Avoid fetching from an already-terminated plan.

Some plan node types don't react well to being called again after
they've already returned NULL.  PortalRunSelect() has long dealt
with this by calling the executor with NoMovementScanDirection
if it sees that we've already run the portal to the end.  However,
commit ba2c6d6ce overlooked this point, so that persisting an
already-fully-fetched cursor would fail if it had such a plan.

Per report from Tomas Barton.  Back-patch to v11, as the faulty
commit was.  (I've omitted a test case because the type of plan
that causes a problem isn't all that stable.)

Discussion: https://postgr.es/m/CAPV2KRjd=ErgVGbvO2Ty20tKTEZZr6cYsYLxgN_W3eAo9pf5sw@mail.gmail.com
This commit is contained in:
Tom Lane 2021-09-09 13:36:31 -04:00
parent b27d0cd314
commit b7056c0a25
1 changed files with 15 additions and 3 deletions

View File

@ -358,6 +358,8 @@ PersistHoldablePortal(Portal portal)
savePortalContext = PortalContext;
PG_TRY();
{
ScanDirection direction = ForwardScanDirection;
ActivePortal = portal;
if (portal->resowner)
CurrentResourceOwner = portal->resowner;
@ -380,10 +382,20 @@ PersistHoldablePortal(Portal portal)
}
else
{
/* We must reset the cursor state as though at start of query */
/* Disallow moving backwards from here */
portal->atStart = true;
portal->atEnd = false;
portal->portalPos = 0;
/*
* If we already reached end-of-query, set the direction to
* NoMovement to avoid trying to fetch any tuples. (This check
* exists because not all plan node types are robust about being
* called again if they've already returned NULL once.) We'll
* still set up an empty tuplestore, though, to keep this from
* being a special case later.
*/
if (portal->atEnd)
direction = NoMovementScanDirection;
}
/*
@ -400,7 +412,7 @@ PersistHoldablePortal(Portal portal)
NULL);
/* Fetch the result set into the tuplestore */
ExecutorRun(queryDesc, ForwardScanDirection, 0L, false);
ExecutorRun(queryDesc, direction, 0L, false);
queryDesc->dest->rDestroy(queryDesc->dest);
queryDesc->dest = NULL;