Quoth Philip Hazel on Wed, Jan 12, 2000:
> On Wed, 12 Jan 2000, Ray Gardener wrote:
> > The result is that when mail is sent between the two systems the extra
> > "displayed name" info is not shown.
>
> Then your systems are broken.
True.
> (And then there are the pathological cases, such as the one
> Vadim quoted.)
It was not pathological enough. I believe that through careful
lawyer-like reading of RFC822 I would be able to construct a
nicer one, if I cared.
> It will only be useful if you can deduce "Ray Gardener" from
> <r.a.gardener@???> because its input is only the "real" address,
> not the rest of the RFC 822 address.
Our (or should I say "ex-our"?) system in cs.huji qualifies
"From:" lines with GECOS, using the "w" flag, based on the
address's local part (only if the domain matches, of course),
pulling the value of GECOS from a database which is rebuilt
nightly. Ephraim will post it on request. Alright, alright, I
can stop being evil now. As I took all the important stuff
(except my dear workstation Trilok) from cs.huji with me when I
left, I can post it.
> But then that is probably the only
> sensible approach to this problem, because deciding what to do with
>
> From: Philip "Q." Hazel (Exim author) <ph10@(this is in
Now we know your initial. The only thing left is to guess your
middle name.
> Cambridge)cus.cam.ac.uk> (there's only one "l" in "Philip")
Is "Philip" the only standard spelling, or are there
alternatives?
> is not a job for a regular expression.
True. Ex-our (cs.huji's) configuration just translates this kind
of addresses to those that look like:
Philip Hazel <ph10@???>
stripping all the comments etc. (i.e., full rewrite). I don't
know whether it's good or bad, but then, I'm not the config
file's priest.
Vadik, feeling silly today (haven't slept enough).
--
If you think C++ is not overly complicated, just what is a protected
abstract virtual base pure virtual private destructor, and when
was the last time you needed one?
-- Tom Cargill, C++ Journal, Fall 1990.
