🦋 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: ... | Logs can be inspected at colabti.org/irclogger/irclogger_log/raku-dev | For MoarVM see #moarvm
Set by lizmat on 26 April 2021.
japhb CBOR::Simple 0.0.7 is out and uploaded to zef/fez. It's mostly a general push towards better performance across the board. 06:29
At HEAD, however, is the first work on packed numarray support, which is 2+ ORDERS OF MAGNITUDE faster than unpacked Num arrays. 06:30
moon-child wow 06:31
japhb Processors REALLY like contiguous blocks of small-power-of-two-sized elements that can be memory managed as one big block. Turns out that MoarVM does too. :-) 06:33
I should note that *unpacked* Num arrays are already faster in CBOR::Simple than in JSON::Fast, so that big difference isn't because the unpacked case is particularly slow. ;-) 06:35
Also, CBOR encoding is smaller than JSON for every case I've tried. Unfortunately, the work that CBOR does to compress integers costs it a bit -- *unpacked* Int arrays are the one place that CBOR::Simple is about 2-3x slower than JSON::Fast, despite producing ~40% smaller encoded output. 06:38
Thus I'm looking forward to getting packed intarray support in. :-)
nwc10 IIRC the packed integer format used for MoarVM bytecode was measurable on a profile of loading 06:39
japhb Yeah, not too surprised to hear that.
sena_kun releasable6, status 11:08
releasable6 sena_kun, Next release in ≈18 days and ≈7 hours. There are no known blockers. 0 out of 14 commits logged
sena_kun, Details: gist.github.com/d1966fcdc6ae6792cd...f70feb2660
Xliff o/ 11:09
tellable6 2021-04-30T20:53:54Z #raku <MasterDuke> Xliff what about `for ^5 { once say "first"; .say }`? i.e., `once` instead of `FIRST`
nwc10 \o 11:10
Xliff MasterDuke: Aha! That's what I was looking for. Thanks!!
R#4256 11:14
R4356
Xliff Regarding github.com/rakudo/rakudo/issues/4356 11:15
I'm here for another hour or so if someone wants me to run a piece of code in the environment in question so they don't have to prep it themselves.
lizmat notable6: weekly 12:36
notable6 lizmat, 1 note: 2021-05-02T17:03:11Z <El_Che>: rakudo-pkg added packages for the newly released fedora 34
lizmat notable6: weekly reset 12:38
notable6 lizmat, Moved existing notes to “weekly_2021-05-03T12:38:11Z”
tbrowder lizmat: module for class DateTime::Julian is looking good. your trick of self.DateTime::new() works like a charm!! 13:19
tellable6 2021-05-02T15:54:05Z #moarvm <lizmat> tbrowder putting it in a DateTime::Julian module would be a good start for code evaluation
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2021/05/03/2021-...ble-comma/ 14:23
MasterDuke lizmat: first link (comma announcement is incorrect)
lizmat wot?
MasterDuke goes to nxadm.github.io/rakudo-pkg/ instead 14:24
lizmat argh.... wordpress!
MasterDuke i had a misplaced closing paren in my first sentence
lizmat MasterDuke++ # fixed 14:25
MasterDuke i'm not subscribed to the mailing list but maybe someone who is could chime in to this thread www.nntp.perl.org/group/perl.perl6...g9878.html  and mention that i believe the problem is in libtommath's mp_get_double and the gmp branch fixes it? 14:37
sena_kun lizmat++ 14:59
lizmat MasterDuke: so done 16:18
nwc10 vrurg: are you the author of commit 91ce9a29 - Add a substitute script for `ln -nfs` Currently it is for AIX only where there is no flag with the same meaning as `-n`/`-h` in GNU and BSD implementations. 18:26
in that I'm curious about AIX - you're using Rakudo on JVM on AIX? Or MoarVM?
MasterDuke lizmat++ 18:36
vrurg nwc10: It wasn't for me. There is a closed issue. 18:38
nwc10 aha, ItchyPlant 18:41
patrickb o/ 19:17
nwc10 \o
Geth rakudo/rakuast: b3cf295238 | (Jonathan Worthington)++ | 3 files
RakuAST node for <?{ ... }> and <!{ ... }>
 19:53
Geth rakudo/rakuast: 5d4f794124 | (Jonathan Worthington)++ | 3 files
RakuAST node for `:my $foo;` style decls in rules
 21:11

Powered by Cro and the Raku® Programming Language.

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

Thank you!