|
🦋 Welcome to the MAIN() IRC channel of the Raku Programming Language (raku.org). Log available at irclogs.raku.org/raku/live.html . If you're a beginner, you can also check out the #raku-beginner channel! Set by lizmat on 6 September 2022. |
|||
|
00:07
apac left,
itaipu joined
01:12
oodani left,
oodani joined
01:27
hulk joined,
kylese left
02:15
hulk left,
kylese joined
02:16
lichtkind left
02:41
dg left
02:46
dg joined
03:11
LainIwakura joined
03:23
arkiuat left
03:31
apogee_ntv left,
apogee_ntv joined
03:36
arkiuat joined
|
|||
| patrickb | I've heard of the V language for the first time. vlang.io/ Seems pretty popular. Am I the only one that missed it so far? | 03:50 | |
| Voldenet | vlang is too close to golang for my taste – if I wanted golang I'd just use golang | 04:35 | |
| It's difficult to like golang – it feels very repetitive because of how error handling works and that extra : in assignment is ugly to look at | 04:50 | ||
| things like [][]string{} are also sufficiently weird, when it comes to compiled languages I prefer nimlang or ziglang – they just feel pleasant to look at in comparison | 04:55 | ||
| elcaro | perryprog: You can bench it for yourself -- pastebin.com/czca7sMW | 05:08 | |
|
05:17
Aedil joined
05:37
Pixi left
05:45
melezhik joined
05:47
Sgeo_ joined
|
|||
| Voldenet | without benching, I'd expect substr to be a lot faster and grammar be the slowest | 05:49 | |
|
05:50
Sgeo left
|
|||
| Voldenet | I've benchmarked, rotor was surprisingly slow | 05:54 | |
| I'm amazed but it seems that `sub { DateTime.new(year => .substr(0,4), month => .substr(4,2), day => .substr(6,2), hour => .substr(8,2), minute => .substr(10,2), second => .substr(12,2)) given $s }` is not significantly slower | 05:56 | ||
|
05:56
Pixi joined
05:57
kaskal left
05:58
kaskal joined
|
|||
| elcaro | Yeah, I wouldn't expect it to be. Anytime I need to cut up a string fast, I reach for substr :/ | 06:09 | |
| Voldenet | I mean, when you use named parameters vs positional parameters | 06:14 | |
| .substr is plain allocate + copy part of the string, I doubt it can get any faster, the only fastest way would be if maybe no copy was done at all | 06:16 | ||
|
07:03
[Coke] left
07:10
[Coke] joined
07:29
Aedil left
07:32
ynzoqn joined
07:34
ynzoqn left
07:55
melezhik left
07:57
librasteve_ joined
08:19
dakkar joined
08:29
arkiuat left
08:34
Sgeo_ left
08:57
arkiuat joined
09:02
arkiuat left
09:30
arkiuat joined
09:35
arkiuat left
09:46
LainIwakura left
|
|||
| librasteve | I just noticed that Num.FatRat takes a 2nd argument to converts the number to a FatRat with the precision $epsilon. Likely this is what I need - have closed the issue for now and will try it... | 09:51 | |
|
10:03
LainIwakura joined
10:05
arkiuat joined
10:11
LainIwakura left
10:15
arkiuat left
10:24
LainIwakura joined
10:29
lichtkind joined
10:43
arkiuat joined
10:47
arkiuat left
11:01
arkiuat joined
|
|||
| simon_sibl | probably late to the party but the new Raku website looks amazing ! | 11:04 | |
|
11:08
arkiuat left
11:20
arkiuat joined
11:25
arkiuat left
|
|||
| librasteve | @simon_sibl ... thanks! all written in raku, of course (harcstack.org) | 11:25 | |
|
11:37
crnlskn joined
11:38
arkiuat joined
11:43
lichtkind left
11:44
LainIwakura left
11:47
arkiuat left
11:49
apac joined
|
|||
| simon_sibl | I'm not much of a web person but the harc stack looks pretty good I should try to learn and make a little project with it | 11:55 | |
|
11:56
dgl joined
11:58
dg left,
dgl is now known as dg
12:00
arkiuat joined
12:04
arkiuat left
|
|||
| mc2 | ls | 12:20 | |
| oops sorry | |||
| sjn | ls: cannot open directory '.': Permission denied | 12:28 | |
|
12:29
arkiuat joined
12:34
arkiuat left
12:51
Guest96 joined
|
|||
| simon_sibl | lmao, that happens to me also more than it should xD | 12:59 | |
|
13:03
arkiuat joined
13:06
stanrifkin joined
13:08
arkiuat left
|
|||
| librasteve | ww? | 13:23 | |
| crnlskn | bash: ww: command not found | 13:25 | |
| scnr | |||
| librasteve | ww == 'wrong window' (I learned that from lizmat) | 13:26 | |
| crnlskn | oh, I am aware, i was continuing the previous joke :) | ||
| librasteve | clear | 13:27 | |
| crnlskn | also my fedora install here suggested 'w' as a valid command, which made me realize that 'ww?' could be read as a regex and then *would* math 'w' | ||
| er s/math/match | |||
|
13:30
arkiuat joined
13:35
arkiuat left
14:04
arkiuat joined
14:10
arkiuat left
14:16
Guest96 left,
Sgeo joined
14:30
arkiuat joined
14:34
arkiuat left
14:58
melezhik joined
|
|||
| melezhik | . | 14:58 | |
|
15:03
arkiuat joined
|
|||
| librasteve | o/ | 15:47 | |
| Voldenet: you wrote | |||
| sub tiniest-fat-rat($n = 2) { with $n / 2 { return $n if .FatRat == 0; samewith($n / 2); } }; say tiniest-fat-rat.FatRat.nude; #(1 9223372036854775808) | 15:52 | ||
| evalable6 | (1 9223372036854775808) | ||
| librasteve | which has had me scratching my head ... then I wrote: | ||
|
15:52
NucleusBiffBot joined,
NucleusBiffBot left
|
|||
| say $n.WHAT; ... it's always a (Rat) | 15:54 | ||
| then | |||
| $*RAT-OVERFLOW = FatRat; | |||
| and it recurses forever | 15:55 | ||
| or you can go | |||
| sub tiniest-fat-rat(FatRat $n = <2>.FatRat) { | |||
| perryprog | "tiniest fat rat" oh my goshhh you have to be doing this on purpose | 15:57 | |
| librasteve | ;-) | ||
|
16:10
human-blip left
|
|||
| coleman | "fatrat" is the hostname of the server that serves raku.org and the docs :D | 16:12 | |
|
16:12
human-blip joined
|
|||
| coleman | We don't have many hosts but I'm going to try and find another class name or type or role for the next one. I had to decomission "baggy". RIP baggy. | 16:13 | |
|
16:14
LainIwakura joined
|
|||
| lizmat | PSA: it looks like irclogs.raku.org and Geth will be down on Mon 15 Sep from 0600 - 0800 UTC due to electricity works at their data center | 16:19 | |
|
16:20
LainIwakura left
16:32
dakkar left
16:59
Aedil joined
17:07
melezhik left
17:30
crnlskn left
|
|||
| arkiuat | I still need to go to old.raku.org to find the information about the IRC channel bots, but I'm pretty sure someone already made an issue for that | 17:52 | |
| Is there an instance of Commitable running at present? github.com/Raku/whateverable/wiki/Committable | 17:59 | ||
| m: for "2025-05-25T17:00:00" -> $d { for DateTime.new($d).in-timezone(-6*60*60), DateTime.new($d ~ "-06:00") { .julian-date.say } } | 18:04 | ||
| camelia | 2460821.208333 2460821.458333 |
||
| arkiuat | so it's not just that .julian-date was ignoring timezone information, but that it was only sometimes doing so, depending on how that information was supplied | ||
| I'm not sure how to run this again the fix that lizmat already put in | 18:05 | ||
| this goes back to a discussion on this channel back on May 25, and lizmat's fix I'm referring to is github.com/rakudo/rakudo/pull/5886 | 18:07 | ||
| oh, never mind, I just figured out what I'm doing wrong. | 18:08 | ||
| m: for "2025-05-25T17:00:00" -> $d { for DateTime.new($d).in-timezone(-6*60*60), DateTime.new($d ~ "-06:00") { .say } } | 18:09 | ||
| camelia | 2025-05-25T11:00:00-06:00 2025-05-25T17:00:00-06:00 |
||
| arkiuat | One is at 11 AM in TZ-6 and the other is at 5pm in TZ-6 | ||
| but while I'm doing blasts from the past, we also had a couple of discussions about [Z] where I expressed the opinion that it was the matrix-transpose operator, and it turns out there is a section in the Traps doc all about this particular issue: docs.perl6.org/language/traps#Usin...t_of_lists | 18:13 | ||
| weird, how did that "perl6" get in there? I meant docs.raku.org/language/traps#Using...t_of_lists | 18:14 | ||
|
18:23
human-blip left
18:25
human-blip joined
18:40
crnlskn joined
18:48
crnlskn left
|
|||
| [Coke] | docs.raku.org/language/traps#Using...t_of_lists is the URL on the site. | 19:17 | |
|
19:36
apac left
|
|||
| librasteve | arkiuat: yes the bot doc is an open issue …. PRs like that very simple to do and always welcome! (my proposal is a new box on the community page) | 19:41 | |
|
19:49
melezhik joined
|
|||
| melezhik | . | 19:49 | |
| changed theme to light one and added systemd examples, also added some clarity on Sparrow essentials on the start page - sparrowhub.io | 19:51 | ||
|
19:58
melezhik left
20:33
arkiuat left
20:42
stanrifkin left
20:44
stanrifkin joined,
stanrifkin left
20:46
arkiuat joined
20:50
arkiuat left
20:52
mc2 left
20:57
Aedil left
21:05
arkiuat joined
21:10
arkiuat left
21:22
arkiuat joined
21:27
arkiuat left
21:33
arkiuat joined
22:05
apac joined
|
|||
| Voldenet | > perryprog │ "tiniest fat rat" oh my goshhh you have to be doing this on purpose | 22:14 | |
| Yes. :D | |||
|
22:19
apac left
|
|||
| Voldenet | obviously the largest fat rat the more scary it becomes, so the tiniest one is the least scary | 22:19 | |
| and "$*RAT-OVERFLOW" makes me imagine the pool of rats :) | 22:21 | ||
| back to the topic, you're absolutely right, when tested on Num it behaves differently | |||
| m: sub tiniest-fat-rat(Num() $n = 2) { with $n / 2 { return $n if .FatRat == 0; samewith($n / 2); } }; say tiniest-fat-rat.FatRat.nude; # roughly 1e-6 precision | |||
| camelia | (1 524288) | ||
| Voldenet | and in fact changing the precision would let us store numbers with a lot higher precision due to how Num() actually works | 22:23 | |
| m: my $p = 1e-300; sub tiniest-fat-rat(Num() $n = 2) { with $n / 2 { return $n if .FatRat($p) == 0; samewith($n / 2); } }; say tiniest-fat-rat.FatRat($p).nude; | 22:24 | ||
| camelia | (1 66969287949141707559276565566250113160087800731595850465234399273146940695308507655824898675980991132974667057347071676574196580355769627724903609841866092524591048592651443658881716281639819636737213638456540468647387132921242297244784649662981643… | ||
| Voldenet | "arbitrary" as long as it's larger than 1e-308 of course | 22:25 | |
| but I still think that parsing Str again in `NumStr -> FatRatStr/FatRat` conversion is a better idea | 22:28 | ||
|
23:31
arkiuat left
|
|||