On Tue, 2 Feb 2010, John Hodrien wrote:
Only I was wrong. That's fixed that part of the problem. Once the packages
are in, they've got the wrong provides. To the extent that a package doesn't
provide itself, if you see what I mean.
I had a fish in /usr/share/rhn/server/importLib/headerSource.py but I'm not
exactly sure what I'm looking for.
I can see it go into _populateDependencyInformation and it seems to run
through the provides with empty name/version/flags.
I've tried the odd horrible hack with only partial joy. Help!
Here's a dirty horrible wrong hack that's just me making up code just for
getting these packages in and working:
if tag == 'provides' and f == 'name' and ic == 0:
v = sanitizeList(self['name'])
ic = len(v)
Bleurgh! My initial guess is that this won't cause too much painp
Example package that can't be pushed for me with 0.7:
http://www.comp.leeds.ac.uk/johnh/ooobasis3.1-core01-3.1.1-9420.x86_64.rpm
Even after it's pushed, it's not workable as the signing the package with
rpmsign doesn't produce a package that can be verified with -K.
Are these just crud packages?
I'll bugzilla, but I thought I'd share the grimness on the list.
jh
_______________________________________________
Spacewalk-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/spacewalk-list
