Date: Fri, 24 Oct 1997 16:26:35 +0930 From: Mike Smith <mike@smith.net.au> To: Michael Schuster <Michael.Schuster@utimaco.co.at> Cc: "hackers@FreeBSD.ORG" <hackers@FreeBSD.ORG> Subject: Re: .zip vs. .tar.gz [was: zipfs filesystem anyone ? ] Message-ID: <199710240656.QAA01921@word.smith.net.au> In-Reply-To: Your message of "Fri, 24 Oct 1997 08:07:04 %2B0200." <34503B08.16D941F7@utimaco.co.at>
next in thread | previous in thread | raw e-mail | index | archive | help
> Mike Smith <mike@smith.net.au> wrote: > > >You are correct; gzipped tarfiles are organised the wrong way around > >(metadata inside the compressed envelope), while zipfiles keep the > >metadata outside. > > If you like, you can first compress all your files and then tar them, > doing something like (not tested, just FTTOMH (from the top of my > head:-)) > > find . -type f <other criteria> | xargs gzip > tar cf . This leaves me with a tree full of compressed files, and when I unpack the tarball I have to uncompress them. And if the tarball contains files that are meant to be compressed, they'll get uncompressed, which I don't want either. > The whole difference obviously comes from different approaches: Zip does > compression and archiving in one, whereas tar and gzip build on the > typical UNIX way of doing things: one tool for one purpose. That way, > your tools will be good, and you can upgrade one without having to touch > the other. If tar was smart, it would use the external compression tool to compress the data for each file as it read it, rather than compressing the output stream. You would still lose, as the tar format does not have a central directory. mike
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199710240656.QAA01921>