ultrasphinx 1.5.3 → 1.6

data/lib/ultrasphinx/search/internals.rb

@@ -9,66 +9,79 @@ module Ultrasphinx
       
       def build_request_with_options opts
       
-        request = Sphinx::Client.new
-      
-        request.SetServer(
-          Ultrasphinx::CLIENT_SETTINGS['server_host'], 
-          Ultrasphinx::CLIENT_SETTINGS['server_port']
-        )
-        
-        # Force extended query mode
-        request.SetMatchMode(Sphinx::Client::SPH_MATCH_EXTENDED) 
-      
-        offset, limit = opts['per_page'] * (opts['page'] - 1), opts['per_page']
-        
-        request.SetLimits offset, limit, [offset + limit, MAX_MATCHES].min
+        request = Riddle::Client.new
+        request.instance_eval do          
+          @server = Ultrasphinx::CLIENT_SETTINGS['server_host']
+          @port = Ultrasphinx::CLIENT_SETTINGS['server_port']          
+          @match_mode = :extended # Force extended query mode
+          @offset = opts['per_page'] * (opts['page'] - 1)
+          @limit = opts['per_page']
+          @max_matches = [@offset + @limit, MAX_MATCHES].min
+        end
+          
+        # Sorting
+        sort_by = opts['sort_by']
+        unless sort_by.blank?
+          if opts['sort_mode'].to_s == 'relevance'
+            # If you're sorting by a field you don't want 'relevance' order
+            raise UsageError, "Sort mode 'relevance' is not valid with a sort_by field"
+          end
+          request.sort_by = sort_by.to_s
+        end
         
-        if SPHINX_CLIENT_PARAMS['sort_mode'][opts['sort_mode']]
-          request.SetSortMode SPHINX_CLIENT_PARAMS['sort_mode'][opts['sort_mode']], opts['sort_by'].to_s
+        if sort_mode = SPHINX_CLIENT_PARAMS['sort_mode'][opts['sort_mode']]
+          request.sort_mode = sort_mode
         else
           raise UsageError, "Sort mode #{opts['sort_mode'].inspect} is invalid"
-        end
-      
-        if weights = opts['weights']
-          # Order the weights hash according to the field order for Sphinx, and set the missing fields to 1.0
-          request.SetWeights(Fields.instance.types.select{|n,t| t == 'text'}.map(&:first).sort.inject([]) do |array, field|
+        end        
+
+        # Weighting
+        weights = opts['weights']
+        if weights.any?
+          # Order according to the field order for Sphinx, and set the missing fields to 1.0
+          request.weights = (Fields.instance.types.select{|n,t| t == 'text'}.map(&:first).sort.inject([]) do |array, field|
             array << (weights[field] || 1.0)
           end)
         end
+        
+        # Class names
+        unless Array(opts['class_names']).empty?
+          request.filters << Riddle::Client::Filter.new(
+            'class_id', 
+            (opts['class_names'].map do |model| 
+              MODELS_TO_IDS[model.to_s] or 
+                MODELS_TO_IDS[model.to_s.constantize.base_class.to_s] or 
+                raise UsageError, "Invalid class name #{model.inspect}"
+            end), 
+            false)
+        end          
 
-        unless opts['class_names'].compact.empty?
-          request.SetFilter('class_id', (opts['class_names'].map do |model| 
-            MODELS_TO_IDS[model.to_s] or 
-            MODELS_TO_IDS[model.to_s.constantize.base_class.to_s] or 
-            raise UsageError, "Invalid class name #{model.inspect}"
-          end))
-        end
-      
-        # Extract ranged raw filters 
-        # Some of this mangling might not be necessary
-        opts['filters'].each do |field, value|          
+        # Extract raw filters 
+        # XXX We should coerce based on the Field values, not on the class
+        Array(opts['filters']).each do |field, value|          
           field = field.to_s
           unless Fields.instance.types[field]
-            raise Sphinx::SphinxArgumentError, "field #{field.inspect} is invalid"
+            raise UsageError, "field #{field.inspect} is invalid"
           end
           begin
             case value
-              when Fixnum, Float, BigDecimal, NilClass, Array
-                request.SetFilter field, Array(value)
+              when Integer, Float, BigDecimal, NilClass, Array
+                # Just bomb the filter in there
+                request.filters << Riddle::Client::Filter.new(field, Array(value), false)
               when Range
-                min, max = [value.begin, value.end].map do |x|
-                  x._to_numeric
-                end
+                # Make sure ranges point in the right direction
+                min, max = [value.begin, value.end].map {|x| x._to_numeric }
                 raise NoMethodError unless min <=> max and max <=> min
                 min, max = max, min if min > max
-                request.SetFilterRange field, min, max
+                request.filters << Riddle::Client::Filter.new(field, min..max, false)
               when String
+                # XXX Hack to move text filters into the query
                 opts['parsed_query'] << " @#{field} #{value}"
               else
                 raise NoMethodError
             end
           rescue NoMethodError => e
-            raise Sphinx::SphinxArgumentError, "filter value #{value.inspect} for field #{field.inspect} is invalid"
+            raise UsageError, "filter value #{value.inspect} for field #{field.inspect} is invalid"
           end
         end
         
@@ -77,7 +90,7 @@ module Ultrasphinx
       
       def get_subtotals(original_request, query)
         request = original_request._deep_dup
-        request.instance_eval { @filters.delete_if {|f| f['attr'] == 'class_id'} }
+        request.instance_eval { @filters.delete_if {|filter| filter.attribute == 'class_id'} }
         
         facets = get_facets(request, query, 'class_id')
         
@@ -91,26 +104,37 @@ module Ultrasphinx
         request, facet = original_request._deep_dup, original_facet        
         facet += "_facet" if Fields.instance.types[original_facet] == 'text'            
         
-        raise UsageError, "Field #{original_facet} does not exist or was not configured for faceting" unless Fields.instance.types[facet]
-
+        unless Fields.instance.types[facet]
+          if facet == original_facet
+            raise UsageError, "Field #{original_facet} does not exist" 
+          else
+            raise UsageError, "Field #{original_facet} is a text field, but was not configured for text faceting"
+          end
+        end
+        
         # Set the facet query parameter and modify per-page setting so we snag all the facets
-        request.SetGroupBy(facet, Sphinx::Client::SPH_GROUPBY_ATTR, '@count desc')
-        limit = self.class.client_options['max_facets']
-        request.SetLimits 0, limit, [limit, MAX_MATCHES].min
+        request.instance_eval do
+          @group_by = facet
+          @group_function = :attr
+          @group_clauses = '@count desc'
+          @offset = 0
+          @limit = Ultrasphinx::Search.client_options['max_facets']
+          @max_matches = [@limit, MAX_MATCHES].min
+        end
         
         # Run the query
         begin
-          matches = request.Query(query)['matches']
-        rescue Sphinx::SphinxInternalError
-          raise ConfigurationError, "Index is out of date. Run 'rake ultrasphinx:index'"
+          matches = request.query(query, UNIFIED_INDEX_NAME)[:matches]
+        rescue DaemonError
+          raise ConfigurationError, "Index seems out of date. Run 'rake ultrasphinx:index'"
         end
                 
         # Map the facets back to something sane
         facets = {}
         matches.each do |match|
-          match = match.last['attrs'] # :(
-          raise ResponseError if facets[match['@groupby']]
-          facets[match['@groupby']] = match['@count']
+          attributes = match[:attributes]
+          raise DaemonError if facets[attributes['@groupby']]
+          facets[attributes['@groupby']] = attributes['@count']
         end
                 
         # Invert hash's, if we have them
@@ -155,69 +179,76 @@ module Ultrasphinx
         end
         FACET_CACHE[facet]
       end
-
-      def reify_results(sphinx_ids)
-    
-        # Order by position and then toss the rest of the data
-        sphinx_ids = sphinx_ids.sort_by do |key, value| 
-          value['index'] or raise ConfigurationError, "Your Sphinx client is not properly patched."
-        end.map(&:first)
-    
-        # Inverse-modulus map the sphinx ids to the table-specific ids
-        ids = Hash.new([])
-        sphinx_ids.each do |id|
-          ids[MODELS_TO_IDS.invert[id % MODELS_TO_IDS.size]] += [id / MODELS_TO_IDS.size] # yay math
+      
+      # Inverse-modulus map the Sphinx ids to the table-specific ids
+      def convert_sphinx_ids(sphinx_ids)    
+        sphinx_ids.sort_by do |item| 
+          item[:index]
+        end.map do |item|
+          class_name = MODELS_TO_IDS.invert[item[:doc] % MODELS_TO_IDS.size]
+          raise DaemonError, "Impossible Sphinx document id #{item[:doc]} in query result" unless class_name
+          [class_name, item[:doc] / MODELS_TO_IDS.size]
         end
-        raise Sphinx::SphinxResponseError, "impossible document id in query result" unless ids.values.flatten.size == sphinx_ids.size
-    
-        # Fetch them for real
+      end
+
+      # Fetch them for real
+      def reify_results(ids)
         results = []
-        ids.each do |model, id_set|
-          klass = model.constantize
-          
-          finder = self.class.client_options['finder_methods'].detect do |method_name|
+        
+        ids.each do |klass_name, id|
+        
+          # What class and class method are we using to get the record?
+          klass = klass_name.constantize
+          finder = Ultrasphinx::Search.client_options['finder_methods'].detect do |method_name|
             klass.respond_to? method_name
           end
           
-          # Ultrasphinx.say "using #{klass.name}.#{finder} as finder method"
-    
-          begin
-            # XXX Does not use Memcached's multiget
-            results += case instances = id_set.map { |id| klass.send(finder, id) }
-              when Hash
-                instances.values
-              when Array
-                instances
-              else
-                Array(instances)
+          # Load it
+          record = begin
+            # XXX Does not use Memcached's multiget, or MySQL's, for that matter
+            klass.send(finder, id)
+          rescue ActiveRecord::RecordNotFound => e
+            if Ultrasphinx::Search.client_options['ignore_missing_records']
+              # XXX Should maybe adjust the total_found count, etc
+            else
+              raise(e)
             end
-          rescue ActiveRecord::ActiveRecordError => e
-            raise Sphinx::SphinxResponseError, e.inspect
-          end
-        end
-    
-        # Put them back in order
-        results.sort_by do |r| 
-          raise Sphinx::SphinxResponseError, "Bogus ActiveRecord id for #{r.class}:#{r.id}" unless r.id
-          
-          model_index = MODELS_TO_IDS[r.class.base_class.name]
-          raise UsageError, "#{r.class.base_class} is not an indexed class. Maybe you indexed an STI child class instead of the base class?" unless model_index
-          
-          index = (sphinx_ids.index(sphinx_id = r.id * MODELS_TO_IDS.size + model_index))
-          raise Sphinx::SphinxResponseError, "Bogus reverse id for #{r.class}:#{r.id} (Sphinx:#{sphinx_id})" unless index
+          end  
           
-          index / sphinx_ids.size.to_f
+          # Add it to the list. Cache_fu does funny things with returned record organization.
+          results += record.is_a?(Hash) ? record.values : Array(record)                
         end
-        
-        # Add an accessor for absolute search rank for each record
-        results.each_with_index do |r, index|
+    
+        # Add an accessor for absolute search rank for each record (does anyone use this?)
+        results.each_with_index do |result, index|
           i = per_page * (current_page - 1) + index
-          r._metaclass.send('define_method', 'result_index') { i }
+          result._metaclass.send('define_method', 'result_index') { i }
         end
         
         results        
       end  
-
+      
+      def perform_action_with_retries
+        tries = 0
+        begin
+          yield
+        rescue NoMethodError,
+            Riddle::VersionError,
+            Riddle::ResponseError,
+            Errno::ECONNREFUSED, 
+            Errno::ECONNRESET, 
+            Errno::EPIPE => e
+          tries += 1
+          if tries <= Ultrasphinx::Search.client_options['max_retries']
+            say "restarting query (#{tries} attempts already) (#{e})"            
+            sleep(Ultrasphinx::Search.client_options['retry_sleep_time']) 
+            retry
+          else
+            say "query failed"
+            raise DaemonError, e.to_s
+          end
+        end
+      end
       
       def strip_bogus_characters(s)
         # Used to remove some garbage before highlighting
@@ -225,7 +256,7 @@ module Ultrasphinx
       end
       
       def strip_query_commands(s)
-        # XXX Hack for query commands, since sphinx doesn't intelligently parse the query in excerpt mode
+        # XXX Hack for query commands, since Sphinx doesn't intelligently parse the query in excerpt mode
         # Also removes apostrophes in the middle of words so that they don't get split in two.
         s.gsub(/(^|\s)(AND|OR|NOT|\@\w+)(\s|$)/i, "").gsub(/(\w)\'(\w)/, '\1\2')
       end 

data/lib/ultrasphinx/ultrasphinx.rb

@@ -34,16 +34,8 @@ module Ultrasphinx
 
   MAX_WORDS = 2**16 # maximum number of stopwords built  
   
-  EMPTY_SEARCHABLE = "__empty_searchable__"
-  
   UNIFIED_INDEX_NAME = "complete"
 
-  SPHINX_VERSION = if `which indexer` =~ /\/indexer\n/m
-      `indexer`.split("\n").first[7..-1]
-    else
-      "unknown"
-    end 
-
   CONFIG_MAP = {
     # These must be symbols for key mapping against Rails itself
     :username => 'sql_user',

data/test/integration/app/app/models/geo/state.rb

@@ -1,5 +1,6 @@
 class Geo::State < ActiveRecord::Base
-  has_many :"geo/addresses"
+  has_many :addresses, :class_name => "Geo::Address"
   
   is_indexed :concatenate => [{:class_name => 'Geo::Address', :field => 'name', :as => 'address_name'}]
+    #:fields => [{:field => 'abbreviation', :as => 'company_name'}],
 end

data/test/integration/app/app/models/person/user.rb

@@ -3,7 +3,8 @@ class User < ActiveRecord::Base
   has_one   :address, :class_name => "Geo::Address"
 
   is_indexed :fields => ['login', 'email', 'deleted'], 
-    :include => [{:class_name => 'Seller', :field => 'company_name', :as => 'company'}],
+    :include => [{:class_name => 'Seller', :field => 'company_name', :as => 'company'},
+      {:class_name => 'Seller', :field => 'sellers_two.company_name', :as => 'company_two', 'association_sql' => 'LEFT OUTER JOIN sellers AS sellers_two ON users.id = sellers_two.user_id', 'function_sql' => "REPLACE(?, '6', ' replacement ')"}],
     :conditions => 'deleted = 0'  
   
 end

data/test/integration/app/config/environment.rb

@@ -6,6 +6,4 @@ Rails::Initializer.run do |config|
   config.load_paths << "#{RAILS_ROOT}/app/models/person" # moduleless model path
 end
 
-Ultrasphinx::Search.client_options['with_subtotals'] = true
-
 # Dependencies.log_activity = true

data/test/integration/app/config/ultrasphinx/development.conf

@@ -1,7 +1,7 @@
 
-# Auto-generated at Mon Oct 08 17:02:18 -0400 2007.
+# Auto-generated at Wed Nov 14 03:32:12 -0500 2007.
 # Hand modifications will be overwritten.
-# /Users/eweaver/Desktop/projects/chow/vendor/plugins/ultrasphinx/test/integration/app/config/ultrasphinx/default.base
+# /Users/eweaver/Desktop/projects/fauna/ultrasphinx/trunk/test/integration/app/config/ultrasphinx/default.base
 indexer {
   mem_limit = 256M
 }
@@ -34,7 +34,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM states
-sql_query = SELECT (states.id * 4 + 0) AS id, CAST(GROUP_CONCAT(addresses.name SEPARATOR ' ') AS CHAR) AS address_name, 0 AS capitalization, 'Geo::State' AS class, 0 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '__empty_searchable__' AS empty_searchable, '' AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM states LEFT OUTER JOIN addresses ON states.id = addresses.state_id WHERE states.id >= $start AND states.id <= $end GROUP BY id
+sql_query = SELECT (states.id * 4 + 0) AS id, CAST(GROUP_CONCAT(DISTINCT addresses.name SEPARATOR ' ') AS CHAR) AS address_name, 0 AS capitalization, 'Geo::State' AS class, 0 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, '' AS company_two, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '' AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM states LEFT OUTER JOIN addresses AS addresses ON states.id = addresses.state_id WHERE states.id >= $start AND states.id <= $end GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id
@@ -65,7 +65,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM sellers
-sql_query = SELECT (sellers.id * 4 + 1) AS id, '' AS address_name, sellers.capitalization AS capitalization, 'Seller' AS class, 1 AS class_id, '' AS company, sellers.company_name AS company_name, CRC32(sellers.company_name) AS company_name_facet, '' AS content, UNIX_TIMESTAMP(sellers.created_at) AS created_at, 0 AS deleted, '' AS email, '__empty_searchable__' AS empty_searchable, '' AS login, sellers.mission_statement AS mission_statement, '' AS name, '' AS state, sellers.user_id AS user_id FROM sellers WHERE sellers.id >= $start AND sellers.id <= $end GROUP BY id
+sql_query = SELECT (sellers.id * 4 + 1) AS id, '' AS address_name, sellers.capitalization AS capitalization, 'Seller' AS class, 1 AS class_id, '' AS company, sellers.company_name AS company_name, CRC32(sellers.company_name) AS company_name_facet, '' AS company_two, '' AS content, UNIX_TIMESTAMP(sellers.created_at) AS created_at, 0 AS deleted, '' AS email, '' AS login, sellers.mission_statement AS mission_statement, '' AS name, '' AS state, sellers.user_id AS user_id FROM sellers WHERE sellers.id >= $start AND sellers.id <= $end GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id
@@ -96,7 +96,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM addresses
-sql_query = SELECT (addresses.id * 4 + 2) AS id, '' AS address_name, 0 AS capitalization, 'Geo::Address' AS class, 2 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, CONCAT_WS(' ', addresses.line_1, addresses.line_2, addresses.city, addresses.province_region, addresses.zip_postal_code) AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '__empty_searchable__' AS empty_searchable, '' AS login, '' AS mission_statement, addresses.name AS name, states.name AS state, 0 AS user_id FROM addresses LEFT OUTER JOIN states ON states.id = addresses.state_id WHERE addresses.id >= $start AND addresses.id <= $end GROUP BY id
+sql_query = SELECT (addresses.id * 4 + 2) AS id, '' AS address_name, 0 AS capitalization, 'Geo::Address' AS class, 2 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, '' AS company_two, CONCAT_WS(' ', addresses.line_1, addresses.line_2, addresses.city, addresses.province_region, addresses.zip_postal_code) AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '' AS login, '' AS mission_statement, addresses.name AS name, states.name AS state, 0 AS user_id FROM addresses LEFT OUTER JOIN states AS states ON states.id = addresses.state_id WHERE addresses.id >= $start AND addresses.id <= $end GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id
@@ -127,7 +127,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM users
-sql_query = SELECT (users.id * 4 + 3) AS id, '' AS address_name, 0 AS capitalization, 'User' AS class, 3 AS class_id, sellers.company_name AS company, '' AS company_name, 0 AS company_name_facet, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, users.deleted AS deleted, users.email AS email, '__empty_searchable__' AS empty_searchable, users.login AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM users LEFT OUTER JOIN sellers ON users.id = sellers.user_id WHERE users.id >= $start AND users.id <= $end AND (deleted = 0) GROUP BY id
+sql_query = SELECT (users.id * 4 + 3) AS id, '' AS address_name, 0 AS capitalization, 'User' AS class, 3 AS class_id, sellers.company_name AS company, '' AS company_name, 0 AS company_name_facet, REPLACE(sellers_two.company_name, '6', ' replacement ') AS company_two, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, users.deleted AS deleted, users.email AS email, users.login AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM users LEFT OUTER JOIN sellers AS sellers ON users.id = sellers.user_id LEFT OUTER JOIN sellers AS sellers_two ON users.id = sellers_two.user_id WHERE users.id >= $start AND users.id <= $end AND (deleted = 0) GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id

data/test/integration/app/config/ultrasphinx/development.conf.canonical

@@ -1,7 +1,7 @@
 
-# Auto-generated at Mon Oct 08 15:40:47 -0400 2007.
+# Auto-generated at Wed Nov 14 03:28:33 -0500 2007.
 # Hand modifications will be overwritten.
-# /Users/eweaver/Desktop/projects/chow/vendor/plugins/ultrasphinx/test/integration/app/config/ultrasphinx/default.base
+# /Users/eweaver/Desktop/projects/fauna/ultrasphinx/trunk/test/integration/app/config/ultrasphinx/default.base
 indexer {
   mem_limit = 256M
 }
@@ -34,7 +34,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM states
-sql_query = SELECT (states.id * 4 + 0) AS id, CAST(GROUP_CONCAT(addresses.name SEPARATOR ' ') AS CHAR) AS address_name, 0 AS capitalization, 'Geo::State' AS class, 0 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '__empty_searchable__' AS empty_searchable, '' AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM states LEFT OUTER JOIN addresses ON states.id = addresses.state_id WHERE states.id >= $start AND states.id <= $end GROUP BY id
+sql_query = SELECT (states.id * 4 + 0) AS id, CAST(GROUP_CONCAT(DISTINCT addresses.name SEPARATOR ' ') AS CHAR) AS address_name, 0 AS capitalization, 'Geo::State' AS class, 0 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, '' AS company_two, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '' AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM states LEFT OUTER JOIN addresses AS addresses ON states.id = addresses.state_id WHERE states.id >= $start AND states.id <= $end GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id
@@ -65,7 +65,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM sellers
-sql_query = SELECT (sellers.id * 4 + 1) AS id, '' AS address_name, sellers.capitalization AS capitalization, 'Seller' AS class, 1 AS class_id, '' AS company, sellers.company_name AS company_name, CRC32(sellers.company_name) AS company_name_facet, '' AS content, UNIX_TIMESTAMP(sellers.created_at) AS created_at, 0 AS deleted, '' AS email, '__empty_searchable__' AS empty_searchable, '' AS login, sellers.mission_statement AS mission_statement, '' AS name, '' AS state, sellers.user_id AS user_id FROM sellers WHERE sellers.id >= $start AND sellers.id <= $end GROUP BY id
+sql_query = SELECT (sellers.id * 4 + 1) AS id, '' AS address_name, sellers.capitalization AS capitalization, 'Seller' AS class, 1 AS class_id, '' AS company, sellers.company_name AS company_name, CRC32(sellers.company_name) AS company_name_facet, '' AS company_two, '' AS content, UNIX_TIMESTAMP(sellers.created_at) AS created_at, 0 AS deleted, '' AS email, '' AS login, sellers.mission_statement AS mission_statement, '' AS name, '' AS state, sellers.user_id AS user_id FROM sellers WHERE sellers.id >= $start AND sellers.id <= $end GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id
@@ -96,7 +96,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM addresses
-sql_query = SELECT (addresses.id * 4 + 2) AS id, '' AS address_name, 0 AS capitalization, 'Geo::Address' AS class, 2 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, CONCAT_WS(' ', addresses.line_1, addresses.line_2, addresses.city, addresses.province_region, addresses.zip_postal_code) AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '__empty_searchable__' AS empty_searchable, '' AS login, '' AS mission_statement, addresses.name AS name, states.name AS state, 0 AS user_id FROM addresses LEFT OUTER JOIN states ON states.id = addresses.state_id WHERE addresses.id >= $start AND addresses.id <= $end GROUP BY id
+sql_query = SELECT (addresses.id * 4 + 2) AS id, '' AS address_name, 0 AS capitalization, 'Geo::Address' AS class, 2 AS class_id, '' AS company, '' AS company_name, 0 AS company_name_facet, '' AS company_two, CONCAT_WS(' ', addresses.line_1, addresses.line_2, addresses.city, addresses.province_region, addresses.zip_postal_code) AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, 0 AS deleted, '' AS email, '' AS login, '' AS mission_statement, addresses.name AS name, states.name AS state, 0 AS user_id FROM addresses LEFT OUTER JOIN states AS states ON states.id = addresses.state_id WHERE addresses.id >= $start AND addresses.id <= $end GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id
@@ -127,7 +127,7 @@ sql_host = localhost
 sql_pass = 
 sql_user = root
 sql_query_range = SELECT MIN(id), MAX(id) FROM users
-sql_query = SELECT (users.id * 4 + 3) AS id, '' AS address_name, 0 AS capitalization, 'User' AS class, 3 AS class_id, sellers.company_name AS company, '' AS company_name, 0 AS company_name_facet, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, users.deleted AS deleted, users.email AS email, '__empty_searchable__' AS empty_searchable, users.login AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM users LEFT OUTER JOIN sellers ON users.id = sellers.user_id WHERE users.id >= $start AND users.id <= $end AND (deleted = 0) GROUP BY id
+sql_query = SELECT (users.id * 4 + 3) AS id, '' AS address_name, 0 AS capitalization, 'User' AS class, 3 AS class_id, sellers.company_name AS company, '' AS company_name, 0 AS company_name_facet, REPLACE(sellers_two.company_name, '6', ' replacement ') AS company_two, '' AS content, UNIX_TIMESTAMP('1970-01-01 00:00:00') AS created_at, users.deleted AS deleted, users.email AS email, users.login AS login, '' AS mission_statement, '' AS name, '' AS state, 0 AS user_id FROM users LEFT OUTER JOIN sellers AS sellers ON users.id = sellers.user_id LEFT OUTER JOIN sellers AS sellers_two ON users.id = sellers_two.user_id WHERE users.id >= $start AND users.id <= $end AND (deleted = 0) GROUP BY id
 
 sql_group_column = capitalization
 sql_group_column = class_id

data/test/integration/app/test/fixtures/sellers.yml

@@ -5,7 +5,7 @@
   company_name: <%= "seller#{num}" %>
   capitalization: <%= num * 1.548 %>
   mission_statement: <%= %w(Add value through developing superior products.).sort_by(&:rand).join(" ") %>
-  created_at: <%= (Time.now - num.weeks).to_s :db %>
-  updated_at: <%= (Time.now - num.days).to_s :db %>  
+  updated_at: <%= ((time = Time.parse("Tue Oct 23 04:28:11")) - num.days).to_s(:db) %>  
+  created_at: <%= (time - num.weeks).to_s(:db) %>
 <% end %>