***************
couchdb-lucene
***************

version 0.1 - 06/14/2010

*******
Thanks
*******

http://www.ittvis.com for permitting the initial release of this code to the public domain and giving me work time to do this.

Robert Newson - explaining how the initial java lucene for couchdb works

Itamar Syn-Hershko - helping me through some CLucene problems I was having

and the whole team behind CouchDB


*************
Requirements
*************

CLucene - http://clucene.sourceforge.net/download.shtml#2_3_2
JsonCpp - http://sourceforge.net/projects/jsoncpp/
CouchDB - http://couchdb.apache.org/ (tested with 0.11)
SpiderMonkey (as per CouchDB)

Build with either Visual Studio or Unix Make 

**************
Configuration 
**************

Tested on both Windows and Linux (executable name followed by folder to store clucene indexes, e.g. a database called 'test would be stored in c:/tmp/test in the example below)

[external]
fti=/workspace/fti/Release/fti.exe c:/tmp query

[update_notification]
update_fti=/workspace/fti/Release/fti.exe c:/tmp update

*****
demo
*****

Start CouchDB
in demo folder run python loader.py and this will create a flickr database and put the latest photo entries and tags into CouchDB. Note this is a live feed, I am not responsible for content of this feed.

e.g. curl http://localhost:5984/flickr/_fti/by_tag?q=jer* to find all tags that start with jer*

supports, include_docs=true and skip, limit requests.