On Sat, Dec 30, 2017 at 09:19:05AM -0500, Robert Blayzor wrote:
> On Dec 30, 2017, at 12:38 AM, Justin Pryzby wrote:
> > BTW depending on your requirements, it may be possible to make pg_dump much
> > more efficient. For our data, it's reasonable to assume that a table is
> > "final" if its cons
On Dec 30, 2017, at 12:38 AM, Justin Pryzby wrote:
> For inheritance (available in and before PG10), the parent may be nonempty,
> which works fine, although someone else might find it unintuitive. (Does the
> doc actually say "should" somewhere ?)
Well it doesn’t say should, but says “normally”
