Обсуждение: [PATCH] Make ReScanForeignScan callback optional for FDWs

Hi hackers,

I'd like to propose a patch that makes the ReScanForeignScan callback 
optional for Foreign Data Wrappers, allowing the planner to automatically 
handle non-rescannable FDWs.

# Background

Currently, FDWs are expected to implement ReScanForeignScan to support 
rescanning in scenarios like nested loop joins. However, some FDWs just cannot
implement rescanning:

- Streaming data sources (Kafka, message queues)
- One-time token-based APIs with expensive re-authentication
- Data sources where re-fetching is prohibitively expensive
- ...

Right now, these FDWs either implement a stub ReScanForeignScan that fails 
at runtime, or they buffer all data in BeginForeignScan to support rescan, 
which wastes memory when rescanning isn't needed.

# Proposed Solution

This patch introduces a 'rescannable' field in the Path structure that 
tracks whether each path supports rescanning. The planner uses this 
information to:

1. Auto-detect FDW rescan capability by checking if ReScanForeignScan is 
   provided
2. Automatically insert Material nodes when non-rescannable paths are used 
   as inner paths in nested loops
3. Reject parameterized foreign scan paths if the FDW doesn't support 
   rescan (preventing planning failures)
4. Raise a clear error for correlated subqueries that cannot be handled 
   (these are planned independently and can't use Material nodes)

# Not only FDWs

Beyond enabling non-rescannable FDWs, this mechanism could also be used 
for performance optimization on other nodes. Some operations can technically
support rescan but at significant cost (MergeAppend redo all the sorts,
Aggregation redo all the calculations...). We may could mark such paths as
non-rescannable in some cases to encourage the planner to materialize results
instead.

The patch is attached. If reviewers feel this is a good idea and needs more
discussion or the complexity warrants it, I'm happy to register this for the
next CommitFest. For now, I'm sharing it here to gather initial feedback on the
approach.

-- 
Adam

On Wed, Dec 3, 2025 at 3:45 AM Adam Lee <adam8157@gmail.com> wrote:
> Beyond enabling non-rescannable FDWs, this mechanism could also be used
> for performance optimization on other nodes. Some operations can technically
> support rescan but at significant cost (MergeAppend redo all the sorts,
> Aggregation redo all the calculations...). We may could mark such paths as
> non-rescannable in some cases to encourage the planner to materialize results
> instead.

I'm not 100% sure, but my first reaction is to think that adding a
'rescannable' field to the Path struct is a bad idea. In these other
cases that you mention, I think we'd want to consider the cost of
paths with and without a materialize node and pick the cheaper one. We
don't have a terribly robust model for the cost of rescanning things
today, I think, but improving seems like the right way forward, rather
than making a hard decision that a certain node fully support
rescanning or is totally incapable of it.

My suggestion would be to consider just adding another callback to
FdwRoutine, or maybe giving one of the existing planner callbacks the
additional job of deciding whether rescan is allowed. That's a much
smaller design change. Suddenly making rescannability a property of
every Path node of every kind seems like it could have broad-ranging
ripple effects and I find that pretty scary. Just allowing an FDW to
turn it off if it really can't implement that feels like it might be
reasonable.

--
Robert Haas
EDB: http://www.enterprisedb.com