[Mailman-Users] Still having trouble with an email filter vis-a-vis Chinese crap.

[Mark Sapiro]

On 05/05/2018 02:05 PM, Kenneth G. Gordon wrote:
> On 5 May 2018 at 13:20, Mark Sapiro wrote:
> 
>>> Subject =?utf-8?B?
>>
>> Do you really mean it doesn't contain the ':'?
>  
> I am reasonably sure that SOME do not, possibly those few which are getting through, but 
> since you brought that up (although I DID suspect as much), I have not been able to check 
> on the accuracy of my memory since so far this morning, I have had no more of them show 
> up.
> 
>> You could try
>>
>> ^Subject:?\s*=\?utf-8\?B\?
>>
>> which would match Subject followed by a colon or not and any amount of
>> white space.


However, if the colon is missing, the

Subject =?utf-8?B?...

line is not a header and it can't be matched by header_filter_rules.

I.e., a message like

   To: someone
   From: someone
   Message-ID: <aaa at b.c>
   Subject hi there
   Date: some date

   body

will be parsed as headers

   To: someone
   From: someone
   Message-ID: <aaa at b.c>

and body

   Subject hi there
   Date: some date

   body

so there's no way you can match that 'Subject' with header_filter_rules
because it isn't a header. If that's really what you're looking at, you
would have to implement a custom handler to match that. See
<https://wiki.list.org/x/4030615>.

On the other hand, if some MUA actually displays the decoded line as the
message subject, it almost certainly does have the colon.

-- 
Mark Sapiro <mark at msapiro.net>        The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan