Merge 0f88665 into 492fa46

twitter-archive · Nov 14, 2014 · 8a7236a · 8a7236a
2 parents 492fa46 + 0f88665
commit 8a7236a
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 2 deletions.
diff --git a/src/com/twitter/Regex.java b/src/com/twitter/Regex.java
@@ -80,7 +80,8 @@ private static String join(Collection<?> col, String delim) {
                                                    "\\uff21-\\uff3a\\uff41-\\uff5a" +  // full width Alphabet
                                                    "\\uff66-\\uff9f" +                 // half width Katakana
                                                    "\\uffa1-\\uffdc";                  // half width Hangul (Korean)
-  private static final String HASHTAG_ALPHA_NUMERIC_CHARS = "0-9\\uff10-\\uff19_" + HASHTAG_ALPHA_CHARS;
+  private static final String SPECIAL_CHARS_INCLUDED = "\\u00b7\\u2022\\u30fb\\uff65";
+  private static final String HASHTAG_ALPHA_NUMERIC_CHARS = "0-9\\uff10-\\uff19_" + HASHTAG_ALPHA_CHARS + SPECIAL_CHARS_INCLUDED;
   private static final String HASHTAG_ALPHA = "[" + HASHTAG_ALPHA_CHARS +"]";
   private static final String HASHTAG_ALPHA_NUMERIC = "[" + HASHTAG_ALPHA_NUMERIC_CHARS +"]";
 

diff --git a/test-data/twitter-text-conformance b/test-data/twitter-text-conformance