On 2005-10-25, Simon Riggs <[EMAIL PROTECTED]> wrote: > On Tue, 2005-10-25 at 17:43 -0400, Tom Lane wrote: > >> 1. Tell people they may have to set add_missing_from = true to reload >> a dump that contains such views. >> >> 2. Revert the change to make add_missing_from default as false, and >> wait a few more releases before making it default. >> >> Comments? > > This needs to be (1). > > Whichever version we make the change for, there will always be a point > where the pg_dump would be taken with add_missing_from=true by default > and moving to new version with add_missing_from=false. Putting that off > for another release will still allow that failure condition to exist at > that later time: so we must tell people about it now.
Wild idea: how about having pg_dump include SET add_missing_from = true; in the dump file if, and only if, it is set that way in the server? Perhaps do this only for server versions <8.1... and ensure that dumps from an 8.1 server include the explicit table names... -- Andrew, Supernews http://www.supernews.com - individual and corporate NNTP services ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
