On Monday, September 27, 2021 1:32 PM, vignesh C <vignesh21@gmail.com> wrote:
>Attached v33 patch has the preprocess_pubobj_list review comment fix
>suggested by Alvaro at [1]. The
>v33-0006-Alternate-grammar-for-ALL-TABLES-IN-SCHEMA.patch patch has
>the grammar changes as suggested by Alvaro at [1]. If we agree this is
>better, I will merge this into the 0001 patch.
>[1] - https://www.postgresql.org/message-id/202109241325.eag5g6mpvoup%40alvherre.pgsql
About the schema patch, I think a schema and a table which belongs to this schema shouldn't be specified at the same
time.
But what if someone uses "ALTER TABLE ... SET SCHEMA ..." after "CREATE PUBLICATION"?
For example:
create schema sch1;
create schema sch2;
create table sch2.t (a int);
create publication pub1 for all tables in schema sch1, table sch2.t; alter table sch2.t set schema sch1;
postgres=# \dRp+
Publication pub1
Owner | All tables | Inserts | Updates | Deletes | Truncates | Via root
----------+------------+---------+---------+---------+-----------+------
----------+------------+---------+---------+---------+-----------+----
postgres | f | t | t | t | t | f
Tables:
"sch1.t"
Tables from schemas:
"sch1"
Table t has been output twice.
I think this should not be supported, should we do something for this scenario?
Regards
Tang
