Fix RDBMS indexer indexing UTF-8 words that encode to > 30 chars

(a better fix would be nice)

Author: Richard Jones

CHANGES.txt

@@ -9,6 +9,7 @@ Fixed:
 - Display of Multilinks where linked Class labelprop values are None
 - Fix references to the old * Registration Permissions
 - Fix missing merge of fix to sf bug 1177057
+- Fix RDBMS indexer indexing UTF-8 words that encode to > 30 chars
 
 
 2005-07-18 0.8.4

roundup/backends/indexer_rdbms.py

@@ -56,8 +56,8 @@ def add_text(self, identifier, text, mime_type='text/plain'):
                 for w in re.findall(r'(?u)\b\w{2,25}\b', text)]
         words = {}
         for word in wordlist:
-            if is_stopword(word):
-                continue
+            if is_stopword(word): continue
+            if len(word) > 25: continue
             words[word] = 1
         words = words.keys()