New regular expression to remove special characters (for instance ’).

1 files changed, 2 insertions, 1 deletions

diff --git a/utils.py b/utils.py
index 3b8b376a..fc945f34 100755
--- a/utils.py
+++ b/utils.py
@@ -74,7 +74,8 @@ def remove_html_tags(data):
     Remove HTML tags for the search.
     """
     p = re.compile(r'<[^<]*?/?>')
-    return p.sub('', data)
+    q = re.compile(r'&#[0-9]+;')
+    return p.sub('', q.sub('', data))
 
 def top_words(dic_articles, n=10, size=5):
     """