xapian_db 1.3.5.4 → 1.3.7.4

data/README.rdoc

@@ -32,9 +32,9 @@ I tried hard but I couldn't find such a thing so I decided to write it, based on
 
 == Requirements
 
-* ruby 1.9.2 or newer
+* ruby 2.0.0 or newer
 * rails 3.0 or newer (if you want to use it with rails)
-* xapian-core and xapian-ruby binaries installed
+* xapian-core and xapian-ruby binaries 1.2.x installed
 
 == Installing xapian binaries
 
@@ -47,7 +47,7 @@ Make sure version 1.2.6 or newer is installed.
 If you want to use xapian_db in a Rails app, you need Rails 3 or newer.
 
 For a first look, look at the examples in the examples folder. There's the simple ruby script basic.rb that shows the basic
-usage of XapianDB without rails. In the basic_rails folder you'll find a very simple Rails app unsing XapianDb.
+usage of XapianDB without rails. In the basic_rails folder you'll find a very simple Rails app using XapianDb.
 
 The following steps assume that you are using xapian_db within a Rails app.
 
@@ -64,7 +64,7 @@ You can override these defaults by placing a config file named 'xapian_db.yml' i
     adapter: datamapper # Avaliable adapters: :active_record, :datamapper
     language: de        # Global language; can be overridden for specific blueprints
     term_min_length: 2  # Ignore single character terms
-    enable_query_flags: FLAG_PHRASE, FLAG_SPELLING_CORRECTION
+    enabled_query_flags: FLAG_PHRASE, FLAG_SPELLING_CORRECTION
 
   development:
     database: db/xapian_db/development
@@ -194,6 +194,32 @@ are ordered by relevance and - within the same relevance - by the natural sort o
     end
   end
 
+You can specify a Ruby method for preprocessing the indexed terms. The method needs to be a class method that takes one argument (the terms to be indexed) and
+returns a string. You can configure it globaly in the config or on a blueprint. For example, this setup will make words with accented e characters searchable
+by their accentless form:
+
+  class Util
+    def self.strip_accents(terms)
+      terms.gsub(/[éèêëÉÈÊË]/, "e")
+    end
+  end
+
+  XapianDb::Config.setup do |config|
+    config.indexer_preprocess_callback Util.method(:strip_accents)
+  end
+
+You may use attributes from associated objects in a blueprint; if you do that and an associated object is updated, your objects should be reindexed, too. You can tell XapnaDB about those dependencies like so:
+
+  XapianDb::DocumentBlueprint.setup(:Person) do |blueprint|
+    blueprint.attribue :address, :as => :json
+
+    blueprint.dependency :Address, when_changed: %i(street zip city) do |address|
+      Person.joins{ address }.where{ adresses.id == my{ address.id } }
+    end
+  end
+
+The block you supply to the dependency declaration must return a collection of objects that should get reindexed, too.
+
 If you want to manage the (re)indexing of your objects on your own, turn off autoindexing in your blueprint:
 
   XapianDb::DocumentBlueprint.setup(:Person) do |blueprint|
@@ -246,6 +272,10 @@ You can query attributes:
 
   results = XapianDb.search "name:Foo"
 
+You can force the sort order:
+
+  results = XapianDb.search "name:Foo", order: [:name]
+
 You can query objects of a specific class:
 
   results = Person.search "name:Foo"

data/lib/type_codec.rb

@@ -146,7 +146,7 @@ module XapianDb
       # @return [String] the encoded number
       def self.encode(number)
         case number.class.name
-          when "Fixnum", "Float", "Bignum"
+          when "Fixnum", "Float", "Bignum", "Integer"
             Xapian::sortable_serialise number
           when "BigDecimal"
             Xapian::sortable_serialise number.to_f
@@ -160,7 +160,7 @@ module XapianDb
       # @return [BigDecimal] the decoded number
       def self.decode(encoded_number)
         begin
-          BigDecimal.new(Xapian::sortable_unserialise(encoded_number).to_s)
+          BigDecimal(Xapian::sortable_unserialise(encoded_number).to_s)
         rescue TypeError
           raise ArgumentError.new "#{encoded_number} cannot be unserialized"
         end
@@ -175,7 +175,7 @@ module XapianDb
       def self.encode(number)
         return nil if number.nil?
         case number.class.name
-          when "Fixnum"
+          when "Fixnum", "Integer"
             Xapian::sortable_serialise number
           else
             raise ArgumentError.new "#{number} was expected to be an integer"

data/lib/xapian_db.rb

@@ -105,9 +105,9 @@ module XapianDb
 
   # Update an object in the index
   # @param [Object] obj An instance of a class with a blueprint configuration
-  def self.index(obj, commit=true)
+  def self.index(obj, commit=true, changed_attrs: [])
     writer = @block_writer || XapianDb::Config.writer
-    writer.index obj, commit
+    writer.index obj, commit, changed_attrs: changed_attrs
   end
 
   # Remove a document from the index
@@ -119,11 +119,11 @@ module XapianDb
 
   # Update or delete a xapian document belonging to an object depending on the ignore_if logic(if present)
   # @param [Object] object An instance of a class with a blueprint configuration
-  def self.reindex(object, commit=true)
+  def self.reindex(object, commit=true, changed_attrs: [])
     writer = @block_writer || XapianDb::Config.writer
     blueprint = XapianDb::DocumentBlueprint.blueprint_for object.class.name
     if blueprint.should_index?(object)
-      writer.index object, commit
+      writer.index object, commit, changed_attrs: changed_attrs
     else
       writer.delete_doc_with object.xapian_id, commit
     end
@@ -191,5 +191,4 @@ module XapianDb
       @block_writer = nil
     end
   end
-
 end

data/lib/xapian_db/adapters/active_record_adapter.rb

@@ -49,7 +49,12 @@ module XapianDb
              # add the after commit logic, unless the blueprint has autoindexing turned off
              if XapianDb::DocumentBlueprint.blueprint_for(klass.name).autoindex?
                after_commit do
-                 XapianDb.reindex(self) unless self.destroyed?
+                 unless self.destroyed?
+                   XapianDb.reindex(self, true, changed_attrs: self.previous_changes.keys)
+                   XapianDb::DocumentBlueprint.dependencies_for(klass.name, self.previous_changes.keys).each do |dependency|
+                     dependency.block.call(self).each{ |model| XapianDb.reindex model, true, changed_attrs: self.previous_changes.keys }
+                   end
+                 end
                end
              end
 
@@ -64,7 +69,6 @@ module XapianDb
                XapianDb.reindex_class(klass, options)
              end
            end
-
          end
 
          # Implement the document helper methods on a module
@@ -76,7 +80,7 @@ module XapianDb
 
              # Implement access to the model id
              define_method :id do
-               return @id unless @d.nil?
+               return @id unless @id.nil?
                # retrieve the class and id from data
                klass_name, id = data.split("-")
                @id = id.to_i
@@ -90,11 +94,8 @@ module XapianDb
                klass =  constantize klass_name
                @indexed_object = klass.find(id.to_i)
              end
-
            end
-
          end
-
        end
      end
    end

data/lib/xapian_db/config.rb

@@ -32,7 +32,7 @@ module XapianDb
       end
 
       # Install delegates for the config instance variables
-      [:database, :adapter, :writer, :stemmer, :stopper].each do |attr|
+      [:database, :adapter, :writer, :stemmer, :stopper, :preprocess_terms].each do |attr|
         define_method attr do
           @config.nil? ? nil : @config.instance_variable_get("@_#{attr}")
         end
@@ -183,5 +183,21 @@ module XapianDb
       @_enabled_query_flags.delete flag
     end
 
+    # Set the indexer preprocess callback.
+    # @param [Method] method a class method; needs to take one parameter and return a string.
+    # @example
+    #   class Util
+    #     def self.strip_accents(terms)
+    #       terms.gsub(/[éèêëÉÈÊË]/, "e")
+    #     end
+    #   end
+    #
+    #   XapianDb::Config.setup do |config|
+    #     config.indexer_preprocess_callback Util.method(:strip_accents)
+    #   end
+    def indexer_preprocess_callback(method)
+      @_preprocess_terms = method
+    end
+
   end
 end

data/lib/xapian_db/database.rb

@@ -35,10 +35,19 @@ module XapianDb
       true
     end
 
-    # Delete all docs of a specific class
+    # Delete all docs of a specific class.
+    #
+    # If `klass` tracks its descendants, then docs of any subclasses will be deleted, too.
+    # (ActiveRecord does this by default; the gem 'descendants_tracker' offers an alternative.)
+    #
     # @param [Class] klass A class that has a {XapianDb::DocumentBlueprint} configuration
     def delete_docs_of_class(klass)
       writer.delete_document("C#{klass}")
+      if klass.respond_to? :descendants
+        klass.descendants.each do |subclass|
+          writer.delete_document("C#{subclass}")
+        end
+      end
       true
     end
 
@@ -73,6 +82,12 @@ module XapianDb
       enquiry.query  = query
       sort_indices   = opts.delete :sort_indices
       sort_decending = opts.delete :sort_decending
+      order          = opts.delete :order
+      raise ArgumentError.new "you can't use sort_indices and order, only one of them" if sort_indices && order
+
+      if order
+        sort_indices = order.map{ |attr_name| XapianDb::DocumentBlueprint.value_number_for attr_name.to_sym }
+      end
 
       sorter = Xapian::MultiValueKeyMaker.new
       if sort_indices
@@ -80,7 +95,7 @@ module XapianDb
         enquiry.set_sort_by_key_then_relevance(sorter, sort_decending)
       else
         sorter.add_value DocumentBlueprint.value_number_for(:natural_sort_order)
-        enquiry.set_sort_by_relevance_then_key sorter, true
+        enquiry.set_sort_by_relevance_then_key sorter, false
       end
 
       opts[:spelling_suggestion] = @query_parser.spelling_suggestion

data/lib/xapian_db/document_blueprint.rb

@@ -26,6 +26,8 @@ module XapianDb
     # ---------------------------------------------------------------------------------
     class << self
 
+      attr_reader :blueprints
+
       # Configure the blueprint for a class.
       # Available options:
       # - adapter (see {#adapter} for details)
@@ -50,6 +52,7 @@ module XapianDb
         # We can always do a field search on the name of the indexed class
         @searchable_prefixes << "indexed_class"
         @attributes = @blueprints.values.map { |blueprint| blueprint.attribute_names}.flatten.compact.uniq.sort || []
+        blueprint
       end
 
       # reset the blueprint setup
@@ -66,11 +69,17 @@ module XapianDb
       # Get all configured classes
       # @return [Array<Class>]
       def configured_classes
-       if @blueprints
-         @blueprints.keys.map {|class_name| XapianDb::Utilities.constantize(class_name) }
-       else
-         []
-       end
+        if @blueprints
+          @blueprints.keys.map {|class_name| XapianDb::Utilities.constantize(class_name) }
+        else
+          []
+        end
+      end
+
+      def dependencies_for(klass_name, changed_attrs)
+        @blueprints.values.map(&:dependencies)
+                          .flatten
+                          .select{ |dependency| dependency.dependent_on == klass_name && dependency.interested_in?(changed_attrs) }
       end
 
       # Get the blueprint for a class
@@ -167,7 +176,6 @@ module XapianDb
           end
         end
       end
-
     end
 
     # ---------------------------------------------------------------------------------
@@ -258,15 +266,17 @@ module XapianDb
     # Blueprint DSL methods
     # ---------------------------------------------------------------------------------
 
-    attr_reader :lazy_base_query, :_natural_sort_order
+    attr_reader :lazy_base_query, :_natural_sort_order, :dependencies
 
     # Construct the blueprint
     def initialize
       @attributes_hash      = {}
       @indexed_methods_hash = {}
       @type_map             = {}
+      @dependencies         = []
       @_natural_sort_order  = :id
       @autoindex            = true
+      @indexer_preprocess_callback = nil
     end
 
     # Set the adapter
@@ -377,7 +387,7 @@ module XapianDb
       @ignore_expression = block
     end
 
-# Define a base query to select one or all objects of the indexed class. The reason for a
+    # Define a base query to select one or all objects of the indexed class. The reason for a
     # base query is to optimize the query avoiding th 1+n problematic. The base query should only
     # include joins(...) and includes(...) calls.
     # @param [expression] a base query expression
@@ -400,6 +410,33 @@ module XapianDb
       @_natural_sort_order = name || block
     end
 
+    def dependency(klass_name, when_changed: [], &block)
+      @dependencies << Dependency.new(klass_name.to_s, when_changed, block)
+    end
+
+    # Set the indexer preprocess callback.
+    # @param [Method] method a class method; needs to take one parameter and return a string.
+    # @example
+    #   class Util
+    #     def self.strip_accents(terms)
+    #       terms.gsub(/[éèêëÉÈÊË]/, "e")
+    #     end
+    #   end
+    #
+    #   XapianDb::DocumentBlueprint.setup(:IndexedObject) do |blueprint|
+    #     blueprint.attribute :name
+    #     blueprint.indexer_preprocess_callback Util.method(:strip_accents)
+    #   end
+    def indexer_preprocess_callback(method)
+      @indexer_preprocess_callback = method
+    end
+
+    # Reader for indexer_preprocess_callback.
+    # Returns the terms preprocessing method for this blueprint, the global method from config or nil.
+    def preprocess_terms
+      @indexer_preprocess_callback || XapianDb::Config.preprocess_terms
+    end
+
     # Options for an indexed method
     class IndexOptions
 
@@ -414,7 +451,24 @@ module XapianDb
         @no_split = options[:no_split]
         @block    = options[:block]
       end
+    end
+
+    class Dependency
+
+      attr_reader :dependent_on, :trigger_attributes, :block
 
+      # Constructor
+      # @param [String] klass_name Name of the dependent class
+      # @param [Array] trigger_attributes List of attributes to watch for changes (if empty, triggers on any change)
+      # @option [Block] block Block that is called when changes are detected; the block must return an array of indexeable objects
+      def initialize(klass_name, trigger_attributes, block)
+        @dependent_on, @trigger_attributes, @block = klass_name, trigger_attributes.map(&:to_s), block
+      end
+
+      def interested_in?(changed_attrs)
+        return true if @trigger_attributes.empty?
+        (@trigger_attributes & changed_attrs).any?
+      end
     end
 
     private
@@ -431,7 +485,5 @@ module XapianDb
       @reserved_method_names ||= Xapian::Document.instance_methods
       @reserved_method_names.include?(attr_name.to_sym)
     end
-
   end
-
 end

data/lib/xapian_db/index_writers/beanstalk_worker.rb

@@ -13,7 +13,7 @@ module XapianDb
       def index_task(options)
         klass = constantize options[:class]
         obj   = klass.respond_to?(:get) ? klass.get(options[:id]) : klass.find(options[:id])
-        DirectWriter.index obj
+        DirectWriter.index obj, true, changed_attrs: options[:changed_attrs]
       end
 
       def delete_doc_task(options)
@@ -24,7 +24,6 @@ module XapianDb
         klass = constantize options[:class]
         DirectWriter.reindex_class klass, :verbose => false
       end
-
     end
   end
-end
+end

data/lib/xapian_db/index_writers/beanstalk_writer.rb

@@ -18,8 +18,8 @@ module XapianDb
 
         # Update an object in the index
         # @param [Object] obj An instance of a class with a blueprint configuration
-        def index(obj, commit=true)
-          beanstalk.put( {:task => "index_task", :class => obj.class.name, :id => obj.id }.to_json )
+        def index(obj, commit=true, changed_attrs: [])
+          beanstalk.put( { :task => "index_task", :class => obj.class.name, :id => obj.id, :changed_attrs => changed_attrs }.to_json )
         end
 
         # Remove an object from the index

data/lib/xapian_db/index_writers/direct_writer.rb

@@ -17,11 +17,11 @@ module XapianDb
       class << self
 
         # Update an object in the index
-        # @param [Object] obj An instance of a class with a blueprint configuration
-        def index(obj, commit=true)
-          blueprint = XapianDb::DocumentBlueprint.blueprint_for(obj.class.name)
+        # @param [Object] object An instance of a class with a blueprint configuration
+        def index(object, commit=true, changed_attrs: [])
+          blueprint = XapianDb::DocumentBlueprint.blueprint_for(object.class.name)
           indexer   = XapianDb::Indexer.new(XapianDb.database, blueprint)
-          doc       = indexer.build_document_for(obj)
+          doc       = indexer.build_document_for(object)
           XapianDb.database.store_doc(doc)
           XapianDb.database.commit if commit
         end
@@ -35,10 +35,10 @@ module XapianDb
 
         # Update or delete a xapian document belonging to an object depending on the ignore_if logic(if present)
         # @param [Object] object An instance of a class with a blueprint configuration
-        def reindex(object, commit=true)
+        def reindex(object, commit=true, changed_attrs: [])
           blueprint = XapianDb::DocumentBlueprint.blueprint_for object.class.name
           if blueprint.should_index?(object)
-            index object, commit
+            index object, commit, changed_attrs: changed_attrs
           else
             delete_doc_with object.xapian_id, commit
           end

data/lib/xapian_db/index_writers/no_op_writer.rb

@@ -12,7 +12,7 @@ module XapianDb
 
         # Update an object in the index
         # @param [Object] obj An instance of a class with a blueprint configuration
-        def index(obj, commit=true); end
+        def index(obj, commit=true, changed_attrs: []); end
 
         # Remove an object from the index
         # @param [String] xapian_id The document id
@@ -28,4 +28,4 @@ module XapianDb
       end
     end
   end
-end
+end