https://bugzilla.redhat.com/show_bug.cgi?id=1216170
--- Comment #1 from Roland Grunberg <rgrunber(a)redhat.com> ---
After speaking to Mikolaj, I realized my explanation from above was not
correct.
The resolution order is not 'Reactor, Local, Workspace (XMvn), Remote'. It is
Reactor, Workspace (XMvn), Local, Remote, and more specifically :
1) reactor
2) ./.m2 (local xmvn repo)
3) xmvn system repos
4) maven local repo (~/.m2/repository)
5) remote repos
So it can't have been the case that Tycho's usage of RepositorySystem was
causing the issue. In fact we use it in other places and it does the right
thing.
http://git.eclipse.org/c/tycho/org.eclipse.tycho.git/tree/tycho-core/src/...
After looking more closely I believe the issue is that the first mvn call
extracts the upstream tycho-bundles-external.zip in the reactor cache. The
second call (to xmvn) simply doesn't attempt to re-extract the system
tycho-bundles-external.zip. Now that we've changed tycho-bundles-external.zip
significantly, upstream, and Fedora Tycho can't be using the same instance.
It should be simple enough to just delete the directory and re-extract it every
time.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=kVNN62dGtn&a=cc_unsubscribe