Re: Why grantor is owner in this case? - Mailing list pgsql-general

From Tom Lane
Subject Re: Why grantor is owner in this case?
Date
Msg-id 4759.1261758025@sss.pgh.pa.us
Whole thread Raw
In response to Why grantor is owner in this case?  ("donniehan" <donniehan@126.com>)
Responses Re: Why grantor is owner in this case?  ("donniehan" <donniehan@126.com>)
Re: Why grantor is owner in this case?  ("donniehan" <donniehan@126.com>)
List pgsql-general
"donniehan" <donniehan@126.com> writes:
> I have a question about the grantor. Why the grantor is owner in the following case ?  I think it should be
postgres(dba). 

Grants done by a superuser on an object he doesn't own are treated as
being done by the object owner instead.  Otherwise you end up with
grants that don't have a clear chain of traceability to the owner,
which causes all sorts of un-fun issues for REVOKE.  (I'm too lazy
to come up with the details right now, but if you care you can look
back in the pgsql-hackers archives to find the discussions where this
behavior was agreed on.)

            regards, tom lane

pgsql-general by date:

Previous
From: Tom Lane
Date:
Subject: Re: date_trunc on date is immutable?
Next
From: Tom Lane
Date:
Subject: Re: Out of space making backup