Re: row filtering for logical replication

Amit Kapila <amit.kapila16@gmail.com>

From: Amit Kapila <amit.kapila16@gmail.com>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>
Cc: "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, Peter Smith <smithpb2250@gmail.com>, "tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.com>, Euler Taveira <euler@eulerto.com>, Greg Nancarrow <gregn4422@gmail.com>, vignesh C <vignesh21@gmail.com>, Ajin Cherian <itsajin@gmail.com>, Dilip Kumar <dilipbalaut@gmail.com>, Rahila Syed <rahilasyed90@gmail.com>, Peter Eisentraut <peter.eisentraut@enterprisedb.com>, Önder Kalacı <onderkalaci@gmail.com>, japin <japinli@hotmail.com>, Michael Paquier <michael@paquier.xyz>, David Steele <david@pgmasters.net>, Craig Ringer <craig@2ndquadrant.com>, Amit Langote <amitlangote09@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-01-22T04:21:00Z
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. Release cache tuple when no longer needed

  2. Add some additional tests for row filters in logical replication.

  3. Fix one of the tests introduced in commit 52e4f0cd47.

  4. Allow specifying row filters for logical replication of tables.

  5. Move scanint8() to numutils.c

  6. Replace Test::More plans with done_testing

  7. Reduce relcache access in WAL sender streaming logical changes

  8. Small cleanups related to PUBLICATION framework code

  9. Add a view to show the stats of subscription workers.

  10. Allow publishing the tables of schema.

  11. Doc: improve documentation of CREATE/ALTER SUBSCRIPTION.

  12. Add PublicationTable and PublicationRelInfo structs

  13. Remove unused argument "txn" in maybe_send_schema().

  14. Add prepare API support for streaming transactions in logical replication.

  15. Unify PostgresNode's new() and get_new_node() methods

  16. Use l*_node() family of functions where appropriate

  17. Add support for prepared transactions to built-in logical replication.

  18. Restore the portal-level snapshot after procedure COMMIT/ROLLBACK.

  19. Rename a parse node to be more general

  20. Remove unused column atttypmod from initial tablesync query

  21. SEARCH and CYCLE clauses

On Fri, Jan 21, 2022 at 8:19 PM Alvaro Herrera <alvherre@alvh.no-ip.org> wrote:
>
> If ATTACH PARTITION or CREATE TABLE .. PARTITION OF don't let you
> specify replica identity, I suspect it's because both partitioning and
> logical replication were developed in parallel, and neither gave too
> much thought to the other.
>

I think the reason CREATE TABLE .. syntax form doesn't have a way to
specify RI is that we need to have an index for RI. Consider the below
example:
----
CREATE TABLE parent (a int primary key, b int not null, c varchar)
PARTITION BY RANGE(a);
CREATE TABLE child PARTITION OF parent FOR VALUES FROM (0) TO (250);
CREATE UNIQUE INDEX b_index on child(b);
ALTER TABLE child REPLICA IDENTITY using INDEX b_index;
----

In this, the parent table's replica identity is the primary
key(default) and the child table's replica identity is the b_index. I
think if we want we can come up with some syntax to combine these
steps and allow to specify replica identity during the second step
(Create ... Partition) but not sure if we have a convincing reason for
this feature per se.

>
> I suspect that a better way to attack this problem is to let ALTER TABLE
> ... ATTACH PARTITION and CREATE TABLE .. PARTITION OF specify a replica
> identity as necessary.
>
> My suggestion is to avoid painting us into a corner from which it will
> be impossible to get out later.
>

Apart from the above reason, here we are just following the current
model of how the update/delete behaves w.r.t RI. Now, I think in the
future we can also think of uplifting some of the restrictions related
to RI for filters if we find a good way to have columns values that
are not in WAL. We have discussed this previously in this thread and
thought that it is sensible to have a RI restriction for
updates/deletes as the patch is doing for the first version.

I am not against inventing some new syntaxes for row/column filter
patches but there doesn't seem to be a very convincing reason for it
and there is a good chance that we won't be able to accomplish that
for the current version.

-- 
With Regards,
Amit Kapila.