On Jun 22, 2008, at 15:45 , Norbert Hartl wrote:
On Sun, 2008-06-22 at 12:42 +0200, Adrian Lienhard wrote:
I made 3 more tests pass:
- Fixed Monticello to pass MCSnapshotBrowserTest>>testButtonMutex (Monticello-al.317) - Fix FloatTest to not depend on platform for representation of NaN bit pattern (KernelTests-al.67) - Remove test testUnimplementedNonPrimitiveCalls (Tests-al.25)
Hmmm I'm not sure if it is a good idea to delete this test. It is a kind of test like ReleaseTest>>testUndeclared. It isn't critical but it helps moving to a clean target. I would opt to keep the test. At least I would turn it into a "expected failure" as long as we aren't far enough to state this to be totally clean.
I'm fine with making this test an expected failure if we want to make it pass in the future (in this case the test should exclude some cases where it is ok that there are no implementors, like for tests that compile those methods at runtime). Adrian