lib/regex.rb in twitter-text-1.0.1 vs lib/regex.rb in twitter-text-1.0.2

@@ -24,12 +24,13 @@
           0x205F,          # White_Space # Zs       MEDIUM MATHEMATICAL SPACE
           0x3000,          # White_Space # Zs       IDEOGRAPHIC SPACE
         ].flatten.freeze
     REGEXEN[:spaces] = Regexp.new(UNICODE_SPACES.collect{ |e| [e].pack 'U*' }.join('|'))
 
-    REGEXEN[:extract_mentions] = /(^|[^a-zA-Z0-9_])[@＠]([a-zA-Z0-9_]{1,20})(?!@)/
-    REGEXEN[:extract_reply] = /^(?:#{REGEXEN[:spaces]})*[@＠]([a-zA-Z0-9_]{1,20})/o
+    REGEXEN[:at_signs] = /[@＠]/
+    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\x80-\xff].{0,79}$/
 
     # 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
@@ -40,21 +41,21 @@
     REGEXEN[:auto_link_hashtags] = /(^|[^0-9A-Z&\/]+)(#|＃)([0-9A-Z_]*[A-Z_]+#{HASHTAG_CHARACTERS}*)/io
     REGEXEN[:auto_link_usernames_or_lists] = /([^a-zA-Z0-9_]|^)([@＠]+)([a-zA-Z0-9_]{1,20})(\/[a-zA-Z][a-zA-Z0-9\x80-\xff\-]{0,79})?/
     REGEXEN[:auto_link_emoticon] = /(8\-\#|8\-E|\+\-\(|\`\@|\`O|\<\|:~\(|\}:o\{|:\-\[|\>o\<|X\-\/|\[:-\]\-I\-|\/\/\/\/Ö\\\\\\\\|\(\|:\|\/\)|∑:\*\)|\( \| \))/
 
     # URL related hash regex collection
-    REGEXEN[:valid_preceeding_chars] = /(?:[^\/"':!=]|^|\:)/
-    REGEXEN[:valid_domain] = /(?:[\.-]|[^[:punct:]])+\.[a-z]{2,}(?::[0-9]+)?/i
-    REGEXEN[:valid_url_path_chars] = /[a-z0-9!\*'\(\);:&=\+\$\/%#\[\]\-_\.,~]/i
+    REGEXEN[:valid_preceding_chars] = /(?:[^\/"':!=]|^|\:)/
+    REGEXEN[:valid_domain] = /(?:[\.-]|[^[:punct:]\s])+\.[a-z]{2,}(?::[0-9]+)?/i
+    REGEXEN[:valid_url_path_chars] = /[a-z0-9!\*'\(\);:&=\+\$\/%#\[\]\-_\.,~@]/i
     # Valid end-of-path chracters (so /foo. does not gobble the period).
     #   1. Allow ) for Wikipedia URLs.
     #   2. Allow =&# for empty URL parameters and other URL-join artifacts
     REGEXEN[:valid_url_path_ending_chars] = /[a-z0-9\)=#\/]/i
     REGEXEN[:valid_url_query_chars] = /[a-z0-9!\*'\(\);:&=\+\$\/%#\[\]\-_\.,~]/i
     REGEXEN[:valid_url_query_ending_chars] = /[a-z0-9_&=#]/i
     REGEXEN[:valid_url] = %r{
       (                                                                                     #   $1 total match
-        (#{REGEXEN[:valid_preceeding_chars]})                                               #   $2 Preceeding chracter
+        (#{REGEXEN[:valid_preceding_chars]})                                                #   $2 Preceeding chracter
         (                                                                                   #   $3 URL
           (https?:\/\/|www\.)                                                               #   $4 Protocol or beginning
           (#{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_query_chars]}*#{REGEXEN[:valid_url_query_ending_chars]})? #   $7 Query String