Mail::ClamAV 0.20 patched to build with ClamAV 0.92

[Denis Beauchemin]

Denis Beauchemin a écrit :
> Julian Field a écrit :
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> I have produced a patched version of this Perl module which should 
>> successfully build with ClamAV 0.92 installed.
>>
>> So to upgrade a test system, download the newest ClamAV+SpamAssassin 
>> package (including ClamAV 0.92) from
>> http://www.mailscanner.info/files/4/install-Clam-0.92-SA-3.2.3.tar.gz
>> Unpack and install it as normal. The installation of Mail::ClamAV 
>> will fail, that is to be expected.
>>
>> Now download
>> http://www.mailscanner.info/files/4/Mail-ClamAV-0.20.JKF.tar.gz
>>
>> Unpack and install it with
>> tar xzf Mail-ClamAV-0.20.JKF.tar.gz
>> cd Mail-ClamAV-0.20
>> perl Makefile.PL
>> make
>> make test
>> make install
>>
>> If you're in luck today, that will successfully build, test and install.
>>
>> Now you should be in business, with a new ClamAV and a working 
>> "clamavmodule" virus scanner.
>>
>> Please tell me how you get on with this.
>> All my mods are in the "JKF" subdirectory of the Mail-ClamAV-0.20 
>> directory.
>>
>> Jules
>>
>>
>>   
> Julian,
>
> I noticed your easy Clam+SA install script didn't ask me if I wanted 
> to install SA and it just did...  I don't really mind because that was 
> on a test system, but is there still a DNS performance problem with SA 
> 3.2.3?
>
> I looked at patch 
> http://issues.apache.org/SpamAssassin/show_bug.cgi?id=5589 but 
> couldn't figure out if it was still needed.  If so, I guess it has to 
> be applied to the installation dir and then run the install stuff 
> (perl Makefile.PL; make; make test; make install)?
>
> Thanks!
>
> Denis
>
OK, I experimented a bit with SA 3.2.3 and patch 5589.  Indeed, it has 
to be applied to the initial directory (patch -p0 
<sa-async-5-5589.patch) and then perl Makefile.PL, make, make test and 
make install.

I did not dare installing it because I got many errors in the make test run:
Failed Test           Stat Wstat Total Fail  Failed  List of Failed
-------------------------------------------------------------------------------
t/spamc_optC.t                       9    4  44.44%  2 4 6 8
t/spamc_optL.t                      16   16 100.00%  1-16
t/spamd_protocol_10.t               10    2  20.00%  9-10
34 tests skipped.
Failed 3/141 test scripts, 97.87% okay. 22/1974 subtests failed, 98.89% 
okay.
make: *** [test_dynamic] Error 255

I tried "./spamassassin --lint -D < spam.msg"  and it seemed to work 
fine but I will have to investigate some more before going into 
production.  After all, I am almost on Christmas leave...

May all of you have a nice Christmas holiday (if it applies).  Contrary 
to last year's Christmas, this one will be a nice white Christmas under 
loads of snow!

Denis

-- 
Denis Beauchemin, analyste
Université de Sherbrooke, S.T.I.
T: 819.821.8000x62252 F: 819.821.8045