Re: [exim] Exim misses some attachments.

Top Page
Delete this message
Reply to this message
Author: Jeremy Harris
Date:  
To: exim-users, exim-dev@exim.org
Subject: Re: [exim] Exim misses some attachments.
On 17/12/15 15:40, Mike Brudenell wrote:
> Interesting…
>
> On 17 December 2015 at 09:45, Jeremy Harris <jgh@???> wrote:
>
>> On 17/12/15 09:34, kuncho pencho wrote:
>>> And is demime is deprecated? My exim version is 4.68.
>>
>> Yes, it was deprecated even back then:
>>
>> http://exim.org/exim-html-4.68/doc/html/spec_html/ch-content_scanning_at_acl_time.html
>>
>
> Looking at the *Exim Specification* for 4.86 (sections and page numbers
> from the PDF file):
>
>    - The descriptions of the *$demime_errorlevel*, *$demime_reason and
>    $found_extension* variables on page 124–125 mention the *demime* condition
>    as being obsolete;
>    - But the description of the *demime* ACL condition itself (section
>    42.25, page 348) doesn't!
>    - Chapter 43 *Content scanning at ACL time* mentions *demime* as being
>    deprecated but in the context of building from source (which I guess quite
>    a few people don't do nowadays?) and in quite a buried way;
>    - But then chapter 43 goes on to give many examples of using *demime*,
>    which seems sort of odd; especially as on page 372 it goes so far as to say
>    "If your virus scanner cannot unpack MIME and TNEF containers itself, you
>    should use the demime condition (see section 43.6) before the malware
>    condition."

>
> So to be fair, I can understand why it might be easy to miss that *demime* is
> deprecated. :-)


Agreed on the documentation clarity.

Should we go further for the upcoming 4.87 and remove support for
demime? The makefile WITH_OLD_DEMIME has been present commented *out*
since December 2004. Any builders including it are doing it
deliberately.... but leaving it dangling for over ten years seems a bit
foolish.

[copy exim-dev]
--
Cheers,
Jeremy