Gilles Scokart wrote:
I already thinked to something like that.   And to be complete, the test should 
also do a resolve using maven libraries
and compare that resulting jars.


well, I think they use different algorithms. However, it makes a good regression test too.

However, that would make really a lot of download  :-)


yes and no.

1. You only need to pull the poms down, not the artifacts
2. if you have a local cache (we do, but it doesn't update very often)
3. the maven results could be stored somewhere in one big file




--
Steve Loughran                  http://www.1060.org/blogxter/publish/5
Author: Ant in Action           http://antbook.org/

Reply via email to