--exclude-from works but "exclude from" in rsyncd.conf doesn't ?

[jw schultz]

On Sat, May 10, 2003 at 01:34:30PM -0700, Wayne Davison wrote:
> On Sat, May 10, 2003 at 04:17:11AM -0700, jw schultz wrote:
> > Just to check, the absolute-path stuff doesn't allow path components
> > outside of the module to be reported in any diagnostic/verbose
> > messages, yes?
> 
> That appears to be true since the server doesn't seem to mention any
> of the config-specified items getting added to the exclude list nor any
> matches that happen.  I haven't checked this is detail yet, though.
> 
> >         if (server_exclude_list && check_exclude(fname, server_exclude_list, st)) {
> >                 return 1;
> >         }
> 
> That's the basic idea, yeah.  I chose to go a step farther and clean up
> the exclude code to make the calls more consistent, and also to make it
> possible to easily add files/lines to the new server_exclude_list (the
> old code played favorites with the default exclude_list, but the new
> code does not).  The patch is getting to be pretty big now, but I think
> the code is quite a bit cleaner.  I've tested it a fair bit (including a
> chroot-test this time), and it seems to be working well.  Once again,
> the patch name is the same:
> 
>     http://www.blorf.net/rsync-daemon-exclude.patch

I haven't applied it yet but the impression i get from it is
very positive.  It looks like the functions of the exclude
interface are being made more generalised and that is good.

> 
> Still no attempt to limit the files that the user can upload, though.  I
> won't try to fiddle with that until I get this patch incorporated (after
> a bit more testing).  If anyone tries this out, please let me know how
> it works for you.

A thought came to me as i looked at this:  If a user does
--delete the server_exclude_list should make those files
appear as though they don't exist even if --delete-excluded
hasn't been set.  As your patch stands now delete_excluded
would cause the server_exclude_list to be ignored.  This
should just require a bit of shuffling in
check_exclude_file().

Seems to me that the f argument to check_exclude_file() could
be combined with delete_excluded by the caller saving an
extern and further generalising this function.



-- 
________________________________________________________________
	J.W. Schultz            Pegasystems Technologies
	email address:		jw at pegasys.ws

		Remember Cernan and Schmitt