Ian Abbott wrote:
On 27/11/2006 23:32, Chris Purves wrote:
I have a list of files that I have written to a file and I would like
clamscan to read the list from that file and scan only the files in
the list. Is there a good way to do this?
I have tried
cat filelist | xargs clamscan
This works, except that xarg can only pass about 200 filenames to
clamscan at a time. So for a filelist containing 1000 filenames
clamscan will be started 5 times, creating extra overhead.
You'll also find that it fails for file names containing whitespace,
quote marks, or backslashes. To work around that, you could use
tr '\n' '\0' < filelist | xargs -0 clamscan
or generate the filelist with null-separated filenames in the first
place (perhaps using find's -print0 command) to avoid the tr. This
assumes you are using the GNU versions of find and xargs. The -0 option
of GNU xargs causes it to use null-separated file names as-is and
shell-quote them properly.
Yes, I was using find -print0 | xargs -0 combination when creating my
filelist. I didn't know about the 'tr' command, though. You learn
something new every day.
What I am actually trying to do is have clamscan only scan files that
are new or have not changed since the last scan. I have gotten as far
as creating a filelist containing a list of files that are new or
where the md5sum has changed. The problem I have now is how to get
that information to clamscan efficiently.
It sounds like the ability to read the list of files to scan from a file
(or from standard input) would be a useful feature to add to clamscan
and clamdscan.
Indeed.
--
Chris
_______________________________________________
Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net
http://lurker.clamav.net/list/clamav-users.html
