Re: leaky views, yet again - Mailing list pgsql-hackers

From Stephen Frost
Subject Re: leaky views, yet again
Date
Msg-id 20101005181520.GK26232@tamriel.snowman.net
Whole thread Raw
In response to Re: leaky views, yet again  (Greg Stark <gsstark@mit.edu>)
List pgsql-hackers
* Greg Stark (gsstark@mit.edu) wrote:
> Though I find it unlikely the sales people would have direct access to
> run arbitrary SQL -- let alone create custom functions.

I'm not really sure why..?  Perhaps not quite the same, but I've got
quite a few users who have direct SQL access (though they use ODBC on
their side, typically, there's nothing which forces them to) and I'd
certainly like to have the views that I've created which do row-level
filtering work correctly.  It's not to the point where I've started
using set-returning functions, but it's really not a situation I like
being in. :/

> If the users that have select access on the view don't have DDL access
> doesn't that make them leak-proof for those users?

Yeah..  I feel like we 'fixed' that whole problem with DO in any case..
Thanks,
    Stephen

pgsql-hackers by date:

Previous
From: Greg Stark
Date:
Subject: Re: leaky views, yet again
Next
From: Heikki Linnakangas
Date:
Subject: Re: leaky views, yet again