[Insight-developers] itkConceptChecking.h

[Bill Lorensen]

Luis,

Given the success of "Adopt a Bug", maybe we should initiate a "Cover a Class".

Bill

On Wed, Dec 10, 2008 at 3:39 PM, Luis Ibanez <luis.ibanez at kitware.com> wrote:
>
>
> ---------- <code_coverage_shameless_propaganda> --------------
>
> Fact:
> =====
>
>           Our current builds have zero warnings.
>
>
> Deduction:
> ==========
>
>  This indicates that the code where you have found this warning,
>  has not been compiled as part of the normal test suite.
>
>
>
> Advertisement:
> ==============
>
>
>  Testing 80% of 158,745 lines of code is not good enough   :-)
>
>  since it means that 31,749 lines of code have never been
>  tested...  (maybe not even compiled...).
>
>
>
> ---------- </code_coverage_shameless_propaganda> --------------
>
>
>
>   Luis
>
>
>
> ------------------------
> Stephen Aylward wrote:
>>
>> Hi,
>>
>> I believe there is a bug (or at least an obscurity) in the concept
>> checking code for BracketOperator.   See the line below marked HERE
>>
>> template <typename T1, typename T2=T1, typename T3=T1>
>> struct BracketOperator
>> {
>>  struct Constraints
>>    {
>>    void constraints()
>>      {
>>      a = static_cast<T3>(b [ c ]);
>>      const_constraints(b, c);
>>      }
>>    void const_constraints(const T1& d, const T2& e)
>>      {
>>      a = static_cast<T3>(b [ c ]);
>>         <<<<< HERE
>>      }
>>    T3 a;
>>    T1 b;
>>    T2 c;
>>    };
>>
>> I've put in lines to avoid an unused variable warning for the two args
>> to that function, but it seems as though the unused vars really should
>> be used.    Perhaps there are missing lines:
>> b = d
>> c = e
>> ?
>>
>> Experts?
>>
>> s
>>
>> PS> Discovered this while fixing warnings....<insert Luis' and Bill's
>> speeches about importance of warnings here> :)
>>
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk.org
> http://www.itk.org/mailman/listinfo/insight-developers
>