🦋 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.
00:04 Sgeo joined 00:11 jpn joined, epony joined
aruniecrisps wait @bscan how do you run this file? 00:12
I'm seing that you're providing both $*IN and @ARGS 00:13
00:16 jpn left 00:19 sena_kun left
bscan Something like cat test.raku | raku navigator.raku test.raku The 2nd specification of test.raku isn't truly needed, but it tells the script the name of where the code came from (even though it won't actually read the file) 00:30
01:06 jpn joined 01:11 jpn left 02:20 kylese left, hulk joined
aruniecrisps I'm getting this as a response: Running navigator.raku Received file: test.raku 90d0cb6c-4a53-427b-8d30-b1195895c2df It looks like it's running, but the data being given seemss to be just a UUID 02:29
here's test.raku for reference: say "Hello world!";
03:15 hulk left, kylese joined 03:49 jpn joined 03:53 jpn left 04:01 epony left 04:04 derpydoo left 04:45 dead1 joined 05:03 itaipu left 05:21 itaipu joined 05:22 CIAvash joined 05:39 jpn joined 05:44 jpn left 06:21 jpn joined 06:25 jpn left 07:15 itaipu left 07:41 mrlo left 07:50 epony joined 08:06 teatime left 08:07 JRaspass left, JRaspass joined, teatime joined 08:16 Sgeo left 08:29 epony left 08:46 CIAvash left 08:49 jpn joined 08:54 epony joined 09:16 dakkar joined 09:29 sena_kun joined 10:15 sena_kun left 10:25 sena_kun joined, sena_kun left 11:14 jpn left 11:19 epony left
tbrowder__ hi, is there now in modern Raku any simple way for a routine to detect unknown (unexpected) named arguments? 11:25
11:27 jpn joined
tbrowder__ i think one can iterate through the capture elements and check that way, but lizmat's tricks with sets may be the way to simplify things for a one liner. 11:29
lizmat m: sub a(:$a, :$b, *%_) { dd %_ }; a :foo 11:31
camelia {:foo}
lizmat tbrowder_: is that what you're looking for ? 11:34
11:34 epony joined
tbrowder__ great, that looks like the important part, thanks! 11:47
btw, i've decided that trying to specify a new api number isn't worth the headache 12:12
.oO( with great power comes great responsibility :-)
tbrowder__ and doesn't work like i thought it did when there was an "api" key in the META6 file 12:13
yes, you are right! 12:14
librasteve @aruniecrisps ++ ... I have been thinking about LSP support in the light of the recent CommaIDE announcements. Would you say it would be less effort for the raku community to support a single LSP for use across VSCode, vim and JetBrains (via github.com/ballerina-platform/lsp4intellij for example) than to do your work AND continue to support the Edument IntelliJ custom plugin? 12:39
interested to hear thoughts from others
[as an aside, I suspect that CommaIDE / IntelliJ plugin has richer features (Grammar debugging and so on) vs. a vanilla LSP and deeper integration / more touch points with the IntelliJ platform ... so jumping from Comma Plugin to LSP Plugin for JetBrains IDEs may reduce the features, in return for a more effective path to a maintainable future...] 12:43
lizmat I think going for a better maintainable future is the way to go 12:51
If someone needs Comma's more advanced features, there'll still be there: it's not like Comma will stop working all of a sudden?
12:58 Some-body_ joined 13:00 DarthGandalf left 13:01 Some-body_ is now known as DarthGandalf 13:28 itaipu joined
tbrowder__ .tell Xliff got yr prs, thnx 13:44
tellable6 tbrowder__, I'll pass your message to Xliff
15:17 jpn left 15:35 jpn joined 15:49 jpn left
tbrowder__ lizmat: ref the Capture stuff, is it possible to create something a dev could use to (1) list named args allowed for a given sub and (2) programmatically apply it when the sub is called? 15:54
15:55 dano left, dano joined 16:03 jpn joined 16:07 El_Che left, El_Che joined
aruniecrisps @librasteve i mean an LSP is generally designed to be a protocol that works across different editors, not sure exactly what you mean 16:07
16:15 discord-raku-bot left 16:16 discord-raku-bot joined
tbrowder__ i'm getting failing tests on github for windows workflows testing. 16:26
www.irccloud.com/pastebin/LFSVnp4m/bigtom 16:27
lizmat yeah, that looks familiar 16:29
:-( 16:30
[Coke] is there a dry run option for mi6 release? 16:31
tbrowder__ any ideas about how to proceed? here's the whole failure part:
lizmat [Coke]: mi6 build ? 16:32
tbrowder__ [Coke]: i don't see anything like it..;. 16:33
antononcube @Coke and @tbrowder My calendars are bigger than yours! 16:35
Well, maybe, not bigger, but nicer.
tbrowder__ ;-) 16:36
antononcube And, of course, the size of the calendar is not that important, but how you use it.
tbrowder__ i use mi6 build all the time, esp for doc redoes 16:37
[Coke] I just ran it and thankfully it prompted me for the one thing I wasn't sure it would get right.
antononcube For example, in Jupyter notebook I can make these kind of calendars (with highlights): 16:38
tbrowder__ antononcube; cool for quick lookups and online use! 16:40
antononcube @tbrowder Yes, exactly. I plan to add (hovering) toolips and hyperlinks. 16:42
tbrowder__ i'll leave that arena to you! i probably will never get around to using your LLM interface, but i will try to remember to demo some of yr stuff when i get the chance--new devs ought to jump on it--esp with raku interface 16:43
antononcube @tbrowder Well, the package "Markup::Calendar" can make standalone HTML calendar files. 16:44
tbrowder__ so i think RSC ought to think about pushing/fixing Raku on Windows--that's, unfortunately, where the young devs are 16:45
paying the few windows experts around, or google summer of code maybe 16:46
antononcube And, yes, I started working on "Text::Calendar" and "Markup::Calendar" because of some LLM and data science projects. Hence, the use in Jupyter is a primary use case. 16:47
tbrowder__ i should say "where they appear to be". just visit CodeProject 16:48
16:51 derpydoo joined
bscan I like LSP because it allows people to use their preferred editors. People who currently use neovim, emacs or vscode for every other language generally don't want to switch to CommaIDE just for Raku. 16:53
aruniecrisps @bscan did you get a chance to look at this? 17:00
bscan Ah, yes, that output looks good to me. The script will dump out all classes, grammars, subs, tokens, etc used in your test.raku (of which there are 0). It will also show warnings and syntax errors. Try it with a more complex script. 17:03
You can also add me as a "Discord friend" if you want to side chat. 17:04
17:15 vlad joined
aruniecrisps Is there any documentation on RakuAST? 17:42
antononcube @arun Yes, "for early adopters." 17:44
dev.to/lizmat/series/23109 17:45
18:05 dakkar left 18:22 vlad left 18:24 vlad joined 18:40 jpn left 18:42 sena_kun joined
librasteve arun: there are two versions of the raku Comma IDE - one is a standalone IDE based on IntelliJ, the other is a plugin for IntelliJ IDEA. (IntelliJ is made by JetBrains). 18:49
if there is an LSP for raku, then there is a way to create an LSP-based plugin for IntelliJ also ... therefore if we support the Comma tech, then we could have two parallel IntelliJ plugins ... my question is whether it would be better to "put all the wood behind one arrow" and to focus primarily on the LSP route ... ?? 18:53
aruniecrisps right now it looks ike neither RakuNavigator nor the Comma IDE plugin is at feature parity with PerlNavigator, for me I imagine the first priority would be to first get the base LSP server functionality to a point that whatever frontend we decide to go with for the LSP client we have a solid foundation to use 18:56
i'm playing around with RakuAST to see what I can do, and it seems like we can do a lot with it 18:57
bscan I'd strongly vote LSP since the market share of neovim+vim+emacs+vscode greatly exceeds the market share of JetBrains. Basically, we need a good LSP either way.
(And I agree that the Raku Navigator does not yet count as a great LSP) 18:58
jdv who runs the discord bot? 18:59
aruniecrisps but it doesn't look like .AST parses use statements at least in Raku REPL, so it looks like we can't just do file-name.IO.slurp.AST wholesale, we'd probaly have to parse the use statements before reading the rest of the Raku file
@lizmat any thoughts? 19:00
jdv the discord bot's nick should be shortened. 19:04
19:04 discord-raku-bot left
[Coke] aruniecrisps - what command are you trying to run in the REPL? 19:04
19:04 discord-raku-bot joined
jdv this is what i see: ibb.co/MNtH79h 19:04
[Coke] yah, for thin windows it's a lot. 19:05
jdv its a waste!:) 19:06
[Coke] I'm at 120 wide and even that is a big gap.
jdv who does the discord bot work for?!
[Coke] It's self documenting, but maybe even just "discord" or "bridge" would be ok. 19:07
19:16 xinming left, xinming joined 19:30 vlad left 19:36 jgaz joined
jdv well, it has to be unique to be registered or whatever. rdb123 or whatever would be fine. the self documenting is only important to newbies and they can ask or figure it out given there's a "sub-nick" in the message. 19:42
19:45 jpn joined 19:49 jpn left 20:28 kjp left, kjp joined 21:41 jpn joined 22:14 jgaz left
Voldenet weechat has weechat.bar.nicklist.size_max for this case 22:48
I use different layout settings on mobile and it works quite well 22:49
lizmat aruniecrisps .AST does parse "use" statements, and creates RakuAST::Statement::Use for them 23:01
23:09 jpn left 23:34 Sgeo joined 23:51 sena_kun left