twitter-text 1.1.4 → 1.1.5

data/Rakefile

@@ -9,7 +9,7 @@ require 'digest'
 
 spec = Gem::Specification.new do |s|
   s.name = "twitter-text"
-  s.version = "1.1.4"
+  s.version = "1.1.5"
   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"

data/lib/autolink.rb

@@ -38,7 +38,7 @@ module Twitter
         options),
       options)
     end
-    
+
     # Add <tt><a></a></tt> tags around the usernames and lists in the provided <tt>text</tt>. The
     # <tt><a></tt> tags can be controlled with the following entries in the <tt>options</tt>
     # hash:

data/lib/regex.rb

@@ -1,4 +1,4 @@
-# coding: UTF-8
+# encoding: utf-8
 
 module Twitter
   # A collection of regular expressions for parsing Tweet text. The regular expression
@@ -31,7 +31,12 @@ module Twitter
     REGEXEN[:extract_mentions] = /(^|[^a-zA-Z0-9_])#{REGEXEN[:at_signs]}([a-zA-Z0-9_]{1,20})(?=(.|$))/o
     REGEXEN[:extract_reply] = /^(?:#{REGEXEN[:spaces]})*#{REGEXEN[:at_signs]}([a-zA-Z0-9_]{1,20})/o
 
-    REGEXEN[:list_name] = /^[a-zA-Z\u0080-\u00ff].{0,79}$/
+    major, minor, patch = RUBY_VERSION.split(/\./)
+    if major.to_i >= 1 && minor.to_i >= 9
+      REGEXEN[:list_name] = /^[a-zA-Z\u0080-\u00ff].{0,79}$/
+    else
+      REGEXEN[:list_name] = /^[a-zA-Z\x80-\xff].{0,79}$/
+    end
 
     # Latin accented characters (subtracted 0xD7 from the range, it's a confusable multiplication sign. Looks like "x")
     LATIN_ACCENTS = [(0xc0..0xd6).to_a, (0xd8..0xf6).to_a, (0xf8..0xff).to_a].flatten.pack('U*').freeze
@@ -51,9 +56,11 @@ module Twitter
     #  1. Used in Wikipedia URLs like /Primer_(film)
     #  2. Used in IIS sessions like /S(dfd346)/
     REGEXEN[:wikipedia_disambiguation] = /(?:\([^\)]+\))/i
+    # Allow @ in a url, but only in the middle. Catch things like http://example.com/@user
     REGEXEN[:valid_url_path_chars] = /(?:
       #{REGEXEN[:wikipedia_disambiguation]}|
-      [\.\,]?[a-z0-9!\*';:=\+\$\/%#\[\]\-_,~@]
+      @[^\/]+\/|
+      [\.\,]?[a-z0-9!\*';:=\+\$\/%#\[\]\-_,~]
     )/ix
     # Valid end-of-path chracters (so /foo. does not gobble the period).
     #   1. Allow =&# for empty URL parameters and other URL-join artifacts

data/spec/regex_spec.rb

@@ -1,4 +1,4 @@
-#encoding: UTF-8
+# encoding: utf-8
 require File.dirname(__FILE__) + '/spec_helper'
 
 describe "Twitter::Regex regular expressions" do
@@ -21,4 +21,24 @@ describe "Twitter::Regex regular expressions" do
     end
   end
 
+  describe "matching List names" do
+    it "should match if less than 25 characters" do
+      name = "Shuffleboard Community"
+      name.length.should < 25
+      name.should match(Twitter::Regex::REGEXEN[:list_name])
+    end
+
+    it "should not match if greater than 25 characters" do
+      name = "Most Glorious Shady Meadows Shuffleboard Community"
+      name.length.should > 25
+      name.should match(Twitter::Regex[:list_name])
+    end
+
+    it "should match Japense names less than 25 characters" do
+      name = CGI.unescape("%E4%B9%97")
+      name.length.should < 25
+      name.should match(Twitter::Regex[:list_name])
+    end
+
+  end
 end

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 1
   - 1
-  - 4
-  version: 1.1.4
+  - 5
+  version: 1.1.5
 platform: ruby
 authors: 
 - Matt Sanford
@@ -18,7 +18,7 @@ autorequire: ""
 bindir: bin
 cert_chain: []
 
-date: 2010-07-08 00:00:00 -07:00
+date: 2010-07-22 00:00:00 -07:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency