[Zope-dev] Python 2.7 and the ZTK - three test failures

[Martijn Faassen]

On 07/06/2010 01:27 PM, Charlie Clark wrote:
> Am 06.07.2010, 13:12 Uhr, schrieb Martijn Faassen<faassen at startifact.com>:
>
>> What do you mean, a doctest embedded within another? I'm probably
>> missing something.
>
> No, it's probably me getting the explicit doctest call wrong. It looks to
> my novice eyes like print statement is being passed to a "doctest" method.
> In matters like these it's usually safe to assume I'm wrong! :-)

No, that comment (if you mean that?) is just a way to send configuration 
parameters to the doctest engine. It calls the view.

> hm, I don't think that can be argued with really, particularly given the
> amount of time I've actually studied this and other documents over reading
> the code. But I do think that, whether the module runs as specified, and
> whether the documentation is up to snuff, are of a different nature and,
> consequently, so are their failures. Shouldn't we be testing documentation
> differently?

I think a failure in documentation should ideally show up when running 
the tests. I do agree that documentation and tests have different 
concerns and that mixing them can (but does not necessarily have to) 
lead to poor documentation.

Regards,

Martijn