Re: row filtering for logical replication

Peter Smith <smithpb2250@gmail.com>

From: Peter Smith <smithpb2250@gmail.com>
To: Ajin Cherian <itsajin@gmail.com>
Cc: Amit Kapila <amit.kapila16@gmail.com>, Euler Taveira <euler@eulerto.com>, "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, Greg Nancarrow <gregn4422@gmail.com>, vignesh C <vignesh21@gmail.com>, "tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.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: 2021-12-24T05:34:34Z
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

The current PG docs text for CREATE PUBLICATION (in the v54-0001
patch) has a part that says

+   A nullable column in the <literal>WHERE</literal> clause could cause the
+   expression to evaluate to false; avoid using columns without not-null
+   constraints in the <literal>WHERE</literal> clause.

I felt that the caution to "avoid using" nullable columns is too
strongly worded. AFAIK nullable columns will work perfectly fine so
long as you take due care of them in the WHERE clause. In fact, it
might be very useful sometimes to filter on nullable columns.

Here is a small test example:

// publisher
test_pub=# create table t1 (id int primary key, msg text null);
test_pub=# create publication p1 for table t1 where (msg != 'three');
// subscriber
test_sub=# create table t1 (id int primary key, msg text null);
test_sub=# CREATE SUBSCRIPTION sub1 CONNECTION 'host=localhost
dbname=test_pub application_name=sub1' PUBLICATION p1;

// insert some data
test_pub=# insert into t1 values (1, 'one'), (2, 'two'), (3, 'three'),
(4, null), (5, 'five');
test_pub=# select * from t1;
 id |  msg
----+-------
  1 | one
  2 | two
  3 | three
  4 |
  5 | five
(5 rows)

// data at sub
test_sub=# select * from t1;
 id | msg
----+------
  1 | one
  2 | two
  5 | five
(3 rows)

Notice the row 4 with the NULL is also not replicated. But, perhaps we
were expecting it to be replicated (because NULL is not 'three'). To
do this, simply rewrite the WHERE clause to properly account for
nulls.

// truncate both sides
test_pub=# truncate table t1;
test_sub=# truncate table t1;

// alter the WHERE clause
test_pub=# alter publication p1 set table t1 where (msg is null or msg
!= 'three');

// insert data at pub
test_pub=# insert into t1 values (1, 'one'), (2, 'two'), (3, 'three'),
(4, null), (5, 'five');
INSERT 0 5
test_pub=# select * from t1;
 id |  msg
----+-------
  1 | one
  2 | two
  3 | three
  4 |
  5 | five
(5 rows)

// data at sub (not it includes the row 4)
test_sub=# select * from t1;
 id | msg
----+------
  1 | one
  2 | two
  4 |
  5 | five
(4 rows)

~~

So, IMO the PG docs wording for this part should be relaxed a bit.

e.g.
BEFORE:
+   A nullable column in the <literal>WHERE</literal> clause could cause the
+   expression to evaluate to false; avoid using columns without not-null
+   constraints in the <literal>WHERE</literal> clause.
AFTER:
+   A nullable column in the <literal>WHERE</literal> clause could cause the
+   expression to evaluate to false. To avoid unexpected results, any possible
+   null values should be accounted for.

Thoughts?

------
Kind Regards,
Peter Smith.
Fujitsu Australia