Given the admnistrative requirement for mail to be included in mailLocalAddress,
compliant systems shouldn't be able to tell the difference, should they?
If you're looking for coding advice for your server routing code, I'd wager your first
alternative will run faster and permit more flexibility for special-case systems or
directory entries that don't care to conform to the standard. (And I'm sure there will
be special circumstances where this non-confirmance is appropriate.)
If I understand all this correctly, I don't see a need for the standard to require that
routing consult "mail" given that it already requires consulting "mailLocalAddress"
which will normally include the "mail" address.
Mark
Gregory Neil Shapiro wrote:
> : We need to decide once and for all:
> :
> : Should we use the 'mail' attribute for routing or leave it as a whitepages entry?
>
> mark> As opposed to what?
> ...
> mark> So what are we debating? Whether "mail" is one of the
> mark> "mailLocalAddress", as required in 4.2?
>
> As currently written, the LDAP search is:
>
> (&(objectClass=inetLocalMailRecipient)(mailLocalAddress=%0))
>
> If we include the 'mail' attribute, it would become:
>
> (&(objectClass=inetLocalMailRecipient)(|(mailLocalAddress=%0)(mail=%0))
>
> where %0 is the key.
This archive was generated by hypermail 2.1.2 : Thu Oct 24 2002 - 10:31:46 PDT