>
> On Tue, 24 Feb 1998, Jan Wieck wrote:
>
> > >
> > > On Tue, 24 Feb 1998, Bruce Momjian wrote:
> > >
> > > > > Actually, I'm not married to db_* for views...it was a "quick fix"
> > > > > to ensure that things still worked. Whatever we decide on, both Julie and
> > > > > Peter, at a minimum, need to know relatively soon. I know in Julie's
> > > > > case, she does do a call to pg_user...I let her know tonight that she
> > > > > needs to change it to db_user, for the *current* code...
> > > >
> > > > Good. I didn't want the db_ namespace pollution. I will call it
> > > > pg_user_no_passwd, and make it a view, not a rule. Is that OK with
> > > > everyone?
> > >
> > > Works for me...
> >
> > How? When I create a view the way Bruce explained (update pg_class),
> > my backend crashes on SELECT FROM view during the rewrite. For some
> > reason the rewrite handler cannot get the rule locks correctly.
>
> Ppl are taking me a slight bit too literally :( "Works for
> me"...I like the idea...not necessarily implemented it though :)
I already have the pg_shadow + pg_user-view diff ready. Works
really! Must run a regression test and send it after that
succeeded. It updates createuser, destroyuser and initdb too.
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#======================================== jwieck@debis.com (Jan Wieck) #