Horton

Re: SQL:2011 application time

Paul A Jungwirth <pj@illuminatedcomputing.com>

View thread
From: Paul Jungwirth <pj@illuminatedcomputing.com>
To: Peter Eisentraut <peter@eisentraut.org>
Cc: jian he <jian.universality@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2024-11-22T00:30:06Z
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

Here are updated patches. I left off the final PERIODs patch because it still has some conflicts
with the new NOT NULL constraints work. I'll soon an update with that soon.

On 11/21/24 04:56, Peter Eisentraut wrote:
>> On the publisher side: You can use REPLICA IDENTITY DEFAULT with a temporal PK/UNIQUE index. There 
>> is no validation step, and sending the changes works fine. But REPLICA IDENTITY USING INDEX fails 
>> because the validation step rejects the non-btree index.
> 
> Ok, I have committed the fix for this, and I'll continue working through the rest of the patches.

Thanks!

On 11/13/24 02:11, Peter Eisentraut wrote:
> -       return get_equal_strategy_number_for_am(am);
> +       /* For GiST indexes we need to ask the opclass what strategy number to use. */
> +       if (am == GIST_AM_OID)
> +               return GistTranslateStratnum(opclass, RTEqualStrategyNumber);
> +       else
> +               return get_equal_strategy_number_for_am(am);
>
> This code should probably be pushed into get_equal_strategy_number_for_am().  That function already
> has a switch based on index AM OIDs.  Also, there are other callers of
> get_equal_strategy_number_for_am(), which also might want to become aware of GiST support.

Done. The reason I didn't do this before is because we need the opclass, not just the am. I put an
explanation about that into the function comment. If that's a problem I can undo the change.

> For the new test file, remember to add it to src/test/subscription/meson.build.
>
> Also, maybe add a introductory comment in the test file to describe generally what it's trying to test.

Done.

On 11/13/24 02:50, Peter Eisentraut wrote:
> A quick comment on the patch
>
> v43-0005-Add-UPDATE-DELETE-FOR-PORTION-OF.patch
>
> regarding the code in transformForPortionOfClause() and the additions you made to lsyscache.c:
>
> What you are doing is taking a type OID and a function OID and then converting them back to name and
> namespace and then building a node and then feeding that node back through the parse analysis
> transformation. This all seems quite fishy and cumbersome.  I think instead of building a FuncCall
> and transforming it, try to build a FuncExpr directly.  Then you wouldn't need these new helper
> functions, which would also reduce the surface area of your patch.

Okay, that makes sense. I changed that for the FOR PORTION OF bounds expression. I still need to get
the rangetype name and namespace to look up its constructor function. Even if I built another FuncExpr
myself there, I don't see any way to get the constructor oid without calling SearchSysCache3(PROCNAMEARGSNSP, ...).
So I'm still doing type oid -> type name -> func oid.

> Additional mini-comment:
>
> #include "utils/rangetypes.h"
>
> in src/include/nodes/execnodes.h appears to be unnecessary (but it is then required in src/backend/
> commands/trigger.c).

Done.

On 11/13/24 02:53, Peter Eisentraut wrote:
> I committed a few fixes in this area today.  Has everything here been addressed?

Yes, everything here was addressed in my v41 patches (sent 2024-10-21).

Rebased to 4c4aaa19a6.

Yours,

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