Bruce Momjian <pgman@candle.pha.pa.us> writes:
> I think Jim has some very good points here. What does his
> implementation lack?
Forward compatibility to a future tablespace implementation.
If we do this, we'll be stuck with supporting this feature set,
not to mention this syntax; neither of which have garnered any
support from the assembled hackers.
I went back to look at TODO.detail/tablespaces, and find that it's
badly in need of editing. Much of the discussion there is
back-and-forthing about the question of naming files by OID,
which is now a done deal. But it is clear that people wanted to
have a notion of tablespaces as objects somewhat orthogonal to
databases. I didn't see any support for hard-wiring tablespace
assignments on the basis of "tables here, indexes there", either.
regards, tom lane