> pgadmin has some umm, interesting queries over pg_depends. It sounds
> like this change could complicate those. I doubt it's an
> insurmountable problem of course.
Yeah. But the only real point of the change is cleanliness, and if it's
injecting ugliness into clients then that argument loses a lot of its
force. I looked at pg_dump a bit and it definitely would get uglier
--- not really more complicated, but defaults would get handled a bit
differently from everything else. Seems like a fertile source of bugs.
So maybe we'd better forget the idea :-(
regards, tom lane
In response to
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2008-09-21 18:25:08|
|Subject: pg_settings.sourcefile patch is a security breach|
|Previous:||From: dpage||Date: 2008-09-21 18:09:18|
|Subject: Re: Proposal: move column defaults into pg_attribute along with attacl|