rudge 0.0.7 → 0.0.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 2fce88b22b8874cc54365bcf58e2fcf216fecb8a
-  data.tar.gz: 74b020bbcc59f8974cb3512ce0377d836401a1de
+  metadata.gz: 7b599e54d26036ce43a7a04b7d8d0147faa7ce3a
+  data.tar.gz: 0aa3264aaf613e84b5de6a16f65d27d02d9e076b
 SHA512:
-  metadata.gz: 65bd3060fa479677a8a786e7f6795df02dfb55ecb6367079ce234b8c4a59e512dad804f66840259bd4a5a256ebc700facd406a69f43a08fc8b667d80843a3e39
-  data.tar.gz: 38475cd6f0b53d486f4760f225887e010d73d9efc0acaa58eac6841ed195edb5b4f7b88030fe93ba72266dc24d19826b86c8847be376c33044755ea215a02c0c
+  metadata.gz: 7556b1e408626af47b5218052e408b9bec645e84a6534e1a9c7deb9e29a2e6d4766bbe1d13cb02b36c5f40ddcd12d269ffebcb9b8988c9c2365d7a85671fc0e2
+  data.tar.gz: a8f66f6b00e9f2c508c7051cf254fce4420c2993708a188fdc056aadc64d81b248ae0ec023f6806574c5cad0a818dc92b91cf542821a70174d6677454c352e22

data/lib/rudge.rb

@@ -1,23 +1,29 @@
 require "rudge/abbreviations"
 
 class Rudge
-  # end of sentence marker
-  EOS = "\001"
+  # end of sentence marker (before and after whitespace)
+  EOSBW = "\001"
+  EOSAW = "\002"
 
-  def self.sentences(text)
+  def self.sentences(text, options={})
     text = text.dup
 
-    # initial split after punctuation,
-    # preserves trailing whitespace for the ellipsis correction
-    text.gsub!(/([\.?!](?:\"|\'|\)|\]|\})?)(\s+)/) { $1 << EOS << $2 }
+    # initial sentence markers, before and after whitespace
+    text.gsub!(/([.?!](?:\"|\'|\)|\]|\})?)(\s+)/) { $1 << EOSBW << $2 << EOSAW }
 
-    # correct ellipsis marks and rows of stops
-    text.gsub!(/(\.\.\.*)#{EOS}/) { $1 }
+    # remove sentence markers on ellipsis
+    text.gsub!(/(\.\.\.*)#{EOSBW}(\s+)#{EOSAW}/) { $1 << $2 }
 
-    # correct abbreviations - precompile regexp?
-    text.gsub!(/\s(#{Rudge::Abbreviations.list.join("|")})\.#{EOS}/i) { " " << $1 << "." }
+    # remove sentence markers on abbreviations
+    abbreviations = Rudge::Abbreviations.list.join("|")
+    text.gsub!(/(\s)(#{abbreviations})\.#{EOSBW}(\s+)#{EOSAW}/i) { $1 << $2 << "." << $3 }
 
-    # split on EOS marker, strip gets rid of trailing whitespace
-    text.split(EOS).map { | sentence | sentence.strip }
+    if options[:keep_trailling_whitespace]
+      # split after whitespace, remove EOSBW marker
+      text.split(EOSAW).map { | sentence | sentence.gsub(EOSBW, "") }
+    else
+      # remove initial whitespace, split at markers
+      text.gsub(/\A\s+/, "").split(/#{EOSBW}\s+#{EOSAW}/)
+    end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rudge
 version: !ruby/object:Gem::Version
-  version: 0.0.7
+  version: 0.0.8
 platform: ruby
 authors:
 - bruno coelho
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-02-21 00:00:00.000000000 Z
+date: 2014-10-09 00:00:00.000000000 Z
 dependencies: []
 description: A Ruby Judge! Determines the sentences in a text.
 email: bruno.coelho@gmail.com
@@ -38,7 +38,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.14
+rubygems_version: 2.4.1
 signing_key: 
 specification_version: 4
 summary: Rudge - determines sentences.