|
08:56
sena_kun joined
|
|||
| Geth | roast: 83313e9e9d | (Elizabeth Mattijsen)++ | S12-subset/multi-dispatch.t Add test for #1801 |
09:17 | |
| rakudo/main: 50a68ec906 | (Elizabeth Mattijsen)++ | src/core.c/Main.rakumod Don't assume @*ARGS only contains strings Since it's a dynamic variable, it *can* be set to anything really inside a script. Make sure each element is a string before processing to prevent LTA errors. Fixes #1803 |
09:33 | ||
| roast: 4936ddc404 | (Elizabeth Mattijsen)++ | S06-other/main.t Add test for #1803 |
|||
| rakudo/main: 25640772f7 | (Elizabeth Mattijsen)++ | t/02-rakudo/18-pseudostash.t Add test for #1835 |
09:54 | ||
| rakudo/main: c05f170446 | (Elizabeth Mattijsen)++ | src/core.c/Rat.rakumod Rework Rat <=> Rat candidate First get all the info, then check if denominators are the same. If they are, then compare numerators without needing to multiply. Apart from an efficiency gain on Rats with the same denominator, it also fixes the case of <1/0> <=> <-1/0> which would previously return Same Fixes #1850 |
10:21 | ||
| roast: 6573e19039 | (Elizabeth Mattijsen)++ | S03-operators/spaceship.t Add test for #1850 |
10:28 | ||
| rakudo/main: 3c4cfc43b2 | (Elizabeth Mattijsen)++ | src/core.c/Rat.rakumod Rince repeat c05f170446 for < <= > >= As they suffered from the same issue. This breaks an in my view incorrect test that was based on the incorrect semantics of Rat < Rat in the case of denominator 0: 0/0..0/0 ~~ -1/0..1/0 was: False, is now True as the right hand side is now a "real" range as far as .ACCEPTS semantics is concerned. |
11:02 | ||
| roast: d0defaa049 | (Elizabeth Mattijsen)++ | 2 files Add/fix more tests related to #1850 |
11:03 | ||
| roast: 3706ce6011 | (Elizabeth Mattijsen)++ | S03-smartmatch/range-range.t Fix test in light of Rat < <= > >= Rat fix Specifically github.com/rakudo/rakudo/commit/3c4cfc43b2 |
11:04 | ||
| rakudo/main: d048d9f0b0 | (Elizabeth Mattijsen)++ | t/12-rakuast/xx-fixed-in-rakuast.rakutest Add test for #1898 |
11:57 | ||
|
12:16
finanalyst joined
|
|||
| Geth | rakudo/main: 31795cd2b5 | (Elizabeth Mattijsen)++ | t/12-rakuast/xx-fixed-in-rakuast.rakutest Add test for #1947 |
12:26 | |
|
12:29
finanalyst left
12:36
finanalyst joined
|
|||
| Geth | rakudo/main: f725f5f4bd | (Elizabeth Mattijsen)++ | src/core.c/Hash.rakumod Make { } / hash() (aka empty hash creation) 8x as fast As suggested by Juerd++ in github.com/rakudo/rakudo/issues/1951 |
12:50 | |
| lizmat | 100 issues closed since I started working through them in reverse order :-) | 12:54 | |
| tbrowder | very good! | 13:28 | |
| timo | impressive | ||
| tbrowder | now, how can i as a user completely shut off precompilation? setting the env var RAKUDO_NO_PRECOMPILATION seems not to work | 13:34 | |
| i manually blow away the .precomps but they come back. i have set it in my shell before running and the .precomp comes back | 13:37 | ||
| using ver 2024.09 on debian | 13:38 | ||
| lizmat | ugexe? | ||
| timo | this is with using -I. everywhere, right? | 13:41 | |
| tbrowder | i've filed an issue. | 13:51 | |
| um, i have set the shell env, but the module does use precompiled modules from other | 13:53 | ||
| sources | |||
| and i have used : raku -I. ./prog to no avail | 13:57 | ||
| timo | the "steps to reproduce" are a little vague still, but i'll have a look what's up | 13:59 | |
| tbrowder | i can't find the thread but someone earlier suggested setting one of the phasers to not run somehow. may have been @ugexe ... | ||
| Geth | rakudo/main: 064575a4cc | (Elizabeth Mattijsen)++ | src/core.c/IO/Notification.rakumod Fix handling of IO::Paths in IO::Notification.watch-path The problem is really that IO::Path stringifies to the original string specifification, instead of to an absolute path. Since the method was using Str() coercion, it's handling re IO::Path could be incorrect wrt $*CWD. Fixes #1971 and #1972 |
14:02 | |
|
14:05
dogbert17 left
|
|||
| Geth | roast: c0df819248 | (Elizabeth Mattijsen)++ | S17-channel/basic.t Add test for #1974 |
14:08 | |
| tbrowder | whoa, i may be lying...different behavior now...?? | 14:14 | |
| timo | yeah, without RAKUDO_NO_PRECOMPILATION i see a .precomp folder being created, and with RAKUDO_NO_PRECOMPILATION it doesn't sem to happen | 14:20 | |
| ugexe | $ zef --help 2>&1 | grep precompile | 14:21 | |
| --/precompile-install Skip precompiling during installation | 14:22 | ||
| Geth | rakudo/main: 0ebeee030e | (Elizabeth Mattijsen)++ | t/12-rakuast/xx-fixed-in-rakuast.rakutest Add tests for #1986 |
14:23 | |
| rakudo/main: b94ad1d7b3 | (Elizabeth Mattijsen)++ | t/04-nativecall/18-routine-sig-sanity.t Add test for #1993 |
14:31 | ||
|
15:05
lizmat_ joined
15:06
rba_ joined
15:07
greppable6 joined
15:09
tellable6 joined
15:10
kjp_ joined
15:15
lizmat left,
kjp left,
rba left,
tellable6__ left,
greppable6__ left,
rba_ is now known as rba
17:51
lizmat_ left,
lizmat joined
|
|||
| Geth | rakudo/main: 67a426bbf1 | (Elizabeth Mattijsen)++ | t/12-rakuast/xx-fixed-in-rakuast.rakutest Add test for #2034 |
18:12 | |
| roast: b190eaaf86 | (Elizabeth Mattijsen)++ | S04-statements/sink.t Add test for #2069 |
18:30 | ||
|
21:38
finanalyst left
21:50
lizmat_ joined,
lizmat_ left
22:59
sena_kun left
|
|||