On Thu, Apr 21, 2005 at 09:24:49PM -0500, John Browne wrote: > > Ok, I've been using postgres for a-while now, and am curious how you > guys handle this. What is the best way to make modifications to > tables that have lots of dependent objects, like views? Do you just > do the obvious drop...cascade and manually re-create your views? Do > you keep your "create view" statements in a text file for this purpose > or just pull them from a pg_dump file? Is there a better way to > handle this sort of thing?
I put CREATE statements in a file so I can easily reload them. Sometimes I put views and functions in their own files so they can be reloaded separately, since dropping and recreating them is usually a trivial, non-destructive operation, unlike dropping and recreating tables. I usually put DDL statements in a transaction, for a couple of reasons: so that a mistake doesn't leave me with half-done work (any error will cause the entire transaction to roll back), and to make the changes atomic for the benefit of other transactions. -- Michael Fuhr http://www.fuhr.org/~mfuhr/ ---------------------------(end of broadcast)--------------------------- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
