Відео

Just took a look, they seem nice. Thanks for pointing it out :)

Pretty impressive solutions I'm glad you enjoyed the streams :)

Feel free to try it out, it's pretty fast :)

Je viens de retirer ça au cas ou mais bon, chuis juste un ptit youtuber

Thanks a lot :)

Ahahah, thank you again again Mr. Russia

Let me know if you find it any good :)

J'ai répondu dans l'autre commentaire

Il est possible de créé des albums privé (en ajoutant un mot de passe) J'explique ça à 5:42

I'm getting weird results when I searched nnn.

I heard of it but I guess never had the time to dive in it. yazi just crossed my way earlier. They both seem nice but I don't see a reason to switch for now

@@cacharle nnn is more on the minimalistic side of things. I don't think you'll be wowed by it, after trying yazi. I have a soft spot for nnn because it's easy to modify it, but it's definitely not the most feature packed one.

@@khatakeddb I see what you mean

Double me! 😅

Indeed, that used to be my opinion aswell but some other comments have pointed out that little endian is faster when you want to cast a pointer to int32 to int16 or int8 (just go look for the other comments for a more detailed explanation) So know I think I appreciate little endian more 🙂

I've never seen that before, thanks Seems quite overkill for this usecase but I guess the goal of doing AoC in Haskell is to relearn Haskell so why not :D

I tried it and it works, altho the spacing is a bit strange: (read@Int) doesn't work (read @ Int) aswell but (read @Int) does 😄

Indeed, it works and feels a bit cleaner, thanks!

my contact?

Indeed, another comment mentioned that and it was a nice discovery for me

@@cacharle I've been dealing with a bunch of low level embedded stuff and this is great for it. Thanks for the regular short tutorials they are great. Keep it up!

I do but I'm better at C The newest C++ standards are discouraging tbh. It's a crazy amount of complexity and it seems poorly organized. (the error messages for template related things are also crazy) If I want that kind of complexity, I'd rather use Rust, which is also a complex language but much better organized imo

Oooooooh thank you for that! super interesting

Really? That sounds odd..

It's not a template. It's line copy shortcut. In vscode, I would do it with cmd+c & cmd+v on a line without selecting any part of the line.

Indeed, it was just a copy/paste (yy then p in vim)

@@cacharle can you give use you neovim configuration?

@@IsdaouneAbderrahmane github.com/cacharle/dotfiles Look into config/nvim

I like having more screen space for the code, I'll keep it like this unless too many people complain

@@cacharle i'm gonna be honest, i kinda agree with @zanehisako1416 on this one. The Transparency is not the problem. It's the weird artifacts around you that are a bit distracting. (maybe a greenscreen/bluescreen you can "chroma-key" out would be a nice alternative) Your video was very informative nontheless! keep it up!

It works well enough!

Yes, I do have plan of making more involved videos in which I build some small project The reason I haven't done them yet is simply lack of time

Nice timing, the file you output will definitely be portable with this :D

Could you give me your C code and the output of running `cat -e` on your text file? I do get the `file!1` when I try with: ❯ /usr/bin/cat -e getline_test.txt hello$ from$ $ my$ file!1$ ❯ /usr/bin/cat getline_test.c #include <stdio.h> #include <stdlib.h> int main(void) { FILE *f = fopen("./getline_test.txt", "r"); char *line = NULL; size_t line_size = 0; ssize_t read_size = 0; while ((read_size = getline(&line, &line_size, f)) != -1) printf("%s", line); free(line); fclose(f); return 0; }

My guess would be that you don't have a newline at the end of your file (I explain this at 5:15) If you're trimming the newline at the end of the line. You should add a condition to check if there is one: if (read_size > 0 && line[read_size - 1] == ' ') line[read_size - 1] = '\0';

Now I understand. Thank you!

@@cacharle code: #include <stdio.h> #include <stdlib.h> #include <unistd.h> int main() { FILE *file = fopen("text.txt", "r"); char *line; size_t line_buffer_len = 0; ssize_t read_size = 0; while ((read_size = getline(&line, &line_buffer_len, file)) != -1) { line[read_size - 1] = '\0'; printf("line: '%s', line_buffer_len: '%zu', read_size = '%zu' ", line, line_buffer_len, read_size); } return 0; }

Oh yes, it looks like a modern rewrite of script with some server to upload the recordings Thank you for the recommendation

That's unfortunate. Someone mentioned the flameshot program in another comment

Because it was more minimal and easier to setup. I did try gitea on my home kubernetes cluster a few months ago and I can confirm that I am 100% happy with my cgit + git-shell approach, gitea is annoyingly complex, especially for one person. The only thing my personal git server is missing now is a git lfs integration

I'm glad you're able to take somthing out of the videos. Indeed that permissions make sense in binary, it's just a convention to write them in octal (and it's also just more convenient)

i don't know how helpful this is, but i usually try to read from right to left, so: `int * const x` becomes "x is a const pointer-to int" and `const int * x` becomes "x is a pointer-to const int"

I know you're supposed to read it in a "spiral" direction but I never learned that. Instead you can use this website if you get confused: cdecl.org/

C is not confusing at all 😅