Please do not reply directly to this email. All additional comments should be made in the comments box of this bug report.
Summary: Merge Review: busybox
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=225631
varekova@redhat.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |MODIFIED AssignedTo|nobody@fedoraproject.org |varekova@redhat.com Flag| |fedora-review?
------- Additional Comments From varekova@redhat.com 2007-02-19 10:32 EST ------- Thanks for your comments. The fixed version is busybox-1.2.2-6.fc7. (In reply to comment #1)
- instead of mv the files to reverse the patch, I suggest
patch -R -p1 < %{PATCH0}
changed
- Is DOLFS really used? I can't find it in the sources
removed
- the man page timestamp should be kept with -p
fixed
- buildroot is not the preferred one
fixed
- At least the selinux patch should be proposed upstream. Has it been done?
I'm investigating it.
- the .static patch and the .anaconda are unreadable, although they bring in important changes. I think there should be a comment explaining verbally what is done
- the whole process should also be commented since it is not trivial. For example something along (maybe dispatched where things are done):
# in %prep the .static patch is applied, to have a static busybox # built. The executable is kept as busybox-static. # then the .static patch is reverted and the .anaconda patch is # applied to generate the busybox especially tailored for anaconda.
changed
Suggestion:
/ between $RPM_BUILD_ROOT/%{_mandir} is not useful
use %defattr(-,root,root,-) instead of %defattr(-,root,root)
%patch8 -b .gcc111 -p1 should certainly be %patch8 -b .gcc41 -p1