twitter-text 1.4.5 → 1.4.6

data/README.rdoc

@@ -89,6 +89,7 @@ Thanks to everybody who has filed issues, provided feedback or contributed patch
   * Patrick Ewing - http://github.com/hoverbird
   * Jeff Smick - http://github.com/sprsquish
   * Kenneth Kufluk - https://github.com/kennethkufluk
+  * Keita Fujii - https://github.com/keitaf
 
 * Patches from the community …
   * Jean-Philippe Bougie - http://github.com/jpbougie

data/lib/autolink.rb

@@ -137,12 +137,6 @@ module Twitter
         text = $3
         text = yield(text) if block_given?
         "#{before}<a href=\"#{options[:hashtag_url_base]}#{html_escape(text)}\" title=\"##{html_escape(text)}\" #{target_tag(options)}class=\"#{options[:url_class]} #{options[:hashtag_class]}\"#{extra_html}>#{html_escape(hash)}#{html_escape(text)}</a>"
-      end.gsub(Twitter::Regex[:auto_link_cj_hashtags]) do
-          before = $1
-          hash = $2
-          text = $3
-          text = yield(text) if block_given?
-          "#{before}<a href=\"#{options[:hashtag_url_base]}#{html_escape(text)}\" title=\"##{html_escape(text)}\" #{target_tag(options)}class=\"#{options[:url_class]} #{options[:hashtag_class]}\"#{extra_html}>#{html_escape(hash)}#{html_escape(text)}</a>"
       end
     end
 

data/lib/extractor.rb

@@ -165,13 +165,6 @@ module Twitter
           :hashtag => hash_text,
           :indices => [start_position, end_position]
         }
-      end.scan(Twitter::Regex[:auto_link_cj_hashtags]) do |before, hash, hash_text|
-        start_position = $~.char_begin(2)
-        end_position = $~.char_end(3)
-        tags << {
-          :hashtag => hash_text,
-          :indices => [start_position, end_position]
-        }
       end
       tags.each{|tag| yield tag[:hashtag], tag[:indices].first, tag[:indices].last } if block_given?
       tags

data/lib/regex.rb

@@ -71,17 +71,15 @@ module Twitter
       (0x2F800..0x2FA1F).to_a # Kanji (CJK supplement)
     ].flatten.pack('U*').freeze
 
-    CJ_BOUNDARY = /(?:\A|\z|#{REGEXEN[:spaces]}|「|」|。|\.)/
+    HASHTAG_BOUNDARY = /(?:\A|\z|#{REGEXEN[:spaces]}|「|」|。|\.|!)/
 
     # A hashtag must contain latin characters, numbers and underscores, but not all numbers.
-    HASHTAG_ALPHA = /[a-z_#{LATIN_ACCENTS}#{NON_LATIN_HASHTAG_CHARS}]/io
-    HASHTAG_ALPHANUMERIC = /[a-z0-9_#{LATIN_ACCENTS}#{NON_LATIN_HASHTAG_CHARS}]/io
+    HASHTAG_ALPHA = /[a-z_#{LATIN_ACCENTS}#{NON_LATIN_HASHTAG_CHARS}#{CJ_HASHTAG_CHARACTERS}]/io
+    HASHTAG_ALPHANUMERIC = /[a-z0-9_#{LATIN_ACCENTS}#{NON_LATIN_HASHTAG_CHARS}#{CJ_HASHTAG_CHARACTERS}]/io
 
-    CJ_HASHTAG = /(#{CJ_BOUNDARY})(#|＃)([#{CJ_HASHTAG_CHARACTERS}]+)(?=#{CJ_BOUNDARY})/o
-    NON_CJ_HASHTAG = /(^|[^0-9A-Z&\/\?]+)(#|＃)(#{HASHTAG_ALPHANUMERIC}*#{HASHTAG_ALPHA}#{HASHTAG_ALPHANUMERIC}*)/io
+    HASHTAG = /(#{HASHTAG_BOUNDARY})(#|＃)(#{HASHTAG_ALPHANUMERIC}*#{HASHTAG_ALPHA}#{HASHTAG_ALPHANUMERIC}*)(?=#{HASHTAG_BOUNDARY})/io
 
-    REGEXEN[:auto_link_hashtags] = /#{NON_CJ_HASHTAG}/io
-    REGEXEN[:auto_link_cj_hashtags] = /#{CJ_HASHTAG}/io
+    REGEXEN[:auto_link_hashtags] = /#{HASHTAG}/io
 
     REGEXEN[:auto_link_usernames_or_lists] = /([^a-zA-Z0-9_]|^|RT:?)([@＠]+)([a-zA-Z0-9_]{1,20})(\/[a-zA-Z][a-zA-Z0-9_\-]{0,24})?/o
     REGEXEN[:auto_link_emoticon] = /(8\-\#|8\-E|\+\-\(|\`\@|\`O|\<\|:~\(|\}:o\{|:\-\[|\>o\<|X\-\/|\[:-\]\-I\-|\/\/\/\/Ö\\\\\\\\|\(\|:\|\/\)|∑:\*\)|\( \| \))/

data/spec/autolinking_spec.rb

@@ -269,19 +269,11 @@ describe Twitter::Autolink do
         end
       end
 
-      context "with a hashtag preceded by Japanese" do
-        def original_text; "の#twj_dev"; end
-
-        it "should be linked" do
-          @autolinked_text.should have_autolinked_hashtag('#twj_dev')
-        end
-      end
-
       context "with a hashtag followed by Japanese" do
          def original_text; "#twj_devの"; end
 
         it "should be linked" do
-          @autolinked_text.should have_autolinked_hashtag('#twj_dev')
+          @autolinked_text.should have_autolinked_hashtag('#twj_devの')
         end
       end
 

data/spec/extractor_spec.rb

@@ -228,11 +228,11 @@ describe Twitter::Extractor do
         end
 
         it "should not allow the multiplication character" do
-          @extractor.extract_hashtags("#pre#{[0xd7].pack('U')}post").should == ["pre"]
+          @extractor.extract_hashtags("#pre#{[0xd7].pack('U')}post").should == []
         end
 
         it "should not allow the division character" do
-          @extractor.extract_hashtags("#pre#{[0xf7].pack('U')}post").should == ["pre"]
+          @extractor.extract_hashtags("#pre#{[0xf7].pack('U')}post").should == []
         end
       end
 
@@ -253,7 +253,7 @@ describe Twitter::Extractor do
       extracted_hashtag[:indices].last.should == offset + hashtag.chars.to_a.size + 1
     end
 
-    def no_match_hashtag_in_text(text)
+    def not_match_hashtag_in_text(text)
       extracted_hashtags = @extractor.extract_hashtags_with_indices(text)
       extracted_hashtags.size.should == 0
     end
@@ -283,17 +283,17 @@ describe Twitter::Extractor do
         end
 
         it "should not allow the multiplication character" do
-          match_hashtag_in_text('pre', "#pre#{[0xd7].pack('U')}post")
+          not_match_hashtag_in_text("#pre#{[0xd7].pack('U')}post")
         end
 
         it "should not allow the division character" do
-          match_hashtag_in_text('pre', "#pre#{[0xf7].pack('U')}post")
+          not_match_hashtag_in_text("#pre#{[0xf7].pack('U')}post")
         end
       end
     end
 
     it "should not extract numeric hashtags" do
-      no_match_hashtag_in_text("#1234")
+      not_match_hashtag_in_text("#1234")
     end
   end
 end

data/spec/spec_helper.rb

@@ -85,7 +85,7 @@ end
 
 RSpec::Matchers.define :have_autolinked_hashtag do |hashtag|
   match do |text|
-    @link = Nokogiri::HTML(text).search("a[@href='http://twitter.com/search?q=#{CGI.escape hashtag}']")
+    @link = Nokogiri::HTML(text).search("a[@href='http://twitter.com/search?q=#{hashtag.sub(/^#/, '%23')}']")
     @link &&
     @link.inner_text &&
     @link.inner_text == hashtag
@@ -93,7 +93,7 @@ RSpec::Matchers.define :have_autolinked_hashtag do |hashtag|
 
   failure_message_for_should do |text|
     if @link
-      "Expected link text to be #{hashtag}, but it was #{@link.inner_text}"
+      "Expected link text to be [#{hashtag}], but it was [#{@link.inner_text}] in #{text}"
     else
       "Expected hashtag #{hashtag} to be autolinked in '#{text}', but no link was found."
     end

data/twitter-text.gemspec

@@ -1,6 +1,6 @@
 spec = Gem::Specification.new do |s|
   s.name = "twitter-text"
-  s.version = "1.4.5"
+  s.version = "1.4.6"
   s.authors = ["Matt Sanford", "Patrick Ewing", "Ben Cherry", "Britt Selvitelle", "Raffi Krikorian", "J.P. Cummins", "Yoshimasa Niwa"]
   s.email = ["matt@twitter.com", "patrick.henry.ewing@gmail.com", "bcherry@gmail.com", "bs@brittspace.com", "raffi@twitter.com", "jcummins@twitter.com", "niw@niw.at"]
   s.homepage = "http://twitter.com"

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: twitter-text
 version: !ruby/object:Gem::Version 
-  hash: 13
+  hash: 11
   prerelease: false
   segments: 
   - 1
   - 4
-  - 5
-  version: 1.4.5
+  - 6
+  version: 1.4.6
 platform: ruby
 authors: 
 - Matt Sanford