Zoffix New TPF Grant Proposal has been posted: "Perl 6: Bugfixing and Performance of Rationals Fixing Constraints on Constants": news.perlfoundation.org/2018/04/gra...fixin.html Please leave your comments 00:39
k, modules.perl6.org is now fully back in business, with rebuilt DB, switched to master and re-enabled cron 00:41
AlexDaniel Zoffix++ # everything 00:43
squashable6: next 02:26
squashable6 AlexDaniel, ⚠🍕 Next SQUASHathon in 3 days and ≈7 hours (2018-04-07 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
Zoffix w00t... Zoffix 32.0 has been released \o/ 04:01
ZofBot: grab a piece of upgrade cake, bruh
ZofBot Zoffix, github
AlexDaniel 🍰 \o/ 04:22
Zoffix: congratz! I hope lots of bug fixes and no regressions ;)
Geth nqp: 0edf50db6a | (Elizabeth Mattijsen)++ | tools/build/MOAR_REVISION
Get samcv++ latest strand improvements
 04:41
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...5-g156a7a6
samcv :) yay 04:42
Kaiepi i have to say the optimization work you guys have been doing is really impressive 04:52
samcv Kaiepi: thank you! 04:53
Geth rakudo: 19626e6cb3 | (Elizabeth Mattijsen)++ | tools/build/NQP_REVISION
Bump NQP to get latest strand improvements, samcv++
 04:58
rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....9-g0edf50d
de68bfc109 | (Elizabeth Mattijsen)++ | src/core/List.pm6

  - use nqp ops directly to get random value
  - use lexicals instead of attributes where this makes sense
Zoffix samcv++ 05:06
rakudo: c53fde3eb8 | (Elizabeth Mattijsen)++ | src/core/List.pm6
Make List.roll about 20x faster

  - use nqp ops to create random values
  - use native ints instead of full blown Ints
For some reason, Num.floor was taking up more than half of the CPU. This is now bypassed completely. Using native ints makes up for the rest of the improvement. --profile can be very informative sometimes.		 06:41
[Tux] Rakudo version 2018.03-152-gde68bfc10 - MoarVM version 2018.03-76-g21d6c5470
csv-ip5xs0.897 - 0.911
csv-ip5xs-208.672 - 9.206
csv-parser37.105 - 38.598
csv-test-xs-200.456 - 0.473
test9.439 - 9.620
test-t2.430 - 2.565
test-t --race1.033 - 1.039
test-t-2043.437 - 43.857
test-t-20 --race15.151 - 15.249
 06:51
timotimo holy wow, the List.roll thing 10:06
Geth rakudo: 6a6bb23dc0 | (Elizabeth Mattijsen)++ | 2 files
Make Map/TypedHash.roll about 30% faster

  - get the random number in NQP
  - avoid HLL .rand.floor		 14:11
rakudo: 5981f81b63 | (Elizabeth Mattijsen)++ | src/core/Hash.pm6
Oops, remove trailing garbage, Masterduke++ for spotting

Weird thing is, settings compiled without issue and spectest ran fine as well :-(		 14:24
[TuxCM] samcv, does p6 support (or plan tyo support) iso-6937/2 ? en.wikipedia.org/wiki/ISO/IEC_6937 18:24
samcv i had no plans but i can add support if someone wants it 18:25
[TuxCM] <jidding>I dare you :)</kidding> 18:26
samcv so you don't actually want it hah?
[TuxCM] (I'd use it in tooling)
it currently still is *the* encoding used in inter-local-community government communications 18:27
in the Netherlands
samcv ah ok
[TuxCM] the "problem" is that the diacritical indicators come in front of the base character instead of after the base character 18:28
each supported character is one or two bytes
samcv [TuxCM]: what do you use currently for converting to/from this format? 18:29
hmm this encoding doesn't seem too complex. at least from the wikipedia page. 18:35
FROGGS
.oO( digitus )
 18:37
AlexDaniel releasable6: uptime 19:00
releasable6 AlexDaniel, 1 day, 4 hours, 35 minutes, and 12 seconds, 1556.488281MiB maxrss. This is Rakudo version 2018.03-134-g20495f097 built on MoarVM version 2018.03-56-g85fc758ce implementing Perl 6.c.
AlexDaniel reportable6: uptime 19:01
reportable6 AlexDaniel, 4 days, 1 hour, 11 minutes, and 23 seconds, 1307.859375MiB maxrss. This is Rakudo version 2018.03-134-g20495f097 built on MoarVM version 2018.03-56-g85fc758ce implementing Perl 6.c.
AlexDaniel hmmmm…
only three bots leaked… but that's something new 19:02
pmurias .msg jnthn a big problem with the current parameter interning is that paramaterized types are registered when they are deserialized, and objects are deserialized lazily 21:29
yoleaux pmurias: Sorry, this command is admin-only.
pmurias .tell jnthn a big problem with the current parameter interning is that paramaterized types are registered when they are deserialized, and objects are deserialized lazily 21:30
yoleaux pmurias: I'll pass your message to jnthn.
samcv [Tux]: can you provide me we samples of iso-6937/2 21:33
[Tux] sure 22:02
tux.nl/Files/iso6737.tgz : diac.tst is all allowed characters in 6937, diac.utf is the same in utf-8 22:04
samv, I currently use perl5 and C (both use state machine and translation tables) 22:05
in C it is a huge nested case statement 22:06
Zoffix Any way to get camelia up to date with JVM? What's blocking it building latest and greatest, like it does with moar? 22:50
yoleaux 13:32Z <jdv79> Zoffix: oh yeah. forgot about that. that's about as much interest metacp6an got.
timotimo someone got a windows and could run a very short snippet of perl6 code for me? 22:54
Zoffix timotimo: yeah I can
timotimo cool, try this please: 22:55
first set $path to a random file on your hard drive 22:56
say run("CertUtil.exe", "-hashfile", $path, "SHA256", :out).out.slurp.subst(" ", "");
probably want to :close that slurp, too
Zoffix $path is what? PATH?
timotimo Do not use it for new code. Use .slurp method method instead. 22:57
no, a random file
just to see if the output is as expected
Zoffix timotimo: gist.github.com/zoffixznet/3d4fcc3...e05b0d8083 23:00
timotimo ok so the output is those three lines huh? 23:05
oh, no, the other two line are stderr?
i'm hoping we can use this for GTK::Simple 23:06
Zoffix timotimo: no, that's all on STDOUT. More precisely, there are also two blank lines. Here's a run with :!err: gist.github.com/zoffixznet/3214b04...ce824f7b74 23:07
timotimo shouldn't the subst have removed all the spaces in there? 23:08
it would need :g?
Zoffix It'd need :g for that yeah.
timotimo i was under the impression that the hash would be output as two hex digits at a time with spaces in between
Zoffix timotimo: that was a run on Win10. Here's a run on Win7, that indeed has pairs of hex digits: gist.github.com/zoffixznet/90bb0ad...f4afc932b7 23:09
timotimo m: say "7c29dfc0784ddfa8e5af1fd917581ee2acd954c40136835a64c8edf7d1f96f9f".chars 23:10
camelia 64
timotimo say run("CertUtil.exe", "-hashfile", $path, "SHA256", :out).out.slurp(:close).subst(" ", "", :g).lines.grep(/^[A..F a..f 0..9] ** 64$/).head()
that should do it, right?
Zoffix timotimo: here're runs with `:g` gist.github.com/zoffixznet/9debbca...f6e3fc2232
timotimo: that gives warnings from the regex. But using <:hexdigit> does work: gist.github.com/zoffixznet/3eb1483...f1bc46eeca 23:12
timotimo i forgot to put < > around the [] 23:13
man i'm tired
Zoffix Also can just do `.first` instead of `.grep.head`
timotimo i wasn't sure if .first also has a mode without a matcher 23:15
Zoffix It doesn't but I meant you could use .first(/..../) instead of .grep(/..../).head 23:16
err
timotimo oh, of course
Zoffix It DOES
timotimo well, that's even better
m: say [].first
Zoffix :)
camelia Nil
Zoffix m: say <a b c>.first
camelia a
timotimo thanks for the assist
m: say "foo" ~~ / <hexdigit> / 23:17
camelia No such method 'hexdigit' for invocant of type 'Match'. Did you mean 'xdigit'?
in block <unit> at <tmp> line 1
timotimo m: say "foo" ~~ / <xdigit> /
camelia ｢f｣
xdigit => ｢f｣
timotimo is xdigit the same as :hexdigit?
Zoffix no idea
timotimo anyway, i pushed a gtk-simple 23:19
one day i'll be proficient enough to just write working perl 6 code even when i'm this tired :P 23:23
Zoffix "Unhandled exception: failed to load library 'dynext/perl6_ops_moar.dll'" 23:24
timotimo um
how did you get that to happen? 23:25
Zoffix Full output: gist.github.com/zoffixznet/4780654...6645f9fcc2
something to do with R#1662 maybe
synopsebot R#1662 [open]: github.com/rakudo/rakudo/issues/1662 [⚠ blocker ⚠] Windows Issues with "Replace error-prone file timestamp check by checksum when loading deps"
Zoffix releasable6: next
releasable6 Zoffix, Next release in ≈17 days and ≈19 hours. 4 blockers. 0 out of 155 commits logged (⚠ 9 warnings)
Zoffix, Details: gist.github.com/23e55c8f221407680f...394894d77c 23:26
timotimo it's surprising that you could even run zef if it fails to find the dynext file 23:31
Zoffix .tell bartolin FWIW, this commit used 309 as scale (citing perf loss concerns), but I imagine 17 is good enough, as doubles have at most 18 significant figures. Looking at openjdk's impl of BigInteger.divide, there's even a fast path when scale is < 18. Perhaps something to investigate, to improve perf? github.com/perl6/nqp/commit/b9f195...6650695165 23:32
yoleaux Zoffix: I'll pass your message to bartolin.
Zoffix .tell bartolin err, I meant BigDecimal.divide's impl 23:33
yoleaux Zoffix: I'll pass your message to bartolin.
timotimo Zoffix: btw, isprime takes as second argument the number of miller-rabin tests to do; 100 is extremely wasteful. maybe if people keep using is-prime as a benchmark we should reduce that a little 23:37
just pinging you as a random bystander m) 23:38
timotimo goes to bed 23:41
MasterDuke Zoffix: happy birthday! (it was my daughter's first birthday today, so easy to remember)
Zoffix Hah. Cool. Thanks :) 23:56

Powered by Cro and the Raku® Programming Language.

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

Thank you!