thinking-sphinx 2.0.6 → 2.0.7

data/lib/thinking_sphinx/attribute.rb

@@ -0,0 +1,407 @@
+module ThinkingSphinx
+  # Attributes - eternally useful when it comes to filtering, sorting or
+  # grouping. This class isn't really useful to you unless you're hacking
+  # around with the internals of Thinking Sphinx - but hey, don't let that
+  # stop you.
+  #
+  # One key thing to remember - if you're using the attribute manually to
+  # generate SQL statements, you'll need to set the base model, and all the
+  # associations. Which can get messy. Use Index.link!, it really helps.
+  #
+  class Attribute < ThinkingSphinx::Property
+    attr_accessor :query_source
+
+    SphinxTypeMappings = {
+      :multi     => :sql_attr_multi,
+      :datetime  => :sql_attr_timestamp,
+      :string    => :sql_attr_str2ordinal,
+      :float     => :sql_attr_float,
+      :boolean   => :sql_attr_bool,
+      :integer   => :sql_attr_uint,
+      :bigint    => :sql_attr_bigint,
+      :wordcount => :sql_attr_str2wordcount
+    }
+
+    if Riddle.loaded_version.to_i > 1
+      SphinxTypeMappings[:string] = :sql_attr_string
+    end
+
+    # To create a new attribute, you'll need to pass in either a single Column
+    # or an array of them, and some (optional) options.
+    #
+    # Valid options are:
+    # - :as     => :alias_name
+    # - :type   => :attribute_type
+    # - :source => :field, :query, :ranged_query
+    #
+    # Alias is only required in three circumstances: when there's
+    # another attribute or field with the same name, when the column name is
+    # 'id', or when there's more than one column.
+    #
+    # Type is not required, unless you want to force a column to be a certain
+    # type (but keep in mind the value will not be CASTed in the SQL
+    # statements). The only time you really need to use this is when the type
+    # can't be figured out by the column - ie: when not actually using a
+    # database column as your source.
+    #
+    # Source is only used for multi-value attributes (MVA). By default this will
+    # use a left-join and a group_concat to obtain the values. For better performance
+    # during indexing it can be beneficial to let Sphinx use a separate query to retrieve
+    # all document,value-pairs.
+    # Either :query or :ranged_query will enable this feature, where :ranged_query will cause
+    # the query to be executed incremental.
+    #
+    # Example usage:
+    #
+    #   Attribute.new(
+    #     Column.new(:created_at)
+    #   )
+    #
+    #   Attribute.new(
+    #     Column.new(:posts, :id),
+    #     :as => :post_ids
+    #   )
+    #
+    #   Attribute.new(
+    #     Column.new(:posts, :id),
+    #     :as => :post_ids,
+    #     :source => :ranged_query
+    #   )
+    #
+    #   Attribute.new(
+    #     [Column.new(:pages, :id), Column.new(:articles, :id)],
+    #     :as => :content_ids
+    #   )
+    #
+    #   Attribute.new(
+    #     Column.new("NOW()"),
+    #     :as   => :indexed_at,
+    #     :type => :datetime
+    #   )
+    #
+    # If you're creating attributes for latitude and longitude, don't forget
+    # that Sphinx expects these values to be in radians.
+    #
+    def initialize(source, columns, options = {})
+      super
+
+      @type           = options[:type]
+      @query_source   = options[:source]
+      @crc            = options[:crc]
+      @all_ints       = options[:all_ints]
+
+      @type         ||= :multi    unless @query_source.nil?
+      if @type == :string && @crc
+        @type = is_many? ? :multi : :integer
+      end
+
+      source.attributes << self
+    end
+
+    # Get the part of the SELECT clause related to this attribute. Don't forget
+    # to set your model and associations first though.
+    #
+    # This will concatenate strings and arrays of integers, and convert
+    # datetimes to timestamps, as needed.
+    #
+    def to_select_sql
+      return nil unless include_as_association? && available?
+
+      separator = all_ints? || all_datetimes? || @crc ? ',' : ' '
+
+      clause = columns_with_prefixes.collect { |column|
+        case type
+        when :string
+          adapter.convert_nulls(column)
+        when :datetime
+          adapter.cast_to_datetime(column)
+        when :multi
+          column = adapter.cast_to_datetime(column)   if is_many_datetimes?
+          column = adapter.convert_nulls(column, '0') if is_many_ints?
+          column
+        else
+          column
+        end
+      }.join(', ')
+
+      clause = adapter.crc(clause)                          if @crc
+      clause = adapter.concatenate(clause, separator)       if concat_ws?
+      clause = adapter.group_concatenate(clause, separator) if is_many?
+      clause = adapter.downcase(clause)                     if insensitive?
+
+      "#{clause} AS #{quote_column(unique_name)}"
+    end
+
+    def type_to_config
+      SphinxTypeMappings[type]
+    end
+
+    def include_as_association?
+      ! (type == :multi && (query_source == :query || query_source == :ranged_query))
+    end
+
+    # Returns the configuration value that should be used for
+    # the attribute.
+    # Special case is the multi-valued attribute that needs some
+    # extra configuration.
+    #
+    def config_value(offset = nil, delta = false)
+      if type == :multi
+        multi_config = include_as_association? ? "field" :
+          source_value(offset, delta).gsub(/\s+/m, " ").strip
+        "uint #{unique_name} from #{multi_config}"
+      else
+        unique_name
+      end
+    end
+
+    # Returns the type of the column. If that's not already set, it returns
+    # :multi if there's the possibility of more than one value, :string if
+    # there's more than one association, otherwise it figures out what the
+    # actual column's datatype is and returns that.
+    #
+    def type
+      @type ||= begin
+        base_type = case
+        when is_many?, is_many_ints?
+          :multi
+        when @associations.values.flatten.length > 1
+          :string
+        else
+          translated_type_from_database
+        end
+
+        if base_type == :string && @crc
+          base_type = :integer
+        else
+          @crc = false unless base_type == :multi && is_many_strings? && @crc
+        end
+
+        base_type
+      end
+    end
+
+    def updatable?
+      [:integer, :datetime, :boolean].include?(type) &&
+      unique_name != :sphinx_internal_id &&
+      !is_string?
+    end
+
+    def live_value(instance)
+      object = instance
+      column = @columns.first
+      column.__stack.each { |method|
+        object = object.send(method)
+        return sphinx_value(nil) if object.nil?
+      }
+
+      sphinx_value object.send(column.__name)
+    end
+
+    def all_ints?
+      @all_ints || all_of_type?(:integer)
+    end
+
+    def all_datetimes?
+      all_of_type?(:datetime, :date, :timestamp)
+    end
+
+    def all_strings?
+      all_of_type?(:string, :text)
+    end
+
+    private
+
+    def source_value(offset, delta)
+      if is_string?
+        return "#{query_source.to_s.dasherize}; #{columns.first.__name}"
+      end
+
+      query = query(offset)
+
+      if query_source == :ranged_query
+        query += query_clause
+        query += " AND #{query_delta.strip}" if delta
+        "ranged-query; #{query}; #{range_query}"
+      else
+        query += " WHERE #{query_delta.strip}" if delta
+        "query; #{query}"
+      end
+    end
+
+    def query(offset)
+      base_assoc = base_association_for_mva
+      end_assoc  = end_association_for_mva
+      raise "Could not determine SQL for MVA" if base_assoc.nil?
+
+      relation = Arel::Table.new(base_assoc.table)
+
+      association_joins.each do |join|
+        relation = relation.join(join.relation, Arel::OuterJoin).
+          on(*join.association_join)
+      end
+
+      relation = relation.project "#{foreign_key_for_mva base_assoc} #{ThinkingSphinx.unique_id_expression(adapter, offset)} AS #{quote_column('id')}, #{primary_key_for_mva(end_assoc)} AS #{quote_column(unique_name)}"
+
+      relation.to_sql
+    end
+
+    def query_clause
+      foreign_key = foreign_key_for_mva base_association_for_mva
+      " WHERE #{foreign_key} >= $start AND #{foreign_key} <= $end"
+    end
+
+    def query_delta
+      foreign_key = foreign_key_for_mva base_association_for_mva
+      <<-SQL
+#{foreign_key} IN (SELECT #{quote_column model.primary_key}
+FROM #{model.quoted_table_name}
+WHERE #{@source.index.delta_object.clause(model, true)})
+      SQL
+    end
+
+    def range_query
+      assoc       = base_association_for_mva
+      foreign_key = foreign_key_for_mva assoc
+      "SELECT MIN(#{foreign_key}), MAX(#{foreign_key}) FROM #{quote_table_name assoc.table}"
+    end
+
+    def primary_key_for_mva(assoc)
+      quote_with_table(
+        assoc.table, assoc.primary_key_from_reflection || columns.first.__name
+      )
+    end
+
+    def foreign_key_for_mva(assoc)
+      if ThinkingSphinx.rails_3_1?
+        if assoc.reflection.through_reflection
+          quote_with_table assoc.table, assoc.reflection.through_reflection.foreign_key
+        else
+          quote_with_table assoc.table, assoc.reflection.foreign_key
+        end
+      else
+        quote_with_table assoc.table, assoc.reflection.primary_key_name
+      end
+    end
+
+    def end_association_for_mva
+      @association_for_mva ||= associations[columns.first].detect { |assoc|
+        assoc.has_column?(columns.first.__name)
+      }
+    end
+
+    def base_association_for_mva
+      @first_association_for_mva ||= begin
+        assoc = end_association_for_mva
+        while !assoc.parent.nil?
+          assoc = assoc.parent
+        end
+
+        assoc
+      end
+    end
+
+    def association_joins
+      joins = []
+      assoc = end_association_for_mva
+      while assoc != base_association_for_mva
+        joins << assoc.join
+        assoc = assoc.parent
+      end
+
+      joins
+    end
+
+    def is_many_ints?
+      concat_ws? && all_ints?
+    end
+
+    def is_many_datetimes?
+      is_many? && all_datetimes?
+    end
+
+    def is_many_strings?
+      is_many? && all_strings?
+    end
+
+    def translated_type_from_database
+      case type_from_db = type_from_database
+      when :integer
+        integer_type_from_db
+      when :datetime, :string, :float, :boolean
+        type_from_db
+      when :decimal
+        :float
+      when :timestamp, :date
+        :datetime
+      else
+        raise <<-MESSAGE
+
+Cannot automatically map attribute #{unique_name} in #{@model.name} to an
+equivalent Sphinx type (integer, float, boolean, datetime, string as ordinal).
+You could try to explicitly convert the column's value in your define_index
+block:
+  has "CAST(column AS INT)", :type => :integer, :as => :column
+        MESSAGE
+      end
+    end
+
+    def type_from_database
+      column = column_from_db
+      column.nil? ? nil : column.type
+    end
+
+    def integer_type_from_db
+      column = column_from_db
+      return nil if column.nil?
+
+      case column.sql_type
+      when adapter.bigint_pattern
+        :bigint
+      else
+        :integer
+      end
+    end
+
+    def column_from_db
+      klass = @associations.values.flatten.first ?
+        @associations.values.flatten.first.reflection.klass : @model
+
+      klass.columns.detect { |col|
+        @columns.collect { |c| c.__name.to_s }.include? col.name
+      }
+    end
+
+    def all_of_type?(*column_types)
+      @columns.all? { |col|
+        klasses = @associations[col].empty? ? [@model] :
+          @associations[col].collect { |assoc| assoc.reflection.klass }
+        klasses.all? { |klass|
+          column = klass.columns.detect { |column| column.name == col.__name.to_s }
+          !column.nil? && column_types.include?(column.type)
+        }
+      }
+    end
+
+    def sphinx_value(value)
+      case value
+      when TrueClass
+        1
+      when FalseClass, NilClass
+        0
+      when Time
+        value.to_i
+      when Date
+        value.to_time.to_i
+      when String
+        value.to_crc32
+      else
+        value
+      end
+    end
+
+    def insensitive?
+      @sortable == :insensitive
+    end
+  end
+end

data/lib/thinking_sphinx/auto_version.rb

@@ -0,0 +1,38 @@
+module ThinkingSphinx
+  class AutoVersion
+    def self.detect
+      version = ThinkingSphinx::Configuration.instance.version
+      case version
+      when '0.9.8', '0.9.9'
+        require "riddle/#{version}"
+      when /1.10/
+        require 'riddle/1.10'
+      when /2.0.\d/
+        require 'riddle/2.0.1'
+      else
+        documentation_link = %Q{
+For more information, read the documentation:
+http://freelancing-god.github.com/ts/en/advanced_config.html          
+}
+
+        if version.nil? || version.empty?
+          STDERR.puts %Q{
+Sphinx cannot be found on your system. You may need to configure the following
+settings in your config/sphinx.yml file:
+  * bin_path
+  * searchd_binary_name
+  * indexer_binary_name
+
+#{documentation_link}
+}
+        else
+          STDERR.puts %Q{
+Unsupported version: #{version}
+
+#{documentation_link}
+}
+        end
+      end
+    end
+  end
+end

data/lib/thinking_sphinx/bundled_search.rb

@@ -0,0 +1,44 @@
+module ThinkingSphinx
+  class BundledSearch
+    attr_reader :client
+    
+    def initialize
+      @searches = []
+    end
+    
+    def search(*args)
+      @searches << ThinkingSphinx.search(*args)
+      @searches.last.append_to client
+    end
+    
+    def search_for_ids(*args)
+      @searches << ThinkingSphinx.search_for_ids(*args)
+      @searches.last.append_to client
+    end
+    
+    def searches
+      populate
+      @searches
+    end
+    
+    private
+    
+    def client
+      @client ||= ThinkingSphinx::Configuration.instance.client
+    end
+    
+    def populated?
+      @populated
+    end
+    
+    def populate
+      return if populated?
+      
+      @populated = true
+      
+      client.run.each_with_index do |results, index|
+        searches[index].populate_from_queue results
+      end
+    end
+  end
+end

data/lib/thinking_sphinx/class_facet.rb

@@ -0,0 +1,20 @@
+module ThinkingSphinx
+  class ClassFacet < ThinkingSphinx::Facet
+    def name
+      :class
+    end
+    
+    def attribute_name
+      Riddle.loaded_version.to_i < 2 ? 'class_crc' : 'sphinx_internal_class'
+    end
+    
+    def value(object, attribute_hash)
+      if Riddle.loaded_version.to_i < 2
+        crc = attribute_hash['class_crc']
+        ThinkingSphinx::Configuration.instance.models_by_crc[crc]
+      else
+        attribute_hash['sphinx_internal_class']
+      end
+    end
+  end
+end