"Steven Flatt" <[EMAIL PROTECTED]> writes:
> Would this be considered a bug or is the tablespace info excluded for
> a particular reason?
It's intentional: pg_dump doesn't want it there. In a pg_dump script
we want to do
set default_tablespace = foo_space;
create index foo_idx on foo using btree (a);
because that way the creation command doesn't fail completely if there's
no such tablespace in the target installation.
We could maybe teach pg_dump to strip the clause out of
pg_get_indexdef's output, but that seemed complicated and fragile.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 3: Have you checked our extensive FAQ?
http://www.postgresql.org/docs/faq
