# HG changeset patch
# User Marcin Kuzminski <marcin@python-works.com>
# Date 1355184930 -3600
# Node ID 7d0476e1f1dc5fdd9e4f2a3c3b7b0741fdd5d084
# Parent  2ec4a2a5316777948cafd12f0f4127cbf89438f9
fixes issue #678 Incorrect diff markup when diff contains >, <, or & symbols
- regex by \W did split &amp; and other to 3 tokens, and escaping was broken

diff -r 2ec4a2a53167 -r 7d0476e1f1dc rhodecode/lib/diffs.py
--- a/rhodecode/lib/diffs.py	Tue Dec 11 00:02:23 2012 +0100
+++ b/rhodecode/lib/diffs.py	Tue Dec 11 01:15:30 2012 +0100
@@ -193,6 +193,9 @@
         (?:^\+\+\+[ ](b/(?P<b_file>.+)|/dev/null)(?:\n|$))?
     """, re.VERBOSE | re.MULTILINE)
 
+    #used for inline highlighter word split
+    _token_re = re.compile(r'()(&gt;|&lt;|&amp;|\W+?)')
+
     def __init__(self, diff, vcs='hg', format='gitdiff', diff_limit=None):
         """
         :param diff:   a text in diff format
@@ -274,9 +277,8 @@
         else:
             old, new = next_, line
 
-        oldwords = re.split(r'(\W)', old['line'])
-        newwords = re.split(r'(\W)', new['line'])
-
+        oldwords = self._token_re.split(old['line'])
+        newwords = self._token_re.split(new['line'])
         sequence = difflib.SequenceMatcher(None, oldwords, newwords)
 
         oldfragments, newfragments = [], []