No you understand certainly, I wasn't clear. If we provide a nice api
for comparing things are equal it would be a shame to not provide the
convenience for doubles. Newcomers to the environment would not
necessarily understand the pitfalls of testing equality on doubles.
Thus providing convenience in one area could look very attractive in
the general case if there are no other methods hinting their
appropriate use. I'm sure such a method has already been written in a
repository somewhere . I just thought it would be good to put them in
together. That make sense?

Mike

On Tuesday, July 7, 2009, Oscar Nierstrasz <[email protected]> wrote:
>
> Hi Mike,
>
> I don't understand.  assert:equals: should only be used for exact
> equality tests, not for comparing doubles.
>
> Doubles should never go through this method.
>
> Existing tests won't know about assert:equals:
>
> Did I misunderstand you?
>
> - on
>
> On 7 Jul 2009, at 19:24, Michael Roberts wrote:
>
>> If you are adding this it would be good to either add or review
>> existing implementation that checks the comparison is within a
>> tolerance. This would be for asserting double values are close but not
>> necessarily exact. I would rather we didn't get double comparison
>> going through this method you describe which would be easy to do.
>>
>> Thanks
>> mike
>
>
> _______________________________________________
> Pharo-project mailing list
> [email protected]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>

_______________________________________________
Pharo-project mailing list
[email protected]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Reply via email to