On 21 Oct 2011, "David H. Lipman" <DLipman~nospam~@Verizon.Net>
wrote in comp.mail.pegasus-mail.ms-windows:
> There must be a configuration setting that needs to be set/changed
> in my setup to alter the way the rules are inititated.
I only tried local filters, but I can't make them work. Here's what I
did:
I don't have any Chinese spam messages with encoded subjects in my spam
folder, so I made my own sample. I found a spam usenet post in the
Google archive, that had a subject in Hindi or something. When I
forwarded it to my email, Google encoded it for me like this:
Subject: =?windows-1256?B?RndkOiDj5t7aIOHdys0gx+Hj5sfe2iDH4ePNzObI?=
=?windows-1256?B?5SDmx+Hj4+Tm2uUg5sfh49Td0eUgKOHHIMrTys7P4+Ug?=
=?windows-1256?B?3ewg49rV7ckgx+Hh5Sk=?=
The subject appears in Pegasus as "Fwd: <a bunch of Hindi characters>"
Some of my Expression criteria were:
Subject:*
Subject: *
Subject: Fwd*
Subject: *<a substring of the Hindi characters copied and pasted
directly from the message into the Regular Expression
criteria box.>
Subject: =*
Subject:*windows-1256*
Subject: [=]*
Subject:?[=]*
Subject:*[=]*
Only the first four triggered my filter. This seems to tell me that
Pegasus decodes the subject before evaluating the rule. There seems to
be no way to make Pegasus look at the encoding, only the results. It
doesn't see the character set designation.