xapian-fu 1.4.0 → 1.5.0

data/CHANGELOG.rdoc

@@ -1,3 +1,7 @@
+=== 1.5.0 (23rd May 2012)
+
+* Support for custom term weights when indexing (Mike Ragalie)
+
 === 1.4.0 (13th March 2012)
 
 * Support for indexing Arrays properly

data/README.rdoc

@@ -160,6 +160,20 @@ And any combinations of the above:
 
   db.search("(ruby OR sinatra) -rails xap*")
 
+== Custom term weights
+
+Sometimes you may want to increase the weight of a particular term in
+a document. Xapian supports adding
+{extra weight}(http://trac.xapian.org/wiki/FAQ/ExtraWeight) to a term
+at index time by providing an integer "wdf" (default is 1).
+
+You may set an optional :weights option when initializing a XapianDb.
+The :weights option accepts a Proc or Lambda that will be called with
+the key, value and list of document fields as each term is indexed.
+Your function should return an integer to set the weight to.
+
+  XapianDb.new(:weights => lambda {|k, v, f| k == :title ? 3 : 1}
+
 == Boolean terms
 
 If you want to implement something like [this](http://getting-started-with-xapian.readthedocs.org/en/latest/howtos/boolean_filters.html#searching),

data/lib/xapian_fu/version.rb

@@ -1,3 +1,3 @@
 module XapianFu #:nodoc:
-  VERSION = "1.4.0"
+  VERSION = "1.5.0"
 end

data/lib/xapian_fu/xapian_db.rb

@@ -98,6 +98,26 @@ module XapianFu #:nodoc:
   # and sort them efficiently (without having to resort to storing
   # leading zeros or anything like that).
   #
+  # == Term Weights
+  #
+  # The <tt>:weights</tt> option accepts a Proc or Lambda that sets
+  # custom {term weights}[http://trac.xapian.org/wiki/FAQ/ExtraWeight].
+  #
+  # Your function will receive the term key and value and the full list of
+  # fields, and should return an integer weight to be applied for that term
+  # when the document is indexed.
+  #
+  # In this example,
+  #
+  #   XapianDb.new(:weights => Proc.new do |key, value, fields|
+  #     return 10 if fields.keys.include?('culturally_important')
+  #     return 3  if key == 'title'
+  #     1
+  #   end)
+  #
+  # terms in the title will be weighted three times greater than other terms,
+  # and all terms in 'culturally important' items will weighted 10 times more.
+  #
   class XapianDb # :nonew:
     # Path to the on-disk database. Nil if in-memory database
     attr_reader :dir
@@ -117,6 +137,7 @@ module XapianFu #:nodoc:
     # Whether this db will generate a spelling dictionary during indexing
     attr_reader :spelling
     attr_reader :sortable_fields
+    attr_accessor :weights_function
 
     def initialize( options = { } )
       @options = { :index_positions => true, :spelling => true }.merge(options)
@@ -135,6 +156,7 @@ module XapianFu #:nodoc:
       @store_values << @options[:collapsible]
       @store_values = @store_values.flatten.uniq.compact
       @spelling = @options[:spelling]
+      @weights_function = @options[:weights]
     end
 
     # Return a new stemmer object for this database

data/lib/xapian_fu/xapian_doc.rb

@@ -275,10 +275,12 @@ module XapianFu #:nodoc:
         else
           v = v.to_s
         end
+        # get the custom term weight if a weights function exists
+        weight = db.weights_function ? db.weights_function.call(k, v, fields).to_i : 1
         # add value with field name
-        tg.send(index_method, v, 1, 'X' + k.to_s.upcase)
+        tg.send(index_method, v, weight, 'X' + k.to_s.upcase)
         # add value without field name
-        tg.send(index_method, v)
+        tg.send(index_method, v, weight)
       end
 
       db.boolean_fields.each do |name|

data/spec/xapian_db_spec.rb

@@ -202,6 +202,25 @@ describe XapianDb do
       xdb.size.should == 2
     end
 
+    it "should assign extra weight according to the weights function" do
+      xdb = XapianDb.new(:weights => lambda {|k, v, f| k == :title ? 3 : 1})
+      xdb << { :text => "once upon time", :title => "A story" }
+
+      s = xdb.search("story")
+      terms = s.first.terms
+      terms.select {|t| t.term.match(/story/)}.map(&:wdf).uniq.should == [3]
+      terms.select {|t| t.term.match(/upon/)}.map(&:wdf).uniq.should == [1]
+    end
+
+    it "should use a weight of 1 if no weights function is provided" do
+      xdb = XapianDb.new
+      xdb << { :text => "once upon time", :title => "A story" }
+
+      s = xdb.search("story")
+      terms = s.first.terms
+      terms.map(&:wdf).uniq.should == [1]
+    end
+
     it "should generate boolean terms for multiple values" do
       xdb = XapianDb.new(:dir => tmp_dir, :create => true,
                          :fields => {

metadata

@@ -1,78 +1,60 @@
---- !ruby/object:Gem::Specification 
+--- !ruby/object:Gem::Specification
 name: xapian-fu
-version: !ruby/object:Gem::Version 
-  hash: 7
+version: !ruby/object:Gem::Version
+  version: 1.5.0
   prerelease: 
-  segments: 
-  - 1
-  - 4
-  - 0
-  version: 1.4.0
 platform: ruby
-authors: 
+authors:
 - John Leach
 - Damian Janowski
 autorequire: 
 bindir: bin
 cert_chain: []
-
-date: 2012-03-13 00:00:00 Z
-dependencies: 
-- !ruby/object:Gem::Dependency 
+date: 2012-05-23 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
   name: rspec
-  prerelease: false
-  requirement: &id001 !ruby/object:Gem::Requirement 
+  requirement: &12933040 !ruby/object:Gem::Requirement
     none: false
-    requirements: 
+    requirements:
     - - ~>
-      - !ruby/object:Gem::Version 
-        hash: 19
-        segments: 
-        - 2
-        - 7
-        - 0
+      - !ruby/object:Gem::Version
         version: 2.7.0
   type: :development
-  version_requirements: *id001
-- !ruby/object:Gem::Dependency 
-  name: rake
   prerelease: false
-  requirement: &id002 !ruby/object:Gem::Requirement 
+  version_requirements: *12933040
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: &12928460 !ruby/object:Gem::Requirement
     none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        hash: 3
-        segments: 
-        - 0
-        version: "0"
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
   type: :development
-  version_requirements: *id002
-- !ruby/object:Gem::Dependency 
-  name: rdoc
   prerelease: false
-  requirement: &id003 !ruby/object:Gem::Requirement 
+  version_requirements: *12928460
+- !ruby/object:Gem::Dependency
+  name: rdoc
+  requirement: &12907260 !ruby/object:Gem::Requirement
     none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        hash: 3
-        segments: 
-        - 0
-        version: "0"
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
   type: :development
-  version_requirements: *id003
-description: A library to provide a more Ruby-like interface to the Xapian search engine.
+  prerelease: false
+  version_requirements: *12907260
+description: A library to provide a more Ruby-like interface to the Xapian search
+  engine.
 email: john@johnleach.co.uk
 executables: []
-
 extensions: []
-
-extra_rdoc_files: 
+extra_rdoc_files:
 - README.rdoc
 - LICENSE
 - CHANGELOG.rdoc
-files: 
+files:
 - lib/xapian_fu.rb
 - lib/xapian_fu/xapian_doc_value_accessor.rb
 - lib/xapian_fu/query_parser.rb
@@ -166,42 +148,34 @@ files:
 - spec/query_parser_spec.rb
 homepage: http://github.com/johnl/xapian-fu
 licenses: []
-
 post_install_message: 
-rdoc_options: 
+rdoc_options:
 - --title
 - Xapian Fu
 - --main
 - README.rdoc
 - --line-numbers
-require_paths: 
+require_paths:
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement 
+required_ruby_version: !ruby/object:Gem::Requirement
   none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
-      - 0
-      version: "0"
-required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
-      - 0
-      version: "0"
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
 requirements: []
-
 rubyforge_project: xapian-fu
-rubygems_version: 1.8.10
+rubygems_version: 1.8.11
 signing_key: 
 specification_version: 3
 summary: A Ruby interface to the Xapian search engine
-test_files: 
+test_files:
 - spec/xapian_doc_spec.rb
 - spec/xapian_db_spec.rb
 - spec/stopper_factory_spec.rb
@@ -261,3 +235,4 @@ test_files:
 - spec/xapian_doc_value_accessor_spec.rb
 - spec/build_db_for_value_testing.rb
 - spec/query_parser_spec.rb
+has_rdoc: