Re: [Exim] Interpreting Subject: lines - opinions, please

Top Page
Delete this message
Reply to this message
Author: Philip Hazel
Date:  
To: exim-users
Subject: Re: [Exim] Interpreting Subject: lines - opinions, please
On Mon, 7 Jul 2003, Oliver Eikemeier wrote:

> You would break part of my spam filter:
>
>   deny    condition      = ${if match{$h_subject:}{\N(?i)=\?big5\?[bq]\?[^\?]+\?=\N} {yes}{no}}
>           message        = Sorry, nobody speaks chinese here


Yes, those are the cases I worry about.

BUT: such configs are created by people who on the whole know what they
are doing. They will understand how to change to $rh_xxx: instead.
Indeed, you can make that change now, in advance.

End users are much more likely to have written something simple like

if $h_subject: contains "....."

in a filter, where the ..... happens to contain characters that need
encoding. They don't know about MIME encoding etc. Trying to educate
them to use (say) $mh_xxxx: instead is going to be one long, continuous
battle for ever more.

And the third point is that if we end up with all three of $h_ $rh_ and
$mh_ it is complicated to explain, and the existence of $h_ becomes
anomalous. What would it be useful for?

I don't yet have a firm view on this. That's why I'm collecting opinions
from the list.... Which choice is likely to cause the greater pain and
trouble in the long run?


--
Philip Hazel            University of Cambridge Computing Service,
ph10@???      Cambridge, England. Phone: +44 1223 334714.
Get the Exim 4 book:    http://www.uit.co.uk/exim-book