"Ed L." <pgsql@xxxxxxxxxxxxx> writes:
> Are there
> are any known or obvious gotchas associated with transforming a
> unique index on a non null column into a primary key via this
> sql?
> update pg_index
> set indisprimary = 't'
> where indexrelid = <my non-null unique index oid>
The problem with that is there won't be any pg_constraint entry,
nor any pg_depend entries linking to/from the constraint.
I don't offhand know which bits of logic look at indisprimary
and which pay attention to the pg_constraint entry (and 7.4 is
probably different from current sources on the point anyway).
Things could get a bit weird though, particularly for pg_dump.
Of course, you could gin up the required pg_constraint and pg_depend
entries by hand too, but it's a lot more complex than the above.
If you really wanna do this I'd strongly recommend experimenting
in a scratch database.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
