On Mon, Jul 31, 2006 at 04:59:08PM +0100, Tony Finch wrote:
> On Mon, 31 Jul 2006, Dean Brooks wrote:
> >
> > TF/03 Note in the spec that $authenticated_id is not set for local
> > submissions from trusted users.
> >
> > Thing is, it doesn't appear to function as described above, and
> > instead functions as the updated spec describes. That is, it seems to
> > set authenticated_id even for trusted users (properly IMO) unless it
> > is overridden using -oMai.
>
> The change in the spec is consistent with the change log. I believe it is
> correct. Which bit of the spec are you looking at, and what makes you
> think Exim is setting $authenticated_id for trusted users?
Hmm, in the release version of 4.63, I find this in spec.txt in the
description of $authenticated_id:
When a message is submitted locally (that is, not over a TCP connection)
the value of $authenticated_id is normally the login name of the calling
process. However, a trusted user can override this by means of the -oMai
command line option.
I also compiled and tested 4.63 and $authenticated_id is being set correctly
(again, properly IMO) for trusted users when no command-line options are set.
Why would you want the default for authenticated_id for trusted users
to be blank anyway? This causes confusion since you cannot ever
depend on a consistent default in the absence of command-line
overrides for locally submitted messages.
--
Dean Brooks
dean@???