On November 8, 2019 1:53:13 PM GMT+01:00, Stephan Seitz
<stse+post...@fsing.rootsland.net> wrote:
>On Fr, Nov 08, 2019 at 01:44:53 +0100, Stephan Seitz wrote:
>>Has anyone an idea how I can debug this further? Or why it is only
>>sometimes working?
>
>I think I’ve found the problem. It happens if the subject has non-ASCII
>
>characters to the line looks like:
>
>=?UTF-8?Q?Aw=3A_Weinprobe_gut_=C3=BCberstanden=3F?=
>
>Has anyone an idea how can I change the regex so that it works with
>encoded subject lines? Or can I tell postfix to decode the line, then
>employ the regex und encode the line again?
For such encoded content you need to match the encoded form. See
http://www.postfix.org/BUILTIN_FILTER_README.html
I don't think that postfix is the right tool for this job, on the other hand I
don't really see the problem you're trying to solve.
Be careful not to break dkim signatures when modifying headers. Subject is a
signed header in most cases.
>Shade and sweet water!
>
> Stephan
--
Christian Kivalo
