Horton

Re: SQL:2011 application time

Paul A Jungwirth <pj@illuminatedcomputing.com>

View thread
From: Paul Jungwirth <pj@illuminatedcomputing.com>
To: jian he <jian.universality@gmail.com>
Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2023-09-14T16:11:02Z
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. Rename gist stratnum support function

  2. Remove support for temporal RESTRICT foreign keys

  3. Cache NO ACTION foreign keys separately from RESTRICT foreign keys

  4. Fix NO ACTION temporal foreign keys when the referenced endpoints change

  5. Improve whitespace in without_overlaps test

  6. Tests for logical replication with temporal keys

  7. Support for GiST in get_equal_strategy_number()

  8. Make the conditions in IsIndexUsableForReplicaIdentityFull() more explicit

  9. Replace get_equal_strategy_number_for_am() by get_equal_strategy_number()

  10. Improve internal logical replication error for missing equality strategy

  11. Simplify IsIndexUsableForReplicaIdentityFull()

  12. Fix ALTER TABLE / REPLICA IDENTITY for temporal tables

  13. doc: Update pg_constraint.conexclop docs for WITHOUT OVERLAPS

  14. doc: Add PERIOD to ALTER TABLE reference docs

  15. doc: Add WITHOUT OVERLAPS to ALTER TABLE reference docs

  16. Add temporal FOREIGN KEY contraints

  17. Add temporal PRIMARY KEY and UNIQUE constraints

  18. Add stratnum GiST support function

  19. Avoid crashing when a JIT-inlined backend function throws an error.

  20. Revert temporal primary keys and foreign keys

  21. Fix ON CONFLICT DO NOTHING/UPDATE for temporal indexes

  22. Add test for REPLICA IDENTITY with a temporal key

  23. Use half-open interval notation in without_overlaps tests

  24. Use daterange and YMD in without_overlaps tests instead of tsrange.

  25. Rename pg_constraint.conwithoutoverlaps to conperiod

  26. Fix comment on gist_stratnum_btree

  27. Add missing TAP test name

  28. Improve error handling of HMAC computations

  29. Rename functions to avoid future conflicts

Attachments

Thanks for the thorough review and testing!

Here is a v14 patch with the segfault and incorrect handling of NO 
ACTION and RESTRICT fixed (and reproductions added to the test suite).

A few more comments below on feedback from you and Peter:

On 9/12/23 02:01, jian he wrote:
> hi. some trivial issue:
> 
> in src/backend/catalog/index.c
> /* * System attributes are never null, so no need to check. */
> if (attnum <= 0)
> 
> since you already checked attnum == 0
> so here you can just attnum < 0?

I fixed the "/* *" typo here. I'm reluctant to change the attnum 
comparison since that's not a line I touched. (It was just part of the 
context around the updated comment.) Your suggestion does make sense 
though, so perhaps it should be a separate commit?

> ERROR:  column "valid_at" named in WITHOUT OVERLAPS is not a range type
> 
> IMHO, "named" is unnecessary.

Changed.

> doc/src/sgml/catalogs.sgml
> pg_constraint adds another attribute (column): contemporal, seems no doc entry.

Added.

> also the temporal in oxford definition is "relating to time", here we
> can deal with range.
> So maybe  "temporal" is not that accurate?

I agree if we allow multiple WITHOUT OVERLAPS/etc clauses, we should 
change the terminology. I'll include that with the multiple-range-keys 
change discussed upthread.

On 9/1/23 02:30, Peter Eisentraut wrote:
 > * There is a lot of talk about "temporal" in this patch, but this
 > functionality is more general than temporal.  I would prefer to change
 > this to more neutral terms like "overlaps".

Okay, sounds like several of us agree on this.

 > * The field ii_Temporal in IndexInfo doesn't seem necessary and could
 > be  handled via local variables.  See [0] for a similar discussion:
 >
 > [0]:
 > 
https://www.postgresql.org/message-id/flat/f84640e3-00d3-5abd-3f41-e6a19d33c40b@eisentraut.org

Done.

 > * In gram.y, change withoutOverlapsClause -> without_overlaps_clause
 > for consistency with the surrounding code.

Done.

 > * No-op assignments like n->without_overlaps = NULL; can be omitted.
 > (Or you should put them everywhere.  But only in some places seems
 > inconsistent and confusing.)

Changed. That makes sense since newNode uses palloc0fast. FWIW there is 
quite a lot of other code in gram.y that sets NULL fields though, 
including in ConstraintElem, and it seems like it does improve the 
clarity a little. By "everywhere" I think you mean wherever the file 
calls makeNode(Constraint)? I might go back and do it that way later.

I'll keep working on a patch to support multiple range keys, but I 
wanted to work through the rest of the feedback first. Also there is 
some fixing to do with partitions I believe, and then I'll finish the 
PERIOD support. So this v14 patch is just some minor fixes & tweaks from 
September feedback.

Yours,

-- 
Paul              ~{:-)
pj@illuminatedcomputing.com