> -----Original Message----- > From: news [mailto:[EMAIL PROTECTED] On Behalf Of Bill Barker > Sent: Saturday, 18 February 2006 2:03 PM > To: [email protected] > Subject: [VOTE] Resolving JUnit failures > > > <[EMAIL PROTECTED]> wrote in message > news:[EMAIL PROTECTED] > http://vmgump.apache.org/gump/public/junit/junit/gump_work/bui > ld_junit_junit.html > >CLASSPATH :/opt/jdk1.4/lib/tools.jar ... > > [javac] Compiling 139 source files to > /x1/gump/public/workspace/junit > > [javac] > > > /x1/gump/public/workspace/junit/junit/extensions/ActiveTestSui > te.java:20: > > ><identifier> expected > > [javac] public ActiveTestSuite(Class<? extends TestCase> > theClass) > >{ > > > > > >It seems that JUnit needs Java5, but got only a 1.4... > > Yup. It seems like our options are: > 1) Change vmgump to run with Java 5, and hope that the ~50% > coverage we got last time has been fixed by their respective > projects (it's been awhile, so who knows :). > 2) Change the 196 projects that <depend /> on JUnit to used > packaged-junit (ugh).
+1 > 3) Change JUnit to build from a branch (not as bad as 2, but > still not what Gump is about) > 4) Just hope that the JUnit project comes to their senses, > and restores JVM > 1.4 compatibility (not likely, given how extensive the change > seems to be). -1 Gump (and the projects it supports) needs to recognize that we have two distinct products (the 3.X line and the 4.X line). After all - the 4.X line is bringing a bunch of improvements to testing that would not be realistically achievable without JDK5. /Steve. > If it wasn't for the fact that most projects that adopt Java > 5 also seem to soon after move to exclusively requiring > Maven2, I'd vote 1) (e.g. mina). > If JUnit plans to do that as well, we might as well do 2) or > 3), since Gump support for Maven2 seems a long way off :(. > > And, actually, the inclusion of the [VOTE] is more in the > hopes of getting people to actually read this. So, if > anybody has any better ideas to resolve this, just add them > to the list, and they will be counted. Also, since this is > mostly metadata, I'm considering all votes from ASF > committers binding (although, as always, the Gump PMC could > choose to override this :). > And, needless to be said, I'm volunteering for any grunt-work > that needs to be done to implement it (please vote 1 or 4 :). > > >Jan > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
