xapian-fu 1.6.0 → 1.7.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 97c807d9e1a6c4ce62f3565aaae147f9bca5f2c037ccd39b51d8fda583544451
-  data.tar.gz: 3e12b31653f3cc34914e13597f67af5c632e072fe622447c1da879b92be06ee4
+  metadata.gz: 6c5695db34da1fceed61e5dc6b5b8e997838e15b0bc50e3d3ff0991f8abeba0d
+  data.tar.gz: 3cb11fb4955a85a4ab55a09aa4d0575bb944effa9fd8c1f73ed068002d464c2d
 SHA512:
-  metadata.gz: 47fda9c3e053b1c3e8dd8c9bfecca1ddba6499d9e35d49ad1a37d6b8e4b6710a8f28af88b7d1b4dd927132151df3811f520cdfac30ed8d1b96b3fe62c5abed04
-  data.tar.gz: 628fcdb102e4ab3ed4dcd4fe7832d7a9c15c681085c9bc87251d927e0db9dd305b400617217eaa97dec13bcf9fd73a26811c36e207aa6e9ef86d98a32c242137
+  metadata.gz: 35e471dee5c7cdf52f7370c692c7f15e5da5ad82d0899f243c5dcb9036081ba207fbcf8bd9e0d4b6a6e8f21ed709fe08fc9d058bd04b4c4fbece0773587f4501
+  data.tar.gz: 26b8d54f4cea15bef6580f9a9cf00f4aefd2fc4d1dd97e53d0b600fb72a6b301aa9a9bcbb7b35e0f83b328f928fe375cee05926198ef92e3d22b7fe6325e7ea2

data/CHANGELOG.rdoc

@@ -1,3 +1,8 @@
+=== 1.7.0 (22nd March 2020)
+
+* Add support for setting stemmer_strategy
+* Remove support for Ruby version < 2.1
+
 === 1.6.0 (7th October 2019)
 
 * Support newer versions of xapian

data/README.rdoc

@@ -20,7 +20,7 @@ Hash with full text indexing (and ACID transactions).
 === Xapian Bindings
 
 Xapian Fu requires the Xapian Ruby bindings to be available.  On
-Debian/Ubuntu, you can install the `libxapian-ruby1.8` package to get
+Debian/Ubuntu, you can install the `ruby-xapian` package to get
 them.  Alternatively, you can install the `xapian-ruby` gem, which
 reportedly will also provide them.  You can also just get the
 upstream Xapian release and manually install it.

data/lib/xapian_fu/version.rb

@@ -1,3 +1,3 @@
 module XapianFu #:nodoc:
-  VERSION = "1.6.0"
+  VERSION = "1.7.0"
 end

data/lib/xapian_fu/xapian_db.rb

@@ -38,6 +38,11 @@ module XapianFu #:nodoc:
   # XapianDb to wipe the current on-disk database and start afresh.
   # The default is <tt>false</tt>.
   #
+  # Setting the <tt>:type</tt> option to either :glass or :chert will force that
+  # database backend, if supported. Leave as nil to auto-detect existing
+  # databases and create new databases with the library default (recommended).
+  # Requires xapian >=1.4
+  #
   #   db = XapianDb.new(:dir => '/tmp/mydb', :create => true)
   #
   # == Language, Stemmers and Stoppers
@@ -57,6 +62,10 @@ module XapianFu #:nodoc:
   #
   #   db = XapianDb.new(:language => :italian, :stopper => false)
   #
+  # The <tt>:stopper_strategy</tt> option specifies the default stop strategy
+  # that will be used when indexing and can be: <tt>:none</tt>, <tt>:all</tt> or
+  # <tt>:stemmed</tt>. Defaults to <tt>:stemmed</tt>
+  #
   # == Spelling suggestions
   #
   # The <tt>:spelling</tt> option controls generation of a spelling
@@ -167,6 +176,8 @@ module XapianFu #:nodoc:
     attr_reader :field_options
     attr_accessor :weights_function
     attr :field_weights
+    # The default stopper strategy
+    attr_accessor :stopper_strategy
 
     def initialize( options = { } )
       @options = { :index_positions => true, :spelling => true }.merge(options)
@@ -175,10 +186,23 @@ module XapianFu #:nodoc:
       @db_flag = Xapian::DB_OPEN
       @db_flag = Xapian::DB_CREATE_OR_OPEN if @options[:create]
       @db_flag = Xapian::DB_CREATE_OR_OVERWRITE if @options[:overwrite]
+      case @options[:type]
+      when :glass
+        raise XapianFuError.new("type glass not recognised") unless defined?(Xapian::DB_BACKEND_GLASS)
+        @db_flag |= Xapian::DB_BACKEND_GLASS
+      when :chert
+        raise XapianFuError.new("type chert not recognised") unless defined?(Xapian::DB_BACKEND_CHERT)
+        @db_flag |= Xapian::DB_BACKEND_CHERT
+      when nil
+        # use library defaults
+      else
+        raise XapianFuError.new("type #{@options[:type].inspect} not recognised")
+      end
       @tx_mutex = Mutex.new
       @language = @options.fetch(:language, :english)
       @stemmer = @options.fetch(:stemmer, @language)
       @stopper = @options.fetch(:stopper, @language)
+      @stopper_strategy = @options.fetch(:stopper_strategy, :stemmed)
       @field_options = {}
       setup_fields(@options[:fields])
       @store_values << @options[:store]

data/lib/xapian_fu/xapian_doc.rb

@@ -227,6 +227,27 @@ module XapianFu #:nodoc:
       end
     end
 
+    STOPPER_STRATEGIES = {
+      :none    => 0,
+      :all     => 1,
+      :stemmed => 2
+    }
+
+    def stopper_strategy
+      if @stopper_strategy
+        @stopper_strategy
+      else
+        @stopper_strategy =
+          if ! @options[:stopper_strategy].nil?
+            @options[:stopper_strategy]
+          elsif db
+            db.stopper_strategy
+          else
+            :stemmed
+          end
+      end
+    end
+
     # Return this document's language which is set on initialize, inherited
     # from the database or defaults to :english
     def language
@@ -276,6 +297,7 @@ module XapianFu #:nodoc:
       tg.document = xapian_document
       tg.stopper = stopper if stopper
       tg.stemmer = stemmer
+      tg.set_stopper_strategy(XapianDoc::STOPPER_STRATEGIES.fetch(stopper_strategy, 2))
       tg.set_flags Xapian::TermGenerator::FLAG_SPELLING if db.spelling
       index_method = db.index_positions ? :index_text : :index_text_without_positions
       fields.each do |k,o|

data/spec/xapian_db_spec.rb

@@ -38,6 +38,27 @@ describe XapianDb do
     File.exists?(tmp_dir).should be_true
   end
 
+  it "should create a glass database when type is glass" do
+    pending "this version of xapian doesn't support glass" unless defined?(Xapian::DB_BACKEND_GLASS)
+    glassdir = tmp_dir + 'glass'
+    XapianDb.new(:dir => glassdir, :create => true, :type => :glass).rw
+    File.exists?(glassdir + '/iamglass').should == true
+  end
+
+  it "should create a chert database when type is chert" do
+    pending "this version of xapian doesn't support chert" unless defined?(Xapian::DB_BACKEND_CHERT)
+    chertdir = tmp_dir + 'chert'
+    XapianDb.new(:dir => chertdir, :create => true, :type => :chert).rw
+    File.exists?(chertdir + '/iamchert').should == true
+  end
+
+  it "should raise an exception when type is unrecognised" do
+    chertdir = tmp_dir + 'whatever'
+    lambda {
+      XapianDb.new(:dir => chertdir, :create => true, :type => :whatever).rw
+    }.should raise_error(XapianFu::XapianFuError)
+  end
+
   it "should flush documents to the index when flush is called" do
     xdb = XapianDb.new(:dir => tmp_dir, :create => true)
     xdb.flush

data/spec/xapian_doc_spec.rb

@@ -261,4 +261,38 @@ describe XapianDoc do
     end
   end
 
+  describe "stopper_strategy" do
+    it "should stop all stop words when stopper_strategy is set to :all " do
+      xdb = XapianDb.new(:stopper_strategy => :all )
+      xdoc = xdb.documents.new("She fished for fish").to_xapian_document
+      terms = xdoc.terms.collect { |t| t.term }
+      terms.should_not include "for"
+      terms.should include "fish"
+    end
+
+    it "should stop stemmed words by when stopper_strategy is set to :stemmed " do
+      xdb = XapianDb.new(:stopper_strategy => :stemmed)
+      xdoc = xdb.documents.new("She fished for fish").to_xapian_document
+      terms = xdoc.terms.collect { |t| t.term }
+      terms.should_not include "Zfor"
+      terms.should include "fish"
+    end
+
+    it "should stop no words by when stopper_strategy is set to :none " do
+      xdb = XapianDb.new(:stopper_strategy => :none)
+      xdoc = xdb.documents.new("She fished for fish").to_xapian_document
+      terms = xdoc.terms.collect { |t| t.term }
+      terms.should include "Zfor"
+      terms.should include "for"
+      terms.should include "fish"
+    end
+
+    it "should stop stemmed words by default " do
+      xdb = XapianDb.new
+      xdoc = xdb.documents.new("She fished for fish").to_xapian_document
+      terms = xdoc.terms.collect { |t| t.term }
+      terms.should_not include "Zfor"
+      terms.should include "fish"
+    end
+  end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: xapian-fu
 version: !ruby/object:Gem::Version
-  version: 1.6.0
+  version: 1.7.0
 platform: ruby
 authors:
 - John Leach
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-10-07 00:00:00.000000000 Z
+date: 2020-03-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
@@ -39,6 +39,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: irb
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rdoc
   requirement: !ruby/object:Gem::Requirement
@@ -171,7 +185,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 1.9.3
+      version: 2.1.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
@@ -179,8 +193,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements:
 - libxapian-dev, or the xapian-ruby gem
-rubyforge_project: 
-rubygems_version: 2.7.6.2
+rubygems_version: 3.1.2
 signing_key: 
 specification_version: 4
 summary: A Ruby interface to the Xapian search engine