> Questions > --------- > Question the first: do you guys think it's reasonable to use JUnit for this? > Or is there a better, perhaps more standarized osgi approach that I'm just > not aware of? For instance, I noticed a post by Peter Kriens[2], but I > couldn't actually find this project and it seems a bit old.
Radman is using spring-osgi-test for testing. Features that I find interesting: * I can start test case using either equinox, knopflerfish or felix. * Integration with IDE is useful. I.e. I don't have to run mvn install on dependent projects as it search maven project relative to the running test case before resolving to ~/.m2/repository. (But more to this later) Features that I find it less interesting: * By default, it started up spring related bundles. * Search bundles by relative path to the test case project is inperfect. It only checks for artifactId, instead of groupId and artifactId. This is a known bug, still waiting for spring-osgi dev for confirmation when will this be fixed. * Resolve to maven default cache location [~/.m2/repository] is hardcoded. This is a not so good idea when for example, in windows your repository is on D drive. * Minor issue with IDEA user, where the working path to run the test case should be the project that has the test case path, instead of the location of the .ipr file. > Question the second: given the state of the KF Junit project that I > described above, do you think it's reasonable to import it here and build on > it? May be should try to do spring-osgi-test instead. > Question the third: anybody else interested in this? Or should I just hoard > this privately to myself? I'm very interested with this. I want to be involved with this. Regards, Edward Yakop _______________________________________________ general mailing list [email protected] http://lists.ops4j.org/mailman/listinfo/general
