On 18/03/2008, Eddy Petrișor <eddy.petrisor@xxxxxxxxx> wrote: > On 18/03/2008, Eddy Petrișor <eddy.petrisor@xxxxxxxxx> wrote: > > On 18/03/2008, Ralf Corsepius <rc040203@xxxxxxxxxx> wrote: > > > > > > On Tue, 2008-03-18 at 16:33 +0200, Eddy Petrișor wrote: > > > > On 18/03/2008, Ralf Corsepius <rc040203@xxxxxxxxxx> wrote: > > > > > > > > > > > You will want to work on copies of your SOURCEX's instead of SOURCEX > > > > > directly. > > > > > > > > I will dig into this, since I really dislike having outside scripts doing this. > > > > > > > > > I'd assume you to be pretty close to a functional solution and to be > > > facing issues with details of your implementation. > > > > > > Actually, I am stuck because it seems that the alteration happening in > > the %prep phase either happens too late or happens on some temporary > > files which do not end up in the binary rpm. > > > > I suspect the later, since I see this in the build log: > > > > + . common.inf > > ++ VERS=2 > > ++ SUBVERS=1 > > ++ RELEASE=.i4.22 > > ++ echo 2.1..i4.22 > > ++ tr -s . > > + export THISVER=2.1.i4.22 > > + THISVER=2.1.i4.22 > > + echo 'This version (as sourced from the package) is: 2.1.i4.22' > > This version (as sourced from the package) is: 2.1.i4.22 > > + sed -i 's#^THISVER=.*#THISVER="2.1.i4.22"#' > > /usr/src/redhat/SOURCES/common.pre.sh > > + sed -i 's#^THISVER=.*#THISVER="2.1.i4.22"#' > > /usr/src/redhat/SOURCES/common.post.sh > > + exit 0 > > > Actually the replace is made on the real > /usr/src/redhat/SOURCES/common.post.sh, but it seems those files are > already part of the running spec when I am doing the replacement. > > I didn't saw this because I was doing builds in a chroot by feeding > the rpm inside that chroot a src.rpm. I decided to use some outside script. -- Regards, EddyP ============================================= "Imagination is more important than knowledge" A.Einstein -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
