Robin Lee wrote:
> I am using the planex[1] tool to do massive rpm rebuild.
> It use something like 'rpm -q --specfile' to predict the result
> rpms, and generate a dependency tree. And then use this info to
> build rpms in el7 mock one by one.
>
> I then use the ''%_debugsource_template %{nil}' trick, which can
> satisfy planex by now.
>
> [1] https://github.com/xenserver/planex
Cool, that looks handy. Is planex intended to support
generating srpms on newer systems like Fedora and building
on RHEL/CentOS? That seems like a recipe for other odd
failures, particularly as rpm development has picked up a
bit recently. We're likely to see more changes with each
Fedora release (or at least far more often than they end up
in RHEL/CentOS). :)
The mock --buildsrpm might be better for this than using
rpmbuild directly on the host. Coupled with mock's
--resultdir argument it doesn't look like it would be all
tht hard to replace rpmbuild in planex/cmd/makesrpm.py with
a mock equivalent. Maybe that would even be done
optionally, if the target OS differs from the host OS. That
would help planex guard against the next incompatible
change.
It is still odd that there's no way to override the
_debugsource_packages macro once it's set. I'd say it's
worth asking the rpm folks if that's intentional or not. It
may be a bug that they'll fix.
--
Todd
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Man was made at the end of the week's work when God was tired.
-- Mark Twain
Attachment:
signature.asc
Description: PGP signature
_______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx
