On Sun, Feb 2, 2014 at 10:15 PM, Etsuro Fujita
<fujita.etsuro@lab.ntt.co.jp> wrote:
> Allowing ALTER COLUMN SET STORAGE on foreign tables would make sense if for
> example, "SELECT * INTO local_table FROM foreign_table" did create a new
> local table of columns having the storage types associated with those of a
> foreign table?
Seems like a pretty weak argument. It's not that we can't find
strange corner cases where applying SET STORAGE to a foreign table
doesn't do something; it's that they *are* strange corner cases. The
options as we normally don't understand them just aren't sensible in
this context, and a good deal of work has been put into an alternative
options framework, which is what authors of FDWs ought to be using.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
