On Mon, Jul 16, 2012 at 04:49:46PM +0530, Amit Kapila wrote:
> Code Changes
> ----------------
> I will make changes in following functions to ensure that connoinherit
> should be appropriately set(pass the value as true).
> a. index_constraint_create()
> b. ATAddForeignKeyConstraint()
> c. CreateTrigger().
>
> Other places I have checked seems to be fine.
Looks right.
> Test
> --------
> I will create testcases similar to mentioned in bug report for
> a. unique key case, same as in bug-report
Good.
> b. foreign key case
> c. exclusion constraint case
Test coverage for those is optional.
> > Care to prepare a patch with a test case addition?
> Let me know if above is sufficient or shall I include anything more in
> patch.
I can't think of anything else.
As a side question for the list, should we fix this differently in 9.2 to
avoid forcing an initdb for the next beta? Perhaps have
ATExecDropConstraint() only respect connoinherit for CONSTRAINT_CHECK?