On Fri, 2011-10-14 at 22:06 -0400, Phil Pennock wrote:
> On 2011-10-14 at 16:26 +0100, John Horne wrote:
> > Whilst the use of EXPAND_LISTMATCH_RHS is tempting, I feel that if we
> > enable it then we should make the effort to see if there is some
> > mechanism the developers could provide to make things easier for us.
>
> Yours looks like a reasonable case for EXPAND_LISTMATCH_RHS. You are
> using the conditions safely.
>
> Moving forward: how far can you get using a named list, which references
> the variables you're interested in?
>
As far as I can tell this would work. However, we use various match_*
calls with differing variables. As such each named list, of different
types, would probably be used only once (so no caching benefit). I see
no real advantage (as opposed to using an in-line configuration), in our
case, other than perhaps the config being a bit more readable.
>
> I've been toying with the idea that Exim needs named parameterised
> lookups. Notably, in 60 years of programming language development, this
> idea has been stumbled upon before and has been given a catchy name.
> "Functions".
>
Well that would certainly be useful :-)
However, I'm thinking that in our case we may be able to avoid using
EXPAND_LISTMATCH_RHS by using an 'acl' call. By using fixed variables
(either acl_m_ or acl_c_) so that the first contains the item to be
checked, and the second is the list, we can then push all the ugly
config bits into the acl. Possibly a third parameter to indicate if
wildcards are allowed. Again, I see no reason why this shouldn't work.
John.
--
John Horne Tel: +44 (0)1752 587287
Plymouth University, UK Fax: +44 (0)1752 587001
