Re: Converting shell script to RPM

[Tim Mooney <mooney dogbert cc ndsu NoDak edu>]

In regard to: Re: Converting shell script to RPM, Timur Tabi said (at...:

>Now, I'm even more confused.  Why would I install software during the
>process when I build the RPM file?  After all, the RPM file is used to
>install the software.  If I install the software while building the RPM,
>then what's left to do?  The software is already installed!

The %install stuff in an RPM spec file really means:

	"lay out the files, as I want them to appear, in preparation for
	packaging"

When %install is run, it's run once, when the package is built.  You're
installing the files into your packaging "area".  That area can either be
a temporary area (if you're using a BuildRoot, which I highly recommend)
or it can be into the real directories (e.g. /usr/local/bin, etc), but
in either case you're laying out the files in preparation for rpm to take
over and package them up into an RPM.

Once the package is constructed, the RPM package file is moved to however
many machines you want, and running

	rpm -ivh mypackage

or

	rpm -Uvh mypackage

results in those files being installed on the machine.  The %install isn't
run now, it was run once at build time.  After that, it's not used again.

Tim
-- 
Tim Mooney                              mooney@dogbert.cc.ndsu.NoDak.edu
Information Technology Services         (701) 231-1076 (Voice)
Room 242-J6, IACC Building              (701) 231-8541 (Fax)
North Dakota State University, Fargo, ND 58105-5164