Re: [Bug-tar] CHOWN_MODIFIES_SYMLINK causes unexpected results

[Paul Eggert]

"Howell, Daniel A" <address@hidden> writes:

> these days the
> likelihood is that chown does follow symlinks, so maybe the default
> should be changed?

Yes, that's right.  That change was made on Jan 18 in gnulib, too late
to get into your tar version, but it'll be in the next tar release:

2005-01-18  Jim Meyering  <address@hidden>

        * m4/chown.m4 (gl_FUNC_CHOWN): When cross-compiling, assume that chown
        is sane and DOES follow symlinks.  Besides, testing 20 different
        systems found no broken chown implementations.
        Prompted by a change in rsync's copy of this macro.