Horton

Re: Support logical replication of DDLs

Zheng Li <zhengli10@gmail.com>

View thread
From: Zheng Li <zhengli10@gmail.com>
To: li jie <ggysxcq@gmail.com>
Cc: vignesh C <vignesh21@gmail.com>, Peter Smith <smithpb2250@gmail.com>, Ajin Cherian <itsajin@gmail.com>, Dilip Kumar <dilipbalaut@gmail.com>, Alvaro Herrera <alvherre@alvh.no-ip.org>, "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, Amit Kapila <amit.kapila16@gmail.com>, Masahiko Sawada <sawada.mshk@gmail.com>, Japin Li <japinli@hotmail.com>, rajesh singarapu <rajesh.rs0541@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-12-19T21:25:59Z
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.

Attachments

Attaching v54 patch set which adds support for:
- CREATE/ALTER/DROP FOREIGN TABLE
- IMPORT FOREIGN SCHEMA, this is captured and replicated as individual
CREATE FOREIGN TABLE command for each FOREIGN TABLE in the SCHEMA.

Note:
DROP FOREIGN TABLE ft1 also generates:
DROP type IF EXISTS ft1;
and
DROP type IF EXISTS ft1[];
These two dropped objects are also captured and replicated to the
subscriber along with the DROP FOREIGN TABLE command which aren't
necessary.

In addition, the patch fixed a bug in deparse_CreateSchemaStmt which
causes a quoted identifier to fail in replication, for example:
 CREATE SCHEMA "S 2"; is replicated as CREATE SCHEMA S 2, which will
fail during apply.
Fix is to change %{name}s -> %{name}I in deparse_CreateSchemaStmt.

On Mon, Dec 19, 2022 at 5:02 AM li jie <ggysxcq@gmail.com> wrote:
>
> I have presented some comments below:

Thanks for the feedback. I'll look into these.

> 1. AT_AddColumn
>
> > + tmpobj = new_objtree_VA("ADD %{objtype}s %{definition}s", 3,
> [ IF NOT EXISTS ] is missing here.
>
......
>
> 9. regress test
>
> The test patch is very useful.
>  I see that the sql case in test_ddl_deparse_regress is similar to the
> one in test_ddl_deparse.
> Why don't we merge the test cases in test_ddl_deparse_regress into
> test_ddl_deparse,
> as the sql case can be completely reused with the sql files in test_ddl_deparse?
> I believe this will make the tests more comprehensive and reduce redundancy.

We have set up test_ddl_deparse_regress as a new module initially to
not interfere with what's being tested by test_ddl_deparse. We could
merge the two test modules if it turns out that we can expand on
test_ddl_deparse to achieve our testing goals and to add more test
cases without breaking what's currently being tested by
test_ddl_deparse.

Regards,
Zheng