On Mon, Jun 06, 2005 at 12:03:54PM +0200, Jos Vos wrote: > On Mon, Jun 06, 2005 at 10:08:48AM +0200, Arjan van de Ven wrote: > > > I guess it comes down to with what you mean with "control". I'm not a > > lawyer, but control means to me "have impact on the result". The goal is > > that the binary building process is fully reproducable (see earlier > > parts of the license about the rationale). All the parts that impact the > > results are included in the src.rpm (with some global settings from > > redhat-rpm-config which is also shipped). The RH buildsystem actually > > calls rpmbuild to build the binary, all it does on top of that is some > > queueing. Queueing does not impact the outcome of the build in any way. > > Although I agree with most of your comments, two remarks: > > - There seem to be a few settings that are different from redhat-rpm-config: > see my thread started with > > https://www.redhat.com/archives/fedora-devel-list/2005-February/msg00523.html > > about rebuilding pango c.s. that was never really answered... OK, I can > imagine what settings it needs, and our local RHEL-rebuild environment > has implemented that ;-), but that's actually a bit of reverse engineering. > At least in FC3 (didn't check FC4t* yet) this was still the issue. rpmbuild --target i386 -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx http://www.redhat.com/mailman/listinfo/fedora-devel-list
