[Mimedefang] Using clamdscan --fdpass

[Steffen Kaiser]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, 24 Jul 2013, Daniel O'Connor wrote:

> On 23/07/2013, at 23:32, David F. Skoll <dfs at roaringpenguin.com> wrote:
>> On Tue, 23 Jul 2013 08:17:14 +0930
>> "Daniel O'Connor" <doconnor at gsoft.com.au> wrote:
>>
>>> I'm aware of that possibility, it doesn't seem very elegant to do so
>>> though.
>>
>> I guess elegance is in the eye of the beholder. :)
>
> As always :)
>
>> I will take a patch using --fdpass, but not as originally posted.  Instead,
>> we should make new virus-scanning routines entity_contains_virus_clamdscan
>> and message_contains_virus_clamdscan that are analogous to the *_clamav
>> versions but that invoke clamdscan --fdpass
>
>
> OK. fair enough, I will work on that.
>
> What about modifying the clamd routines to use IO::FDPass? That would 
> add another dependency to MD though, however I have some prototype code

This would not replace the original *_clamd functions, but the *_clamdscan 
functions could be enabled, if (and only if) IO::FDPass is present. So it 
is no real dependency.

In fact, you could just adopt entity_contains_virus_clamd().

> which uses it and File::Find to scan the files.

- -- 
Steffen Kaiser
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iQEVAwUBUe/xJZ8mjdm1m0FfAQKQVggAsKuMX18InZVI6KBAAbhLsWftjviDUAtG
RnZygTWL9kAczhVtXXthg6SDIhS+8dU37qmsszfU87Z05X66Xzk/O0Xc+pgl8zLt
itDyZxlLqokVvVh3X583SvYYpjNyXxhsjP3yLlE+61ToyXvvg3m0Wnr141SMd6mz
WUFWYCNU12/CDBaD1tXIwGcdJOvgDLESunDprbRuMj6W1EVO9JgM72pp8fvwLK7j
qXrARa2kJis28+ocN9K8KeQmQ0isOuxPU3xPd9V/pz9jbHwrX5GVU9VFhnLxrk15
GrJ+QE70miDpxZOVt00N6+NpmzPY74WOrII9VAmRk9hdtdfVIYgpRg==
=Imlg
-----END PGP SIGNATURE-----