itrigga-core_ext 1.5.0 → 1.5.1

data/Gemfile

@@ -1,24 +1,36 @@
 source "http://rubygems.org"
+gemspec
 # Add dependencies required to use your gem here.
 # Example:
 #   gem "activesupport", ">= 2.3.5"
 
 # Add dependencies to develop your gem here.
 # Include everything needed to run rake, tests, features, etc.
-gem "fastercsv", ">= 1.5.4"
+gem "fastercsv", ">= 1.5.4", :require=>'false'
+gem "bundler"
+
+# HTML-cleaning
+gem "sanitize"
+
+
 
 group :development do
-  gem "rspec", "1.3.0"
-  gem "rspec-rails", "1.3.2"
+  gem "rspec", "2.9.0"
+  gem "rspec-rails", "2.9.0"
 #  gem "bundler", ">= 1.0.0"
   gem "jeweler", "~> 1.6.4"
 end
 
 group :test do
-  gem "rspec", "1.3.0"
-  gem "rspec-rails", "1.3.2"
-#  gem "bundler", ">= 1.0.0"
-#  gem "rcov", ">= 0"
-  gem "fastercsv", ">= 1.5.4"
+  gem "rspec", "2.9.0"
   gem 'test-unit', '1.2.3' # needed by rspec if using ruby 1.9.X
+  gem "rspec-rails", "2.9.0"
+
+  
+  if RUBY_VERSION =~ /^1.8/
+    gem "rcov", ">= 0"
+  else
+    gem 'simplecov', :require => false
+    gem 'simplecov-rcov', :require => false
+  end
 end

data/Rakefile

@@ -35,23 +35,30 @@ Jeweler::Tasks.new do |gem|
 end
 Jeweler::RubygemsDotOrgTasks.new
 
-require 'spec/version'
-require 'spec/rake/spectask'
-require 'spec/ruby'
 
-Spec::Rake::SpecTask.new(:spec) do |spec|
-  spec.spec_files = FileList['spec/**/*_spec.rb']
-  spec.spec_opts = ['--options', 'spec/spec.opts']
-end
+require 'rspec/core/rake_task'
 
+desc 'Default: run specs.'
 task :default => :spec
 
-desc "Run all specs with rcov"
-Spec::Rake::SpecTask.new(:rcov) do |t|
-  t.spec_files = FileList['spec/**/*_spec.rb']
-  t.spec_opts = ['--options', 'spec/spec.opts']
+desc "Run specs"
+RSpec::Core::RakeTask.new do |t|
+  t.pattern = "./spec/**/*_spec.rb" # don't need this, it's default.
+  # Put spec opts in a file named .rspec in root
+end
+
+desc "Generate code coverage"
+RSpec::Core::RakeTask.new(:coverage) do |t|
+  t.pattern = "./spec/**/*_spec.rb" # don't need this, it's default.
   t.rcov = true
-  t.rcov_dir = 'coverage'
-  t.rcov_opts = ['--exclude', "features,kernel,load-diff-lcs\.rb,instance_exec\.rb,lib/spec.rb,lib/spec/runner.rb,^spec/*,bin/spec,examples,/gems,/Library/Ruby,\.autotest,#{ENV['GEM_HOME']}"]
-  t.rcov_opts << '--sort coverage --text-summary --aggregate coverage.data'
+  t.rcov_opts = ['--exclude', 'spec']
+end
+require 'rdoc/task'
+Rake::RDocTask.new do |rdoc|
+  version = File.exist?('VERSION') ? File.read('VERSION') : ""
+
+  rdoc.rdoc_dir = 'rdoc'
+  rdoc.title = "net_helper #{version}"
+  rdoc.rdoc_files.include('README*')
+  rdoc.rdoc_files.include('lib/**/*.rb')
 end

data/VERSION

@@ -1 +1 @@
-1.5.0
+1.5.1

data/itrigga-core_ext.gemspec

@@ -5,7 +5,7 @@
 
 Gem::Specification.new do |s|
   s.name = "itrigga-core_ext"
-  s.version = "1.5.0"
+ s.version = "1.5.1"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Al Davidson", "Anson Kelly"]
@@ -61,6 +61,7 @@ Gem::Specification.new do |s|
     end
   else
     s.add_dependency(%q<fastercsv>, [">= 1.5.4"])
+    s.add_dependency(%q<bundler>, [">= 1.0.0"])
     s.add_dependency(%q<rspec>, ["= 1.3.0"])
     s.add_dependency(%q<rspec-rails>, ["= 1.3.2"])
     s.add_dependency(%q<jeweler>, ["~> 1.6.4"])

data/lib/itrigga/core_ext/string.rb

@@ -1,6 +1,16 @@
 # -*- encoding : utf-8 -*-
+require 'sanitize'
 class String
   
+  def to_plain_text_preserving_links(  )
+    ::Sanitize.clean( self, :elements=>['a'], :attributes=>{'a'=>['href']} ).to_s.strip
+  end
+  
+  def to_plain_text()
+    ::Sanitize.clean( HTMLEntities.new.decode(self) ).to_s.strip
+  end
+  
+  
   def remove_whitespace
     self.gsub(" ","")
   end
@@ -14,14 +24,37 @@ class String
   # Includes duplicates
   #
   def word_count
-    self.to_utf8.scan(/(\w|-|')+/).size
+    #re = /[\p{Word}\p{Punct}]/u # <-- tried several forms of this regex, this appears to be the only one that works correctly with Cyrillic AND Arabic script
+    #re = /\s+/u
+    #to_utf8.scan(re).size
+    
+    split.inject(0) do |sum, word|
+      if word.contains_cjk?
+        sum += word.scan(cjk_regex).size   # => ONLY work in Ruby 1.9. 
+                             #    Search for other methods to do this for 1.8
+      else
+        sum += 1
+      end
+    end
+  end
+  
+  def char_count
+    trim.size
+  end
+  
+  def cjk_regex
+    /\p{Han}|\p{Katakana}|\p{Hiragana}|\p{Hangul}/
+  end
+  
+  def contains_cjk?
+    !!(self =~ /\p{Han}|\p{Katakana}|\p{Hiragana}|\p{Hangul}/)
   end
   
   #
   # Removes starting, trailing whitespace and double spaces
   #
-  def trim
-    self.gsub(/^(.*[^\s])\s+$/, '\1').gsub(/^\s*(.*)$/, '\1')
+  def trim
+    self.gsub(/^(.*[^\s])\s+$/, '\1').gsub(/^\s*(.*)$/, '\1')
   end
   
   def to_utf16le
@@ -30,27 +63,28 @@ class String
   
   def to_active_record_condition
     "%#{self.trim.gsub(/[[:space:]]+/, '%')}%"
-  end  
 
+  end  
+ 
   # need to force encoding for ruby 1.9 otherwise regex fails when comparing string of 2 different encodings
-  # TODO : String.blank? - do we need to do force encoding? is UTF-8 a good default?
+  # TODO : String.blank? - do we need to do force encoding? is UTF-8 a good default?
   def blank?
     begin
       if Gem::Version.new(''+RUBY_VERSION) >= Gem::Version.new("1.9.0") # only for ruby 1.9+
         self.dup.force_encoding("UTF-8") !~ /\S/
-      else
+      else
   	    self !~ /\S/
       end
-    rescue ArgumentError => e
-      if e.message =~ /^(invalid\ byte\ sequence|incompatible\ character\ encodings)/
-	      empty?
-      else
-	      raise e
+    rescue ArgumentError => e
+      if e.message =~ /^(invalid\ byte\ sequence|incompatible\ character\ encodings)/
+	      empty?
+      else
+	      raise e
       end
-    end
+    end
   end
   
   # converts the encoding to UTF-8 regardless of current encoding
-  def to_utf8
+  def to_utf8
     text = self.dup
     # shiny new ruby 1.9 way
     return text if text.encoding.name == "UTF-8" && text.valid_encoding? # already utf-8 yay!    
@@ -76,7 +110,7 @@ class String
         
     # ok so we are out of suggestions. Just return the string and hope that its ok
     text
-    
+    
   end
   
 end

data/spec/spec_helper.rb

@@ -6,8 +6,27 @@ RAILS_ROOT = File.expand_path(File.join(File.dirname(__FILE__), '..'))
 $LOAD_PATH.unshift(File.join(RAILS_ROOT, 'lib'))
 $LOAD_PATH.unshift(File.dirname(__FILE__))
 
+require 'rubygems'
+require 'bundler/setup'
+
+require 'rubygems'
+require 'bundler'
+begin
+   Bundler.setup(:default, :development, :test)
+ rescue Bundler::BundlerError => e
+   $stderr.puts e.message
+   $stderr.puts "Run `bundle install` to install missing gems"
+   exit e.status_code
+ end
+require 'test/unit'
+require 'rspec'
+
 require 'itrigga/core_ext'
 
 # Requires supporting files with custom matchers and macros, etc,
 # in ./support/ and its subdirectories.
 Dir["RAILS_ROOT}/support/**/*.rb"].each {|f| require f}
+
+RSpec.configure do |config|
+  
+end

data/spec/string_spec.rb

@@ -23,6 +23,81 @@ describe String do
         str.word_count.should == 2
       end
     end
+    
+    # these examples generated with http://generator.lorem-ipsum.info/
+    context "when string is in Arabic" do
+      before{ @string =<<-END
+        من وقد جحافل ليرتفع بولندا, تعد أم العالم العمليات اقتصادية. مما ثم غضون عشوائية الدنمارك, كل يبق وبعد التاريخ،, تم الفاشي والألمانية، للإمبراطورية أضف. شبح واعتلاء بالهجوم و. الإطلاق الربيع، والإتحاد عن على, تسبب وبداية بتخصيص وضم من. تم بها وبدأت واستمر الإنجليز،, كنقطة انتباه لان تم.
+
+       قد فقد دخول."
+END
+      }
+      it "should count the right number of words" do 
+        @string.word_count.should == 50
+      end
+    end
+    
+    context "when string is in Cyrillic" do
+      before{ @string = "Йн пэр аффэрт ыпикурэи интылльэгэбат! Эи ючю витаэ экшырки адвыржаряюм. Эож тебиквюэ патриоквюы ан, вим дёко омниюм конкльюдатюрквюэ но! Ыт ыюм нонумй альяквюам, ыам ед эррэм фюйзчыт, вяш ку эквюедым." }
+      it "should count the right number of words" do 
+        @string.word_count.should == 30
+      end
+    end
+    
+    context "when string is in Hebrew" do
+      before{ @string = "אל סדר וקשקש תקשורת. רבה לעתים מיותר ומהימנה את, הרוח חשמל כלשהו ויש אל. לעריכת האטמוספירה שער על. הטבע אנגלית שיתופית בקר גם, צעד או מושגי אודות. כתב אם שונה." }
+      it "should count the right number of words" do 
+        @string.word_count.should == 30
+      end
+    end
+    
+    
+    context "when string contains Chinese" do
+      # see http://stackoverflow.com/questions/12488565/how-to-count-words-in-a-multi-language-text-using-ruby-javascript
+      before{ @string = 'The last Olympics was held in  北京' }
+      it "should count the english words as words, plus one word for each Chinese character" do 
+        @string.word_count.should == 8
+      end
+    end
+  end
+  
+  describe "contains_cjk?" do
+    context "when the string contains Chinese characters" do
+      it "should be true" do
+        'The last Olympics was held in 北京'.contains_cjk?.should == true
+      end
+    end
+    context "when the string does not contain Chinese characters" do
+      it "should be false" do
+        'The last Olympics was held in London'.contains_cjk?.should == false
+      end
+    end
   end
   
+  
+  describe "to_plain_text_preserving_links" do
+    context "when it contains a tags" do
+      let(:content){ 'here is <a href="http://foo.com/">foo</a>' }
+      
+      it "preserves the a tags" do
+        content.to_plain_text_preserving_links.should == 'here is <a href="http://foo.com/">foo</a>'
+      end
+      
+      context "and other tags" do
+        let(:content){ '<b>here is <a href="http://foo.com/"><i>foo</i></a></b><img src="http://foo.com/bar.jpg">' }
+        
+        it "removes any non-a tags" do
+          content.to_plain_text_preserving_links.should == 'here is <a href="http://foo.com/">foo</a>'
+        end
+      end
+      context "and the a tags contain attributes other than href" do
+        let(:content){ '<b>here is <a rel="nofollow" href="http://foo.com/" title="Go to foo">foo</a></b><img src="http://foo.com/bar.jpg">' }
+        
+        it "removes any non-href attributes but leaves the href attributes unaffected" do
+          content.to_plain_text_preserving_links.should == 'here is <a href="http://foo.com/">foo</a>'          
+        end
+        
+      end
+    end
+  end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: itrigga-core_ext
 version: !ruby/object:Gem::Version
-  version: 1.5.0
+  version: 1.5.1
   prerelease: 
 platform: ruby
 authors:
@@ -14,7 +14,7 @@ date: 2012-05-08 00:00:00.000000000Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: fastercsv
-  requirement: &70102764111580 !ruby/object:Gem::Requirement
+  requirement: &70191121319620 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -22,10 +22,10 @@ dependencies:
         version: 1.5.4
   type: :runtime
   prerelease: false
-  version_requirements: *70102764111580
+  version_requirements: *70191121319620
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &70102764111100 !ruby/object:Gem::Requirement
+  requirement: &70191121319140 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - =
@@ -33,10 +33,10 @@ dependencies:
         version: 1.3.0
   type: :development
   prerelease: false
-  version_requirements: *70102764111100
+  version_requirements: *70191121319140
 - !ruby/object:Gem::Dependency
   name: rspec-rails
-  requirement: &70102764110600 !ruby/object:Gem::Requirement
+  requirement: &70191121318640 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - =
@@ -44,10 +44,10 @@ dependencies:
         version: 1.3.2
   type: :development
   prerelease: false
-  version_requirements: *70102764110600
+  version_requirements: *70191121318640
 - !ruby/object:Gem::Dependency
   name: jeweler
-  requirement: &70102764110120 !ruby/object:Gem::Requirement
+  requirement: &70191121318160 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -55,7 +55,7 @@ dependencies:
         version: 1.6.4
   type: :development
   prerelease: false
-  version_requirements: *70102764110120
+  version_requirements: *70191121318160
 description: ! "Adds methods for \n    Hash: \n    - symbolize_keys\n    - recursive_symbolize_keys\n
   \   Array:\n    - to_csv (requires FasterCSV gem if ruby 1.8)\n    - select_with_index\n
   \   - to_auto_complete_list\n    - active_record_type_condition_rendering\n  "