Bill Page wrote:
>
> On 19 January 2015 at 06:06, Ralf Hemmecke <[email protected]> wrote:
> >> Yes. 'xftest...' means failing test.
> >
> > OK.
> >
> > Waldek wrote:
> >> expected failures are known bugs, so no news.
> >
>
> I am not familiar with the details of particular test frameworks but
> it seems to me that one should be able to distinguish between "known
> bugs" and "expected failures". A test that is expected to fail might
> represent a proven limitation of an given algorithm, not a bug at all.
> In fact it would be considered a bug if such a test passed. Of course
> it might be possible in principle to convert such a test to a passing
> test with a suitable change of logic. If a particular test framework
> makes the assumption that this is always possible and only implements
> passing tests as successful, this should be clearly stated.
I think we should make terminology cleared: failed test means that
program behaves in incorrect (different than wanted) way. In
particular, if operation should signal error then passing test
means that error happended and failed test means that
no error was signalled ('testLibraryError' is doing this).
If operation is supposed to produce "failed", then passing test
means that we got "failed", etc. So failed test is different
thing than failed operation. Concerning bugs, failed test
may mean unimplemented feature. Normally I do not add such
test to the testsuite, but there are few tests of this kind.
--
Waldek Hebisch
[email protected]
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.
