Fix an error in the original coding of holdable cursors: PersistHoldablePortal

thought that it didn't have to reposition the underlying tuplestore if the
portal is atEnd.  But this is not so, because tuplestores have separate read
and write cursors ... and the read cursor hasn't moved from the start.
This mistake explains bug #2970 from William Zhang.

Note: the coding here is pretty inefficient, but given that no one has noticed
this bug until now, I'd say hardly anyone uses the case where the cursor has
been advanced before being persisted.  So maybe it's not worth worrying about.

diff --git a/src/backend/commands/portalcmds.c b/src/backend/commands/portalcmds.c
index be5ca1105902619419d8b92f7e401661d5f425a4..c5d453807ebc770ecbb3718e0c6d9c0ae9ccc7b2 100644 (file)
--- a/src/backend/commands/portalcmds.c
+++ b/src/backend/commands/portalcmds.c
@@ -363,7 +363,20 @@ PersistHoldablePortal(Portal portal)
         */
        MemoryContextSwitchTo(portal->holdContext);
 
-       if (!portal->atEnd)
+       if (portal->atEnd)
+       {
+               /* we can handle this case even if posOverflow */
+               HeapTuple       tup;
+               bool            should_free;
+
+               while ((tup = tuplestore_gettuple(portal->holdStore, true,
+                                                                                 &should_free)) != NULL)
+               {
+                       if (should_free)
+                               pfree(tup);
+               }
+       }
+       else
        {
                long            store_pos;