Small performance-improvement and bug-fix for indexer:

In rare cases find will return duplicate ids. The previous check that
re-initialized the value would never match (id at that point was python's
built-in function id). It's faster to not re-process an id when it has
been seen.

Author: Ralf Schlatterbeck

roundup/backends/indexer_common.py

@@ -1,4 +1,4 @@
-#$Id: indexer_common.py,v 1.10 2008-08-18 07:03:44 richard Exp $
+#$Id: indexer_common.py,v 1.11 2008-09-11 19:41:07 schlatterbeck Exp $
 import re, sets
 
 from roundup import hyperdb
@@ -86,8 +86,9 @@ def search(self, search_terms, klass, ignore={}):
         propdefs = klass.getprops()
         for resid in klass.find(**propspec):
             resid = str(resid)
-            if not nodeids.has_key(id):
-                nodeids[resid] = {}
+            if resid in nodeids:
+                continue # we ignore duplicate resids
+            nodeids[resid] = {}
             node_dict = nodeids[resid]
             # now figure out where it came from
             for linkprop in propspec.keys():