lingua 0.6.0 → 0.6.1

data/.gitignore

@@ -1,8 +1,6 @@
 ## MAC OS
 .DS_Store
 
-*.gemspec
-
 ## TEXTMATE
 *.tmproj
 tmtags

data/VERSION

@@ -1 +1 @@
-0.6.0
+0.6.1

data/lib/lingua/en/sentence.rb

@@ -29,7 +29,7 @@ module Lingua
       ABBR_DETECT = /(?:\s(?:(?:(?:\w\.){2,}\w?)|(?:\w\.\w)))/ unless defined?(ABBR_DETECT)
 
       # Finds punctuation that ends paragraphs.
-      PUNCTUATION_DETECT = /([\.?!](?:\"|\'|\)|\]|\})?)(\s+)/ unless defined?(PUNCTUATION_DETECT)
+      PUNCTUATION_DETECT = /((?:[\.?!]|[\r\n]+)(?:\"|\'|\)|\]|\})?)(\s+)/ unless defined?(PUNCTUATION_DETECT)
 
       CORRECT_ABBR = /(#{ABBR_DETECT})#{EOS}(\s+[a-z0-9])/
 

data/lingua.gemspec

@@ -0,0 +1,64 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE DIRECTLY
+# Instead, edit Jeweler::Tasks in Rakefile, and run the gemspec command
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = %q{lingua}
+  s.version = "0.6.1"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["David Balatero"]
+  s.date = %q{2010-05-17}
+  s.description = %q{Provides sentence splitting, syllable, and text-quality algorithms.}
+  s.email = %q{dbalatero@gmail.com}
+  s.extra_rdoc_files = [
+    "LICENSE",
+     "README.rdoc"
+  ]
+  s.files = [
+    ".document",
+     ".gitignore",
+     "LICENSE",
+     "README.rdoc",
+     "Rakefile",
+     "VERSION",
+     "lib/lingua.rb",
+     "lib/lingua/en/paragraph.rb",
+     "lib/lingua/en/readability.rb",
+     "lib/lingua/en/sentence.rb",
+     "lib/lingua/en/syllable.rb",
+     "lib/lingua/en/syllable/guess.rb",
+     "lingua.gemspec",
+     "spec/lingua/en/paragraph_spec.rb",
+     "spec/lingua/en/readability_spec.rb",
+     "spec/lingua/en/sentence_spec.rb",
+     "spec/spec.opts",
+     "spec/spec_helper.rb"
+  ]
+  s.homepage = %q{http://github.com/dbalatero/lingua}
+  s.rdoc_options = ["--charset=UTF-8"]
+  s.require_paths = ["lib"]
+  s.rubygems_version = %q{1.3.6}
+  s.summary = %q{This is a maintained version of Ruby's Lingua port.}
+  s.test_files = [
+    "spec/lingua/en/paragraph_spec.rb",
+     "spec/lingua/en/readability_spec.rb",
+     "spec/lingua/en/sentence_spec.rb",
+     "spec/spec_helper.rb"
+  ]
+
+  if s.respond_to? :specification_version then
+    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
+      s.add_development_dependency(%q<rspec>, [">= 1.2.9"])
+    else
+      s.add_dependency(%q<rspec>, [">= 1.2.9"])
+    end
+  else
+    s.add_dependency(%q<rspec>, [">= 1.2.9"])
+  end
+end
+

data/spec/lingua/en/sentence_spec.rb

@@ -101,6 +101,25 @@ describe Lingua::EN::Sentence do
       end
     end
 
+    describe "short sentences w/ line breaks" do
+      before(:each) do
+        @doc = <<-EOF
+        So how does the 401(k) plan work?  Let's see -
+
+        The 401(k) consists of - first, asking your employer to set aside a portion (upto 15% of your total income) in keeping with the plan.
+        EOF
+        @sentences = klass.sentences(@doc)
+      end
+
+      it "should find 3 sentences" do
+        @sentences.should have(3).things
+      end
+
+      it "should stop at line breaks" do
+        @sentences[1].should == "Let's see -"
+      end
+    end
+
     describe "sentences with URLs and abbreviation" do
       before(:each) do
         text = "Many of these leading names now have their own website, e.g.  http://www.kaptest.com/. Hello, e.g. you don't know what you mean. I'm so angry about what you said about the U.S.A. or the u.S. or the U.S.S.R. ok."

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 0
   - 6
-  - 0
-  version: 0.6.0
+  - 1
+  version: 0.6.1
 platform: ruby
 authors: 
 - David Balatero
@@ -14,7 +14,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-04-20 00:00:00 -07:00
+date: 2010-05-17 00:00:00 -07:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -53,6 +53,7 @@ files:
 - lib/lingua/en/sentence.rb
 - lib/lingua/en/syllable.rb
 - lib/lingua/en/syllable/guess.rb
+- lingua.gemspec
 - spec/lingua/en/paragraph_spec.rb
 - spec/lingua/en/readability_spec.rb
 - spec/lingua/en/sentence_spec.rb