Horton

Re: Converting SetOp to read its two inputs separately

Richard Guo <guofenglinux@gmail.com>

View thread View on postgr.es
From: Richard Guo <guofenglinux@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers@lists.postgresql.org, David Rowley <dgrowleyml@gmail.com>
Date: 2024-11-14T09:32:57Z
Lists: pgsql-hackers 
On Thu, Nov 14, 2024 at 6:28 PM Richard Guo <guofenglinux@gmail.com> wrote:
> 1. We need to teach set_operation_ordered_results_useful() that sorted
> input paths are also useful for INTERSECT/EXCEPT, so that we can have
> setop_pathkeys set for the subqueries.

BTW, I noticed a typo in the comment for function
set_operation_ordered_results_useful().

  * set_operation_ordered_results_useful
  *     Return true if the given SetOperationStmt can be executed by utilizing
  *     paths that provide sorted input according to the setop's targetlist.
- *     Returns false when sorted paths are not any more useful then unsorted
+ *     Returns false when sorted paths are not any more useful than unsorted
  *     ones.

Thanks
Richard

Commits

  1. Get rid of old version of BuildTupleHashTable().

  2. Use ExecGetCommonSlotOps infrastructure in more places.

  3. Improve planner's handling of SetOp plans.

  4. Convert SetOp to read its inputs as outerPlan and innerPlan.

  5. Fix typo in header comment for set_operation_ordered_results_useful

  6. Allow planner to use Merge Append to efficiently implement UNION