This plugin SHOULD NOT be used anymore, please use Sunspot

acts_as_solr Rails plugin

This plugin adds full text search capabilities and many other nifty features from Apache's Solr to any Rails model. It was based on the first draft by Erik Hatcher.

Current Release

The current stable release is v1.0 and was released on 20-05-2009.

This plugin is intended as a way to "upgrade" and older or incompatible version of acts_as_solr you might be using. It's unlikely that I'm going to add new features to this specific fork and if you're starting up a new project and want to use Solr you should definitely look for http://github.com/outoftime/sunspot/

Changes

Please refer to the CHANGE_LOG

Installation

script/plugin install git://github.com/mauricio/acts_as_solr.git

Requirements

• Java Runtime Environment(JRE) 1.5 aka 5.0 or higher http://java.sun.com/

Features

• Fully compatible with will_paginate view helpers and :page/:per_page options
• Using Solr 1.3.0

Things to be done

• Examples using the "did you mean?" feature

Mantainer

Maurício Linhares (mauricio dot linhares AT gmail dot com)

Original Authors

• Erik Hatcher
• Thiago Jackiw

Release Information

Released under the MIT license.

Migrating from older versions

If you're migrating from older acts_as_solr versions (that probably came from another repo and not http://github.com/mauricio/acts_as_solr/tree/master ) all you have to do is to copy the "jetty/solr/conf" folder to your "RAILS_ROOT/config/solr". Config files now live in your application and not the plugin.

More info

http://wiki.github.com/mauricio/acts_as_solr

Basic Usage

# Just include the line below to any of your ActiveRecord models:
class Book < ActiveRecord::Base
acts_as_solr
end
With this acts_as_solr will index all fields in your active_record model
that are not it's primary key, crated_at or updated_at fields.
Or if you want, you can specify only the fields that should be indexed:
class Book < ActiveRecord::Base
acts_as_solr :fields => [:name, :author]
end
The "fields" that are going to be indexed don't need to be database fields,
they can be simple methods on your model:
class Book < ActiveRecord::Base
belongs_to :author

acts_as_solr :fields => [:name, :author_name]

def author_name
  self.author.name if self.author
end
end
And this is all there is to it, you don't need to get your hands dirty on
Solr schema or document files, only if the default config doesn't suit your
needs. After every save/destroy the Solr index for that object is going to be
updated, so all you need to do is configure your active_record object as above
start the Solr server (rake solr:start) and be done with it.
If you're starting up from an already populated database, you'll have to build
and index based on that data. To do this, start the Solr server (rake solr:start)
and then call the reindex task (rake solr:rebuild_index), this will rebuild
the index based on all objects currently stored on your database.
Then to find instances of your model, just do:
books = Book.find_by_solr(query)
"query" is a string representing your query
You can find out more about Lucene's query sintax here:
http://lucene.apache.org/java/2_4_1/queryparsersyntax.html
Finding using pagination:
books = Book.find_by_solr( query, :page => 2, :per_page => 10 )
The object returned by a find_by_solr call is an ActsAsSolr::SearchResults
object where you can find out more about the response Solr sent back to you and
the real active_record objects returned by your query.
This object works just like any other Enumerable (as an Array) and you can
call any Enumerable method on it. It's also completely compatible with
will_paginate view helpers.
books = Book.find_by_solr( query, :page => 2, :per_page => 10 )
books.each do |book| # this "book" variable is a real Book object from the database
puts "#{book.name} - #{book.author_name} - #{book.year}"
end
Please see ActsAsSolr::ActsMethods for a complete info