greeb 0.1.0.rc3 → 0.1.0.rc4

data/README.md

@@ -132,6 +132,11 @@ systematic and awesome.
 
 ## Build Status [<img src="https://secure.travis-ci.org/eveel/greeb.png"/>](http://travis-ci.org/eveel/greeb)
 
+If you're using [Rubinius](http://rubini.us) please note that it has the
+incompatible `StringScanner` implementation. More information can be
+provided under the following link:
+<https://github.com/rubinius/rubinius/issues/1808>.
+
 ## Dependency Status [<img src="https://gemnasium.com/eveel/greeb.png?travis"/>](https://gemnasium.com/eveel/greeb)
 
 ## Copyright

data/lib/greeb/segmentator.rb

@@ -33,6 +33,15 @@ class Greeb::Segmentator
     @sentences
   end
 
+  # Subsentences memoization method.
+  #
+  # @return [Set<Greeb::Entity>] a set of subsentences.
+  #
+  def subsentences
+    detect_subsentences! unless @subsentences
+    @subsentences
+  end
+
   # Extract tokens from the set of sentences.
   #
   # @param sentences [Array<Greeb::Entity>] a list of sentences.
@@ -48,6 +57,21 @@ class Greeb::Segmentator
     ]
   end
 
+  # Extract subsentences from the set of sentences.
+  #
+  # @param sentences [Array<Greeb::Entity>] a list of sentences.
+  #
+  # @return [Hash<Greeb::Entity, Array<Greeb::Entity>>] a hash with
+  #   sentences as keys and subsentences arrays as values.
+  #
+  def subextract *sentences
+    Hash[
+      sentences.map do |s|
+        [s, subsentences.select { |ss| ss.from >= s.from and ss.to <= s.to }]
+      end
+    ]
+  end
+
   protected
     # Implementation of the sentence detection method. This method
     # changes the `@sentences` ivar.
@@ -84,6 +108,41 @@ class Greeb::Segmentator
       nil.tap { @sentences << rest if rest.from and rest.to }
     end
 
+    # Implementation of the subsentence detection method. This method
+    # changes the `@subsentences` ivar.
+    #
+    # @return [nil] nothing.
+    #
+    def detect_subsentences!
+      @subsentences = SortedSet.new
+
+      rest = tokens.inject(new_subsentence) do |subsentence, token|
+        if !subsentence.from and SENTENCE_DOESNT_START.include?(token.type)
+          next subsentence
+        end
+
+        subsentence.from = token.from unless subsentence.from
+
+        next subsentence if subsentence.to and subsentence.to > token.to
+
+        if [:punct, :spunct].include? token.type
+          subsentence.to = tokens.
+            select { |t| t.from >= token.from }.
+            inject(token) { |r, t| break r if t.type != token.type; t }.
+            to
+
+          @subsentences << subsentence
+          subsentence = new_subsentence
+        elsif :separ != token.type
+          subsentence.to = token.to
+        end
+
+        subsentence
+      end
+
+      nil.tap { @subsentences << rest if rest.from and rest.to }
+    end
+
   private
     # Create a new instance of {Greeb::Entity} with `:sentence` type.
     #
@@ -92,4 +151,12 @@ class Greeb::Segmentator
     def new_sentence
       Greeb::Entity.new(nil, nil, :sentence)
     end
+
+    # Create a new instance of {Greeb::Entity} with `:subsentence` type.
+    #
+    # @return [Greeb::Entity] a new entity instance.
+    #
+    def new_subsentence
+      Greeb::Entity.new(nil, nil, :subsentence)
+    end
 end

data/lib/greeb/version.rb

@@ -5,5 +5,5 @@
 module Greeb
   # Version of Greeb.
   #
-  VERSION = '0.1.0.rc3'
+  VERSION = '0.1.0.rc4'
 end

data/spec/segmentator_spec.rb

@@ -89,8 +89,10 @@ module Greeb
 
       subject { Segmentator.new(@tokenizer) }
 
+      let(:sentences) { subject.sentences }
+
       it 'should be extracted' do
-        subject.extract(*subject.sentences).must_equal({
+        subject.extract(*sentences).must_equal({
           Entity.new(0,  6, :sentence) => [
             Entity.new(0, 5, :letter),
             Entity.new(5, 6, :punct)
@@ -108,5 +110,22 @@ module Greeb
         })
       end
     end
+
+    describe 'subsentence extractor' do
+      before { @tokenizer = Tokenizer.new('Hello, I am JC Denton.') }
+
+      subject { Segmentator.new(@tokenizer) }
+
+      let(:sentences) { subject.sentences }
+
+      it 'should extract subsentences' do
+        subject.subextract(*sentences).must_equal({
+          Entity.new(0,  22, :sentence) => [
+            Entity.new(0, 6, :subsentence),
+            Entity.new(7, 22, :subsentence)
+          ]
+        })
+      end
+    end
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: greeb
 version: !ruby/object:Gem::Version
-  version: 0.1.0.rc3
+  version: 0.1.0.rc4
   prerelease: 6
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-07-10 00:00:00.000000000 Z
+date: 2012-07-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -112,7 +112,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: -2527935574265859361
+      hash: 1130932854600612903
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements: