Older versions of tar couldn't ungunzip (or bzip2-d). Thus the pipe.
Older version didn't have gunzip capability built-in not because it wasn't
possible, but because it fits UNIX's traditional notion of "smaller is
beautiful". By keeping the tarring capability and compression capability
separate, tar could still be useful even if new compression methods came
out. Indeed, while tar has been alive and well, at least three
compression algorithms have existed -- compress, gzip, bzip2 -- and
probably others I've never used.
People kinda got tired of remembering how to pipe and stuff so GNU's tar
has everything built-in.
I don't think anyone unzips first then pipes to tar; you usually untar
then pipe that to unzip.
-Mark
On Tue, 13 Mar 2001, Harry Souders wrote:
> On Thu, Mar 08, 2001 at 08:49:00AM -0800, Mytho X wrote:
> > Then bzcat kernel-2.2.18pre21 | tar xvf -
>
> I've been wondering this for some time.
>
> Why the piped tar?
>
> I often seen people un-gzipping and piping to tar. Or perhaps tar'ing
> and then piping to gzip. Why is this better than, say?
>
> [file you want to un-gzip and untar]
> tar -xzf myfile.tar.gz
>
>
---
Mark K. Kim
http://www.cbreak.org/mark/
PGP key available upon request.
