00:19 japhb left 00:25 japhb joined 05:48 kjp left 05:49 kjp joined 05:51 kjp left, kjp joined 07:10 ilogger2_ left 07:11 ilogger2 joined 10:06 librasteve_ left 10:30 librasteve_ joined 10:42 sjn left, Nemokosch joined
Nemokosch bisectable: dd [>>max<<] {a => 1, b => 2}, {b => 3, c => 4}, {c => 5, d => 6} 10:43
bisectable6 Nemokosch, Will bisect the whole range automagically because no endpoints were provided, hang tight
10:43 sjn joined
bisectable6 Nemokosch, Output on all releases: gist.github.com/6ea0436affbd063ad7...a842ee299d 10:43
Nemokosch, Bisecting by output (old=2025.01 new=2025.02) because on both starting points the exit code is 1 10:44
Nemokosch, bisect log: gist.github.com/332b493c8a0ac9f385...51bee7be38 10:47
Nemokosch, Output on all releases and bisected commits: gist.github.com/c61ede64e2053a9a73...8f48cfe8af
11:15 Nemokosch left
lizmat hmmm do we have a reason why .CREATE is actually calling the Mu.CREATE method, as opposed to just codegenning an nqp::create() op ? 12:22
just like .WHAT will codegen an nqp::what op 12:23
aha... nqp::create doesn't work well on contained values 13:13
making it call the CREATE method, makes it work on the decont
[Tux] Rakudo v2026.01-14-g33f68d2cd (v6.d) on MoarVM 2026.01-4-g0742ba286
csv-ip5xs0.257 - 0.262
csv-ip5xs-201.035 - 1.069
csv-parser1.073 - 1.080
csv-test-xs-200.118 - 0.119
test1.797 - 1.801
test-t0.436 - 0.438
test-t --race0.287 - 0.301
test-t-205.623 - 5.711
test-t-20 --race1.373 - 1.388
14:44
csv-test-xs 0.017 - 0.017
tux.nl/Talks/CSV6/speed4-20.html / tux.nl/Talks/CSV6/speed4.html tux.nl/Talks/CSV6/speed.log
[Coke] .seen finanalyst 15:54
tellable6 [Coke], I saw finanalyst 2025-10-11T17:45:45Z in #raku-dev: <finanalyst> oops sorry, wrong window
[Coke] m: dd my $s=(1...5); 17:38
camelia $s = $((1, 2, 3, 4, 5).Seq)
[Coke] m: dd my $s=1...5;
camelia (1, 2, 3, 4, 5).Seq
nemokosch ... apparently has lower precedence than = (the scalar version of it, at least) 17:39
[Coke] (trying to figure out why the parens are used in the doc example on Seq.
lizmat because in the () case, you're assigning an item ? 17:44
nemokosch the important difference in the output is the lack of $s = in the second case - $s is in fact equal to 1, and the sequence construction only happens afterwards 17:45
[Coke] ok, my underlying task here was to figure out where "method skip" was defined on a Seq. that's not fun.
lizmat probably on Any in Anyiterable-methods 17:46
[Coke] Yup, on Any. ... that seems... unreasonably high up the chain.
lizmat m: dd 42.skip 17:47
camelia ().Seq
[Coke] (and the docs have it on Seq itself)
lizmat m: dd (42,666).skip
camelia (666,).Seq
lizmat thing is any item can be considered a one element list
[Coke] OK. moving 'method skip' in the docs (and making a lack of finding anything at all a failure for the check-signatures test)
lizmat therefore it lives that high up
nemokosch "any item can be considered a one element list" is the kind of Perlism that could have a design issue 17:48
it's tricky, too, because some of the behavior is obtained with .list while some other with .iterator 17:49
18:26 librasteve_ left 18:29 notna joined 18:55 librasteve_ joined 19:35 notna left 23:06 librasteve_ left