[Insight-developers] Deceptive Coverage Numbers

[Luis Ibanez]

Gaetan,

Thanks for the clarification.

I guess we are back to "Plan A" then.

Karthik is currently working on it.



    Luis


----------------------
Gaëtan Lehmann wrote:
> 
> Hi Luis,
> 
> Le 16 avr. 09 à 18:22, Luis Ibanez a écrit :
> 
>>
>> Hi Bill,
>>
>> Good catch...
>>
>> Visual inspection is probably the best idea at this point,
>> since it also provides a code review and will prompt many
>> improvements.
>>
>> Another option that comes to mind is to wrap this classes,
>> (for some common instantiations) in which case, all the
>> templated methods will be instantiated.
> 
> 
> Wrapping only ensures that the code builds - already a nice beginning  - 
> but I'm afraid it won't help to enhance the code coverage.
> Also, with wrapitk, itk::Statistics::Histogram is not build currently,  
> because of the nested namespace (It's ok in the unstable version).
> 
> Regards,
> 
> Gaëtan
> 
> 
>>
>>
>>
>>   Luis
>>
>>
>> ----------------------
>> Bill Lorensen wrote:
>>
>>> Luis,
>>> Take a look at the coverage for this file:
>>> http://www.cdash.org/CDash/viewCoverageFile.php?buildid=314097&fileid=10830502 
>>>
>>> Note that several methods do not show up as program lines and thus do
>>> not show up as uncovered lines. This is not limited to the new
>>> statistics coverage, we have the same issues in ITK.
>>> I believe it is because the template code is never referenced and
>>> therefore is not considered code. I think the only remedy for now  is a
>>> visual inspection.
>>> Bill
>>
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at 
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ITK FAQ at: 
>> http://www.itk.org/Wiki/ITK_FAQ
>>
>> Follow this link to subscribe/unsubscribe:
>> http://www.itk.org/mailman/listinfo/insight-developers
> 
>