Re: Per-Database Roles - Mailing list pgsql-hackers

From Stephen Frost
Subject Re: Per-Database Roles
Date
Msg-id -1705304821428018513@unknownmsgid
Whole thread Raw
In response to Re: Per-Database Roles  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On May 22, 2012, at 12:18, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Another objection is that it wouldn't scale up nicely to multiple levels
> of catalog hierarchy.  But maybe local/global is enough.

That would be a huge improvement and this wouldn't get in the way of
any solution to the global oid conflict issue coming along in the
future..

We still have the issue of name conflict between the global roles and
the local roles, right?  Unless we allow those to overlap and pick
whatever is "closest"? Then there is the question of role membership
and if we would allow that to go between local and global or what the
semantics of that would look like..

Thanks,

Stephen

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Readme of Buffer Management seems to have wrong sentence
Next
From: Peter Eisentraut
Date:
Subject: should postgres -C print units?