A different approach to fuzzy finding
After growing frustrated by various fuzzy finders not being as accurate as I expected, I designed a new fuzzy finder called zf. I have mentioned zf in passing many times, but until now I haven’t taken the time to write about it in detail. In short, zf is a new terminal fuzzy finder with a ranking algorithm designed specifically for filtering filepaths. It has been rewarding to develop and use, and perhaps you will enjoy it too!
The rest of this post describes what motivated me to create zf, what makes zf different from other fuzzy finders, and some things I learned along the way.
In the summer of 2021 I noticed many of my fuzzy file searches didn’t sort the file I wanted as the first result. Motivated by frustration1 I thought, “surely I can make something better,” and I began thinking of alternate algorithms for fuzzy finding.
I theorized that prioritizing matches on file names would yield the best results. I figured that in the majority of cases, the queried name would be unique among all other file names. If this is most often the case, then it makes sense to optimize the algorithm to match based on the file name.
