It looks like surefire 2.4 fixes this issue. We were holding our tests to surefire 2.2 for some reason (presumably a bug in 2.3); I've removed that restriction, and the counts look much better now (at least on the 1.0.x branch; the trunk build is still running).
The removal of the 2.2 restriction means that running 'mvn compile -Dtest=false' will now report test failures, since no test called 'false' will be found. Instead, we can run 'mvn compile -DskipTests'. -Patrick On Jan 31, 2008 7:52 AM, Patrick Linskey <[EMAIL PROTECTED]> wrote: > There's a bug in surefire that mis-counts bugs somehow. > > -Patrick > > > On Jan 31, 2008 7:32 AM, David Ezzio <[EMAIL PROTECTED]> wrote: > > Actually, miscounted a few, but even so, the total that I now get (819) > > is quite different from what TeamCity reports. > > > > > > David Ezzio wrote: > > > How is that in TeamCity, it indicates that 39,575 tests were run while > > > locally I see 529 tests were run (for the test goal)? > > > > > > > > > -- > Patrick Linskey > 202 669 5907 > -- Patrick Linskey 202 669 5907
