Re: Extended Statistics set/restore/clear functions.

jian he <jian.universality@gmail.com>

From: jian he <jian.universality@gmail.com>
To: Corey Huinker <corey.huinker@gmail.com>
Cc: Michael Paquier <michael@paquier.xyz>, Tomas Vondra <tomas@vondra.me>, pgsql-hackers@lists.postgresql.org, tgl@sss.pgh.pa.us
Date: 2025-11-19T07:49:56Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →
  1. Add test doing some cloning of extended statistics data

  2. Add test for pg_restore_extended_stats() with multiranges

  3. Add support for "mcv" in pg_restore_extended_stats()

  4. Include extended statistics data in pg_dump

  5. Add support for "dependencies" in pg_restore_extended_stats()

  6. Add test for MAINTAIN permission with pg_restore_extended_stats()

  7. Add pg_restore_extended_stats()

  8. Add routine to free MCVList

  9. Improve pg_clear_extended_stats() with incorrect relation/stats combination

  10. Add pg_clear_extended_stats()

  11. Introduce routines to validate and free MVNDistinct and MVDependencies

  12. Fix typo in stat_utils.c

  13. Move attribute statistics functions to stat_utils.c

  14. Improve error messages of input functions for pg_dependencies and pg_ndistinct

  15. Improve test output of extended statistics for ndistinct and dependencies

  16. Fix some compiler warnings

  17. Add input function for data type pg_dependencies

  18. Add input function for data type pg_ndistinct

  19. Rework output format of pg_dependencies

  20. Rework output format of pg_ndistinct

  21. Fix comments of output routines for pg_ndistinct and pg_dependencies

  22. Move code specific to pg_dependencies to new file

  23. Move code specific to pg_ndistinct to new file

  24. Document some structures in attribute_stats.c

  25. Fix FATAL message for invalid recovery timeline at beginning of recovery

Attachments

On Tue, Nov 18, 2025 at 4:52 PM Corey Huinker <corey.huinker@gmail.com> wrote:
> v15:
>
> - catches duplicate object keys cited above
> - enforces attnum ordering (ascending positive numbers followed by descending negative numbers, no zeros allowed), which means we get duplicate attnum detection for free
> - attnum validation is now done as soon as the attnum is parsed
> - tests refactored to put attnums in proper order
> - unfortunately, this means that one of the error cases from stats_import.sql (attnum = 0) is now an error rather than something that can be soft-excluded.
> - didn't enforce combinatorical completeness for dependencies because not all combinations are guaranteed to be there.
> - didn't enforce combinatorical completeness for ndistinct because I'm not convinced we should.
>

hi.

some of the switch->default, default don't have ``break``.

+ for (int i = 0; i < nitems; i++)
+ {
+ MVNDistinctItem *item = parse_state.distinct_items->elements[i].ptr_value;

exposing the ptr_value seems not a good idea, we can foreach_ptr
the attached patch using foreach_ptr.

in function pg_ndistinct_in some errsave can change to ereturn.
(I didn't do this part, though).

+ /*
+ * The attnum cannot be zero a negative number beyond the number of the
+ * possible expressions.
+ */
+ if (attnum == 0 || attnum < (0-STATS_MAX_DIMENSIONS))
+ {
+ errsave(parse->escontext,
+ errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed pg_ndistinct: \"%s\"", parse->str),
+ errdetail("Invalid \"%s\" element: %d.",
+ PG_NDISTINCT_KEY_ATTRIBUTES, attnum));
+ return JSON_SEM_ACTION_FAILED;
+ }
This part had no coverage tests, so I added a few.


as mentioned before
+ errsave(parse->escontext,
+ errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed pg_ndistinct: \"%s\"", parse->str),
+ errdetail("The \"%s\" key must contain an array of at least %d "
+  "and no more than %d attributes.",
+  PG_NDISTINCT_KEY_NDISTINCT, 2, STATS_MAX_DIMENSIONS));
here PG_NDISTINCT_KEY_NDISTINCT, should be PG_NDISTINCT_KEY_ATTRIBUTES.


Please check the attached minor miscellaneous changes.

--
jian
https://www.enterprisedb.com/