|
00:19
yewscion joined
00:21
yewscion left
00:32
Manifest0 joined
01:32
swaggboi left
01:36
Manifest0 left
01:45
swaggboi joined
02:12
Voldenet left
02:13
Xliff left
02:15
Voldenet joined
02:45
kylese left,
kylese joined
03:15
kylese left,
kylese joined,
eseyman left
03:17
eseyman joined
03:46
Chanakan5591 joined,
Chanakan left
|
|||
| guifa | what's the best way to package an executable (not native call) with a module to run with Proc::Async? | 03:49 | |
| IIRC, using resources/ isn't optimal because %?RESOURCES<executable>.absolute isn't guaranteed to point to the location of the file | 03:50 | ||
|
03:58
zetaaaa left
|
|||
| guifa | hrm, I'm managing to get a "failed to spawn" error | 04:09 | |
|
04:53
Aedil joined
05:18
guifa left
06:39
Tirifto left
06:54
Tirifto joined
06:57
hellwolf left
07:06
Sgeo left
08:51
hellwolf joined
08:58
Manifest0 joined
09:51
sena_kun joined
09:58
teatwo joined
09:59
tea3po joined
10:00
teatime left
10:01
teatime joined
10:02
teatwo left
10:04
tea3po left
10:07
sena_kun left
10:37
MoC joined
10:58
dg left
10:59
dg joined
11:26
Chanakan5591 left
|
|||
| tbrowder | guifa: my naive answer is put it in bin/ | 11:28 | |
| tellable6 | tbrowder, I'll pass your message to guifa | ||
| tbrowder | otherwise pack it inside a sub using "run" | 11:30 | |
|
11:34
MoC left
|
|||
| tbrowder | Raku Christmas Advent is rapidly approaching.... | 12:55 | |
| :chr | 12:57 | ||
| 🎄 | |||
| 🎄 🎄 🎄 🎄 | 12:59 | ||
| lizmat | tbrowder: I think we need to approach possible writers (the usual suspects) directly | 13:03 | |
| antononcube | I am trying to decide which of three posts to submit as a Raku Advent post. | 13:06 | |
| tbrowder | maybe all 3? | ||
| antononcube | Haha, sure. | 13:10 | |
| (1) Chess positions plots and corresponding LLM personas, (2) Overview of graph algorithms and plotting, (3) Raku for Data Science and LLMs. | 13:11 | ||
| The last one is 80% written, It started a request from Steve Roe for the London workshop. | 13:12 | ||
| The Chess one — I made it last year actually, but I never completely finished it. | 13:13 | ||
| @ab5tract With the new IntelliJ I get the following error: > Plugin 'Comma for Raku' (version '2.0') is not compatible with the current version of the IDE, because it requires build 242.* or older but the current build is IU-243.21565.193 | 14:18 | ||
|
14:18
Sgeo joined
|
|||
| ab5tract | Strange, it’s supposed to be set as a minimum version. Thanks for reporting, I will look into it | 14:22 | |
| antononcube | Good luck! | 14:35 | |
|
14:36
snonux left
14:37
snonux joined
15:07
[Coke] joined
|
|||
| [Coke] returns. | 15:07 | ||
| (everytime that mac mini loses power, the external drive that has my home dir mounts in the wrong place, goes to /fooN instead of /foo) | 15:53 | ||
| antononcube | I plan to make a Raku package that parses/interprets R code using the R library "tidyverse". How should name that package? 1) R::Tidyverse::Grammar 2) DSL::R::Tidyverse | 15:59 | |
| > The tidyverse is an opinionated collection of R packages designed for data science. All packages share an underlying design philosophy, grammar, and data structures. | |||
|
16:29
yewscion joined
17:15
yewscion left
|
|||
| ab5tract | Grammar::R::Tidyverse? | 17:27 | |
| Also, very cool idea! | 17:28 | ||
| antononcube | That is a very often "required feature" during/after my data wrangling DSL translations talks -- translating from Python or R data wrangling code (to whatever else.) Mostly about Python, though... | 17:35 | |
| My have a family of "DSL::*" packages, so I will most likely go with "DSL::R::Tidyverse". | 17:36 | ||
| @ab5tract Do you do data wranling on regular basis? (Say with R.) | 17:42 | ||
| ab5tract | Nope. When I need to wrangle data I usually use Raku, but it’s not a daily thing | ||
| DSL is obviously fine as it is your choice but I associate it differently | 17:44 | ||
| antononcube | Aha! So, I should make "DSL::Raku::Reshaper" first. | ||
| ab5tract | I wouldn’t call R or Python a DSL, but maybe this tidyverse totally is | ||
| antononcube | Yeah, I consider R "tidyverse" and Python "pandas" to be DSLs in this case. | 17:46 | |
| ab5tract | My take is that I wouldn’t name a module that can handle Puppet recipes DSL::Ruby::Puppet | 17:47 | |
| I’d just say DSL::Puppet | |||
| But that’s me :) | 17:48 | ||
| antononcube | I see. Hmm, I might be too verbose on those package names. But it is easy to loose track, I think. | ||
| This diagram shows the languages and packages addressed: | 17:49 | ||
| raw.githubusercontent.com/antononc...-state.png | |||
| Someone who is new to data science / data wrangling might get lost in the all different names and flovors. | 17:50 | ||
| ab5tract | Ah, that’s a significant web there | ||
| antononcube | 👍 | 17:51 | |
| ab5tract | Still, I just personally don’t see R::Base being a DSL | ||
| Might as well just hop one layer up and use R::* if you are going to implement a significant portion of the language | 17:52 | ||
| antononcube | Right. That works only as actions class name | ||
| Ok. That can be specialized "DSL::R::DataQueryWorkflows::Tidyverse". | 17:54 | ||
| That actually, is the more consistent naming given my previously made / named DSL packages. | 17:55 | ||
| ab5tract | Consistency++ | 18:00 | |
| Though it’s quite a mouthful :) | 18:01 | ||
| antononcube | Yeah, those names are long. | 18:05 | |
| But there is an umbrella package "DSL::Translators" that rules them all! | 18:07 | ||
| ab5tract | you can load a module from another module such that your distribution ships a `Tidyverse` so that users could just `use Tidyverse` | ||
| via `needs` | |||
| anyway, just throwing that out there as an option | |||
| antononcube | Ha! And Google knows it: | 18:08 | |
| cdn.discordapp.com/attachments/633...dc0be& | |||
| Yeah, ok. Ideally, "use DSL::Translators" should be sufficiently short, but shortcuts like "Tidyverse" or "Pandas" would be nice. | 18:10 | ||
|
18:29
hellwolf left
|
|||
| ab5tract | you could have `use DSL::Translators` make the short-names available or even immediately bring them into scope | 18:33 | |
| antononcube: github.com/ab5tract/comma-plugin/i...2477141418 | |||
| tl;dr -- in a rush, pushed a potential fix, please check if you can otherwise I will verify later myself | 18:34 | ||
|
18:50
yewscion joined
19:03
zetaaaa joined
19:16
hellwolf joined
19:57
teatwo joined
19:59
teatime left
20:00
hellwolf left
20:01
hellwolf joined
20:18
Aedil left
21:04
sena_kun joined
|
|||
| librasteve | coming to this a bit late, but I think you (anton) should find a good root name (Dan, Red, Cro eg) and then do your own subdirs off that - my inspiration is Cro (and jnthn knows waht he is doing) with many distributions and packages under the Cro::X::Y ... all the way down to class Cro::WebApp::Template::Compiled | 21:21 | |
| (ie a class call Compiled in a dir tree in a module called Cro::WebApp | 21:22 | ||
| ) | |||
| antononcube | @librasteve Yeah, I think I came to the same conclusion. | 21:38 | |
| ab5tract | None of which precludes offering short name alternatives , but I digress :) | 21:50 | |
|
21:57
zetaaaa left
|
|||
| timo | how does #raku like my hex dumper asciinema.org/a/kN82JXZd4hJWBIAzvQ6CcxJmb | 22:36 | |
| teatwo | sexy | 22:39 | |
| does the underlining mean soimething | |||
| I thought maybe it meant ascii but I'm far away from my screen and can't see | |||
| timo | the underlining on the hex bytes is for printable ascii range | 22:54 | |
|
23:01
guifa joined
23:07
sena_kun left
|
|||
| teatwo | nice | 23:21 | |
|
23:35
yewscion left
|
|||
