterDuke joined
MasterDuke i almost have a companion PR to github.com/Raku/nqp/pull/703 for Rakudo, but it currently dies during `make install` with `Type check failed in binding to parameter '<anon>'; expected Any but got BOOTStrArray (BOOTStrArray) in block <unit> at /home/dan/Source/perl6/rakudo/tools/build/clean-target-files.raku line 1` 19:16
but assuming whatever the fix is doesn't slow the optimization down any, it takes `my @a = ^1_000_000; my $b; my $s := now; $b := @a.shift while @a; say now - $s; say $b` from ~0.9s down to ~0.15s 19:17
lizmat the problem in Raku is that @a can be anything adhering to the Positional role 19:27
so there may well be nothing to do nqp::elems on
MasterDuke yeah, i have it call &prefix:<+> instead 19:30
i get the same error if i have it call .elems also, though that's slightly slower than &prefix:<+> 19:31
lizmat and calling .Bool on it? or is that the current situation? 19:37
and didn't we just find an issue with calling .Bool on things ?
MasterDuke i'm not calling .Bool 19:41
but fwiw, changing to calling .Bool gives same error
lizmat meh 19:48
MasterDuke: I think there's something screwy going on with github.com/rakudo/rakudo/blob/mast...ng.pm6#L55 20:15
I guess the .Bool comes from the || 20:16
will look at this later tonight or tomorrow 20:17
afk for a bit&
vrurg @a could as well be bound to a Proxy. 20:38
20:53 patrickb left 20:57 patrickb joined 21:04 MasterDuke left 21:08 MasterDuke joined
MasterDuke i think the problem line is `unit sub MAIN(Str $folder, *@files where @files.so, Bool :$v);` 21:09
lizmat: that comment is 4 years old, is it still true? 21:12
21:13 patrickb left 21:31 MasterDuke left 21:35 MasterDuke joined
MasterDuke interesting, turns out it was just the `if` optimization that was the problem, if i only do it for `while` it's fine 21:55
yep, tests pass 21:59
22:12 JRaspass joined 22:40 dogbert11 joined 22:41 dogbert11 left 22:42 dogbert11 joined, dogbert17 left 22:45 dogbert11 left 22:46 dogbert11 joined 22:55 JRaspass left 23:31 vrurg left 23:53 vrurg joined