Hi Surafel,
Thank you for your review!
But I have not finish fixing a patch yet.
> v Use <optional> tage in documentation
ok.
> v Don’t modified existing test case add new one instead
These existing tests I modified are the results of commands "SELECT
pg_get_triggerdef(oid, true) ... ".
I modified them in order to match with pg_get_functiondef().
The keyword 'OR REPLACE' is always added in CREATE FUNCTION.
Does not it make sense?
> v Comment in pg_constraint.c is extended make it short
> v Error message can be more guider if it tells about general rule
ok.
> v Wrong result in concurrency case
Thank you for the detailed example case. I appreciate it.
The issue does not seem to be resolved easily because I have to
modify pg_get_constraintdef_worker() in ruleutils.c.
And it takes some more time to modify pg_get_constraintdef_worker().
So, I am moving this patch to "Returned with Feedback".
Regards,
Okano Naoki
Fujitsu
