exa VS fd

Even though ls is a good-old standby, I prefer the aliases for ll and la that rely on exa. These aliases default to showing important information like git status and can even prevent me from needing to launch subsequent, exploratory commands.

https://github.com/ogham/exa already exists. Haven't looked too closely at feature differences though.

exa is installed and acting as replacement for good old ls ... https://the.exa.website/

I've used a program called exa. (You can get it from cargo.) To get it to replace the "ls" command, I simply created an alias to "ls" command in my .zshrc file.

This page doesn’t actually list any features or explain why you should use exa instead of ls. The homepage [0] is more useful in this regard.

0: https://the.exa.website/

Tools that make my life easier would be fzf, fd, exa, micro, ripgrep, bat etc.

I recently came across exa and only a few months prior learnt about micro (and have switched from nano).

Exa

I found a directory that seems to control output for exa, but I didn't dig too much further than that: https://github.com/ogham/exa/tree/master/src/output

I love using https://github.com/ogham/exa

I'm aware of them, but not familiar. I always use tools like Filelight, ncdu, and fd and do my own cleaning.

This is what I use:

https://github.com/sharkdp/fd

Which you can install as a binary, or via cargo. fd is spectacular.

To that you can add: https://github.com/junegunn/fzf

Which you can bind to a key in your shell for convenience.

Tools like skim work great with others like fd to provide additional functionality when it comes to locating files and filtering through search results. I also use skim's sk command pretty heavily in other scripts when interactivity is required.

Alt to find: https://github.com/sharkdp/fd and fzf https://github.com/lotabout/skim

Also, at -e 9, cjxl only uses 1 thread, so you could process multiple files in parallel. Unfortunately, if your cpu is slow, you probably don't have much memory and at those resolutions it could be hard to max out your cpu, but maybe you could at least cut the time in half. I usually use https://github.com/sharkdp/fd for that. You could use those options: fd -e png -j 8 -x cjxl -d 0 -e 9 -E 3 --num_threads 0 '{}' '{.}.jxl' Look at your memory usage while transcoding 1 file and adjust the -j flag so that you max out either your memory or cpu.

Btw, I recommend https://github.com/sharkdp/fd, it handles multithreading on its own and seems to also have windows builds. I usually just run:

fd is much faster than find, with IMO better ergonomics. Just like find, it has an --exec flag that you can use instead of piping the output to rm (or --exec-batch, to make a single call to rm).