elastic_searchable 0.5.0 → 0.6.0

data/VERSION

@@ -1 +1 @@
-0.5.0
+0.6.0

data/elastic_searchable.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{elastic_searchable}
-  s.version = "0.5.0"
+  s.version = "0.6.0"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Ryan Sonnek"]
-  s.date = %q{2011-03-04}
+  s.date = %q{2011-03-08}
   s.description = %q{integrate the elastic search engine with rails}
   s.email = %q{ryan@codecrate.com}
   s.extra_rdoc_files = [

data/lib/elastic_searchable/callbacks.rb

@@ -12,10 +12,10 @@ module ElasticSearchable
         self.class.delete_id_from_index_backgrounded self.id
       end
       def update_index_on_create
-        index_in_elastic_search :create
+        reindex :create
       end
       def update_index_on_update
-        index_in_elastic_search :update
+        reindex :update
       end
     end
   end

data/lib/elastic_searchable/index.rb

@@ -53,6 +53,35 @@ module ElasticSearchable
         ['', index_name, action].compact.join('/')
       end
 
+      # reindex all records using bulk api
+      # options:
+      #   :scope - scope the find_in_batches to only a subset of records
+      #   :batch - counter to start indexing at
+      #   :include - passed to find_in_batches to hydrate objects
+      # see http://www.elasticsearch.org/guide/reference/api/bulk.html
+      def reindex(options = {})
+        batch = options.delete(:batch) || 1
+        options[:batch_size] ||= 1000
+        options[:start] ||= (batch - 1) * options[:batch_size]
+        scope = options.delete(:scope) || self
+        scope.find_in_batches(options) do |records|
+          puts "reindexing batch ##{batch}..."
+          batch += 1
+          actions = []
+          records.each do |record|
+            next unless record.should_index?
+            begin
+              doc = record.as_json_for_index.to_json
+              actions << {:index => {'_index' => index_name, '_type' => index_type, '_id' => record.id}}.to_json
+              actions << doc
+            rescue => e
+              puts "Unable to bulk index record: #{record.inspect} [#{e.message}]"
+            end
+          end
+          ElasticSearchable.request :put, '/_bulk', :body => "\n#{actions.join("\n")}\n"
+        end
+      end
+
       private
       def index_name
         self.elastic_options[:index] || ElasticSearchable.default_index
@@ -63,13 +92,13 @@ module ElasticSearchable
     end
 
     module InstanceMethods
-      # index the object in elasticsearch
-      # fire after_index callbacks after operation is complete 
+      # reindex the object in elasticsearch
+      # fires after_index callbacks after operation is complete 
       # see http://www.elasticsearch.org/guide/reference/api/index_.html
-      def index_in_elastic_search(lifecycle = nil)
+      def reindex(lifecycle = nil)
         query = {}
         query.merge! :percolate => "*" if self.class.elastic_options[:percolate]
-        response = ElasticSearchable.request :put, self.class.index_type_path(self.id), :query => query, :body => self.indexed_json_document.to_json
+        response = ElasticSearchable.request :put, self.class.index_type_path(self.id), :query => query, :body => self.as_json_for_index.to_json
 
         self.run_callbacks("after_index_on_#{lifecycle}".to_sym) if lifecycle
         self.run_callbacks(:after_index)
@@ -80,7 +109,7 @@ module ElasticSearchable
         end
       end
       # document to index in elasticsearch
-      def indexed_json_document
+      def as_json_for_index
         self.as_json self.class.elastic_options[:json]
       end
       def should_index?
@@ -92,7 +121,7 @@ module ElasticSearchable
       # can be done automatically when indexing using :percolate => true config option
       # http://www.elasticsearch.org/blog/2011/02/08/percolator.html
       def percolate
-        response = ElasticSearchable.request :get, self.class.index_type_path('_percolate'), :body => {:doc => self.indexed_json_document}.to_json
+        response = ElasticSearchable.request :get, self.class.index_type_path('_percolate'), :body => {:doc => self.as_json_for_index}.to_json
         response['matches']
       end
 

data/test/test_elastic_searchable.rb

@@ -109,6 +109,25 @@ class TestElasticSearchable < Test::Unit::TestCase
     end
   end
 
+  context 'with empty index when multiple database records' do
+    setup do
+      Post.create_index
+      @first_post = Post.create :title => 'foo', :body => "first bar"
+      @second_post = Post.create :title => 'foo', :body => "second bar"
+      Post.clean_index
+    end
+    context 'Post.reindex' do
+      setup do
+        Post.reindex
+        Post.refresh_index
+      end
+      should 'have reindexed both records' do
+        ElasticSearchable.request :get, "/elastic_searchable/posts/#{@first_post.id}"
+        ElasticSearchable.request :get, "/elastic_searchable/posts/#{@second_post.id}"
+      end
+    end
+  end
+
   context 'with index containing multiple results' do
     setup do
       Post.create_index
@@ -182,7 +201,7 @@ class TestElasticSearchable < Test::Unit::TestCase
   context 'activerecord class with optional :if=>proc configuration' do
     context 'when creating new instance' do
       setup do
-        Blog.any_instance.expects(:index_in_elastic_search).never
+        Blog.any_instance.expects(:reindex).never
         @blog = Blog.create! :title => 'foo'
       end
       should 'not index record' do end #see expectations

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: elastic_searchable
 version: !ruby/object:Gem::Version 
-  hash: 11
+  hash: 7
   prerelease: 
   segments: 
   - 0
-  - 5
+  - 6
   - 0
-  version: 0.5.0
+  version: 0.6.0
 platform: ruby
 authors: 
 - Ryan Sonnek
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-03-04 00:00:00 -06:00
+date: 2011-03-08 00:00:00 -06:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency