better regex used avoiding garbage google images

stamparm · stamparm · commit e938331d8e0a · 2010-05-15T22:02:28.000Z
diff --git a/lib/utils/google.py b/lib/utils/google.py
@@ -54,7 +54,7 @@ def __parsePage(self, page):
 
         matches = []
 
-        regExpr = "class=\042?r\042?\076\074a href=\042(http[s]*://.+?)\042\sclass=\042?l\042?"
+        regExpr = "li class=\042?g\042?\076.+?a href=\042(http[s]*://.+?)\042\sclass=\042?l\042?"
         matches = re.findall(regExpr, page, re.I | re.M)
 
         return matches
