Yes, thanks, those are reasonable goals. ---------------------------------------------------------------------------
Daniel Farina wrote: > On Wed, Nov 24, 2010 at 7:21 PM, Bruce Momjian <br...@momjian.us> wrote: > > What are we adding a pl/pgsql dependency for? ?What is the benefit that > > will warrant requiring people who disable plpgsql to enable it for > > restores? > > There are two use cases I want to cover: > > 1) It should be possible to restore a dump made with --clean on an > empty database without error, so it can be run in a transaction and > the error code can be usefully monitored. > > 2) It should be possible a database be dumped and restored by a > non-superuser, again, cleanly, as per 1. > > It was easy enough to change all the "DROP ..." statements to "DROP > ... IF EXISTS", but the ALTER statements have no equivalent, and thus > the only way for a dump created with --clean to run without error is > to ensure that all table and domain constraints exist prior to > restore. > > The obvious mechanisms that have come to mind in this thread are: > > * An IF EXISTS variant of ALTER, at least for TABLE and DOMAIN > (although it may be strange to only support it on a couple of types) > > * Use of anonymous-DO code blocks (the prototype uses this, and this > depends on plpgsql) > > * Bizarre things I can imagine doing that involve creative queries > that, as a side effect, might drop objects that I have not mentioned > because I thought they were too gross to be given serious > consideration. But it might be plpgsql-less, which would be nice. > > Note that in the case where one wants to dump/restore as a > non-superuser that one may not be in a position to conveniently do a > (DROP|CREATE) DATABASE statement to work around the problem. > > -- > fdr -- Bruce Momjian <br...@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. + -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers