https://seb.jambor.dev/posts/improving-shell-workflows-with-fzf/
How to use fzf. Finding patterns in shell workflows and automating them using fzf.
function delete-branches() {
git branch |
grep --invert-match '\*' |
cut -c 3- |
fzf --multi --preview="git log {} --" |
xargs --no-run-if-empty git branch --delete --force
}
This is really slick.
Those of you reading this might also be interested in two_percent, which is a fork of skim, which in turn is a Rust implementation of fzf. two_percent is faster, more efficient and uses less memory than fzf, which is especially noticeable with large inputs.
My most direct use of fzf is to search large result sets for something I can't 100% remember the name or location of, so this actually sounds nice. I've managed to get fzf to slow down a few times and... well, I'm sure as hell not organizing that folder structure.
It's faster and more memory efficient basically. skim also appears to have been abandoned (no updates in over an year), whereas two_percent is being actively developed.
