00:03 ky1 left, ky1 joined 00:22 ky1 left 00:23 ky1 joined 00:32 ky1 left 00:33 ky1 joined 00:42 ky1 left 00:43 ky1 joined 00:52 ky1 left 00:53 ky1 joined 01:02 ky1 left 01:03 ky1 joined 01:12 ky1 left 01:13 ky1 joined 01:22 ky1 left 01:23 ky1 joined 01:28 Manifest0 left 01:29 stanrifkin left 01:30 hulk joined 01:31 kylese left 01:32 ky1 left 01:33 ky1 joined 01:42 ky1 left 01:43 ky1 joined 01:45 Aedil joined 01:53 ky1 left, ky1 joined 02:02 ky1 left 02:03 ky1 joined 02:05 hulk left 02:06 kylese joined 02:13 ky1 left, ky1 joined 02:15 kylese left, kylese joined 02:22 ky1 left 02:23 ky1 joined, Xliff joined 02:30 kylese left 02:32 ky1 left 02:33 ky1 joined 02:34 kylese joined
Xliff \o 02:37
Hass setting function pointers through NativeCall gotten any easier?
timo: I remember you came up with a way to do it, but I can't find your example through the search on logs.liz.nl 02:38
02:42 ky1 left 02:43 ky1 joined
Xliff use NativeCall; sub sprintf-v(Blob, Str, & () ) returns int64 is native is symbol('sprintf') { * }; sub set_func_pointer( \func, &sprint) { my $buf = buf8.allocate(20); my $len = &sprint($buf, '%lld', func); $buf.say; my $b =$buf.subbuf(^$len).decode.Int; say $b; Pointer.new($b) }; sub a { 1 }; say set_func_pointer(&a, &sprintf-v) 02:51
evalable6 Buf[uint8]:0x<31 33 39 38 31 37 39 37 33 35 30 …
Xliff, Full output: gist.github.com/f19f20d37f92696b4f...a569b4f70b
02:53 ky1 left, ky1 joined
Xliff What does "Native callback must be a code handle" mean? 02:57
03:02 ky1 left 03:03 ky1 joined 03:12 ky1 left 03:13 ky1 joined
antononcube @librasteve Here is a Markdown file with lists of packages for doing Data Science and LLMs: github.com/antononcube/RakuForPred...nd-LLMs.md 03:16
Xliff What is the proper way to test for Any? 03:19
Not .defined, I want to know if a container contains the type object Any 03:20
m: say Any =:= Any
camelia True
Xliff m: say Any === Any
camelia True
Xliff m: say Any === Int
camelia False
Xliff m: my $a; say $a === Int
camelia False
Xliff m: my $a; say $a =:= Int
camelia False
Xliff m: my $a; say $a =:= Any 03:21
camelia False
03:22 ky1 left 03:23 ky1 joined 03:33 ky1 left, ky1 joined 03:43 ky1 left, ky1 joined
ugexe m: say 1.isa(Any); say Any.isa(any); say Mu.isa(any); 03:50
camelia ===SORRY!===
Function "any" may not be called without arguments (please use () or whitespace to denote arguments, or &any to refer to the function as a noun, or use .any if you meant to call it as a method on $_)
at <tmp>:1
------> say 1…
ugexe m: say 1.isa(Any); say Any.isa(Any); say Mu.isa(Any);
camelia True
True
False
ugexe yeah that isn't what you asked
03:53 ky1 left, ky1 joined 04:02 ky1 left 04:03 ky1 joined 04:11 teatime left 04:12 ky1 left 04:13 ky1 joined 04:22 ky1 left 04:23 ky1 joined 04:32 ky1 left 04:33 ky1 joined 04:43 ky1 left, ky1 joined 04:52 ky1 left 04:53 ky1 joined 05:03 ky1 left, ky1 joined 05:13 ky1 left, ky1 joined 05:23 ky1 left, ky1 joined 05:32 ky1 left 05:33 ky1 joined 05:42 ky1 left 05:43 ky1 joined 05:52 ky1 left 05:53 ky1 joined 06:03 ky1 left, ky1 joined 06:13 ky1 left, ky1 joined 06:22 ky1 left 06:23 ky1 joined 06:32 ky1 left 06:33 ky1 joined 06:43 ky1 left, ky1 joined 06:53 ky1 left, ky1 joined
librasteve thanks 07:00
07:02 ky1 left 07:03 ky1 joined 07:12 ky1 left 07:13 ky1 joined 07:22 ky1 left 07:23 ky1 joined 07:32 ky1 left 07:33 ky1 joined 07:43 ky1 left, ky1 joined 07:46 sena_kun joined 07:53 ky1 left, ky1 joined 08:02 ky1 left 08:03 ky1 joined 08:12 ky1 left 08:13 ky1 joined 08:22 ky1 left 08:23 ky1 joined
Xliff ugexe: Yeah. Thanks for trying, though. I didn't know about .isa 08:32
08:33 ky1 left
Xliff ugexe: Is that related to .ACCEPTS? 08:33
08:33 ky1 joined
Xliff ugexe: So I'm down to this... 08:33
m: say Any === Any; say 1 === Any; say Mu === Any
camelia True
Cannot resolve caller infix:<===>(Mu:U, Any:U); none of these signatures matches:
($? --> Bool::True)
(\a, \b --> Bool)
(Enumeration:D $a, Enumeration:D $b --> Bool)
(Int:D $a, Int:D $b --> Bool)
(int $a, int $b --…
Xliff m: say Any === Any; say 1 === Any; 08:34
camelia True
False
Xliff say "===" doesn't work on Mu, oddly enough. We won't need to go that far down,
08:43 ky1 left, ky1 joined 08:53 ky1 left, ky1 joined 09:03 ky1 left, ky1 joined 09:12 ky1 left 09:13 ky1 joined 09:22 ky1 left 09:23 ky1 joined 09:27 Sgeo left 09:32 ky1 left 09:33 ky1 joined 09:37 Manifest0 joined 09:42 ky1 left 09:43 ky1 joined
lizmat m: say Mu === Mu 09:45
camelia Cannot resolve caller infix:<===>(Mu:U, Mu:U); none of these signatures matches:
($? --> Bool::True)
(\a, \b --> Bool)
(Enumeration:D $a, Enumeration:D $b --> Bool)
(Int:D $a, Int:D $b --> Bool)
(int $a, int $b --> Bool)…
lizmat heh
m: say IterationEnd === IterationEnd 09:46
camelia Cannot resolve caller infix:<===>(); none of these signatures matches:
($? --> Bool::True)
(\a, \b --> Bool)
(Enumeration:D $a, Enumeration:D $b --> Bool)
(Int:D $a, Int:D $b --> Bool)
(int $a, int $b --> Bool)
(uin…
lizmat heh
09:52 ky1 left 09:53 ky1 joined 10:02 ky1 left 10:03 ky1 joined 10:06 lizmat_ joined 10:09 lizmat left 10:12 ky1 left 10:13 ky1 joined 10:22 ky1 left 10:23 ky1 joined
c.a53 Hello. Are the final release binaries of Comma archived somewhere? 10:27
(I know about the current plugin develpoment, but it still crashes a lot on my Windows machine inside IDEA) 10:28
10:32 ky1 left, lizmat_ left 10:33 lizmat joined, ky1 joined 10:43 ky1 left, ky1 joined 10:53 ky1 left, ky1 joined
antononcube @c.a53 Most likely not. Maybe this can repository can give some hints: github.com/Raku/intellij-ide-plugi...ource-dump 10:54
But, for me, the old CommaIDE was crashing a lot, on macOS. The currently actively developed plugin is producing hangs initially, but those gradually got resolved. 10:56
ab5tract c.a63: what version of the comma plugin have you been running? Stability should be much stronger in the last two or three beta releases 11:02
11:03 ky1 left, ky1 joined
ab5tract It also requires IDEA 2024.2+ 11:09
11:12 ky1 left 11:13 ky1 joined
c.a53 2.0, beta 21 on IDEA 2024.2.3 Ultimate 11:15
When opening a project (previously edited by Comma 2023.04 Community) it tells me to select an SDK and then, before I have any time to navigate through the file structure to find Rakudo, whole IDE freezes and must be killed by task manager. 11:17
Previously I was trying beta 15, created a new project and before having an opportunity to write more than 25 lines it froze during editing in the same way, than again after reopening and few edits. Inside IDE REPL didn't work at all. 11:21
11:22 ky1 left
My current Rakudo version is 2024.08 11:22
(Rakudo Star distribution) 11:23
11:23 ky1 joined
antononcube I have not tried the REPL yet with new comma-plug-in . 11:23
ab5tract Yeah I haven’t focused on the REPL so thanks for raising it to my attention 11:26
What I can say is that old projects are a problem
If at all possible use ‘New Project from Existing Sources’ 11:27
And better yet, it’s an option, is clone the repo and open it that way
c.a53 I use default non unicode locale on Windows 10 (as switching to unicode breaks some other programs I use), so that might be a problem as well (but it works mostly fine in the old standalone Comma 2023.04 installation... OK unless I start inputting strings like "ěšč"). 11:28
ab5tract As far as I know that shouldn’t be a problem
But there are indexes built from an old versions being loaded with different class names by the new plugin 11:29
As well as some other configuration fields that have changed in the project settings files
I try to identify old comma projects and force a clean on them but using New Project from Existing Spurces already does that 11:30
So I highly recommend against opening an old project directly 11:31
Also, I try to respond to tickets or reported issues almost immediately, so please consider documenting your issues on GitHub or sharing them here 11:32
Thanks for trying the new comma and with a bit of luck and time we can nail down any annoyances 11:34
11:42 ky1 left 11:43 ky1 joined
c.a53 Tried it as a new project now, and it allowed me to select the SDK and seems to work. I'll try to report any issues I run into. Thanks for help and all the work put into keeping Comma alive. 11:44
ab5tract Awesome! Ill take a look at the repl situation next 11:45
11:52 ky1 left 11:53 ky1 joined 11:56 sena_kun left 12:02 ky1 left
c.a53 👍 12:03
12:03 ky1 joined 12:22 ky1 left 12:23 ky1 joined 12:32 ky1 left 12:33 ky1 joined
antononcube @ab5tract Personal opinion — yes having a Raku REPL in IntelliJ / Comma is very nice, but I do not see it as high priority. (A moderate one.) 12:39
With the latest refactoring of “Terminal::LineEditor” I can use Raku in Terminal.app (on macOS.) I also use VSCode for Raku explorations / programming / scripts. 12:41
12:43 ky1 left, ky1 joined
BTW, for a long time — until a week ago — in order have a decent Terminal.app “experience” with Raku I had to use “Jupyter::Kernel”. 12:44
@c.a53 Do you know / use AutoHotKey, by any chance? If yes, how easy it is to write a script that reproduces your crashes using IntelliJ and Comma? 12:47
12:52 ky1 left 12:53 ky1 joined 13:02 ky1 left 13:03 ky1 joined
ab5tract antononcube: I agree with you about the built in REPlL. I’ll probably still look into it because I generally want feature parity with the older editions where possible. Or at least removing the indication that it is a feature, until it can be added back in later. 13:03
I also have a long term plan of seeing we can get Jupyter running in IntelliJ’s built in browser 13:05
antononcube That would be nice. I had success running Jupyter notebooks with Raku kernel in IntelliJ in the past, but the most recent changes in IntelliJ seem to have broken that. 13:06
c.a53 @antononcube My knowledge of AutoHotKey is minimal, so I cannot help here. If AutoHotKey was able to record the key presses during editing, then replay them to reproduce the crash... yes, that would be useful. It's hard to remember what precisely I was typing when the crash happened and as the crash is mostly associated with complete IDE freeze, it's usually impossible to use the IntelliJs issue reporting dialogue after 13:10
the crash (it may display, but cannot be interacted with).
antononcube Yeah, I never successfully used IntelliJ's crash dialogue. (Tried many time with CommaIDE.) 13:11
13:12 ky1 left 13:13 ky1 joined 13:18 xinming left 13:19 xinming joined 13:22 ky1 left 13:23 ky1 joined 13:32 ky1 left 13:33 ky1 joined 13:42 ky1 left 13:43 ky1 joined 13:53 ky1 left, ky1 joined 14:02 ky1 left 14:03 ky1 joined 14:22 ky1 left 14:23 ky1 joined 14:32 ky1 left 14:33 ky1 joined 14:52 ky1 left 14:53 ky1 joined
holmdunc JetBrains aren't packing the parachutes properly! 14:53
El_Che need to test the new comma. Hopefully will the plugin road limit the breakage with intellij updates 14:54
tellable6 2024-10-24T22:29:33Z #raku-dev <jdv> El_Che the 2024.10 release happened
El_Che ok, kicking the builds 15:01
15:02 ky1 left 15:03 ky1 joined 15:12 ky1 left 15:13 ky1 joined
El_Che t/02-rakudo/15-gh_1202.t and t/12-rakuast/xx-fixed-in-rakuast.rakutest are complaining :) 15:14
15:22 ky1 left 15:23 ky1 joined
El_Che but it looks they are floppers 15:26
15:32 ky1 left 15:33 ky1 joined 15:42 ky1 left 15:43 ky1 joined
El_Che github.com/nxadm/rakudo-pkg/releas...g/v2024.10 15:46
published, packaes in the respective repositories
15:52 ky1 left 15:53 ky1 joined 16:03 ky1 left, ky1 joined 16:13 ky1 left, ky1 joined 16:33 ky1 left, ky1 joined 16:43 ky1 left, ky1 joined 16:52 ky1 left 16:53 ky1 joined 17:03 ky1 left, ky1 joined 17:13 ky1 left, ky1 joined 17:22 ky1 left 17:23 ky1 joined 17:33 ky1 left, ky1 joined 17:42 ky1 left 17:43 ky1 joined 17:52 ky1 left 17:53 ky1 joined 18:01 maylay_ left 18:03 ky1 left, ky1 joined 18:06 maylay joined 18:12 ky1 left 18:13 ky1 joined 18:22 ky1 left 18:23 ky1 joined 18:32 ky1 left 18:33 ky1 joined 18:42 Aedil left 18:43 ky1 left, ky1 joined 18:53 ky1 left, ky1 joined 19:02 ky1 left 19:03 ky1 joined 19:12 ky1 left 19:13 ky1 joined 19:15 Sgeo joined 19:23 ky1 left, ky1 joined 19:32 ky1 left 19:33 ky1 joined 19:42 ky1 left 19:43 ky1 joined 19:52 ky1 left 19:53 ky1 joined, cm left 20:02 ky1 left 20:03 ky1 joined 20:12 ky1 left 20:13 ky1 joined 20:20 stanrifkin joined 20:22 ky1 left 20:23 ky1 joined 20:32 ky1 left 20:33 ky1 joined 20:42 ky1 left 20:43 ky1 joined 20:52 rapidcow joined, ky1 left 20:53 ky1 joined
librasteve El_Che: I recommend the new comma - but as you mention, it will need active development to keep pace with the current IntelliJ build ... before ab5tracts' rescue, it would often be necessary to have an older IntelliJ version just to run the Comma plugin in parallel with the current version for new developments 20:54
20:56 rapidcow left
antononcube @librasteve I still keep and use the most recent "old" CommaIDE for macOS. 21:02
21:02 ky1 left 21:03 ky1 joined
librasteve me too - in general it is the best experience for pure raku code dev and I would love it if we could continue to enjoy both a standalone CommaIDE and an IntelliJ IDEA Comma plugin - my workflow is to use CommaIDE for pleasure and IntelliJ + Comma for work (and this is useful for hybrid projects - raku + Rust for example) 21:04
but I would rather we (ie @ab5stract) ensured that the Comma plugin is kept in synch with the latest IntelliJ (which I suspect is quite demanding) rather than having both CommaIDE and the Comma plugin lag the current version 21:06
antononcube Right, I mentioned this a few times -- I use CLion with Comma-plugin for NativeCall projects.
Here is an example: 21:07
cdn.discordapp.com/attachments/633...225e2&
librasteve yes - I agree ... CommaIDE is a "very nice to have" but the workman tool is the Comma Plugin 21:08
did you get Jupyter raku kernel to run in IntellllliJ ?
21:08 sena_kun joined
antononcube Before the latest updates of IntelliJ yes. Maybe they still do in PyCharm. (I will check.) 21:09
librasteve hmmm - whats the latest version number that works will Jupyter raku? (maybe I will downgrade, maybe that will kill Comma Plugin) 21:12
21:12 ky1 left 21:13 ky1 joined
antononcube I am not sure. I have to say I am getting too lazy to specify the "managed Jupyter server" in IntelliJ... 21:16
My biggest problem with IntelliJ Jupyter with Raku-kernel is that I can plot graphics using "JavaScripdt::D3".
... cannot plot... 21:17
librasteve yeah - ok probably better to run jupyter outside the IDE in that case I guess 21:18
(too many movng parts)?
antononcube @librasteve Just checked -- with the latest IntelliJ Jupyter with Raku-kernel can be run: 21:22
cdn.discordapp.com/attachments/633...14bc8&
21:22 dustinm`_ left, ky1 left 21:24 ky1 joined
librasteve ok - I will try 21:25
El_Che yes, the IDE was nice, but I suppose most of us are not pure Raku devs and need support for other languages as well 21:32
21:32 ky1 left 21:33 ky1 joined 21:37 dustinm` joined
antononcube @El_Che VS Code is doing that nicely. I wanted to use Raku with Emacs, but the Raku-mode for Emacs is too basic. 21:39
21:42 ky1 left 21:43 ky1 joined
librasteve &afk 21:47
timo would be neat if someone were to build a tree-sitter grammar for raku, though i'm not sure if tree sitter is powerful enough to make some of the more advanced things not a royal pain 21:49
and it's probably not easy to make tree-sitter grammars react to customizations like custom operators, and especially not to slangs 21:54
22:03 ky1 left, ky1 joined 22:10 cm joined 22:13 ky1 left, ky1 joined 22:23 ky1 left, ky1 joined 22:42 ky1 left 22:43 ky1 joined 22:52 ky1 left 22:53 ky1 joined 22:59 sena_kun left 23:02 ky1 left 23:03 ky1 joined 23:12 ky1 left 23:13 ky1 joined 23:22 ky1 left 23:23 ky1 joined 23:32 ky1 left 23:33 ky1 joined 23:42 ky1 left 23:43 ky1 joined 23:44 Xliff left 23:52 ky1 left 23:53 ky1 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!