ugexe i'm curious how long it takes other languages to compile one million conditionals 00:41
03:03 MasterDuke joined
MasterDuke some random thoughts about about sorted_keys. here's a graph of it's running time imgur.com/uVOuWic 03:12
the shape seems correct, but the actual run times might need some work 03:13
my test was copying out `sorted_keys()` from the nqp source and then running it against a hash composed of keys that were `'lowered_lex_' ~ $i`, where $i was 0..n 03:18
it spent 33% of it's time in MVM_string_compare. my test got quite a bit faster if i changed the keys to `nqp::indexingoptimized('lowered_lex_' ~ $i)` 03:20
it got faster if i changed the keys to `'ll_' ~ $i`. some of that is probably because there's less string to compare against, but also now almost all of them were in-situ-8 strings, instead of just "regular" 8-bit strings 03:29
maybe `sorted_keys()` should be implemented as a merge sort instead of a heap sort, since that should reduce the number of comparisons performed 03:36
03:40 kjp left 03:41 kjp joined
MasterDuke it looks like MVM_string_compare could be sped up by adding specialized versions of the final comparison (if/when chars are found to differ at some index less than the scan length) for each string type 03:44
from ~470ms to ~370ms for `'lowered_lex_' ~ $i` and from ~400ms to ~330ms for `'ll_' ~ $i` (all using `nqp::indexingoptimized`) 03:46
ugexe nice 04:21
Geth rakudo/azure_improvements_squashed: c0819d38f7 | (Patrick Böker)++ | azure-pipelines.yml
Allow re-runs by renaming artifacts on each retry
 08:14
rakudo/azure_improvements_squashed: 6d7ea08904 | (Patrick Böker)++ | azure-pipelines.yml
different var accces
 08:31
08:48 sena_kun joined
Geth rakudo/azure_improvements_squashed: e7928e839c | (Timo Paulssen)++ (committed by Patrick Böker) | 2 files
Reimagine Azure CI

  - Prettier output (using the JUnit test result formatter)
  - Restructure making better use of parameters
  - Disable OOMing Win_JMV
  - Speed up CI on Windows by working around the compiler search
  - More granular output using `echo` commands
  - add a nodelay & blocking spec test run
  - persist build artifacts
  - Speed up testing by running tests in parallel
  - Bump the OS versions to latest as of now		 09:00
rakudo/main: 4 commits pushed by (Patrick Böker)++, (Timo Paulssen)++
patrickb timo: It is done! ^
timo hooray!
ab5tract patrickb++ ! 09:18
09:57 sena_kun left 10:00 sena_kun joined 10:06 sena_kun left 10:49 gfldex left
Geth PrettyDump/pd-fix-and-improvement: 447beb12a1 | (Patrick Böker)++ | lib/PrettyDump.rakumod
Fix `pd` to print the results out, as it should
 10:49
PrettyDump/pd-fix-and-improvement: 3010484511 | (Patrick Böker)++ | lib/PrettyDump.rakumod
Make `pd` return it's input

This allows putting pd in the middle of a pipeline.
PrettyDump: patrickbkr++ created pull request #2:
Pd fix and improvement		 10:56
11:25 librasteve_ left
lizmat notable6: weekly 12:07
notable6 lizmat, 2 notes: 2024-11-28T19:32:04Z <antononcube>: youtu.be/S_3e7liz4KM ; 2024-12-01T18:10:32Z <antononcube>: rakuforprediction.wordpress.com/20...aku-set-1/
lizmat notable6: weekly reset
notable6 lizmat, Moved existing notes to “weekly_2024-12-02T12:07:43Z”
Geth roast: 59ebbc1666 | (Christian Bartolomäus)++ | S32-hash/adverbs.t
[JVM] Fudge failing tests for X::Adverb

As described in github.com/rakudo/rakudo/issues/5709 these tests don't work anymore, because the wrong multi is selected.		 12:12
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2024/12/02/2024-49-advention/ 12:48
13:15 gfldex joined
timo lizmat: i think the CI part should read that patrickb finished the work i started on re-imagining the CI stuff 13:51
and the big integer issue was in debug helper code that is meant to be run from gdb or lldb, not in the moar remote debugger :)
lizmat so fixed, thanks for the headsup! 13:53
|Tux| Rakudo v2024.10-65-gdd45859fe (v6.d) on MoarVM 2024.10-40-gac58731d4
csv-ip5xs0.261 - 0.262
csv-ip5xs-201.107 - 1.109
csv-parser1.485 - 1.488
csv-test-xs-200.143 - 0.144
test1.860 - 1.863
test-t0.398 - 0.400
test-t --race0.274 - 0.275
test-t-204.728 - 4.736
test-t-20 --race1.232 - 1.244
 14:48
tux.nl/Talks/CSV6/speed4-20.html / tux.nl/Talks/CSV6/speed4.html tux.nl/Talks/CSV6/speed.log
14:51 MasterDuke left
Geth rakudo: usev6++ created pull request #5710:
[JVM] Add extra check for GLOBAL being null		 14:58
URI-Encode/main: 664569aaf9 | (Elizabeth Mattijsen)++ | 15 files
CI test for update as Raku community module
 15:18
URI-Encode/main: c37a137667 | (Elizabeth Mattijsen)++ | 2 files
0.1
 15:20
Test-When/main: 3cb9cde0bc | (Elizabeth Mattijsen)++ | 18 files
CI test for update as Raku community module
 15:53
rakudo: usev6++ created pull request #5711:
Move test file that dies on JVM to t/09-moar		 15:58
Test-When/main: 86c4267ffb | (Elizabeth Mattijsen)++ | 2 files
1.1
 15:59
Test-Stream/main: b80d50935b | (Elizabeth Mattijsen)++ | 60 files
CI test for update as Raku community module
 16:52
rakudo/lower-bound-for-revision-gating: 0a03eab1af | ab5tract++ | t/02-rakudo/12-multi-revision-gated.t
Add test names, remove unnecessary :err matcher
 16:56
Test-Stream/main: 0b3a5b26e4 | (Elizabeth Mattijsen)++ | 4 files
CI tweaks
 17:01
rakudo: usev6++ created pull request #5712:
Fix type when binding attribute for RealPolymod		 17:13
Test-Stream/main: 12bc5af330 | (Elizabeth Mattijsen)++ | 5 files
0.1
 17:30
roast: f87c8ae15b | (Christian Bartolomäus)++ | S10-packages/precompilation.t
[JVM] Unfudge now passing test

I'm not sure when this started to work. Probably at some point during the last 14 months.		 17:52
Test-Notice/main: fd8d825322 | (Elizabeth Mattijsen)++ | 14 files
CI test for update as Raku community module
 18:15
Test-Notice/main: 44314408fe | (Elizabeth Mattijsen)++ | 2 files
1.1
 18:17
[Coke] bartolin++ 18:20
Geth Terminal-Width/main: 55d3bb5cbe | (Elizabeth Mattijsen)++ | 14 files
CI test for update as Raku community module
 18:31
Terminal-Width/main: f0f41f7b79 | (Elizabeth Mattijsen)++ | 3 files
CI fix

Just try loading the module in CI environment, as there probably isn't a TTY available here anyway		 18:38
Terminal-Width/main: 75fdd303a1 | (Elizabeth Mattijsen)++ | 2 files
1.1
 18:40
18:55 sena_kun joined 19:22 librasteve_ joined 20:28 sjn left 20:49 sena_kun left
Geth rakudo/adverb-the-map-vairants: 92b05ea7ed | ab5tract++ | 2 files
Provide adverbial forms of map

This commit opens a different way of specifying map variants. This is mostly for cosmetic reasons and operates under the impression that named argument adverbs, had they been available, would have been used over the smashedcase versions we currently offer.
  * [flatmap](github.com/Raku/problem-solving/issues/430)		 22:38
rakudo: ab5tract++ created pull request #5713:
Provide adverbial forms of map		 22:39
ab5tract anybody else ever wish we could specify adverbs a la `$x.method:adverb: |@args` ? 23:00
23:27 donaldh_ joined 23:28 donaldh left
timo interesting 23:51
ab5tract only my makers can explain why I prefer `.map:flat:` over `.flatmap:` as much as I do ... and I'm pretty sure my poor old parents would just look at me in utter confusion if I asked 23:56
timo map:<flat> :) 23:57
ab5tract (though if I were really trying to rock my druthers, I'd ask Santa for `.map:flat &block`, meaning that adding an adverb would alread trigger "colon mode"
timo m: role Test { method map:<flat>(|c) { say "flat"; self.flatmap(|c) }; method map:<duck>(|c) { say "duck"; self.duckmap(|c) } }; my $in = [1, [2, "a", 3, "b"], 4, "x"] but Test; $in.map:<flat>({ say "^$_^" }); $in.map:<duck>({ say "quack $_" }); 23:59
camelia flat
^1^
^2 a 3 b^
^4^
^x^
duck
quack 1
quack 2 a 3 b
quack 4
quack x
ab5tract timo: intriguing! unfortunately it doesn't seem to dispatch as expected
oh .... wow
I have never seen a method named that way!

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!