🦋 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.
00:00 reportable6 left, reportable6 joined 01:00 linkable6 left, evalable6 left 01:01 evalable6 joined 01:02 linkable6 joined 01:24 razetime joined 01:34 _________ left 02:12 ProperNoun joined 02:46 razetime left 03:02 razetime joined
guifa jdv librasteve thanks! 03:35
SmokeMachine that's.....crazy lol
03:37 kst joined
guifa re RSC: I was wondering to what extent we should leave :P5 in there. One thing to consider is that PCRE is still the most common engine. Maybe instead of :P5 just deprecate in favor of :pcre or similar 03:37
03:52 razetime left 04:35 rf left 05:10 _________ joined 05:11 jpn left
SmokeMachine I've been thinking on my old EEL (EventExpressionLanguage (github.com/FCO/EventExpressionLanguage)) to make it a slang for Raku (after getting exited with guifa's BF's talk) (instead of something a Raku script would parse). And I'm trying to rethink its syntax. So, does this make any sense for any of you? gist.github.com/FCO/274fcdce93b2c5...1e3dfafb0c 05:17
05:17 guifa_ joined 05:31 siavash joined 06:00 reportable6 left 06:01 reportable6 joined 07:01 bloatable6 left, quotable6 left, notable6 left, linkable6 left, shareable6 left, squashable6 left, coverable6 left, unicodable6 left, nativecallable6 left, evalable6 left, statisfiable6 left, bisectable6 left, releasable6 left, reportable6 left, tellable6 left, benchable6 left, committable6 left, sourceable6 left, coverable6 joined, notable6 joined, bisectable6 joined 07:02 evalable6 joined, benchable6 joined, sourceable6 joined, tellable6 joined 07:03 linkable6 joined, statisfiable6 joined, reportable6 joined, bloatable6 joined, quotable6 joined, unicodable6 joined, squashable6 joined, committable6 joined, shareable6 joined, nativecallable6 joined, releasable6 joined 07:19 siavash left 07:21 siavash joined 07:44 jpn joined 08:10 jpn left 08:16 tea3po left, tea3po joined 08:48 jpn joined 08:56 jpn left
xinming releasable6: status 09:11
releasable6 xinming, Next release will happen when it's ready. There are no known blockers. 0 out of 64 commits logged (⚠ 167 warnings)
xinming, Details: gist.github.com/6d15049a9bb73b1cfd...d82b2c6a77
09:11 sena_kun joined
lizmat xinming: there won't be a 2023.07 release 09:28
on account of vacation :-)
09:30 razetime joined 09:32 Sgeo left, razetime left 09:43 jpn joined 09:48 jpn left 10:05 ab5tract joined 10:06 siavash left 10:08 siavash joined
xinming lizmat: got it, Will waiting for 2023.08 the 10:20
then
lizmat you could download the repo and built the latest :-) 10:21
is there a specific feature you need after 2023.06 ?
10:24 siavash left, siavash joined 10:34 siavash2 joined 10:36 siavash left 10:37 jpn joined 10:43 jpn left
xinming lizmat: Nope, It's just always go for newest version habbit. 10:45
lizmat ok :-) 10:46
11:14 jpn joined
snonux i am with whatever my distro sticks to :-) as long the version is not 100 years old! 11:15
11:19 jpn left 12:00 reportable6 left 12:02 reportable6 joined 12:09 jpn joined 12:11 siavash2 left
lizmat weekly: github.com/Raku/problem-solving/issues/378 12:13
notable6 lizmat, Noted! (weekly)
12:14 timo left, jpn left 12:24 mark22k left, poohman joined 12:25 timo joined 12:26 mark22k joined 12:32 timo left 12:38 timo joined
nemokosch Is there any chance the decision won't get blocked by niner yet again? 12:48
I mean, in the case of any other decision than "let it all remain the way it is"
13:04 phogg joined 13:10 phogg left, phogg joined
lizmat nemokosch: I started a new issue, let's not start we preconceived notions 13:41
13:58 jpn joined 14:03 jpn left
antononcube @lizmat Are "App::Rak" Markdown documents in your articles repository upto date with "App::Rak" itself? 14:17
lizmat they should be 14:18
antononcube @lizmat Meaning, have you developed/changed new functionalities not reflected in the articles.
@lizmat Ok - thanks!
@lizmat I plan to file an issue for "App::Rak" about language extensions. Guess which? 🙂 14:28
lizmat haz no inspiration atm 14:29
nemokosch App::Rak is useful and it could get some love regarding improvements, I'm used to encountering bugs 14:30
antononcube Sorry -- no need to entertain my escapades ...
nemokosch however, the code is such much "not me" that I wouldn't want to work with it, honestly
so much*
14:30 timo left
antononcube @nemokosch I am trying to generate "App::Rak" usage code using the provided documentation. I am using several NLP techniques, (and technologies.) 14:31
lizmat OOC, what would make the code more "you" nemokosch?
14:32 timo joined
nemokosch if it weren't so monolithic, for starters 14:32
antononcube @lizmat Obviously, interlaced with a certain type of comments.
lizmat well, it's actually dualithic, wth 14:33
with "rak" providing the low-level functionality
nemokosch actually, it does consist of two parts: the CLI and the underlying library - but the cut is often not too obvious
the CLI part itself is quite thick and does quite a lot of imperative magic to set the scene up 14:34
lizmat nemokosch are you saying there are too many comments, or is it about the content of the comments ?
nemokosch no, I'm saying that the App::Rak part itself does a lot, and it's often not obvious what is done in the underlying library and what is downright transformed inside the CLI part 14:35
lizmat well, having a CLI that takes 143 options, is bound to be a little complicated 14:36
antononcube @lizmat That is why natural language command have to be used. 🙂
nemokosch iirc it was a common sight that there was some shared variable which at some point got set imperatively 14:37
this is part of the reason there were strange side-effect bugs, like certain options interacted in unexpected ways 14:38
14:56 timo left 15:02 timo joined
guifa_ lizmat : if you want I can start to work on a proof of concept Perl regex module. Then we can play around with deciding whether it's better to keep Perl/PCRE in core or figure out a pluggable manner 15:15
lizmat guifa_: that would be excellent!
guifa_ I just need to finish up one or two more small things on the ECMA262 and thankfully 95% of the word transfers over since JS basically uses PCRE 15:16
work*
lizmat but allowing for pluggable quotelangs will be on my list of todo :-)
guifa_ tbh though, Perl --> Raku can be a simple transpiling, but for how captures are numbered, and that's solvable with a minor edit to the match class i think 15:17
15:17 timo left 15:21 timo joined 15:35 jpn joined 15:38 timo left 15:44 timo joined 15:54 rf joined 15:57 jpn left 16:03 timo left 16:05 timo joined
antononcube @guifa I plan to complain (publicaly) in one of my blog posts about your neglet of "Intl::Token::Number". What reference name do you prefer: 1) alabamenhu, or 2) guifa ? 16:11
lizmat antononcube did you provide a PR ? 16:13
antononcube @lizmat I was planning to ask : do you prefer PR or Issue? I guess PR...
I will implement my proposed extensions and make the PR within an hour. 16:14
lizmat ++antononcube 16:16
antononcube @guifa Of course, I can just mention the name of Raku module. 🙂 16:17
16:23 razetime joined
guifa guifa works. I'm gonna update it soon though I was just waiting for RakuAST 16:25
just so many of them 16:26
antononcube @guifa Right, that kind of excuses are going to be mentioned too...
[Coke] (blin), ok, found docker/dblin in the repo. I get the same error, No build for start point 16:27
(running it in docker I would expect eliminates the OS issue)
.. ugh. and if I specify --old and --new I instead get a bus error! 16:28
Looks like depending on what old and new I have, I either get the "no build" or a bus error. 16:29
lizmat meh
antononcube @lizmat I will join @nemokosch in the "App::Ral" is too monolithic complaints -- the file Rak.rakumod is so big that Comma has 0.5 lag while editing it. (On a fairly powerful MacBookPro.) 16:33
@lizmat This observation might make you more "entrenched" in vi. 🙂 16:34
lizmat antononcube now imagine having "rak" in there as well
[Coke] Where does whateverable keep the builds?
(so I can pre-vet that i'm picking one expected to work) 16:35
lizmat 3700 lines in App::Rak, 1408 lines in rak
antononcube @lizmat I see!
lizmat and that's without the 2149 + 899 lines of rakudoc 16:36
antononcube I'm open to suggestions on how to break it up
but I'm not seeing it atm 16:38
antononcube @lizmat Aghh -- refactoring is hard. I might have some suggestions after I make the natural DSL code generation for "App::Rak".
lizmat sure :-)
antononcube @lizmat Often I make DSLs just for that -- to figure the interaction between components and the corresponding architecture. 16:39
[Coke] opens github.com/Raku/Blin/issues/32 - if someone not on a mac m2 could duplicate that, it'd be helpful.
antononcube @lizmat I am just going to make the PR without testing. (I do not dare... and I need to finish a blog post.) 16:40
16:42 poohman left
@lizmat Just created the PR request : github.com/lizmat/App-Rak/pull/42 16:46
16:54 timo left 16:56 timo joined
nemokosch On a different note: do y'all have an idea about GUI Raku content that could be tried on Windows? Simply for demonstrative purposes 16:57
16:59 jpn joined
antononcube "GUI Raku content" -- what do you mean? A Raku workflow that requires interactivity? 17:00
17:09 jpn left
rf I actually like monolithic applications, especially ones as straight-forward as Rak tbh! 17:09
antononcube @rf Learned programming with FORTRAN and Pascal, you did, hmm? 17:14
(Trying out a LLM Yoda persona...) 17:15
Some might like this: i.imgur.com/x9fpDH0.png 17:17
Geth ecosystem/main: 52b5bb2061 | (Elizabeth Mattijsen)++ | META.list
Remove Shell::Command

It will live in zef soon		 17:50
18:00 reportable6 left, reportable6 joined 18:01 Sgeo joined 18:16 tjr joined
nemokosch I meant some GUI program written in Raku 18:22
18:29 razetime left 18:39 timo left
Geth ecosystem/main: c9f69e32de | (Elizabeth Mattijsen)++ | META.list
Remove "if", it will live in zef soon
 19:00
rf @antonocube: No actually I learned with C and worked in lots of programs with files >5-10k lines, so that may have something to do with it 19:02
19:06 timo joined
nemokosch Anyway, what's so bad about Pascal 😡 19:09
rf Turbo Pascal is great :) 19:11
nemokosch >your Pascal:
>turboflex Pascal: 19:12
19:13 timo left 19:23 timo joined
antononcube @nemokosch There is some terminal menu package/module. But that is CLI, not GUI. There is GNU plotting package/module that might work on Windows — interactive, but “half” GUI. 19:32
nemokosch my whole reasoning is that if you are an absolute outsider beginner, this is stuff you'd like to see working. you are most probably a Windows user who wants to at least know that graphical action is possible 19:40
20:13 teatwo joined 20:16 tea3po left
antononcube That is why I think “notebook solutions” for Raku a really important. 20:19
20:19 teatwo left, teatwo joined 20:20 jpn joined 20:25 jpn left 20:26 teatwo left, teatwo joined 20:28 teatwo left, teatwo joined 20:33 jpn joined
guifa Don't we have a Jupyter thing working now for Raku? 20:37
20:43 ab5tract left
SmokeMachine github.com/bduggan/raku-jupyter-kernel 20:46
20:47 jpn left 20:52 bigdata joined 20:58 ab5tract joined
antononcube @guifa Yes, there are (at least) two notebook solutions for Raku right now: 1) "Jupyter::Kernel" with the Jupypter framework, and 2) "Text::CodeProcessing" and "RakuMode" for Mathematica. 21:14
Especially for integrating LLMs into Raku workflows using a notebook is the best. At least, a good REPL is needed. 21:15
@guifa Oh, and "power" tokens <local-number> would really help. 21:17
21:30 ab5tract left
nemokosch Yeah but I'm not thinking of a notebook solution 21:30
Sorry to break it down for you but I don't think any layperson would be impressed by the way a notebook document looks 😅 21:31
it's not even a real application
it reminds me of being a 5th grader in school and thinking "why the hell do we need to suffer with Comenius Logo when even Paint is a much better image editor" 21:32
21:35 bigganso23456 joined 21:36 bigganso23456 left
guifa_ at work they use Jupyter all over the place for python. Regardless whether I find it useful or not, clearly other folk do 22:00
lizmat: one thing that's been getting me a lot. Is there a reason some of the AST nodes are so picky about what kind of positional you pass them? 22:05
22:06 bigdata left
lizmat guifa_: example ? 22:12
guifa_ I had had
RakuAST::Regex::CharClassElement::Enumeration.new( elements => @foo.map(*.GENERATE) ) 22:13
lizmat ah, yes
guifa_ where GENERATE produces RakuAST::Regex::CharClassEnumerationElement::Character objects (in this case, at least)
lizmat add a .List after that
guifa_ the only way it would take it is if I did .Array. .List or Slips didn't work
lizmat .List didn't ? 22:14
antononcube I understand that -- I mentioned noteboooks "in ralation to..."
guifa_ Actually looks like .List does too, must have had another issue somewhere else too
lizmat method new(List :$elements!, Bool :$negated) {
is the signature
now, remember this is basically NQP 22:15
guifa_ yeah fair
lizmat so, it doesn't know how to handle a .Seq
that's really a HLL construct that NQP doesn't know about
antononcube @guifa How scalable is your wrapper solution in "Intl::Token::Number" ? In other words, should I try to understand it, or you plan to replace it with som RakuAST based code? 22:16
lizmat guifa_
and slurpies only work on positionals in nqp
guifa_ actually, wait, can confirm that .List does not work. .List ends up with an empty enumeration, .Array fills it properly 22:17
lizmat that feels... weird
22:18 bigdata joined
guifa_ antononcube: I mean, that one will probably mostly end up becoming a RakuAST translation of what's there. So if you understand how it works, that's probably mostly how the RAST version will work 22:18
basically, it grabs data from CLDR for digits, decimals, etc. Then pieces it together in a standardized format, currently without regard to whether in some locale the ordering may be different 22:20
22:20 sena_kun left
guifa_ so it looks for something akin to <negative>? [<digit> | <grouper>]+ [<decimal> <digit>+]? [<exponential> <plus-or-minus> <digit>+]? 22:21
where each of those are determined by values pulled from CLDR rather than hard coded. That's where RakuAST will speed things along, because I'll compile individual regexen per locale which will hard code those in and hsould be much more effiicent 22:22
antononcube @guifa Thanks for the detailed explanations!
guifa_ oh yeah I can also do the role bit much nicer now hahah 22:35
23:15 ProperNoun left 23:16 ProperNoun 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!