From 64f56e2b4c75ee03f43bb253309fd61fe1f5c3a8 Mon Sep 17 00:00:00 2001
From: cedricbonhomme <devnull@localhost>
Date: Tue, 6 Jul 2010 09:54:43 +0200
Subject: Regular expression improvement.

---
 utils.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/utils.py b/utils.py
index 482a59a7..91f61d72 100755
--- a/utils.py
+++ b/utils.py
@@ -81,8 +81,8 @@ def clear_string(data):
     """
     p = re.compile(r'<[^<]*?/?>')
     q = re.compile(r'&#[0-9]+;')
-    r = re.compile(r's+')
-    return p.sub('', q.sub('', r.sub('', data)))
+    r = re.compile(r'\s')
+    return p.sub('', q.sub('', r.sub(' ', data)))
 
 def top_words(dic_articles, n=10, size=5):
     """
-- 
cgit