twitter-text 1.2.5 → 1.3.0

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    twitter-text (1.2.5)
+    twitter-text (1.3.0)
       actionpack
 
 GEM

data/README.rdoc

@@ -29,7 +29,15 @@ usernames, lists, hashtags and URLs.
 
     html = auto_link("link @user, please #request")
   end
-
+  
+  # For Ruby on Rails you want to add this to app/helpers/application_helper.rb
+  module ApplicationHelper
+    include Twitter::Autolink
+  end
+  
+  # Now the auto_link function is available in every view. So in index.html.erb:
+  <%= auto_link("link @user, please #request") %>
+  
 === Usernames
 
 Username extraction and linking matches all valid Twitter usernames but does
@@ -79,6 +87,7 @@ Thanks to everybody who has filed issues, provided feedback or contributed patch
   * Ben Cherry - http://github.com/bcherry
   * Patrick Ewing - http://github.com/hoverbird
   * Jeff Smick - http://github.com/sprsquish
+  * Kenneth Kufluk - https://github.com/kennethkufluk
 
 * Patches from the community …
   * Jean-Philippe Bougie - http://github.com/jpbougie

data/lib/autolink.rb

@@ -139,17 +139,9 @@ module Twitter
 
       text.gsub(Twitter::Regex[:valid_url]) do
         all, before, url, protocol, domain, path, query_string = $1, $2, $3, $4, $5, $6, $7
-        if !protocol.blank? # || domain =~ Twitter::Regex[:probable_tld_domain]
+        if !protocol.blank?
           html_attrs = tag_options(options.stringify_keys) || ""
-          full_url = ((protocol =~ Twitter::Regex[:www] || protocol.blank?) ? "http://#{url}" : url)
-          "#{before}<a href=\"#{html_escape(full_url)}\"#{html_attrs}>#{html_escape(url)}</a>"
-        elsif all =~ Twitter::Regex[:probable_tld_domain]
-          before_tld, tld_domain = $1, $2
-
-          html_attrs = tag_options(options.stringify_keys) || ""
-          full_url = "http://#{tld_domain}"
-          prefix = (before_tld == before ? before : "#{before}#{before_tld}")
-          "#{prefix}<a href=\"#{html_escape(full_url)}\"#{html_attrs}>#{html_escape(tld_domain)}</a>"
+          "#{before}<a href=\"#{html_escape(url)}\"#{html_attrs}>#{html_escape(url)}</a>"
         else
           all
         end

data/lib/extractor.rb

@@ -127,15 +127,7 @@ module Twitter
           start_position = valid_url_match_data.char_begin(3)
           end_position = valid_url_match_data.char_end(3)
           urls << {
-            :url => ((protocol =~ Twitter::Regex[:www] || protocol.blank?) ? "http://#{url}" : url),
-            :indices => [start_position, end_position]
-          }
-        elsif all =~ Twitter::Regex[:probable_tld_domain]
-          tld_domain = $2
-          start_position = valid_url_match_data.char_begin(1) + $~.char_begin(2)
-          end_position = valid_url_match_data.char_begin(1) + $~.char_end(2)
-          urls << {
-            :url => "http://#{tld_domain}",
+            :url => url,
             :indices => [start_position, end_position]
           }
         end

data/lib/regex.rb

@@ -56,12 +56,7 @@ module Twitter
     REGEXEN[:valid_preceding_chars] = /(?:[^-\/"':!=A-Z0-9_@＠]|^|\:)/i
     REGEXEN[:valid_domain] = /(?:[^[:punct:]\s][\.-](?=[^[:punct:]\s])|[^[:punct:]\s]){1,}\.[a-z]{2,}(?::[0-9]+)?/i
 
-    # For protocol-less URLs, we'll accept them if they end in one of a handful of likely TLDs
-    REGEXEN[:probable_tld_domain] = /^(.*?)((?:[a-z0-9_\.\-]+)\.(?:com|net|org|gov|edu))$/i
-
-    REGEXEN[:www] = /www\./i
-
-    REGEXEN[:valid_general_url_path_chars] = /[a-z0-9!\*';:=\+\$\/%#\[\]\-_,~]/i
+    REGEXEN[:valid_general_url_path_chars] = /[a-z0-9!\*';:=\+\,\$\/%#\[\]\-_~]/i
     # Allow URL paths to contain balanced parens
     #  1. Used in Wikipedia URLs like /Primer_(film)
     #  2. Used in IIS sessions like /S(dfd346)/
@@ -70,22 +65,27 @@ module Twitter
     REGEXEN[:valid_url_path_chars] = /(?:
       #{REGEXEN[:wikipedia_disambiguation]}|
       @#{REGEXEN[:valid_general_url_path_chars]}+\/|
-      [\.\,]?#{REGEXEN[:valid_general_url_path_chars]}
+      [\.,]#{REGEXEN[:valid_general_url_path_chars]}+|
+      #{REGEXEN[:valid_general_url_path_chars]}+
     )/ix
     # Valid end-of-path chracters (so /foo. does not gobble the period).
     #   1. Allow =&# for empty URL parameters and other URL-join artifacts
-    REGEXEN[:valid_url_path_ending_chars] = /[a-z0-9=#\/]/i
+    REGEXEN[:valid_url_path_ending_chars] = /[a-z0-9=_#\/]|#{REGEXEN[:wikipedia_disambiguation]}/io
     REGEXEN[:valid_url_query_chars] = /[a-z0-9!\*'\(\);:&=\+\$\/%#\[\]\-_\.,~]/i
-    REGEXEN[:valid_url_query_ending_chars] = /[a-z0-9_&=#]/i
+    REGEXEN[:valid_url_query_ending_chars] = /[a-z0-9_&=#\/]/i
     REGEXEN[:valid_url] = %r{
       (                                                                                     #   $1 total match
         (#{REGEXEN[:valid_preceding_chars]})                                                #   $2 Preceeding chracter
         (                                                                                   #   $3 URL
-          ((?:https?:\/\/|www\.)?)                                                          #   $4 Protocol or beginning
+          (https?:\/\/)                                                                     #   $4 Protocol
           (#{REGEXEN[:valid_domain]})                                                       #   $5 Domain(s) and optional post number
-          (/#{REGEXEN[:valid_url_path_chars]}*
-            #{REGEXEN[:valid_url_path_ending_chars]}?
-          )?                                                                                #   $6 URL Path
+          (/
+            (?:
+              #{REGEXEN[:valid_url_path_chars]}+#{REGEXEN[:valid_url_path_ending_chars]}|   # 1+ path chars and a valid last char
+              #{REGEXEN[:valid_url_path_chars]}+#{REGEXEN[:valid_url_path_ending_chars]}?|  # Optional last char to handle /@foo/ case
+              #{REGEXEN[:valid_url_path_ending_chars]}                                      # Just a # case
+            )?
+          )?                                                                                #   $6 URL Path and anchor
           (\?#{REGEXEN[:valid_url_query_chars]}*#{REGEXEN[:valid_url_query_ending_chars]})? #   $7 Query String
         )
       )

data/spec/autolinking_spec.rb

@@ -467,10 +467,9 @@ describe Twitter::Autolink do
       context "with a url lacking the protocol" do
         def original_text; "I like www.foobar.com dudes"; end
 
-        it "links to the original text with the full href" do
+        it "does not link at all" do
           link = Nokogiri::HTML(@autolinked_text).search('a')
-          link.inner_text.should == 'www.foobar.com'
-          link.first['href'].should == 'http://www.foobar.com'
+          link.should be_empty
         end
       end
 

data/spec/test_urls.rb

@@ -6,14 +6,13 @@ module TestUrls
     "http://google.com/#foo",
     "http://google.com/#search?q=iphone%20-filter%3Alinks",
     "http://twitter.com/#search?q=iphone%20-filter%3Alinks",
+    "http://somedomain.com/index.php?path=/abc/def/",
     "http://www.boingboing.net/2007/02/14/katamari_damacy_phon.html",
     "http://somehost.com:3000",
     "http://xo.com/~matthew+%-x",
     "http://en.wikipedia.org/wiki/Primer_(film)",
     "http://www.ams.org/bookstore-getitem/item=mbk-59",
     "http://chilp.it/?77e8fd",
-    "www.foobar.com",
-    "WWW.FOOBAR.COM",
     "http://tell.me/why",
     "http://longtlds.info",
     "http://✪df.ws/ejp",
@@ -28,6 +27,8 @@ module TestUrls
   INVALID = [
     "http://no-tld",
     "http://tld-too-short.x",
+    "www.foobar.com",
+    "WWW.FOOBAR.COM",
     "http://-doman_dash.com"
   ] unless defined?(TestUrls::INVALID)
 

data/twitter-text.gemspec

@@ -1,6 +1,6 @@
 spec = Gem::Specification.new do |s|
   s.name = "twitter-text"
-  s.version = "1.2.5"
+  s.version = "1.3.0"
   s.authors = ["Matt Sanford", "Patrick Ewing", "Ben Cherry", "Britt Selvitelle", "Raffi Krikorian"]
   s.email = ["matt@twitter.com", "patrick.henry.ewing@gmail.com", "bcherry@gmail.com", "bs@brittspace.com", "raffi@twitter.com"]
   s.homepage = "http://twitter.com"

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: twitter-text
 version: !ruby/object:Gem::Version 
-  hash: 21
+  hash: 27
   prerelease: false
   segments: 
   - 1
-  - 2
-  - 5
-  version: 1.2.5
+  - 3
+  - 0
+  version: 1.3.0
 platform: ruby
 authors: 
 - Matt Sanford
@@ -19,7 +19,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-11-18 00:00:00 -08:00
+date: 2010-12-06 00:00:00 -08:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency