isti115

#FSharp https://github.com/isti115/advent-of-code/tree/master/2024/day-01

I usually choose to learn a new language each year using these great little puzzles, this time it's FSharp. This naturally means that my solutions will be poorly written at first, since I'm not yet familiar with the language and make do with what I have. For example I'm pretty sure that there should be a better way to parse today's input instead of this monstrosity:

let pairs = lines |> Seq.map (fun l -> Regex.Matches(l, "(\d+)\s+(\d+)"))

let numpairs =
    pairs |> Seq.map (Seq.head >> _.Groups >> Seq.tail >> Seq.map (_.Value >> int))

let numtups = numpairs |> Seq.map (fun p -> (Seq.head p, p |> Seq.tail |> Seq.head))

~~Also, I was pretty surprised that I couldn't find the unzip function for sequences. I would've expected that to be present after using some other functional languages, such as Haskell and Scala. 🤔~~

Edit: Scratch that, I just need to convert the sequence into a list first... 🤦‍♂️ It actually makes complete sense. https://fsharp.github.io/fsharp-core-docs/reference/fsharp-collections-listmodule.html#unzip Also, I have managed to clean up the parsing a little bit.

Thanks for your input! To me it seems like Nemo only counts the direct descendants and doesn't recurse, which makes it less useful for this purpose, but still nice to know!

Thank you for the idea! I didn't know about the --inodes flag before, this seems like a viable solution for systems where I can't / don't want to install additional software!

Oh, wow, thank you! I had ncdu installed, but it was an older version, which didn't yet have this feature. Now that I updated to the newest (Zig based 🎉) release this looks perfect for my needs!

Thanks! 😀 I've been meaning to post about that as well in order to keep feeding Lemmy with content, but couldn't find either an active alternative to ErgoMechKeyboards or Battlestations. 🤔

Well, I'd suggest asking around your nearest toddler, that's how I managed to borrow this one. 😀 Probably it has been inherited through several generations, that might explain it's wisdom!

Thanks for the advice, but lo and behold, the issue was actually caused by a newly introduced generic parameter not being defined after a dependency update. 🙃 Oh, the wonders of the JS/TS ecosystem...

Whoa, I just checked out some images, that one is on another level as well for sure. 😀 If he's as good at fighting bugs as criminals, you're in good ~~hands~~ wings!

Let me preface this by clarifying that I don't claim to have the one and only right explanation that everyone should accept, I'd just like to point out that this theory also exists: https://hermeneutics.stackexchange.com/a/43799

Yes, quite the role reversal between the parts from yesterday. 😀

Yepp, "Lateral with Tom Scott, Episode 58: Straight-line sports" in case anyone's looking for it. https://lateralcast.com/

River is actually really nice! I have been a Sway user for multiple years before, but as it reached parity with i3, it was deemed "feature complete" (which is reasonable, as it was made to be a drop-in replacement), so I switched to Hyprland, but after making a small contribution I found the development too chaotic, so I started looking for something else.

I have been really happy with River. It still has some small shortcomings (e.g. minor visual glitches with fullscreening, and some less used options still missing), but the developers are very responsive on IRC (which I'm actually not a huge fan of, this was the first time I had to use it, but it's OK for what it is) and it being written in Zig is a huge selling point for me, as I don't have to write any C or C++ to extend it (my first PR is already on the way) when something I'd like to do is not yet possible.

If you're willing to spend a while getting to know it and setting up the environment (it relies on external programs for some functionality that is built-in in other compositors, such as monitor management), then I'd definitely say that it's worth a shot!