Hi,
On Saturday, Nov 7, 2020 2:06 PM Dilip Kumar <dilipbalaut@gmail.com> wrote:
> The patch looks fine to me however I feel that in the test case there are a lot
> of duplicate statement which can be reduced e.g.
> +-- 1. Overwrite existing regular trigger with regular trigger
> (without OR REPLACE)
> +create trigger my_trig
> + after insert on my_table
> + for each row execute procedure funcA(); create trigger my_trig
> + after insert on my_table
> + for each row execute procedure funcB(); -- should fail drop trigger
> +my_trig on my_table;
> +
> +-- 2. Overwrite existing regular trigger with regular trigger (with OR
> +REPLACE) create trigger my_trig
> + after insert on my_table
> + for each row execute procedure funcA(); insert into my_table values
> +(1); create or replace trigger my_trig
> + after insert on my_table
> + for each row execute procedure funcB(); -- OK insert into my_table
> +values (1); drop trigger my_trig on my_table;
>
> In this test, test 1 failed because it tried to change the trigger function without
> OR REPLACE, which is fine but now test 2 can continue from there, I mean we
> don't need to drop the trigger at end of the
> test1 and then test2 can try it with OR REPLACE syntax. This way we can
> reduce the extra statement execution which is not necessary.
OK. That makes sense.
Attached the revised version.
The tests in this patch should not include redundancy.
I checked the tests of trigger replacement for partition tables as well.
Here, I did not and will not delete the comments with numbering from 1 to 8 so that
other developers can check if the all cases are listed up or not easily.
Best,
Takamichi Osumi
