3 ms·
Ask HN: Full-text search of Unix files?
I have a lot of notes organized as text files in a directory hierarchy on my Unix machine. I would like to index and search these notes, using full-text search and approximate matching (so grep does not suffice). I prefer a command-line utility. I am looking for a simple application that works out-of-the-box, no fuss, no muss, so something like Lucene seems overblown. Any suggestions?
- timf 18y agoSee if Beagle fits for you (it can do stemming, for example). http://beagle-project.org/Searching_Data http://beagle-project.org/Searching_Data There's a commandline interface to it. I'm not a beagle user myself, but it looks like something you should at least check out.
- gtani 18y agoedit(I'm assuming, since i'm always wallowing in python or rails, that google desktop, regain, don't have the control over stemming and tokenization that you want...) here's regain http://regain.sourceforge.net/ http://regain.sourceforge.net/ to my knowledge, Aksyonoff's sphinx only hits mysql and postgres. The ferret lib for ruby lists this funcitonality, i haven't tried it http://datanoise.com/articles/2006/9/5/fun-with-ferret http://datanoise.com/articles/2006/9/5/fun-with-ferret
- lazyant 18y agoA few years ago I wrote a little script for what you're describing or similar (but very simple, no indexing) using a Perl module for fuzzy/approximate matching; I think it was this one: http://search.cpan.org/~jhi/String-Approx-3.26/Approx.pm http://search.cpan.org/~jhi/String-Approx-3.26/Approx.pm