Xapit Server

IMPORTANT: This project is currently vaporware.

For an alternative solution (also currently vaporware) see Xapit Sync.

github.com/ryanb/xapit-sync/tree/master

Xapit

Xapit is a Ruby library for interacting with Xapian, a full-text search engine.

github.com/ryanb/xapit/tree/master

How it Works

Xapit Server provides a Rack application which hosts a writable Xapian database and allows interaction through a REST API. Xapit will use this remote database for all search queries and record updates.

There are several advantages to this solution.

• all records will be updated instantly so you do not have to worry about syncing the database.
• the database is only loaded into memory once, not separately for each Rails process. This is very important if you have a large database.
• allows the Xapian database to be stored on a separate machine than the application.
• completely independent of Rails, so you can use it with other Ruby applications.

Here are some disadvantages when compared to Xapit Sync.

• additional security issues (if Rack server is exposed).
• requires a separate daemon process to constantly be running (and monitored).
• more overall traffic to/from the external process.
• greater overhead to perform search requests.
• more difficult to test in staging/development.

Setup

To use this plugin you simply have to setup a rack server. If you are using Phusion Passenger it may look like this.

  <VirtualHost *:80>
    ServerName xapit-server.example.com
    DocumentRoot /path/to/xapit-server/public
  </VirtualHost>

Then configure Xapit to use this server by passing the URL to the database path.

  # in config/initializers/setup_xapit.rb
  Xapit::Config.setup(:database_path => "http://xapit-server.example.com")

Implementation

This solution requires a more complex implementation than Xapit Sync. Not all Xapian classes are easily serialized. This may require Xapit to take extra steps to minimize the amount of traffic to the REST API.