Hacker News new | past | comments | ask | show | jobs | submit login

I made one related to this xkcd:

https://xkcd.com/1360/

Several of us, which I found out by reading comments here, have each developed on our own our little utility to do the same thing: clean up / deduplicate / get rid of old files.

And it's not just a matter of using fdupes or jdupes: I created a database where I tag files based on their cryptographic hash (and size, for speed). For example a file can be tagged as "file with that hash can always be deleted" or "file with that hash can always be renamed to xxx".

I then can run my "bezerker" on entire filesystems and it'll delete files tagged for deletion.

I used the Blake3 hash for speed but I also saved which hash is used, so if something much faster comes up, I can switch (I'd still need to use Blake3 for the hashes already in the DB but that's no issue).

I did this because I got tired of having for example the wife and mother-in-law coming with yet another USB stick, with yet another "directory inside a directory inside a directory with family pictures and movies from 2017" or whatever and having to delete for the ten times that same blurry picture or that same little 3 seconds movie taken by mistake etc.

So, basically, instead of lamenting, in that xkcd 1360 style, about old files and old directories, I made my own little utility to get rid of the problem once and for all.

Works wonder but as you asked: it's really for my own use and I kept it private.

That said I take it there's at least some interest in a public / open-source utility doing that: otherwise there wouldn't be several of us here on HN (and certainly elsewhere too) developing our own solution for this problem.




Consider applying for YC's Summer 2025 batch! Applications are open till May 13

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: