On 02-Dec-98 at 09:27:07 Philip Hazel wrote:
> On Tue, 1 Dec 1998, Alan Thew wrote:
>
>> .username.@wanadoo.fr
>>
>> exim objects but sendmail (apparently) does not. exim 1.73 appears to
>> have allowed this... I'm not wishing/asking for exim to do this but
>> asking for opinion.
>
> This is allowed in the next release, as a result of somebody previously
> pointing out that sendmail violates the standard in this way.
>
Hmm. Now I'm a bit confused. I see that the testing version 2.054 changlog
has an entry stating that leading dots are (in that version) allowed.
However, (a) that violates RFC 821 (from my reading of it) with respect to
local parts (despite what else exim may/may not be allowing) - personal
opinion is that I'd rather exim sticks to the standards. Just because it is
perhaps not standard in other respects (allowing ".." in the local part by
the looks of it) doesn't mean we want to compound it further surely? And (b)
if it was allowed in version 1.73 but then not, I assume there was some
reason for the change - i.e. someone noticed/complained. It just seems to be
going back on what was in practice for some time (version 1.73 being
'current' some time ago). Perhaps someone will complain again that it *is*
allowed?
I'd also make the point that whilst allowing a leading dot still allows exim
to be a 'drop-in' replacement for sendmail, do we want to follow it to the
extent of non-standard practices? Should it diverge from sendmail in that it
is 'correct'?
John.
-----------------------------------------------------------------------------
John Horne E-mail: J.Horne@???
Academic and Information Services Phone : +44 (0) 1752 - 233914
University of Plymouth, UK Fax : +44 (0) 1752 - 233919
--
*** Exim information can be found at
http://www.exim.org/ ***
