On 5 December 2013 10:06, Pavel Stehule <pavel.stehule@gmail.com> wrote: >> I think #1175 is close to being ready for commit. Pavel, will you >> produce an updated patch based on our last discussion? I'll set this >> patch to waiting on author. > > > I expected so your version was a final. I have no problem to do other > enhancing (by me) , but I don't fully understand to your last proposal. Can > you specify it more, please? >
Well I was basically proposing that does_not_exist_skipping() be enhanced to report on non-existent types that form part of the object specification. I think this would affect the CAST, FUNCTION, AGGREGATE and OPERATOR cases, but should be a fairly trivial extension to the code that you've already added.
ok, updated patch is in attachment
It should be possible to make the notice text from DROP...IF EXISTS consistent with the error text from a plain DROP. For example consider cases like func_name(no_such_schema.typename) and func_name(existing_schema.no_such_type).
yes, it is possible, but I wouldn't enhance a scope of this patch, if it is possible, please. I would to finish a --if-exist option for pg_dump a pg_restore before (in this release cycle).
We can continue in cleaning DROP obj messages in next release, or maybe in last commitfest - this functionality is not critical and is not complex.