Right, that was what I thought. Thanks for the clarification. I got
seriously confused by the debacle around the ECF contribution because I
couldn't figure out how that could have worked before. As it turned
out, it did still almost work (since the platform uses requirements
that are open ended) but then failed (since the p2.sdk feature still
has fixed includes). The Galileo Builder was never to blame, the new
p2.sdk feature was. It would have failed with the old builder too.
- thomas
Pascal Rapicault wrote:
The install of F2 will fail because the user has F1 installed and
F1 and F2 both have a strict requirement on a different version of a
singleton bundle.
 Thomas
Hallgren ---05/07/2009 12:25:22 PM---What will happen given the
following scenario:
What will happen given the following scenario:
Feature F1 includes bundle B using the explicit version range
[1.0.0.a,1.0.0.a] This bundle is a singleton.
Feature F1 is installed into my profile.
Feature F2 also includes bundle B, but using the explicit version range
[1.0.0.b,1.0.0.b]. This feature is not installed.
Now I try to install F2. Will that install succeed and actually break
F1? Or will it fail since F1 is locking B at a specific version?
The background to this question is the current problems with ECF and
the
Galileo builder. https://bugs.eclipse.org/bugs/show_bug.cgi?id=275327
- thomas
_______________________________________________
p2-dev mailing list
p2-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/p2-dev
_______________________________________________
p2-dev mailing list
p2-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/p2-dev
|
