🦋 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 available at irclogs.raku.org/raku-dev/live.html | For MoarVM see #moarvm
Set by lizmat on 8 June 2022.
00:00 reportable6 left 00:01 reportable6 joined 00:28 MasterDuke joined
MasterDuke benchable6: 2022.06,2022.01,2022.12 srand 1; my @a; @a = roll(100, (^1000).List) for ^1_000; 00:28
benchable6 MasterDuke, starting to benchmark the 3 given commits
MasterDuke, ¦2022.06: «0.6693» ¦2022.01: «Cannot find this revision (did you mean “2022.12”?)» ¦2022.12: «0.4073»
MasterDuke, No new data found
MasterDuke benchable6: 2022.06,2022.07,2022.12 srand 1; my @a; @a = roll(100, (^1000).List) for ^1_000; 00:29
benchable6 MasterDuke, starting to benchmark the 3 given commits
MasterDuke, ¦2022.06: «0.6751» ¦2022.07: «0.6352» ¦2022.12: «0.6144»
MasterDuke, No new data found
MasterDuke benchable6: 2022.06,2022.07,2022.12 srand 1; my @a; @a = roll(100, ^1000) for ^1_000;
benchable6 MasterDuke, starting to benchmark the 3 given commits
MasterDuke, ¦2022.06: «0.6741» ¦2022.07: «0.6131» ¦2022.12: «0.6149»
MasterDuke, No new data found
MasterDuke huh, locally the Range version is a bit slower 00:30
for 10k iterations, Range version is ~3s and List version is ~1.5s 00:31
with Rakudo 2022.12
oh, it looks like i just got a couple lucky spesh runs in a row. with MVM_SPESH_BLOCKING=1 they both are ~3s 00:36
benchable6: 2022.06,2022.07,2022.12 srand 1; my @a; my $b = ^1000; @a = roll(100, $b) for ^1_000; 00:41
benchable6 MasterDuke, starting to benchmark the 3 given commits
MasterDuke, ¦2022.06: «0.2079» ¦2022.07: «0.2112» ¦2022.12: «0.1933»
MasterDuke, No new data found
MasterDuke that's the problem, the range is getting expanded into a list 00:42
00:42 Xliff left
MasterDuke ah, `multi sub roll($n, +values)`, so since Range does Iterable, it gets expanded. so you have to use the method form of roll or explicitly put the range in a scalar and then use that 00:52
01:57 epony joined 02:26 frost48 joined 02:28 frost48 is now known as frost, frost left, frost joined
MasterDuke m: my $a = ^4; my @b = roll(3, $a); say @b   # whoops, that doesn't really work either 02:42
camelia [^4 ^4 ^4]
MasterDuke m: my $a = ^4; say 3.&roll($a); say $a.&roll(3)   # hm, looks like you can't use the sub form at all to do this 02:45
camelia (^4 ^4 ^4)
(3 3 3 3)
03:48 Kaipei joined 04:03 frost left 04:05 frost joined 04:40 frost left 05:09 frost joined 05:13 frost joined 06:00 reportable6 left, reportable6 joined 06:14 Kaipei left 06:15 Kaipei joined 06:36 epony left 06:46 epony joined 07:05 frost left 07:43 sena_kun joined 08:06 sena_kun left 08:07 sena_kun joined 10:20 sena_kun left 11:07 epony left, sena_kun joined 11:28 epony joined
Geth rakudo/rakuast: 77818c60b4 | (Stefan Seifert)++ | 4 files
RakuAST: implement support for LEAVE phasers
 11:29
nine_ This one ^^^ makes 14 more spectest files pass and definitely was one of the easier ones.
Nemokosch 👏 11:31
Geth rakudo/rakuast: 9b8a160e42 | (Stefan Seifert)++ | 4 files
RakuAST: implement support for LEAVE phasers
 11:38
11:45 nine_ is now known as nine 12:00 reportable6 left 12:02 reportable6 joined 12:16 frost joined 12:24 guifa joined 12:35 frost left
lizmat notable6: weekly 14:32
notable6 lizmat, No notes for “weekly”
Nemokosch lizmat: what will the merge of the rakuast branch mean in practise? things remain more or less the same, except you don't need to build a separate branch to allow the rakuast version with that environment variable? 15:43
lizmat good question: jnthn has stipulated that RakuAST module support should be available through a "use experimental :rakuast" pragma, and I think we still need to figure out the best way to do that 15:46
but mostly you wouldn't notice anything, most likely 15:47
just that if in 6.e.PREVIEW and doing a "use experimental :rakuast" would give you access to the RakuAST:: module tree
so you can build code in an object oriented manner, rather than needing to use EVAL 15:48
16:05 melezhik joined 16:15 melezhik left
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2022/12/05/2022-...mas-again/ 17:02
18:00 reportable6 left 18:02 reportable6 joined 18:10 sena_kun left 18:54 sena_kun joined
Geth ¦ Template6: 2colours self-assigned curly braces are inescapable github.com/raku-community-modules/.../issues/21 20:14
20:26 sena_kun left 20:35 sena_kun joined 20:40 ggoebel joined
ggoebel the raku-advent.blog site currently doesn't provide a link to see all 2022 posts... just "Full list of 2021 Raku Advent Blog Posts". Someone with privileges to do so, might want to fix that. 20:41
Nemokosch that link does exist though, apparently, just isn't "provided" 20:42
ggoebel right... raku-advent.blog/category/2022/ works
the url exists... a link to it doesn't 20:43
I always enjoy reading the articles. thanks to the folks taking the time to write them :-) 20:44
21:18 vrurg_ joined 21:21 vrurg left 22:19 sena_kun left
Nemokosch better late than never.... what does "spesh" mean actually? 22:44
ggoebel jnthn coined the term spesh. I believe it refers to a little bit of specialization. Though it may be overloaded with other meanings. 23:10
Nemokosch Hopefully vrurg can tell something about that later on 🥺 23:12
ggoebel p6weekly.wordpress.com/2014/04/07/...al-things/ 23:15
Nemokosch a good 8 years 23:19
ggoebel From www.jnthn.net/papers/2014-yapceu-p...rmance.pdf slide #48: Enter Spesh 23:21
Spesh is the name for
MoarVM's "type specializer"
(Why? If we called it "spec"
everyone would say it wrong,
or try to Google "Perl 6 spec")
Nemokosch when parrot still existed 23:25
I for one tried to google "Raku spesh" and that didn't help but anyway - thank you for enlightening me 23:30
23:33 kjp left 23:34 kjp joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!