sentence_extractor 0.2.0 → 0.2.1

data/lib/sentence_extractor/extractor.rb

@@ -45,7 +45,8 @@ module SentenceExtractor
       while match != nil
         sentence_end_point = remaining_text =~ /#{@reg_exp}/  # check for sentences
         if sentence_end_point
-          sentences << remaining_text[0..sentence_end_point].strip  # add new sentence to array
+          sentence = remaining_text[0..sentence_end_point].strip # strip sentence of leading and trailing spaces
+          sentences << sentence if (!@first_letter_capitilization || first_letter_uppercase?(sentence))  # add new sentence to array
           remaining_text = remaining_text[sentence_end_point+1..remaining_text.size]  # set the rest of the text to be processed.
         else
           match = nil
@@ -54,6 +55,14 @@ module SentenceExtractor
       sentences
     end
     
+    def first_letter_uppercase? sentence
+      if sentence[0..0] == sentence[0..0].upcase
+        return true
+      else
+        return false
+      end
+    end
+    
    end
    
 end

data/lib/sentence_extractor/version.rb

@@ -1,3 +1,3 @@
 module SentenceExtractor
-  VERSION = "0.2.0"
+  VERSION = "0.2.1"
 end

data/spec/extractor_spec.rb

@@ -26,4 +26,18 @@ describe SentenceExtractor::Extractor do
     sen_exc.extract_sentences("Hello. How are you? Good thanks! ").should eql(["Hello.", "How are you?", "Good thanks!"])
   end
   
+  it "Should leave in sentences that don't start with capital letters" do
+    sen_exc = SentenceExtractor::Extractor.new()
+    sen_exc.extract_sentences("hello. How are you? good thanks! ").should eql(["hello.", "How are you?", "good thanks!"])
+  end
+  
+  it "Should remove sentences that don't start with capital letters" do
+    sen_exc = SentenceExtractor::Extractor.new(language = "en", delimiters = [".", "!","?"], first_letter_capitilization = true)
+    sen_exc.extract_sentences("hello. How are you? good thanks! ").should eql(["How are you?"])
+  end
+  
+  
+  
+  
+  
 end

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: sentence_extractor
 version: !ruby/object:Gem::Version 
-  hash: 23
+  hash: 21
   prerelease: 
   segments: 
   - 0
   - 2
-  - 0
-  version: 0.2.0
+  - 1
+  version: 0.2.1
 platform: ruby
 authors: 
 - Seb Glazebrook
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2012-03-23 00:00:00 Z
+date: 2012-03-26 00:00:00 Z
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: rspec