> dynclasses and spawnw have been busted on win32 for some time now.
> with this patch, these annoying test failures will be skipped on
> win32.
Good, just commit it ;-)
> ~jerry
leo
Just my opinion on this: I'd rather see failures for things that need
to be fixed. Tests should only be skipped if they are not applicable
on the current system, or the test busts everything. Others may feel
different, of course.
Ron
[ skip dynclasses .t on Win32 ]
> Just my opinion on this: I'd rather see failures for things that need
> to be fixed. Tests should only be skipped if they are not applicable
> on the current system, or the test busts everything. Others may feel
> different, of course.
The old problem of skip, todo or let it fail.
Maybe we need pugs' test error classification scheme
> Ron
leo
> The old problem of skip, todo or let it fail.
>
> Maybe we need pugs' test error classification scheme
I'll see what I can do with Parrot::Test.
-- c
All failing TODO tests are successes in the eyes of the test harness and
do not appear in the short summary report. All succeeding TODO tests
are unexpected successes and raise a happy note in the short summary
report.
Adding TODO tests is a good way to write tests for things you know don't
work without making more failures than you want to record.
If there are no objections in a couple of days, I'll apply the patch
myself.
-- c
> I've just added a feature to Parrot::Test to allow you to add todo =>
> 'reason' to the end of all test function calls. This will mark the
> tests as TODO in a way that Test::Builder understands.
> If there are no objections in a couple of days, I'll apply the patch
> myself.
I just applied this patch. TODO away!
-- c