Horton

Re: Support logical replication of DDLs

Zheng Li <zhengli10@gmail.com>

View thread
From: Zheng Li <zhengli10@gmail.com>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>, Japin Li <japinli@hotmail.com>
Cc: Dilip Kumar <dilipbalaut@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, rajesh.rs0541@gmail.com
Date: 2022-03-17T16:22:46Z
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 a run_as_owner option to subscriptions.

  2. Refactor pgoutput_change().

  3. Print the correct aliases for DML target tables in ruleutils.

  4. Fix object identity string for transforms

  5. Add grantable MAINTAIN privilege and pg_maintain role.

  6. Get rid of recursion-marker values in enum AlterTableType

  7. Release cache tuple when no longer needed

  8. Empty search_path in logical replication apply worker and walsender.

  9. Refactor format_type APIs to be more modular

  10. Use wrappers of PG_DETOAST_DATUM_PACKED() more.

Hello Alvaro,

> I think this is a pretty interesting and useful feature.
>
> Did you see some old code I wrote towards this goal?
> https://www.postgresql.org/message-id/20150215044814.GL3391@alvh.no-ip.org
> The intention was that DDL would produce some JSON blob that accurately
> describes the DDL that was run; the caller can acquire that and use it
> to produce working DDL that doesn't depend on runtime conditions.  There
> was lots of discussion on doing things this way.  It was ultimately
> abandoned, but I think it's valuable.

Thanks for pointing this out. I'm looking into it.


Hello Japin,
>The new patch change the behavior of publication, when we drop a table
>on publisher, the table also be dropped on subscriber, and this made the
>regression testa failed, since the regression test will try to drop the
>table on subscriber.

>IMO, we can disable the DDLs replication by default, which makes the
>regression test happy.  Any thoughts?

Good catch, I forgot to run the whole TAP test suite. I think for now
we can simply disable DDL replication for the failing tests when
creating publication: $node_publisher->safe_psql('postgres',
    "CREATE PUBLICATION tap_pub FOR ALL TABLES WITH (ddl='')");
I'll fix it in the next version of patch. I'll also work on breaking
down the patch into smaller pieces for ease of review.

Regards,
Zheng