🦋 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.
SmokeMachine vrurg: thanks! 00:18
jdv i did a blin y'day and it was slim/clean 00:21
tellable6 2022-02-16T02:42:27Z #raku-dev <vrurg> jdv do you know if Blin runs against zef ecosystem modules? I just realized that Config::BINDish must be failing for as long back as since 2021.10. But it's never been reported.
jdv vrurg: i do not
MasterDuke: github.com/rakudo/rakudo/issues/4811 00:25
vrurg: it appears blin only looks at p6c. i'll look into "fixing that 00:32
"
no, that's wrong. it should be including all ecos. 00:33
vrurg jdv: Looks like it. I think I found the cause of my problem later. 00:34
With regard to blin, it looks like there're a couple of issues with the today's NQP bump. 00:35
jdv it likely ends up in "AlwaysFail" - just a count though tere
ok. i'll do another run. can't run it for every change. sometimes it takes many hours... 00:38
vrurg I'll see if I can do it on my side. 00:41
jdv started. best case ~2h. could be up to 9h ime so far. 00:43
vrurg Maybe next time I could do it myself. Last time I tried it behaved strange and I haven't tried anymore. 00:48
jdv cool. i had a few tweaks in my build cause some things were broken. 00:55
i'll try to get that ironed out soon
s/had/have/ 00:56
vrurg Ah, now I remember what was it wrong for me. Nearly every module is marked as "AlwaysFail". :( 02:04
jdv vrurg: github.com/rakudo/rakudo/issues/4813 03:01
nite
MasterDuke jdv: thanks 08:58
Geth nqp: 83d86b3f6e | (Daniel Green)++ | tools/templates/MOAR_REVISION
Bump MoarVM for more unsigned fixes
09:15
rakudo: 430d3dd423 | (Daniel Green)++ | tools/templates/NQP_REVISION
Bump NQP for more unsigned fixes
09:16
Geth rakudo: 6383014298 | (Elizabeth Mattijsen)++ | 2 files
Fix issue with native array.splice

Spotted in blin run
10:59
Geth rakudo: 6d04cc395b | (Elizabeth Mattijsen)++ | 3 files
Fix issue with Net::Curl
13:50
melezhik lizmat I am talking about average time for ado build testspec which is about 20-25 minutes now - dev.azure.com/Rakudo/rakudo/_build...324385256b 16:02
Files=1351, Tests=117115, 1336 wallclock secs (28.78 usr 5.63 sys + 2318.16 cusr 180.80 csys = 2533.37 CPU)
if one run test spec on their more powerful machine of course the time could be less, but for ado build it's about 20 minutes or more now 16:03
github.com/melezhik/fastspec aims to beat that to maybe 10 minutes a build maybe even less, I am working on this right now 16:04
basically the idea to run test spec in parallel  on sparky cluster - github.com/melezhik/fastspec/blob/...le#L24-L47 splitting tests by chunks and run every chunk in parrallel 16:05
sparky would use azure container instance with per seconds price  - github.com/melezhik/fastspec/blob/.../price.txt 16:06
sparky would run all required instance in cluster, run tests on them and them kill them - github.com/melezhik/fastspec/blob/...owfile#L15
github.com/melezhik/fastspec/blob/...owfile#L98 16:07
I am prototyping right now, I maybe will be able to show something in the next week or so
here is another link - azure.microsoft.com/en-us/pricing/...instances/ 16:08
I know that Ado uses linux VMs of 7GB with 2 cores 16:09
Geth rakudo: f1eccb396f | Coke++ | docs/windows.md
Note mimalloc issue in windows config

Also, remove unneeded specification of tools (config defaults things correctly)
16:34
MasterDuke [Coke]: that should be `--no-mimalloc` (note the '-') 16:36
[Coke] bah 16:57
Geth rakudo: 3bde138cbb | Coke++ | docs/windows.md
fix typo

  MasterDuke++
16:58
[Coke] Thanks!
MasterDuke np
[Tux] Rakudo v2022.02-98-g3bde138cb (v6.d) on MoarVM 2022.02-36-ga1a45e57e
csv-ip5xs0.830 - 0.846
csv-ip5xs-205.198 - 5.397
csv-parser3.464 - 3.665
csv-test-xs-200.413 - 0.414
test6.449 - 6.690
test-t1.477 - 1.528
test-t --race0.981 - 1.135
test-t-2021.112 - 21.554
test-t-20 --race6.919 - 7.361
18:37
lizmat m: my $a; my $b is default(42); ($a,$b) = "foo".split(":"); say $b 19:07
camelia 42
lizmat m: my $a; my $b is default(42); ($a,$b) = "foo".split(":"); say $b
camelia 42
lizmat m: my ($a, $b is default(42)) = "foo".split(":"); say $b
camelia (Any)
lizmat shouldn't those need to be equivalent ?