ultrasphinx 1.5 → 1.5.1

data/CHANGELOG

@@ -1,4 +1,6 @@
 
+v1.5.1. Various bugfixes.
+
 v1.5. API change. Change layout of base files to allow overriding of more options, see examples/default.base. Allow sorting on text fields (use the 'sortable' key).
 
 v1.4. New is_indexed 'fields' => {'function_sql'} key for custom field mangling; support setting textual keys in the 'filters'. 

data/README

@@ -7,6 +7,8 @@ Ruby on Rails configurator and client to the Sphinx full text search engine.
 
 Copyright 2007 Cloudburst, LLC. Licensed under the AFL 3. See the included LICENSE file. Some portions copyright Dmytro Shteflyuk and Alexey Kovyrin, distributed under the Ruby License, and used with permission. Some portions copyright PJ Hyett and Mislav Marohnić, distributed under the MIT license, and used with permission.
 
+The public certificate for the gem is at http://rubyforge.org/frs/download.php/25331/evan_weaver-original-public_cert.pem.
+
 == Requirements
 
 * MySQL (or Postgres, experimental)
@@ -96,7 +98,7 @@ See Ultrasphinx::Spell.
   
 == Available Rake tasks
 
-These rake tasks are made available to your Rails app:
+These Rake tasks are made available to your Rails app:
 
 <tt>ultrasphinx:configure</tt>:: Rebuild the configuration file for this particular environment.
 <tt>ultrasphinx:index</tt>:: Reindex the database and send an update signal to the search daemon.

data/lib/ultrasphinx/configure.rb

@@ -8,13 +8,15 @@ module Ultrasphinx
   
         Dir["#{RAILS_ROOT}/app/models/**/*.rb"].each do |filename|
           next if filename =~ /\/(\.svn|CVS|\.bzr)\//
-          begin
-            open(filename) {|file| load filename if file.grep(/is_indexed/).any?}
-          rescue Object => e
-            say "warning; possibly critical autoload error on #{filename}"
-            say e.inspect
-          end
-        end 
+          open(filename) do |file| 
+            begin
+              load filename if file.grep(/is_indexed/).any?
+            rescue Object => e
+              say "warning; possibly critical autoload error on #{filename}"
+              say e.inspect
+            end
+          end 
+        end
   
         # Build the field-to-type mappings.
         Fields.instance.configure(MODEL_CONFIGURATION)
@@ -26,7 +28,7 @@ module Ultrasphinx
 
         load_constants
               
-        puts "Rebuilding Ultrasphinx configurations for #{ENV['RAILS_ENV']} environment" 
+        puts "Rebuilding Ultrasphinx configurations for #{RAILS_ENV} environment" 
         puts "Available models are #{MODEL_CONFIGURATION.keys.to_sentence}"
         File.open(CONF_PATH, "w") do |conf|
         
@@ -64,7 +66,7 @@ module Ultrasphinx
         # Tentatively supporting Postgres now
         connection_settings = klass.connection.instance_variable_get("@config")
 
-        adapter_defaults = ADAPTER_DEFAULTS[connection_settings[:adapter]]
+        adapter_defaults = ADAPTER_DEFAULTS[ADAPTER]
         raise ConfigurationError, "Unsupported database adapter" unless adapter_defaults
 
         conf = [adapter_defaults]                  
@@ -127,8 +129,6 @@ module Ultrasphinx
       
       def build_query(klass, column_strings, join_strings, condition_strings)
 
-        connection_settings = klass.connection.instance_variable_get("@config")
-
         ["sql_query =", 
           "SELECT", 
           column_strings.sort_by do |string| 
@@ -141,7 +141,7 @@ module Ultrasphinx
           condition_strings.uniq.map do |condition| 
             "AND #{condition}"
           end,
-          ("GROUP BY id" if connection_settings[:adapter] == 'mysql') # XXX should be somewhere more obvious
+          ADAPTER_SQL_FUNCTIONS[ADAPTER]['group_by']
         ].flatten.join(" ")
       end
       

data/lib/ultrasphinx/core_extensions.rb

@@ -50,34 +50,17 @@ class Hash
       end._flatten_once]
   end
   
+  def _delete(*args)
+    args.map do |key|
+      self.delete key
+    end    
+  end
+  
   def _to_conf_string(section = nil)
     inner = self.map do |key, value|
       "  #{key} = #{value}"
     end.join("\n")
     section ? "#{section} {\n#{inner}\n}\n" : inner
   end
-  
-  def _deep_stringify_keys
-    Hash[*(self.map do |key, value|
-#      puts "#{key.inspect}, #{value.inspect}"
-      z = [key.to_s,
-        case value
-          when Hash
-            value._deep_stringify_keys
-          when Array
-            value.map do |subvalue|
-              if subvalue.is_a? Hash or subvalue.is_a? Array
-                subvalue._deep_stringify_keys
-              else
-                subvalue
-              end
-            end
-          else
-            value
-        end
-      ]
-#      p z
-#      z
-    end._flatten_once)]
-  end
+
 end

data/lib/ultrasphinx/fields.rb

@@ -55,7 +55,7 @@ module Ultrasphinx
     
     def cast(source_string, field)
       if types[field] == "date"
-        "UNIX_TIMESTAMP(#{source_string})"
+        "#{ADAPTER_SQL_FUNCTIONS[ADAPTER]['timestamp']}#{source_string})"
       elsif source_string =~ /GROUP_CONCAT/
         "CAST(#{source_string} AS CHAR)"
       else
@@ -93,7 +93,7 @@ module Ultrasphinx
             entry['as'] = entry['field'] unless entry['as']
             
             unless klass.columns_hash[entry['field']]
-              ActiveRecord::Base.logger.warn "ultrasphinx: WARNING: field #{entry['field']} is not present in #{model}"
+              Ultrasphinx.say "WARNING: field #{entry['field']} is not present in #{model}"
             else
               save_and_verify_type(entry['as'], klass.columns_hash[entry['field']].type, entry['sortable'], klass)
             end
@@ -116,7 +116,7 @@ module Ultrasphinx
             save_and_verify_type(entry['as'], 'text', entry['sortable'], klass)
           end          
         rescue ActiveRecord::StatementInvalid
-          ActiveRecord::Base.logger.warn "ultrasphinx: WARNING: model #{model} does not exist in the database yet"
+          Ultrasphinx.say "WARNING: model #{model} does not exist in the database yet"
         end  
       end
       

data/lib/ultrasphinx/is_indexed.rb

@@ -120,23 +120,28 @@ If the associations weren't just <tt>has_many</tt> and <tt>belongs_to</tt>, you
 =end
   
     def self.is_indexed opts = {}    
-      opts = opts._deep_stringify_keys
+      opts = HashWithIndifferentAccess.new(opts)
           
       opts.assert_valid_keys ['fields', 'concatenate', 'conditions', 'include']
       
-      Array(opts[:fields]).each do |field|
-        field.assert_valid_keys ['field', 'as', 'facet', 'function_sql', 'sortable'] if field.is_a? Hash
+      Array(opts['fields']).each do |entry|
+        if entry.is_a? Hash
+          entry.stringify_keys!
+          entry.assert_valid_keys ['field', 'as', 'facet', 'function_sql', 'sortable']
+        end
       end
       
-      Array(opts[:concatenate]).each do |concat|
-        concat.assert_valid_keys ['class_name', 'conditions', 'field', 'as', 'fields', 'association_name', 'association_sql', 'facet', 'function_sql', 'sortable']
-        raise Ultrasphinx::ConfigurationError, "You can't mix regular concat and group concats" if concat['fields'] and (concat['field'] or concat['class_name'] or concat['association_name'])
-        raise Ultrasphinx::ConfigurationError, "Group concats must not have multiple fields" if concat['field'].is_a? Array
-        raise Ultrasphinx::ConfigurationError, "Regular concats should have multiple fields" if concat['fields'] and !concat['fields'].is_a?(Array)
+      Array(opts['concatenate']).each do |entry|
+        entry.stringify_keys!
+        entry.assert_valid_keys ['class_name', 'conditions', 'field', 'as', 'fields', 'association_name', 'association_sql', 'facet', 'function_sql', 'sortable']
+        raise Ultrasphinx::ConfigurationError, "You can't mix regular concat and group concats" if entry['fields'] and (entry['field'] or entry['class_name'] or entry['association_name'])
+        raise Ultrasphinx::ConfigurationError, "Group concats must not have multiple fields" if entry['field'].is_a? Array
+        raise Ultrasphinx::ConfigurationError, "Regular concats should have multiple fields" if entry['fields'] and !entry['fields'].is_a?(Array)
       end
       
-      Array(opts[:include]).each do |inc|
-        inc.assert_valid_keys ['class_name', 'field', 'as', 'association_sql', 'facet', 'function_sql', 'sortable']
+      Array(opts['include']).each do |entry|
+        entry.stringify_keys!
+        entry.assert_valid_keys ['class_name', 'field', 'as', 'association_sql', 'facet', 'function_sql', 'sortable']
       end
       
       Ultrasphinx::MODEL_CONFIGURATION[self.name] = opts

data/lib/ultrasphinx/search.rb

@@ -6,11 +6,11 @@ Command-interface Search object.
 
 == Basic usage
   
-To set up a search, instantiate an Ultrasphinx::Search object with a hash of parameters. Only the <tt>'query'</tt> key is mandatory.
+To set up a search, instantiate an Ultrasphinx::Search object with a hash of parameters. Only the <tt>:query</tt> key is mandatory.
   @search = Ultrasphinx::Search.new(
-    'query' => @query, 
-    'sort_mode' => 'descending', 
-    'sort_by' => 'created_at'
+    :query => @query, 
+    :sort_mode => 'descending', 
+    :sort_by => 'created_at'
   )
     
 Now, to run the query, call its <tt>run</tt> method. Your results will be available as ActiveRecord instances via the <tt>results</tt> method. Example:  
@@ -34,21 +34,21 @@ A Sphinx::SphinxInternalError will be raised on invalid queries. In general, que
 
 The hash lets you customize internal aspects of the search.
 
-<tt>'per_page'</tt>:: An integer. How many results per page.
-<tt>'page'</tt>:: An integer. Which page of the results to return.
-<tt>'class_name'</tt>:: An array or string. The class name of the model you want to search, an array of model names to search, or <tt>nil</tt> for all available models.    
-<tt>'sort_mode'</tt>:: 'relevance' or 'ascending' or 'descending'. How to order the result set. Note that 'time' and 'extended' modes are available, but not tested.  
-<tt>'sort_by'</tt>:: A field name. What field to order by for 'ascending' or 'descending' mode. Has no effect for 'relevance'.
-<tt>'weight'</tt>:: A hash. Text-field names and associated query weighting. The default weight for every field is 1.0. Example: <tt>'weight' => {'title' => 2.0}</tt>
-<tt>'filter'</tt>:: A hash. Names of numeric or date fields and associated values. You can use a single value, an array of values, or a range. (See the bottom of the ActiveRecord::Base page for an example.)
-<tt>'facets'</tt>:: An array of fields for grouping/faceting. You can access the returned facet values and their result counts with the <tt>facets</tt> method.
+<tt>:per_page</tt>:: An integer. How many results per page.
+<tt>:page</tt>:: An integer. Which page of the results to return.
+<tt>:class_names</tt>:: An array or string. The class name of the model you want to search, an array of model names to search, or <tt>nil</tt> for all available models.    
+<tt>:sort_mode</tt>:: <tt>'relevance'</tt> or <tt>'ascending'</tt> or <tt>'descending'</tt>. How to order the result set. Note that <tt>'time'</tt> and <tt>'extended'</tt> modes are available, but not tested.  
+<tt>:sort_by</tt>:: A field name. What field to order by for <tt>'ascending'</tt> or <tt>'descending'</tt> mode. Has no effect for <tt>'relevance'</tt>.
+<tt>:weights</tt>:: A hash. Text-field names and associated query weighting. The default weight for every field is 1.0. Example: <tt>:weights => {'title' => 2.0}</tt>
+<tt>:filters</tt>:: A hash. Names of numeric or date fields and associated values. You can use a single value, an array of values, or a range. (See the bottom of the ActiveRecord::Base page for an example.)
+<tt>:facets</tt>:: An array of fields for grouping/faceting. You can access the returned facet values and their result counts with the <tt>facets</tt> method.
 
 Note that you can set up your own query defaults in <tt>environment.rb</tt>: 
   
   Ultrasphinx::Search.query_defaults = {
-    'per_page' => 10,
-    'sort_mode' => 'relevance',
-    'weight' => {'title' => 2.0}
+    :per_page => 10,
+    :sort_mode => 'relevance',
+    :weights => {'title' => 2.0}
   }
 
 = Advanced features
@@ -76,12 +76,12 @@ You need to set the <tt>content_methods</tt> key on Ultrasphinx::Search.excerpti
 There are some other keys you can set, such as excerpt size, HTML tags to highlight with, and number of words on either side of each excerpt chunk. Example (in <tt>environment.rb</tt>):
   
   Ultrasphinx::Search.excerpting_options = {
-    'before_match' => '<strong>', 
-    'after_match' => '</strong>',
-    'chunk_separator' => "...",
-    'limit' => 256,
-    'around' => 3,
-    'content_methods' => [['title'], ['body', 'description', 'content'], ['metadata']] 
+    :before_match => '<strong>', 
+    :after_match => '</strong>',
+    :chunk_separator => "...",
+    :limit => 256,
+    :around => 3,
+    :content_methods => [['title'], ['body', 'description', 'content'], ['metadata']] 
   }
   
 Note that your database is never changed by anything Ultrasphinx does.
@@ -94,40 +94,40 @@ Note that your database is never changed by anything Ultrasphinx does.
     include Parser
     
     cattr_accessor :query_defaults  
-    self.query_defaults ||= {
-      'query' => nil,
-      'page' => 1,
-      'class_name' => nil,
-      'per_page' => 20,
-      'sort_by' => 'created_at',
-      'sort_mode' => 'relevance',
-      'weight' => nil,
-      'filter' => nil,
-      'facets' => nil
-    }
+    self.query_defaults ||= HashWithIndifferentAccess.new({
+      :query => nil,
+      :page => 1,
+      :class_names => nil,
+      :per_page => 20,
+      :sort_by => 'created_at',
+      :sort_mode => 'relevance',
+      :weights => nil,
+      :filters => nil,
+      :facets => nil
+    })
     
     cattr_accessor :excerpting_options
-    self.excerpting_options ||= {
-      'before_match' => "<strong>", 'after_match' => "</strong>",
-      'chunk_separator' => "...",
-      'limit' => 256,
-      'around' => 3,
+    self.excerpting_options ||= HashWithIndifferentAccess.new({
+      :before_match => "<strong>", :after_match => "</strong>",
+      :chunk_separator => "...",
+      :limit => 256,
+      :around => 3,
       # results should respond to one in each group of these, in precedence order, for the excerpting to fire
-      'content_methods' => [['title', 'name'], ['body', 'description', 'content'], ['metadata']] 
-    }
+      :content_methods => [['title', 'name'], ['body', 'description', 'content'], ['metadata']] 
+    })
     
     cattr_accessor :client_options
-    self.client_options ||= { 
-      'with_subtotals' => false, 
-      'max_retries' => 4,
-      'retry_sleep_time' => 3,
-      'max_facets' => 100,
-      'finder_methods' => ['get_cache', 'find']
-    }
+    self.client_options ||= HashWithIndifferentAccess.new({ 
+      :with_subtotals => false, 
+      :max_retries => 4,
+      :retry_sleep_time => 3,
+      :max_facets => 100,
+      :finder_methods => ['get_cache', 'find']
+    })
     
     # mode to integer mappings    
-    SPHINX_CLIENT_PARAMS = { 
-      'sort_mode' => {
+    SPHINX_CLIENT_PARAMS = HashWithIndifferentAccess.new({ 
+      :sort_mode => HashWithIndifferentAccess.new({
         'relevance' => Sphinx::Client::SPH_SORT_RELEVANCE, 
         'descending' => Sphinx::Client::SPH_SORT_ATTR_DESC, 
         'ascending' => Sphinx::Client::SPH_SORT_ATTR_ASC, 
@@ -135,10 +135,10 @@ Note that your database is never changed by anything Ultrasphinx does.
         'extended' => Sphinx::Client::SPH_SORT_EXTENDED,
         'desc' => Sphinx::Client::SPH_SORT_ATTR_DESC, # legacy compatibility
         'asc' => Sphinx::Client::SPH_SORT_ATTR_ASC
-      }
-    }
+      })
+    })
     
-    LEGACY_QUERY_KEYS = ['raw_filters'] #:nodoc:
+    LEGACY_QUERY_KEYS = ['raw_filters', 'filter', 'weight', 'class_name'] #:nodoc:
     
     INTERNAL_KEYS = ['parsed_query'] #:nodoc:
 
@@ -146,7 +146,7 @@ Note that your database is never changed by anything Ultrasphinx does.
       # reading the conf file makes sure that we are in sync with the actual sphinx index,
       # not whatever you happened to change your models to most recently
       unless File.exist? CONF_PATH
-        Ultrasphinx.say "configuration file not found for #{ENV['RAILS_ENV'].inspect} environment"
+        Ultrasphinx.say "configuration file not found for #{RAILS_ENV.inspect} environment"
         Ultrasphinx.say "please run 'rake ultrasphinx:configure'"
       else
         begin  
@@ -178,12 +178,12 @@ Note that your database is never changed by anything Ultrasphinx does.
     
     #  Returns the query string used.
     def query
-      # redundant with method_missing
+      # Redundant with method_missing
       @options['query']
     end
     
     def parsed_query #:nodoc:
-      # redundant with method_missing
+      # Redundant with method_missing
       @options['parsed_query']
     end
     
@@ -193,19 +193,24 @@ Note that your database is never changed by anything Ultrasphinx does.
       @results
     end
     
+    # Returns the facet map for this query, if facets were used.
     def facets
       raise UsageError, "No facet field was configured" unless @options['facets']
       run?(true)
       @facets
     end      
-      
-    
+          
     # Returns the raw response from the Sphinx client.
     def response
       @response
     end
     
-    # Returns a hash of total result counts, scoped to each available model. This requires extra queries against the search daemon right now. Set <tt>Ultrasphinx::Search.client_options['with_subtotals'] = true</tt> to enable the extra queries. Most of the overhead is in instantiating the AR result sets, so the performance hit is not usually significant.
+    def class_name #:nodoc:
+      # Legacy accessor
+      @options['class_name']
+    end
+    
+    # Returns a hash of total result counts, scoped to each available model. This requires extra queries against the search daemon right now. Set <tt>Ultrasphinx::Search.client_options[:with_subtotals] = true</tt> to enable the extra queries. Most of the overhead is in instantiating the AR result sets, so the performance hit is not usually significant.
     def subtotals
       raise UsageError, "Subtotals are not enabled" unless self.class.client_options['with_subtotals']
       @subtotals
@@ -239,14 +244,7 @@ Note that your database is never changed by anything Ultrasphinx does.
     def per_page
       @options['per_page']
     end
-    
-    # Clear the associated facet caches. They will be rebuilt on your next <tt>run</tt> or <tt>excerpt</tt>.
-    def clear_facet_caches
-      Array(@options['facets']).each do |facet|
-        FACET_CACHE.delete(facet)
-      end
-    end
-    
+        
     # Returns the last available page number in the result set.  
     def page_count
       (total_entries / per_page) + (total_entries % per_page == 0 ? 0 : 1)
@@ -268,16 +266,20 @@ Note that your database is never changed by anything Ultrasphinx does.
     end    
     
     # Builds a new command-interface Search object.
-    def initialize opts = {}
-      
-      opts = opts._deep_stringify_keys
-            
+    def initialize opts = {} 
+      opts = HashWithIndifferentAccess.new(opts)            
       @options = self.class.query_defaults.merge(opts._deep_dup._coerce_basic_types)
 
-      @options['filter'] ||= @options['raw_filters'] || {} # XXX legacy name
-
+      # Legacy compatibility
+      @options['filters'] ||= @options['filter'] || @options['raw_filters'] || {}
+      @options['class_names'] ||= @options['class_name']
+      @options['weights'] ||= @options['weight']
+            
+      @options._delete(*LEGACY_QUERY_KEYS)
+      
+      # Coerce some special types
       @options['query'] = @options['query'].to_s
-      @options['class_name'] = Array(@options['class_name'])
+      @options['class_names'] = Array(@options['class_names'])
       
       @options['parsed_query'] = if query.blank? 
          "@empty_searchable #{EMPTY_SEARCHABLE}"
@@ -288,7 +290,7 @@ Note that your database is never changed by anything Ultrasphinx does.
       @results, @subtotals, @facets, @response = [], {}, {}, {}
         
       extra_keys = @options.keys - (SPHINX_CLIENT_PARAMS.merge(self.class.query_defaults).keys + LEGACY_QUERY_KEYS + INTERNAL_KEYS)
-      logger.warn "Discarded invalid keys: #{extra_keys * ', '}" if extra_keys.any?
+      say "discarded invalid keys: #{extra_keys * ', '}" if extra_keys.any? and RAILS_ENV != "test" 
     end
     
     # Run the search, filling results with an array of ActiveRecord objects. Set the parameter to false if you only want the ids returned.
@@ -297,12 +299,12 @@ Note that your database is never changed by anything Ultrasphinx does.
       @paginate = nil # clear cache
       tries = 0
 
-      logger.info "** ultrasphinx: searching for #{@options.inspect}"
+      say "searching for #{@options.inspect}"
 
       begin
               
         @response = @request.Query(parsed_query)
-        logger.info "** ultrasphinx: search returned, error #{@request.GetLastError.inspect}, warning #{@request.GetLastWarning.inspect}, returned #{total_entries}/#{response['total_found']} in #{time} seconds."  
+        say "search returned, error #{@request.GetLastError.inspect}, warning #{@request.GetLastWarning.inspect}, returned #{total_entries}/#{response['total_found']} in #{time} seconds."  
 
         @subtotals = get_subtotals(@request, parsed_query) if self.class.client_options['with_subtotals']
         
@@ -317,11 +319,11 @@ Note that your database is never changed by anything Ultrasphinx does.
                                 
       rescue Sphinx::SphinxResponseError, Sphinx::SphinxTemporaryError, Errno::EPIPE => e
         if (tries += 1) <= self.class.client_options['max_retries']
-          logger.warn "** ultrasphinx: restarting query (#{tries} attempts already) (#{e})"
+          say "restarting query (#{tries} attempts already) (#{e})"
           sleep(self.class.client_options['retry_sleep_time']) if tries == self.class.client_options['max_retries']
           retry
         else
-          logger.warn "** ultrasphinx: query failed"
+          say "query failed"
           raise e
         end
       end
@@ -386,8 +388,8 @@ Note that your database is never changed by anything Ultrasphinx does.
       end
     end
   
-    def logger #:nodoc:
-      RAILS_DEFAULT_LOGGER
+    def say msg #:nodoc:
+      Ultrasphinx.say msg
     end
     
   end

data/lib/ultrasphinx/search/internals.rb

@@ -24,20 +24,20 @@ module Ultrasphinx
         request.SetLimits offset, limit, [offset + limit, MAX_MATCHES].min
         request.SetSortMode SPHINX_CLIENT_PARAMS['sort_mode'][opts['sort_mode']], opts['sort_by'].to_s
       
-        if weights = opts['weight']
+        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|
             array << (weights[field] || 1.0)
           end)
         end
       
-        unless opts['class_name'].compact.empty?
-          request.SetFilter 'class_id', opts['class_name'].map{|m| MODELS_TO_IDS[m.to_s]}
+        unless opts['class_names'].compact.empty?
+          request.SetFilter 'class_id', opts['class_names'].map{|m| MODELS_TO_IDS[m.to_s]}
         end        
       
         # Extract ranged raw filters 
         # Some of this mangling might not be necessary
-        opts['filter'].each do |field, value|
+        opts['filters'].each do |field, value|
           begin
             case value
               when Fixnum, Float, BigDecimal, NilClass, Array
@@ -104,31 +104,33 @@ module Ultrasphinx
         facets = facets.dup
       
         if Fields.instance.types[facet] == 'text'        
-          unless FACET_CACHE[facet]
-            # Cache the reverse CRC map for the textual facet if it hasn't been done yet
-            # XXX not necessarily optimal since it requires a direct DB hit once per mongrel
-            Ultrasphinx.say "caching crc reverse map for text facet #{facet}"
-            
-            Fields.instance.classes[facet].each do |klass|
-              # you can only use a facet from your own self right now; no includes allowed
-              field = (MODEL_CONFIGURATION[klass.name]['fields'].detect do |field_hash|
-                field_hash['as'] == facet
-              end)['field']
-          
-              klass.connection.execute("SELECT #{field} AS value, CRC32(#{field}) AS crc FROM #{klass.table_name} GROUP BY #{field}").each_hash do |hash|
-                (FACET_CACHE[facet] ||= {})[hash['crc'].to_i] = hash['value']
-              end                            
-            end
-          end
-          
-          # Apply the map
+          # Apply the map, rebuilding if the cache is missing or out-of-date
           facets = Hash[*(facets.map do |crc, value|
+            rebuild_facet_cache(facet) unless FACET_CACHE[facet] and FACET_CACHE[facet].has_key?(crc)
             [FACET_CACHE[facet][crc], value]
           end.flatten)]
         end
         
         facets        
       end
+      
+      def rebuild_facet_cache(facet)
+        # Cache the reverse CRC map for the textual facet if it hasn't been done yet
+        # XXX not necessarily optimal since it requires a direct DB hit once per mongrel
+        Ultrasphinx.say "caching CRC reverse map for text facet #{facet}"
+        
+        Fields.instance.classes[facet].each do |klass|
+          # you can only use a facet from your own self right now; no includes allowed
+          field = (MODEL_CONFIGURATION[klass.name]['fields'].detect do |field_hash|
+            field_hash['as'] == facet
+          end)['field']
+      
+          klass.connection.execute("SELECT #{field} AS value, CRC32(#{field}) AS crc FROM #{klass.table_name} GROUP BY #{field}").each_hash do |hash|
+            (FACET_CACHE[facet] ||= {})[hash['crc'].to_i] = hash['value']
+          end                            
+        end
+        FACET_CACHE[facet]
+      end
 
       def reify_results(sphinx_ids)
     
@@ -153,7 +155,7 @@ module Ultrasphinx
             klass.respond_to? method_name
           end
           
-          logger.debug "** ultrasphinx: using #{klass.name}.#{finder} as finder method"
+          # Ultrasphinx.say "using #{klass.name}.#{finder} as finder method"
     
           begin
             # XXX Does not use Memcached's multiget
@@ -173,8 +175,13 @@ module Ultrasphinx
         # Put them back in order
         results.sort_by do |r| 
           raise Sphinx::SphinxResponseError, "Bogus ActiveRecord id for #{r.class}:#{r.id}" unless r.id
-          index = (sphinx_ids.index(sphinx_id = r.id * MODELS_TO_IDS.size + MODELS_TO_IDS[r.class.base_class.name]))
+          
+          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
+          
           index / sphinx_ids.size.to_f
         end
         

data/lib/ultrasphinx/search/parser.rb

@@ -58,6 +58,7 @@ module Ultrasphinx
 
       def query_to_token_stream(query)      
         # First, split query on spaces that are not inside sets of quotes or parens
+        
         query = query.scan(/[^"() ]*["(][^")]*[")]|[^"() ]+/) 
       
         token_stream = []
@@ -68,7 +69,17 @@ module Ultrasphinx
           # recurse for parens, if necessary
           if subtoken =~ /^(.*?)\((.*)\)(.*?$)/
             subtoken = query[index] = "#{$1}(#{parse $2})#{$3}"
-          end       
+          end 
+          
+          # reappend missing closing quotes
+          if subtoken =~ /(^|\:)\"/
+            subtoken = subtoken.chomp('"') + '"'
+          end
+          
+          # strip parentheses within quoted strings
+          if subtoken =~ /\"(.*)\"/
+            subtoken.sub!($1, $1.gsub(/[()]/, ''))
+          end
           
           # add to the stream, converting the operator
           if !has_operator

data/lib/ultrasphinx/ultrasphinx.rb

@@ -59,10 +59,28 @@ sql_query_pre = SET NAMES utf8
     'postgresql' => %(
 type = pgsql
   )}
+  
+  ADAPTER_SQL_FUNCTIONS = {
+    'mysql' => {
+      'group_by' => 'GROUP BY id',
+      'timestamp' => 'UNIX_TIMESTAMP('
+    },    
+    'postgresql' => {
+      'group_by' => '',
+      'timestamp' => 'EXTRACT(EPOCH FROM '
+    }      
+  }
+  
+  ADAPTER = ActiveRecord::Base.connection.instance_variable_get("@config")[:adapter]
      
   # Logger.
   def self.say msg
-    STDERR.puts "** ultrasphinx: #{msg}"
+    msg = "** ultrasphinx: #{msg}"
+    if defined? RAILS_DEFAULT_LOGGER
+      RAILS_DEFAULT_LOGGER.warn msg
+    else
+      STDERR.puts msg
+    end
   end
   
   # Configuration file parser.
@@ -100,13 +118,16 @@ type = pgsql
   # Complain if the database names go out of sync.
   def self.verify_database_name
     if File.exist? CONF_PATH
-      if options_for(
-        "source #{MODEL_CONFIGURATION.keys.first.tableize}", 
-        CONF_PATH
-      )['sql_db'] != ActiveRecord::Base.connection.instance_variable_get("@config")[:database]
-        say "warning; configured database name is out-of-date"
-        say "please run 'rake ultrasphinx:configure'"
-      end rescue nil
+      begin
+        if options_for(
+          "source #{MODEL_CONFIGURATION.keys.first.tableize}", 
+          CONF_PATH
+        )['sql_db'] != ActiveRecord::Base.connection.instance_variable_get("@config")[:database]
+          say "warning; configured database name is out-of-date"
+          say "please run 'rake ultrasphinx:configure'"
+        end 
+      rescue Object
+      end
     end
   end
         

data/test/test_helper.rb

@@ -1,4 +1,6 @@
 
+RAILS_ENV = "test"
+
 def silently
   old_stdout, $stdout = $stdout, StringIO.new
   yield
@@ -8,8 +10,6 @@ end
 RAILS_ROOT = File.dirname(__FILE__)
 $LOAD_PATH << "#{RAILS_ROOT}/../lib" << RAILS_ROOT
 
-RAILS_ENV = "test"
-
 require 'rubygems'
 require 'initializer'
 require 'active_support'

data/test/unit/parser_test.rb

@@ -81,7 +81,13 @@ describe "parser" do
     '(800) 555-LOVE',
     
     'Bend, OR',
-    'Bend, OR'
+    'Bend, OR',
+
+    '"(traditional)"',
+    '"traditional"',   
+    
+    'cuisine:"american (traditional"',
+    '@cuisine "american traditional"'
     
   ].in_groups_of(2).each do |query, result|
     it "should parse" do

data/ultrasphinx.gemspec

@@ -1,14 +1,14 @@
 
-# Gem::Specification for Ultrasphinx-1.5
+# Gem::Specification for Ultrasphinx-1.5.1
 # Originally generated by Echoe
 
 Gem::Specification.new do |s|
   s.name = %q{ultrasphinx}
-  s.version = "1.5"
-  s.date = %q{2007-09-19}
+  s.version = "1.5.1"
+  s.date = %q{2007-09-25}
   s.summary = %q{Ruby on Rails configurator and client to the Sphinx fulltext search engine.}
   s.email = %q{}
-  s.homepage = %q{http://blog.evanweaver.com/pages/code#ultrasphinx}
+  s.homepage = %q{http://blog.evanweaver.com/files/doc/fauna/ultrasphinx/}
   s.rubyforge_project = %q{fauna}
   s.description = %q{Ruby on Rails configurator and client to the Sphinx fulltext search engine.}
   s.has_rdoc = true
@@ -28,7 +28,7 @@ end
 # Echoe.new("ultrasphinx") do |p|
 #   p.project = "fauna"
 #   p.summary = "Ruby on Rails configurator and client to the Sphinx fulltext search engine."
-#   p.url = "http://blog.evanweaver.com/pages/code#ultrasphinx"  
+#   p.url = "http://blog.evanweaver.com/files/doc/fauna/ultrasphinx/"  
 #   p.docs_host = "blog.evanweaver.com:~/www/bax/public/files/doc/"  
 #   p.rdoc_pattern = /is_indexed.rb|search.rb|spell.rb|ultrasphinx.rb|^README|TODO|CHANGELOG|^LICENSE/
 #   p.dependencies = "chronic"

metadata

@@ -3,13 +3,13 @@ rubygems_version: 0.9.4
 specification_version: 1
 name: ultrasphinx
 version: !ruby/object:Gem::Version 
-  version: "1.5"
-date: 2007-09-19 00:00:00 -04:00
+  version: 1.5.1
+date: 2007-09-25 00:00:00 -04:00
 summary: Ruby on Rails configurator and client to the Sphinx fulltext search engine.
 require_paths: 
 - lib
 email: ""
-homepage: http://blog.evanweaver.com/pages/code#ultrasphinx
+homepage: http://blog.evanweaver.com/files/doc/fauna/ultrasphinx/
 rubyforge_project: fauna
 description: Ruby on Rails configurator and client to the Sphinx fulltext search engine.
 autorequire: