00:21 oodani left 00:25 oodani joined 00:54 marcprux joined, marcprux left 03:11 derpydoo joined 03:33 apogee_ntv left 03:34 apogee_ntv joined 04:13 Aedil joined 05:32 Sgeo_ left 06:04 disbot3 joined 06:08 disbot2 left 06:12 apac joined 06:36 apac left 06:49 abraxxa joined 07:31 abraxxa1 joined 07:34 abraxxa left 08:06 lichtkind joined 08:12 lichtkind_ joined 08:15 lichtkind left 08:23 dakkar joined
disbot3 <simon_sibl> I am not a very web oriented dev, would hArc stack be a good tool to use to make a website that allow to download and upload files ? (for the download I would simply list the files, clicking on them would trigger the download) and upload via a button I guess 08:26
<simon_sibl> but every action would require a user interaction right ? even refreshing the files list ? 08:27
08:56 librasteve_ left
disbot3 <jubilatious1_98524> I see [dead] also. Sounds like a line from a movie. 08:58
09:05 abraxxa1 left 09:24 abraxxa joined 09:28 abraxxa1 joined 09:32 abraxxa left 09:34 abraxxa2 joined
disbot3 <holmdunc> They can be saved from [death] by sufficiently karma'd users - just did it 09:36
<holmdunc> cdn.discordapp.com/attachments/633...8f9c4&
09:37 abraxxa1 left 09:52 Aedil left 10:00 apac joined
disbot3 <melezhik.> On my experience HN moderators sometimes are way to strict on multiple posts from the same source , author - they might see thus as self promotion. It’s sad though 10:04
<melezhik.> Oh . I will keep your contact in my bookmarks 😀 10:06
10:36 apac left
disbot3 <antononcube> Self promotion with Raku is somewhat of an oxymoron. HN moderators might be “just” Perl fans. 11:30
<antononcube> I want to make two packages that connect Raku to locally run services. One is Ollama, the H2O. 11:32
<antononcube> What are good names for those? I do not think “WWW::Ollama” is a good one. Here are some variants: 1. Ollama 2. Ollama::Client 3. H2O 4. H2O::Client 5. H2O::AI::Client 11:33
<simon_sibl> How can I make a variable from my main file available on a rakumod file ? I tried $*var but it still doesn’t find it 11:37
<antononcube> Using our might do what you want. I am not sure what “main file” is. 11:38
11:39 abraxxa joined, abraxxa1 joined
disbot3 <simon_sibl> I tried our as well 11:41
11:41 abraxxa2 left
disbot3 <simon_sibl> Main file is basically the file that “use” all the modules and do stuff 11:41
<simon_sibl> But one of the model needs to access a variable created in the main file
11:43 abraxxa left
disbot3 <antononcube> Then make a sub that accesses it. 11:43
11:48 Aedil joined
Geth raku-mode: 5e36e7f76c | quanrong++ | raku-indent.el
Indent method calls when they happen on a new line
 11:51
raku-mode: d06baaa2e8 | Altai-man++ (committed using GitHub Web editor) | raku-indent.el
Merge pull request #62 from ttn-ttn/master

Indent method calls when they happen on a new line
disbot3 <simon_sibl> doesnt work as well, the sub is invisible to the module somehow 12:03
<simon_sibl> the module I "use" cannot access the main file
<simon_sibl> alright, for some reason when in my main file its $node and the module $*node it works 12:09
<simon_sibl> makes no sense to me xD
<antononcube> 🤷‍♂️
[Coke] $* is a dynamic var. 12:12
docs.raku.org/type/Variable#trait_is_dynamic 12:13
disbot3 <simon_sibl> yes but why must the name be different between my main file and my module for it to work ? 12:24
[Coke] you're using the dynamic nature in the module. You could use $*a in both 12:29
(and probably should to avoid confusion)
I'm actually surprised it works with $a in main - you can't do it all in one file: 12:40
m: my $a = 3; sub b { say $*a}; b
camelia Dynamic variable $*a not found
in sub b at <tmp> line 1
in block <unit> at <tmp> line 1
[Coke] m: my $*a = 3; sub b { say $*a}; b
camelia 3
[Coke] m: my $a = 3; sub b { say $a}; b 12:41
camelia 3
[Coke] (just has to match, there..)
Voldenet if you want to access variable (a global one) inside a module, export getter and setter to it 12:43
12:49 derpydoo left
Voldenet avoid dynamic variables for as long as possible - they can be useful exactly in cases where you can't avoid it 12:50
sub query ($q) { try &*logger($q); }; my &*logger = -> $q { say "running query $q" }; query("select"); # example, you need contextual logger, but the framework only supports constant logger, so you have to manually set request context 12:51
evalable6 running query select
Voldenet but otherwise it's a mess to maintain :\ 12:52
disbot3 <simon_sibl> it does not work when I name it $*a in both 13:06
Voldenet may I ask how do you define your dynamic variable in a module? 13:23
if you do `my $*a` in a module, I'm not sure if that can work 13:24
example glot.io/snippets/hblj3xbt2f 13:25
when you do `my $*a` in a module, you define dynamic variable for <load> subroutine, which is not very useful 13:28
though the syntax may hint something else… :) 13:29
m: sub load-foo { my $*x = 1; bar => -> { $*x } }; my $foo = load-foo(); $foo<bar>() # shorter example 13:33
camelia Dynamic variable $*x not found
in block <unit> at <tmp> line 1
disbot3 <simon_sibl> let me create a git repo for this project, will be easier xD 🙏 13:34
Voldenet you can reduce it to glot example probably, it's a lot easier to tinker with that way 13:35
but i'm not a big fan of dynvars either way, they're extremely confusing 13:36
disbot3 <simon_sibl> its very small for now anyway xD github.com/4zv4l/dfs the main file is dfs.raku and the module I use the $*node is Web.rakumod 13:39
<simon_sibl> I can try to create a smaller example
<simon_sibl> well for some reason in the small example it works xD 13:41
Voldenet for some reason my head really hurts when trying to figure out what is $*node :P 13:45
13:48 arkiuat joined
Voldenet it may work in current version, but it assumes a lot of things about how things execute 13:49
arkiuat simon_sibl, when you used our, did you try just declaring the our-var in a module and setting it in the mainline? when you say the variable needs to be "created" in the mainline do you mean declared, or initialized, or something else? 13:51
of course you'd need to either export the our-var from the module or else fully qualify the package name of the var when referring to it in the mainline
disbot3 <simon_sibl> so the main file does the link between the 2 modules, but 1 of them need that variable since it contains information 13:53
<simon_sibl> the main create a "Node" (server) and I made another module for the web server, but the web server needs info from the Node, so I need to be able to access that variable
arkiuat and you can't pass it as an argument? 13:55
disbot3 <simon_sibl> from the main I directly call the SITE.serve (hArc stack), and even with a function, that variable is required in multiple Class and sub within that Web module, wouldnt be practical 13:57
arkiuat I haven't really started learning Air yet. Maybe librasteve can help. 14:00
Voldenet so, I have tried and I can use dynamic variables in this example just fine glot.io/snippets/hblk2nojo0
disbot3 <simon_sibl> I am a very beginner as well xD just trying to have a base working for now, then I can improve it
<jubilatious1_98524> It lives! 14:02
Voldenet erm, I had to add -I. to run command, because fork eats it glot.io/snippets/hblk4twrxq
disbot3 <simon_sibl> thats even better, because I moved the use Web and use Node because I thought the error was because Web was used before I create $*node 14:03
Voldenet $*node is being dynamically created by main.raku, then when you call anything from it, it has $*node
disbot3 <antononcube> @jubilatious1_98524 Thanks! 14:04
Voldenet I still would say to not use dynvars and pass things explicitly: glot.io/snippets/hblk8cdh72 14:06
disbot3 <simon_sibl> but in my code it still doesnt work, the website starts, but when it tries to load and do the for loop on $*node it dies 14:07
Voldenet you can avoid dynvars, so avoid it, it will save you from headaches later 14:08
disbot3 <simon_sibl> yep, just did like your last example 14:10
<simon_sibl> thank you ! 🙏
14:14 melezhik_ joined
arkiuat sub sleep, sleep-timer, and sleep-until are all three documented (and presumably implemented) in class Date. The first two take a Real $seconds argument, and the last takes an Instant. 14:15
Does anyone know why these are in Date? DateTime, Instant, or even role Dateish all seem more appropriate.
melezhik_ . 14:16
14:19 melezhik_ left
Voldenet I'm betting that probably Date was made earlier and it made sense at the time 14:20
14:21 melezhik_ joined
melezhik_ please vote for yet another Raku post if you like it - news.ycombinator.com/item?id=45425774 - thanks 14:21
Voldenet sleep has 0 sense in the context and sleep-until coerces to Instant 14:22
but hey, note that they are subs inside Date.rakumod, they're not part of Date 14:23
lizmat right... so they could very well live somewhere else 14:24
I guess the doc just followed the file structure
14:25 abraxxa1 left 14:31 melezhik_ left, melezhik_ joined 14:33 apac joined 14:37 melezhik_ left
arkiuat yeah, I suspected it was some sort of historical reasons 14:38
but it might be helpful to document sleep-until under Instant instead of Date, and if we do that, then might as well move sleep & sleep-timer into DateTime, unless it's felt better to keep all three under a single rubric 14:39
I guess I should have brought this up on #raku-doc instead :D 14:40
14:51 apac left
[Coke] (why is sleep-until not also a method on Instant, I wonder) 15:09
disbot3 <librasteve> @holmdunc thanks for the tip on karma ... ChatGPT tells me you need karma ~30 to have powers of resurrection 15:22
15:24 melezhik_ joined 15:26 melezhik_ left 15:28 apac joined 15:30 Aedil left 15:33 Aedil joined
disbot3 <antononcube> @librasteve My H2O is going to beat your Polaris. 15:38
tbrowder while on docs, is there an ascii version of the unicode negated set operators? if so, i can't find them in docs. for instance, there is '(elem) for is an element of, but i see a unicode for not an element of, but no ascii equivalent. how about '(not elem)' or '(!elem)'? 15:48
hm, i guess because the parsing is very different from the others... 15:55
hm, '(noelem)' 15:56
ugh
arkiuat [Coke] yes, I was wondering that too 15:57
15:57 arkiuat left
Voldenet sleep-until coerces, so its usable with DateTime 16:04
which is a lot more useful in the context
# so this is possible sleep-until DateTime.now.later(second => 5) 16:05
16:23 arkiuat joined
arkiuat Voldenet, so it seems the reasonable thing to do is move (at least the doc for) all three from Date into DateTime 16:24
tbrowder, the docs suggest what you're looking for is !(elem). The operator with a ! prepended is mentioned as "equivalent to" for each of the negated set operators with no ascii version 16:28
16:38 dakkar left 16:43 arkiuat left, arkiuat joined
tbrowder thanks, i missed that. it needs to be added in a few missing places. sounds like another docs pr. 16:44
…in the making
16:48 apac left 16:57 human-blip left 16:59 human-blip joined 17:08 apac joined 17:13 leah2 left 17:18 nine left, nine joined 17:19 dg left 17:26 merp joined 17:29 dg joined 17:34 leah2 joined 17:35 melezhik_ joined 17:37 melezhik_ left 18:02 melezhik_ joined 18:04 leah2 left 18:11 melezhik_ left 18:12 melezhik_ joined 18:13 melezhik_ left, melezhik_ joined 18:15 melezhik_ left, melezhik_ joined
disbot3 <librasteve> o/ 18:19
18:19 leah2 joined 18:21 melezhik_ left, melezhik_ joined 18:26 melezhik_ left 18:27 melezhik_ joined 18:32 melezhik_ left 18:33 melezhik_ joined 18:34 coleman left 18:36 melezhik joined, coleman joined 18:38 melezhik_ left
disbot3 <librasteve> @simon_sibl -- I took a look at your code (the Voldenet variant) - I see a few things that are reminiscent of Air and have a couple of tips at this point ... generally in Raku methods with names in CAPS do something special / system level ... so for example if you make a custom Air::Component, then you need to give it a method HTML ... and then that is magically called by the Air module to render it. Same with sub SITE ... you 18:39
need to export that sub so that the Air library can launch your site when you go raku -I. service.raku
<librasteve> So - yes please do write these methods / subs - but leave it to the Air library to call them 18:40
<librasteve> My advice is to start with a simple example (eg github.com/librasteve/Air-Examples...nimal.raku from github.com/librasteve/Air-Examples) and then try adding your Node / Web classes to that 18:41
<librasteve> please do ask here if you have any questions and I will try my best to help 18:42
arkiuat librasteve++
disbot3 <librasteve> @simon_sibl - I just saw you have done that on the cro-irc ... checking now... 18:44
19:06 Aedil left 19:10 tonyo joined
tonyo . 19:10
19:26 itaipu left 19:42 itaipu joined 19:50 arkiuat left 20:02 arkiuat joined 20:07 arkiuat left 20:27 arkiuat joined 20:32 arkiuat left 20:44 arkiuat joined
Voldenet That makes sense, but then how do you pass variables to that SITE? I think it'd make sense to pass *%_ 20:46
m: sub SITE(:$x) { say $x }; sub air-serve(*%_) { SITE(|%_) }; air-serve(:x(5))
camelia 5
Voldenet it vastly simplifies the whole thing
20:49 arkiuat left 20:56 melezhik left 21:03 arkiuat joined 21:07 arkiuat left 21:14 apac left 21:21 arkiuat joined, lichtkind_ left 21:25 arkiuat left 21:37 melezhik_ joined 21:40 melezhik_ left 21:46 arkiuat joined
tbrowder m: use Test; my $s1 = set(1..3); my $s2 = set(1..6); my $res = $s1 (<=) $s2 21:56
camelia ( no output )
tbrowder m: use Test; my ($s1,$s2) = set(1..3), set(1..6); my $res = $s1 (<=) $s2; is $res, True 22:05
camelia ok 1 -
22:28 Sgeo joined 22:41 summerisle left 22:43 summerisle joined 22:44 perryprog_ joined 22:46 perryprog left, perryprog_ is now known as perryprog 23:37 melezhik_ joined 23:41 melezhik_ left

Powered by Cro and the Raku® Programming Language.

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

Thank you!