Date: Sun, 08 Feb 2015 12:57:20 -0500 From: Lowell Gilbert <freebsd-questions-local@be-well.ilk.org> To: Manish Jain <bourne.identity@hotmail.com> Cc: freebsd-questions@freebsd.org Subject: Re: why does tar archive directory differently based on command line? Message-ID: <44bnl4qoan.fsf@lowell-desk.lan> In-Reply-To: <BLU436-SMTP2585638FB0568F935D69E10F6260@phx.gbl> (Manish Jain's message of "Sun, 8 Feb 2015 13:23:12 %2B0530") References: <BLU436-SMTP2585638FB0568F935D69E10F6260@phx.gbl>
next in thread | previous in thread | raw e-mail | index | archive | help
Manish Jain <bourne.identity@hotmail.com> writes: > I am having problems trying to understand why does tar archive > differently with the following 2 commands, presuming xyz is a > sub-directory : > > tar -c -f - ./xyz | gzip > xyz.tar.gz//xyz nested one level under . and .. > tar -c -f - xyz | gzip > xyz.tar.gz//no . or .. > > With the first invocation, I get a top-level nesting that just has . and .. > With the second, I get a top-level nesting that has xyz, which is much > preferable Right. tar(1) stores the path to the file. This is essential functionality, so if you want a particular path stored, you provide that path in the way you invoke tar in the first place. > The first invocation leads to an absurdity that you have to cd into > . (sounds recursively impossible) to actually get to xyz It's neither absurd nor recursively impossible (whatever that means). If you were planning to extract the tar file from one level higher up the destination tree, it would be exactly what you would want. Perhaps you would be happier with a GUI application for creating your tar files. There are many fine choices, and you probably have one (or more) already installed on your system.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44bnl4qoan.fsf>