On Sat, Nov 6, 2010 at 6:09 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> Since we now have PL/pgsql by default, we could possibly fix pg_dump
> --clean by emitting a DO block, although the syntax for checking
> existence of a table is none too pretty, and it would make pg_dump
> --clean rely for correctness on plpgsql being installed, which might
> be none too desirable. It would actually be sort of spiffy to be able
> to have some of the PL/pgsql control constructs available in straight
> SQL, but I'm not expecting that to happen any time in the forseeable
> future.
I have hacked up a little set of work-in-progress pg_dump patches as a
proof of concept. It uses the PGSQL and DO mechanics, thanks for
making me aware of the latter...
https://github.com/fdr/postgres/tree/pg_dump-if-exists
I will be giving it a try on some more schemas in the near future. On
my small, plain but not-entirely-a-toy schema it works great.
For some reason I cannot seem to get it to exercise the ALTER DOMAIN
check, even though I created a domain with a check constraint...hmm,
I'll have to stare at what can cause it to exercise that later.
fdr