On Tue, 2002-05-14 at 01:42, Tom Lane wrote: > [EMAIL PROTECTED] (Neil Conway) writes: > > I'd prefer this method -- IMHO the readibility of dump scripts isn't > > a top priority (or if it is, we're not doing very well in that regard > > any). I think dump scripts should be as verbose as is necessary to > > ensure that they can't be misinterpreted.
I agree with Neil on this. > Perhaps instead of "readability" I should have said "editability". > The thought that is lurking behind this is that you might want to > retarget a dump script to be reloaded in some other schema. If the > dump is cluttered with umpteen thousand copies of the schema name > that's going to be difficult. sed -e 's/ old_schema\./ new_schema./g' I don't think you should allow the dump to be ambiguous for the sake of making rarely used actions slightly more convenient. > Ideally I'd like the dumped object definitions to contain *no* explicit > references to their containing schema. This would allow, for example, > a pg_restore mode that loads the objects into a different schema. Provide a command line option to pg_restore to do an automatic edit of the schema name (-E old_schema,new_schema). People using "psql <dump" would have to edit the dump. -- Oliver Elphick [EMAIL PROTECTED] Isle of Wight http://www.lfix.co.uk/oliver GPG: 1024D/3E1D0C1C: CA12 09E0 E8D5 8870 5839 932A 614D 4C34 3E1D 0C1C "Yea, though I walk through the valley of the shadow of death, I will fear no evil, for thou art with me; thy rod and thy staff they comfort me." Psalms 23:4
signature.asc
Description: This is a digitally signed message part
