This is a DataMapper plugin for Ferret.
For a single process site, use the ferret index directly:
DataMapper.setup :search, "ferret:///path/to/index"
For a multi-process site, use the distributed index by running ‘ferret start` inside your project’s directory and then setting up the :search repository:
DataMapper.setup :search, "ferret:///tmp/ferret_index.sock"
require ‘rubygems’ require “dm-core” require “dm-is-searchable”
DataMapper.setup(:default, “sqlite3::memory:”) DataMapper.setup(:search, “ferret://#{Pathname(__FILE__).dirname.expand_path.parent + ”index“}”)
class Image
include DataMapper::Resource property :id, Serial property :title, String is :searchable # this defaults to :search repository, you could also do # is :searchable, :repository => :ferret
end
class Story
include DataMapper::Resource property :id, Serial property :title, String property :author, String repository(:search) do # We only want to search on id and title. properties(:search).clear property :id, Serial property :title, String end is :searchable
end
Image.auto_migrate! Story.auto_migrate! image = Image.create(:title => “Oil Rig”); story = Story.create(:title => “Big Oil”, :author => “John Doe”) }
puts Image.search(:title => “Oil Rig”).inspect # => [<Image title=“Oil Rig”>]
# For info on this, see DM::Repository#search and DM::Adapters::FerretAdapter#search. puts repository(:search).search(‘title:“Oil”’).inspect # => { Image => [“1”], Story => [“1”] }
image.destroy
puts Image.search(:title => “Oil Rig”).inspect # => []