> I’m new to rpm’s and spent most of yesterday at work coming up with > a simple binary rpm file. My prep section unpacks the tar file. > Which contains 2 binary executables, a config file, and a shared > object. The build section does nothing. My install section creates > a directory in the opt/ folder and copies over the 4 files. > > My question is: I’ve noticed that in almost every example I see on > the net for binary rpms, people are calling ./configure ./make and > ./make install. Doesn’t that seem overly complicated? Why not just > a couple copy commands? All we are trying to do with a binary rpm > is place some executable files in the person’s file system, right? >From what you've said you seem to be placing pre-compiled binaries, at which point your method is fine. However, if you have access to the source, building it in the RPM is considered ideal and is where the configure && make && make install steps originate from. -greg_______________________________________________Rpm-list mailing listRpm-list@xxxxxxxxxxxxxxxxx://lists.rpm.org/mailman/listinfo/rpm-list
