Re: Row security violation error is misleading

On 21 April 2015 at 22:21, Dean Rasheed <dean(dot)a(dot)rasheed(at)gmail(dot)com> wrote:
> On 21 April 2015 at 20:50, Stephen Frost <sfrost(at)snowman(dot)net> wrote:
>> Thanks a lot for this. Please take a look at the attached.
>
> I've given this a quick read-through, and it looks good to me. The
> interaction of permissive and restrictive policies from hooks matches
> my expections, and it's a definite improvement having tests for RLS
> hooks.
>
> The only thing I spotted was that the file comment for
> test_rls_hooks.c needs updating.
>

So re-reading this, I spotted what looks like another (pre-existing)
bug. In process_policies() there's a loop over all the policies,
collecting quals and with_check_quals, then a test at the end to use
the USING quals for the WITH CHECK quals if there were no
with_check_quals. I think we want to instead do that test inside the
loop -- i.e., for each policy, if there is no with_check_qual *for
that policy*, use it's USING qual instead.

Regards,
Dean