Re: Expected behaviour of constructors (was Re: Motion P1788.1/M005 to avoid non-conforming interval standards)
>> Am 17. August 2016 09:47:20 MESZ, schrieb Walter Mascarenhas <walter.mascarenhas@xxxxxxxxx>:
>>> Dear all,
>>>
>>> I plan to finish an implementation of 1788.1 "soon", and it would be
>>> helpful to have an explicit list of test cases describing the expected
>>> behaviour of constructors with arguments of the form
>>>
>>> strings for for inf and sup +
>>> expected result (including exceptions raised)
>>> mathematical relation between inf and sup
> On 17.08.2016 13:20, Dmitry Nadezhin wrote:
>> Dear Walter,
>>
>> You can take much code from the reference implementation of 1788 written
>> by Marco Nehmeier. It is on github:
>> https://github.com/nehmeier/libieeep1788
On 18.08.2016 00:12, Oliver Heimlich wrote:
> Checking for raised exceptions / signals is a feature that is still
> missing in ITF1788.
Following our discussion, I have added support for IEEE Std 1788-2015
exceptions into the test framework [1]. Also, I have converted
additional tests (especially constructors and NaI tests) from
libieeep1788 into portable unit test format [2] such that you can easily
test any interval arithmetic library.
The implementation of b-textToInterval is wrong (outdated) in
libieeep1788, because it wrongly accepts decorated interval literals, e.
g., “[1.0, 1.0]_com”, see also issue #29 on Github. I have fixed the
unit tests in these cases.
Oliver
[1] https://github.com/oheim/ITF1788
[2] https://sourceforge.net/p/octave/interval/ci/default/tree/test/
(look at libieeep1788_class.itl for constructors)