00:35 pamplemousse__ joined 00:47 Geth left 00:48 Geth joined 00:49 pamplemousse__ left
AlexDaniel sena_kun: fantastic promising news :) github.com/hythm7/Pakku/issues/7#i...-699717770 04:01
tellable6 AlexDaniel, I'll pass your message to sena_kun
04:43 linkable6 left, evalable6 left, tellable6 left 04:44 evalable6 joined 04:45 linkable6 joined, tellable6 joined 05:15 jmerelo joined
jmerelo Just a reminder that Hacktoberfest is coming, and we have created a single reference page to show all your issues that are hacktoberfest-friendly github.com/Raku/hacktoberfest 06:20
06:55 Altai-man joined 07:16 domidumont joined 07:37 maggotbrain left 08:01 Altai-man left 08:04 maggotbrain joined
Geth ¦ problem-solving: JJ assigned to jnthn Issue What is the philosophy of the Raku language github.com/Raku/problem-solving/issues/235 09:11
rakudo/rakuast: c4a758c3e0 | (Elizabeth Mattijsen)++ | 2 files
Add Deparsing support for while / until / repeat

and adapt / add associated tests
10:15
rakudo/rakuast: 5aa04d87d8 | (Elizabeth Mattijsen)++ | src/core.c/Any.pm6
Add RakuAST::Node support to dd

Basically, since RakuAST::Node.raku doesn't really show anything useful at the moment, and since it probably never will, it makes more sense to deparse the given AST and note that. So instead of:
   note $ast.DEPARSE;
... (6 more lines)
10:39
MasterDuke any objection to merging github.com/rakudo/rakudo/pull/3930 ? 11:15
11:20 leont joined
Geth rakudo: 7192a99e69 | (Daniel Green)++ | src/core.c/Real.pm6
Remove unnecessary branch in .truncate

This fixes the difference between `True.truncate` (which returns True) and `False.truncate` (which returned 0) so they both return self.
11:28
rakudo: e679775009 | MasterDuke17++ (committed using GitHub Web editor) | src/core.c/Real.pm6
Merge pull request #3930 from MasterDuke17/simplify_truncate

Remove unnecessary branch in .truncate
11:59 pamplemousse__ joined 12:09 sena_kun joined 12:46 jmerelo left 13:05 Xliff joined
Kaiepi m: my blob8:D $blob .= new: 0b10101010; $blob.read-ubits: -1, 0xDADFACE 13:14
camelia Can only read from position 0..7 in buffer '$blob', you tried: -1
in block <unit> at <tmp> line 1
Kaiepi would X::OutOfRange not fit as a typed exception for this?
13:20 domidumont left 13:23 domidumont joined
Xliff If no one has the time to look into this one: github.com/rakudo/rakudo/issues/3922 13:28
13:28 domidumont left
Xliff Could someone at least take a gander and give me pointers as to where I can start investigating it? 13:28
It's been 11 days since I submitted it, so.... 13:29
sena_kun Xliff, I know nothing about moar/rakudo, but I wonder if me suggesting grepping where error occurs and adding some prints there can shed some light is stupid? 13:31
tellable6 2020-09-28T04:01:17Z #raku-dev <AlexDaniel> sena_kun: fantastic promising news :) github.com/hythm7/Pakku/issues/7#i...-699717770
Xliff sena_kun: Absolutely not stupid.
So I guess that's where I will start.
sena_kun Xliff, if the error is _always_ reproducible, that's a basic measure I'd attempted to. If not a full solution, you'd have some model on hands, like "So internally this and that happens", that can be very helpful for devs as initial mundane steps are already done. 13:32
Xliff However if this dead-ends (as I suspect) with Stashes, then I am at my limit. 13:33
lizmat perhaps nine has an idea 13:36
notable6: weekly 13:37
notable6 lizmat, 4 notes: gist.github.com/9b4331182ba327f7d9...c9d0276db4
13:42 domidumont joined
Xliff lizmat: When is nine usually available? 13:59
nine: github.com/rakudo/rakudo/issues/3922 -- Your thoughts? Might this be a Stash-related issue?
lizmat nine is available when he's not terribly busy at $work 14:00
Xliff lizmat: Sounds familiar.
lizmat or flying a glider :-) 14:01
Geth rakudo: a26beb66e0 | (Elizabeth Mattijsen)++ | lib/BUILDPLAN.rakumod
Make BUILD(ALL)PLAN output a bit clearer

  - fix case when there's only a TWEAK method (was showing as BUILD)
  - mention the package where the TWEAK / BUILD method comes from
14:22
14:30 MasterDuke left
Geth rakudo/rakuast: a44fb7cc50 | (Elizabeth Mattijsen)++ | 17 files
Add standard vim footer
14:33
14:40 MasterDuke joined 14:56 Altai-man joined 14:58 sena_kun left 15:30 |Tux| joined
nine Waiting for repl.it... 15:42
Ah, there's also a zip file 15:43
16:00 pamplemousse__ left
Xliff nine: Thoughts? 16:06
16:16 samcv joined
lizmat notable6: weekly reset 16:29
notable6 lizmat, Moved existing notes to “weekly_2020-09-28T16:29:15Z”
16:42 domidumont left
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2020/09/28/2020-...-releaser/ 16:50
Altai-man lizmat++ 16:52
17:06 pamplemousse__ joined 17:47 toddr joined 18:57 sena_kun joined 18:59 Altai-man left
Geth nqp: usev6++ created pull request #674:
Improve streaming decoder on the JVM backend
19:20
20:11 pamplemousse__ left
AlexDaniel lizmat: “300000x as fast for prime numbers” actually 355.56x as fast according to the commit message 20:24
$ tools/speedup.raku 12.8 0.036
355.56x as fast
lizmat yup, you're right 20:25
fixed
gfldex It has become quite hard to stress a big CPU with .is-prime. 20:26
AlexDaniel gfldex: not sure what you mean 20:27
gfldex I had to start with fairly big numbers to get my threadripper to 1200% load. 20:28
Less then 100 places seams to be pretty memory bound now.
dogbert17 threadripper, impressive 20:32
Xliff gfldex: What specs were you running on? 20:37
gfldex AMD Ryzen Threadripper 1900X 8-Core Processor 20:38
.say for ((1 ~ (0..9).roll(100).join).Int .. ∞).hyper(:batch(1000)).grep(*.is-prime);
that stays at 4 cores for quite some time 20:39
Xliff gfldex: Have a test case I can run on an Intel(R) Core(TM) i9-7900X CPU @ 3.30GHz 20:40
10 cores with hyperthreading.
Ooh! Let me drop the GUI.
MasterDuke assuming the switch to gmp working on goes through, you'll probably have to increase the size again 20:41
20:42 Xliff left
sena_kun gets 800% after some seconds and soon it becomes 1500% 20:42
with 24 threads
Still, CPU is at just around 65%. 20:43
20:48 Xliff joined
Xliff \o -- Gone oldskool. 20:48
epic5 client.
Geth rakudo/rakuast: 581f492235 | (Elizabeth Mattijsen)++ | 2 files
Add Deparsing support for loop / for

and adapt / add associated tests
20:49
lizmat and that concludes my hacking for today& 20:50
Xliff Unfortunately, I can't easily paste the results. 20:51
gfldex: How long does this run on your machine? 20:53
OK. This was fun. 20:57
Oh, and I didn't suspect there were this number of primes this far up the numbering system. 20:58
gfldex++
20:58 Xliff left
timotimo primes get less dense the further up you go, right? 20:59
21:02 sena_kun left
MasterDuke but there are always more that are only two apart. primes are weird 21:09
21:15 Xliff joined 23:19 leont left 23:40 ggoebel left 23:45 ggoebel joined