On Tuesday 20 May 2014 15:56:16 Richard W.M. Jones wrote: > On Tue, May 20, 2014 at 03:33:31PM +0200, Pino Toscano wrote: > > Resolve the given path within the chroot, so scrub can be invoked > > outside the chroot on an already-resolved path. > > Given that realpath is used, its availability is checked manually, > > since scrub-file already depends on the "scrub" feature. Slightly > > ugly, but on the other hand realpath is generally available > > nowadays, so the check should not be failing. > > > > Add few tests in scrub-file for this and other similar issues. > > The realpath test was added (commit a86eb0e0d2c67e2) at a time when we > thought it would be a good idea to have the daemon run on Windows. > Since no one thinks that's a good idea any longer, I think you could > prepend this commit with one which removes tests for realpath / > HAVE_REALPATH and hard-codes optgroup_realpath_available() == 1.
I didn't know the story behind this, and went for the conservative route. Interesting enough, realpath has been unconditionally used already, so I will pull the plug to the conditional usage in some other parts, then. > Is it possible to make this change more generic so in future it could > be applied to other functions? > > - Would it make sense to have a 'sysroot_realpath' library function to > replace 'sysroot_path' in certain functions? Good idea. Do you have also an idea which daemon functions, other than scrub-files and realpath itself, might need such handling? -- Pino Toscano _______________________________________________ Libguestfs mailing list [email protected] https://www.redhat.com/mailman/listinfo/libguestfs
