I have heard about Lucene a lot, that it's one of the best search engine libraries in Java. Is there any similar (as powerful) library for Ruby?
5 Answers
Well, there's Ferret, which is a port of Lucene to Ruby. Also, Lucene is very easy to use from JRuby, if that's an option for you.
Depending on your needs, you might also want to take a look at Solr, which is a higher-level front-end built on Lucene. There is a Ruby interface, solr-ruby, that interacts with Solr via HTTP.

- 39,616
- 7
- 83
- 82
-
A few people have recommended [sunspot](http://sunspot.github.io) to me, which builds upon RSolr, but I have not personally tried it. – Dave Sag Jul 30 '15 at 00:58
Ferret is what you're looking for:
"Ferret is a high-performance, full-featured text search engine library written for Ruby. It is inspired by Apache Lucene Java project."

- 272,464
- 47
- 358
- 399
-
2Well you surely can put the high-performance in question. I find it deadly slow. – Lothar Aug 23 '09 at 15:29
-
1Can you share any performance numbers? Like x seconds to return a search over y documents/rows/whatever? – Mike Buckbee Aug 23 '09 at 19:17
-
1I have no personal experience with Ferret, but judging by what I've read and this page: http://ferret.davebalmain.com/trac/wiki/MyFirstBenchmark it used to be slow but is now comparable with Java Lucene. There are performance numbers on that page. YMMV. – RichieHindle Aug 23 '09 at 19:24
I would try one of them in combination with sphinx.
Thinking Sphinx http://freelancing-god.github.com/ts/en/rails3.html
Riddle http://riddle.freelancing-gods.com/
http://blog.evanweaver.com/files/doc/fauna/ultrasphinx/files/README.html
CLucene is a cross-platform C++ port of Lucene. It can be wrapped and used also from every high-level language (there are also a few legacy Swift projects you could start with). See:
http://sourceforge.net/projects/clucene
http://clucene.git.sourceforge.net/git/gitweb.cgi?p=clucene/clucene;a=summary

- 4,472
- 1
- 30
- 37
unfortunately, in most cases, ferret is not what you're looking for, it's got recurring issues with re-indexing speed, index corruption and segfaults on the server. I think most people are going to SOLR, sphinx, and Xapian. I recall seeing some Tsearch / postgres apps mentioned, Tsearch seems to be a industrial-strength solution
Take a look here