kuahyeow-sunspot 0.9.8 → 0.10.3

data/lib/sunspot/search/highlight.rb

@@ -0,0 +1,38 @@
+module Sunspot
+  class Search
+    # 
+    # A Highlight represents a single highlighted fragment of text from a
+    # document. Depending on the highlighting parameters used for search, there
+    # may be more than one Highlight object for a given field in a given result.
+    #
+    class Highlight
+      HIGHLIGHT_MATCHER = /@@@hl@@@(.*?)@@@endhl@@@/ #:nodoc:
+      
+      #
+      # The name of the field in which the highlight appeared.
+      #
+      attr_reader :field_name
+            
+      def initialize(field_name, highlight) #:nodoc:
+        @field_name = field_name.to_sym
+        @highlight = highlight.to_s.strip
+      end
+      
+      #
+      # Returns the highlighted text with formatting according to the template given in &block.
+      # When no block is given, <em> and </em> are used to surround the highlight.
+      #
+      # ==== Example
+      #
+      #   search.highlights(:body).first.format { |word| "<strong>#{word}</strong>" }
+      #
+      def format(&block)
+        block ||= proc { |word| "<em>#{word}</em>" }
+        @highlight.gsub(HIGHLIGHT_MATCHER) do
+          block.call(Regexp.last_match[1])
+        end
+      end
+      alias_method :formatted, :format
+    end
+  end
+end

data/lib/sunspot/search/hit.rb

@@ -1,5 +1,11 @@
 module Sunspot
   class Search
+    # 
+    # Hit objects represent the raw information returned by Solr for a single
+    # document. As well as the primary key and class name, hit objects give
+    # access to stored field values, keyword relevance score, and geographical
+    # distance (for geographical search).
+    #
     class Hit
       SPECIAL_KEYS = Set.new(%w(id type score)) #:nodoc:
 
@@ -16,15 +22,34 @@ module Sunspot
       # is not from a keyword search.
       #
       attr_reader :score
+      #
+      # For geographical searches, this is the distance between the search
+      # centerpoint and the document's location. Otherwise, it's nil.
+      # 
+      attr_reader :distance
 
       attr_writer :instance #:nodoc:
 
-      def initialize(raw_hit, search) #:nodoc:
+      def initialize(raw_hit, highlights, search) #:nodoc:
         @class_name, @primary_key = *raw_hit['id'].match(/([^ ]+) (.+)/)[1..2]
         @score = raw_hit['score']
+        @distance = raw_hit['geo_distance'].to_f if raw_hit['geo_distance']
         @search = search
         @stored_values = raw_hit
         @stored_cache = {}
+        @highlights = highlights
+      end
+      
+      #
+      # Returns all highlights for this hit when called without parameters.
+      # When a field_name is provided, returns only the highlight for this field.
+      #
+      def highlights(field_name = nil)
+        if field_name.nil?
+          highlights_cache.values.flatten 
+        else
+          highlights_cache[field_name.to_sym]
+        end
       end
 
       # 
@@ -41,7 +66,7 @@ module Sunspot
       def stored(field_name)
         @stored_cache[field_name.to_sym] ||=
           begin
-            field = Sunspot::Setup.for(@class_name).field(field_name)
+            field = setup.field(field_name)
             field.cast(@stored_values[field.indexed_name])
           end
       end
@@ -58,9 +83,31 @@ module Sunspot
         @instance
       end
 
-      def inspect
+      def inspect #:nodoc:
         "#<Sunspot::Search::Hit:#{@class_name} #{@primary_key}>"
       end
+
+      private
+
+      def setup
+        @setup ||= Sunspot::Setup.for(@class_name)
+      end
+
+      def highlights_cache
+        @highlights_cache ||=
+          begin
+            cache = {}
+            if @highlights
+              @highlights.each_pair do |indexed_field_name, highlight_strings|
+                field_name = indexed_field_name.sub(/_[a-z]+$/, '').to_sym
+                cache[field_name] = highlight_strings.map do |highlight_string|
+                  Highlight.new(field_name, highlight_string)
+                end
+              end
+            end
+            cache
+          end
+      end
     end
   end
 end

data/lib/sunspot/session.rb

@@ -15,7 +15,7 @@ module Sunspot
       # For testing purposes
       #
       def connection_class #:nodoc:
-        @connection_class ||= RSolr::Connection
+        @connection_class ||= RSolr
       end
     end
 
@@ -29,10 +29,11 @@ module Sunspot
     # connection. Usually you will want to stick with the default arguments
     # when instantiating your own sessions.
     #
-    def initialize(config = Configuration.build, connection = nil)
+    def initialize(config = Configuration.build, connection = nil, master_connection = nil)
       @config = config
       yield(@config) if block_given?
       @connection = connection
+      @master_connection = master_connection
       @updates = 0
     end
 
@@ -41,23 +42,24 @@ module Sunspot
     #
     def new_search(*types)
       types.flatten!
+      if types.empty?
+        raise(ArgumentError, "You must specify at least one type to search")
+      end
       setup =
         if types.length == 1
           Setup.for(types.first)
         else
           CompositeSetup.for(types)
           end
-      Search.new(connection, setup, Query::Query.new(types, setup, @config))
+      Search.new(connection, setup, Query::Query.new(types), @config)
     end
 
     #
     # See Sunspot.search
     #
     def search(*types, &block)
-      options = types.last.is_a?(Hash) ? types.pop : {}
       search = new_search(*types)
       search.build(&block) if block
-      search.query.options = options
       search.execute!
     end
 
@@ -83,7 +85,7 @@ module Sunspot
     #
     def commit
       @updates = 0
-      connection.commit
+      master_connection.commit
     end
 
     # 
@@ -133,7 +135,7 @@ module Sunspot
       classes.flatten!
       if classes.empty?
         @updates += 1
-        Indexer.remove_all(connection)
+        Indexer.remove_all(master_connection)
       else
         @updates += classes.length
         for clazz in classes
@@ -186,16 +188,43 @@ module Sunspot
     def connection
       @connection ||=
         begin
-          connection = self.class.connection_class.new(
-            RSolr::Adapter::HTTP.new(:url => config.solr.url)
+          connection = self.class.connection_class.connect(
+            :url => config.solr.url,
+            :adapter => config.http_client
           )
-          connection.adapter.connector.adapter_name = config.http_client
+          connection.message.adapter =
+            RSolr::Message::Adapter.const_get(
+              Util.camel_case(config.xml_builder.to_s)
+            ).new
           connection
         end
     end
 
+    # 
+    # Retrieve the Solr connection to the master for this session, creating one
+    # if it does not already exist.
+    #
+    # ==== Returns
+    #
+    # Solr::Connection:: The connection for this session
+    #
+    def master_connection
+      @master_connection ||=
+        begin
+          if config.master_solr.url && config.master_solr.url != config.solr.url
+            master_connection = self.class.connection_class.new(
+              RSolr::Adapter::HTTP.new(:url => config.master_solr.url)
+            )
+            master_connection.adapter.connector.adapter_name = config.http_client
+            master_connection
+          else
+            connection
+          end
+        end
+    end
+
     def indexer
-      @indexer ||= Indexer.new(connection)
+      @indexer ||= Indexer.new(master_connection)
     end
   end
 end

data/lib/sunspot/setup.rb

@@ -57,6 +57,20 @@ module Sunspot
       @dynamic_field_factories_cache[field_factory.name] = field_factory
     end
 
+    # 
+    # The coordinates field factory is used for populating the coordinate fields
+    # of documents during index, but does not actually generate fields (since
+    # the field names used in search are static).
+    #
+    def set_coordinates_field(name)
+      @coordinates_field_factory = FieldFactory::Coordinates.new(name)
+    end
+
+    # 
+    # Add a document boost to documents at index time. Document boost can be
+    # static (the same for all documents of this class), or extracted on a per-
+    # document basis using either attribute or block extraction as per usual.
+    #
     def add_document_boost(attr_name, &block)
       @document_boost_extractor =
         if attr_name
@@ -77,6 +91,9 @@ module Sunspot
       @dsl.instance_eval(&block)
     end
 
+    # 
+    # Return the Field with the given (public-facing) name
+    #
     def field(field_name)
       if field_factory = @field_factories_cache[field_name.to_sym]
         field_factory.build
@@ -88,17 +105,27 @@ module Sunspot
       end
     end
 
-    def text_field(field_name)
-      if field_factory = @text_field_factories_cache[field_name.to_sym]
-        field_factory.build
-      else
-        raise(
-          UnrecognizedFieldError,
-          "No text field configured for #{@clazz.name} with name '#{field_name}'"
-        )
-      end
+    # 
+    # Return one or more text fields with the given public-facing name. This
+    # implementation will always return a single field (in an array), but
+    # CompositeSetup objects might return more than one.
+    #
+    def text_fields(field_name)
+      text_field = 
+        if field_factory = @text_field_factories_cache[field_name.to_sym]
+          field_factory.build
+        else
+          raise(
+            UnrecognizedFieldError,
+            "No text field configured for #{@clazz.name} with name '#{field_name}'"
+          )
+        end
+      [text_field]
     end
 
+    # 
+    # Return the DynamicFieldFactory with the given base name
+    #
     def dynamic_field_factory(field_name)
       @dynamic_field_factories_cache[field_name.to_sym] || raise(
         UnrecognizedFieldError,
@@ -106,11 +133,17 @@ module Sunspot
       )
     end
 
+    # 
+    # Return all attribute fields
+    #
     def fields
       field_factories.map { |field_factory| field_factory.build }
     end
 
-    def text_fields
+    # 
+    # Return all text fields
+    #
+    def all_text_fields
       text_field_factories.map { |text_field_factory| text_field_factory.build }
     end
 
@@ -148,6 +181,7 @@ module Sunspot
     def all_field_factories
       all_field_factories = []
       all_field_factories.concat(field_factories).concat(text_field_factories).concat(dynamic_field_factories)
+      all_field_factories << @coordinates_field_factory if @coordinates_field_factory
       all_field_factories
     end
 
@@ -173,6 +207,9 @@ module Sunspot
       Util.full_const_get(@class_name)
     end
 
+    # 
+    # Get the document boost for a given model
+    #
     def document_boost_for(model)
       if @document_boost_extractor
         @document_boost_extractor.value_for(model)

data/lib/sunspot/text_field_setup.rb

@@ -0,0 +1,29 @@
+module Sunspot
+  # 
+  # A TextFieldSetup encapsulates a regular (or composite) setup, and exposes
+  # the #field() method returning text fields instead of attribute fields.
+  #
+  class TextFieldSetup #:nodoc:
+    def initialize(setup)
+      @setup = setup
+    end
+
+    # 
+    # Return a text field with the given name. Duck-type compatible with
+    # Setup and CompositeSetup, but return text fields instead.
+    #
+    def field(name)
+      fields = @setup.text_fields(name)
+      if fields
+        if fields.length == 1
+          fields.first
+        else
+          raise(
+            Sunspot::UnrecognizedFieldError,
+            "The text field with name #{name} has incompatible configurations for the classes #{@setup.type_names.join(', ')}"
+          )
+        end
+      end
+    end
+  end
+end

data/lib/sunspot/type.rb

@@ -147,7 +147,7 @@ module Sunspot
           end
         end
 
-        def cast(string)
+        def cast(string) #:nodoc:
           time = Time.xmlschema(string)
           Date.civil(time.year, time.mon, time.mday)
         end
@@ -183,15 +183,15 @@ module Sunspot
 
     module ClassType
       class <<self
-        def indexed_name(name)
+        def indexed_name(name) #:nodoc:
           'class_name'
         end
 
-        def to_indexed(value)
+        def to_indexed(value) #:nodoc:
           value.name
         end
 
-        def cast(string)
+        def cast(string) #:nodoc:
           Sunspot::Util.full_const_get(string)
         end
       end

data/lib/sunspot/util.rb

@@ -66,7 +66,7 @@ module Sunspot
       #
       def full_const_get(string)
         string.split('::').inject(Object) do |context, const_name|
-          context.const_get(const_name)
+          context.const_defined?(const_name) ? context.const_get(const_name) : context.const_missing(const_name)
         end
       end
 
@@ -160,5 +160,31 @@ module Sunspot
         destination
       end
     end
+
+    class Coordinates #:nodoc:
+      def initialize(coords)
+        @coords = coords
+      end
+
+      def lat
+        if @coords.respond_to?(:[])
+          @coords[0]
+        else
+          @coords.lat
+        end.to_f
+      end
+
+      def lng
+        if @coords.respond_to?(:[])
+          @coords[1]
+        elsif @coords.respond_to?(:lng)
+          @coords.lng
+        elsif @coords.respond_to?(:lon)
+          @coords.lon
+        elsif @coords.respond_to?(:long)
+          @coords.long
+        end.to_f
+      end
+    end
   end
 end

data/solr/solr/conf/schema.xml

@@ -1,50 +1,64 @@
-<?xml version='1.0' encoding='utf-8' ?>
-<schema name='sunspot' version='0.9'>
+<?xml version="1.0" encoding="UTF-8"?>
+<schema version="0.9" name="sunspot">
   <types>
-    <fieldtype class='solr.TextField' name='text' positionIncrementGap='100'>
+    <fieldtype class="solr.TextField" positionIncrementGap="100" name="text">
       <analyzer>
-        <tokenizer class='solr.StandardTokenizerFactory' />
-        <filter class='solr.StandardFilterFactory' />
-        <filter class='solr.LowerCaseFilterFactory' />
+        <tokenizer class="solr.StandardTokenizerFactory"/>
+        <filter class="solr.StandardFilterFactory"/>
+        <filter class="solr.LowerCaseFilterFactory"/>
       </analyzer>
     </fieldtype>
-    <fieldtype class='solr.RandomSortField' name='rand'></fieldtype>
-    <fieldtype class='solr.BoolField' name='boolean' omitNorms='true' />
-    <fieldtype class='solr.SortableFloatField' name='sfloat' omitNorms='true' />
-    <fieldtype class='solr.DateField' name='date' omitNorms='true' />
-    <fieldtype class='solr.SortableIntField' name='sint' omitNorms='true' />
-    <fieldtype class='solr.StrField' name='string' omitNorms='true' />
+    <fieldtype class="solr.RandomSortField" name="rand"/>
+    <fieldtype name="boolean" class="solr.BoolField" omitNorms="true"/>
+    <fieldtype name="sfloat" class="solr.SortableFloatField" omitNorms="true"/>
+    <fieldtype name="date" class="solr.DateField" omitNorms="true"/>
+    <fieldtype name="sint" class="solr.SortableIntField" omitNorms="true"/>
+    <fieldtype name="string" class="solr.StrField" omitNorms="true"/>
+    <fieldtype name="sdouble" class="solr.SortableDoubleField" omitNorms="true"/>
+    <fieldtype name="slong" class="solr.SortableLongField" omitNorms="true"/>
   </types>
   <fields>
-    <field indexed='true' multiValued='false' name='id' stored='true' type='string' />
-    <field indexed='true' multiValued='true' name='type' stored='false' type='string' />
-    <field indexed='true' multiValued='false' name='class_name' stored='false' type='string' />
-    <field indexed='true' multiValued='true' name='text' stored='false' type='text' />
-    <dynamicField indexed='true' multiValued='true' name='*_text' stored='false' type='text' />
-    <dynamicField indexed='true' name='random_*' stored='false' type='rand' />
-    <dynamicField indexed='true' multiValued='false' name='*_b' stored='false' type='boolean' />
-    <dynamicField indexed='true' multiValued='false' name='*_f' stored='false' type='sfloat' />
-    <dynamicField indexed='true' multiValued='false' name='*_d' stored='false' type='date' />
-    <dynamicField indexed='true' multiValued='false' name='*_i' stored='false' type='sint' />
-    <dynamicField indexed='true' multiValued='false' name='*_s' stored='false' type='string' />
-    <dynamicField indexed='true' multiValued='true' name='*_bm' stored='false' type='boolean' />
-    <dynamicField indexed='true' multiValued='true' name='*_fm' stored='false' type='sfloat' />
-    <dynamicField indexed='true' multiValued='true' name='*_dm' stored='false' type='date' />
-    <dynamicField indexed='true' multiValued='true' name='*_im' stored='false' type='sint' />
-    <dynamicField indexed='true' multiValued='true' name='*_sm' stored='false' type='string' />
-    <dynamicField indexed='true' multiValued='false' name='*_bs' stored='true' type='boolean' />
-    <dynamicField indexed='true' multiValued='false' name='*_fs' stored='true' type='sfloat' />
-    <dynamicField indexed='true' multiValued='false' name='*_ds' stored='true' type='date' />
-    <dynamicField indexed='true' multiValued='false' name='*_is' stored='true' type='sint' />
-    <dynamicField indexed='true' multiValued='false' name='*_ss' stored='true' type='string' />
-    <dynamicField indexed='true' multiValued='true' name='*_bms' stored='true' type='boolean' />
-    <dynamicField indexed='true' multiValued='true' name='*_fms' stored='true' type='sfloat' />
-    <dynamicField indexed='true' multiValued='true' name='*_dms' stored='true' type='date' />
-    <dynamicField indexed='true' multiValued='true' name='*_ims' stored='true' type='sint' />
-    <dynamicField indexed='true' multiValued='true' name='*_sms' stored='true' type='string' />
+    <field name="id"         type="string"  indexed="true" stored="true"  multiValued="false" />
+    <field name="type"       type="string"  indexed="true" stored="false" multiValued="true" />
+    <field name="class_name" type="string"  indexed="true" stored="false" multiValued="false" />
+    <field name="text"       type="text"    indexed="true" stored="false" multiValued="true" />
+    <field name="lat"        type="sdouble" indexed="true" stored="true"  multiValued="false" />
+    <field name="long"       type="sdouble" indexed="true" stored="true"  multiValued="false" />
+    <dynamicField name="*_text"   type="text"    indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_texts"  type="text"    indexed="true" stored="true"  multiValued="true" />
+    <dynamicField name="random_*" type="rand"    indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="_local*" type="sdouble" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_b" type="boolean" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_f" type="sfloat" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_d" type="date" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_i" type="sint" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_s" type="string" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_e" type="sdouble" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_l" type="slong" indexed="true" stored="false" multiValued="false" />
+    <dynamicField name="*_bm" type="boolean" indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_fm" type="sfloat" indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_dm" type="date" indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_im" type="sint" indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_sm" type="string" indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_em" type="sdouble" indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_lm" type="slong" indexed="true" stored="false" multiValued="true" />
+    <dynamicField name="*_bs" type="boolean" indexed="true" stored="true" multiValued="false" />
+    <dynamicField name="*_fs" type="sfloat" indexed="true" stored="true" multiValued="false" />
+    <dynamicField name="*_ds" type="date" indexed="true" stored="true" multiValued="false" />
+    <dynamicField name="*_is" type="sint" indexed="true" stored="true" multiValued="false" />
+    <dynamicField name="*_ss" type="string" indexed="true" stored="true" multiValued="false" />
+    <dynamicField name="*_es" type="sdouble" indexed="true" stored="true" multiValued="false" />
+    <dynamicField name="*_ls" type="slong" indexed="true" stored="true" multiValued="false" />
+    <dynamicField name="*_bms" type="boolean" indexed="true" stored="true" multiValued="true" />
+    <dynamicField name="*_fms" type="sfloat" indexed="true" stored="true" multiValued="true" />
+    <dynamicField name="*_dms" type="date" indexed="true" stored="true" multiValued="true" />
+    <dynamicField name="*_ims" type="sint" indexed="true" stored="true" multiValued="true" />
+    <dynamicField name="*_sms" type="string" indexed="true" stored="true" multiValued="true" />
+    <dynamicField name="*_ems" type="sdouble" indexed="true" stored="true" multiValued="true" />
+    <dynamicField name="*_lms" type="slong" indexed="true" stored="true" multiValued="true" />
   </fields>
   <uniqueKey>id</uniqueKey>
   <defaultSearchField>text</defaultSearchField>
-  <solrQueryParser defaultOperator='AND' />
-  <copyField dest='text' source='*_text' />
+  <solrQueryParser defaultOperator="AND"/>
+  <copyField dest="text" source="*_text"/>
 </schema>