[Insight-developers] Multiple valgrind defects

[Bill Lorensen]

Lots of questions. I need to think more about this.

In the meantime, can you fix the valgrind error?

Bill

On Sun, Oct 23, 2011 at 9:38 AM, Arnaud Gelas
<Arnaud_Gelas at hms.harvard.edu>wrote:

> Bill,
>
> The main goal of the new framework is to provide an extensible framework,
> where new terms can be easily added and contributed by the community.
> Enumerated all possible terms is not a solution.
>
> Briefly, here is a summary
>
> 1) N level sets, N PDEs
> 2) each PDE_{i} has M_{i} terms
> 3) each term can depend on the N level sets, but contribute to a single
> PDE_{j}
> 4) each level set has an ID which could be an integer (fine if you have
> 1000 level sets) or a std::string (can be more appropriate if you have 3-4
> level sets for representing organs or tissues).
>
> Each term needs the level set container, and the current level set "id".
>
> When constructing one term, the current level set id needs to be
> initialized...
>
> Should we create an extra intermediate layer which would specialize
> LevelSetEquationTermContainerBase for integer and for std::string (then
> classes which were inheriting from LevelSetEquationTermContainerBase would
> inherit form this one) ?
> Or is there another possible mechanism to be used ?
>
> Thanks,
> Arnaud
>
>
> On Oct 23, 2011, at 8:45 AM, Bill Lorensen wrote:
>
> I'm not sure how it is used. Are there a known number and types of terms?
> If there are, theb it should probably be an enumerated type.
>
> On Fri, Oct 21, 2011 at 8:03 AM, Arnaud Gelas <
> arnaud_gelas at hms.harvard.edu> wrote:
>
>> Bill,
>>
>> Thanks for pointing this out!
>>
>> A lot of them are in the new level set framework, which is actually only
>> one uninitialized member value in LevelSetEquationtermContainerBase.
>> The type of this member variable could be either a integer type or
>> std::string.
>>
>> I am not sure what would be the best initialization to handle both cases?
>> Any recommendation?
>>
>> Thanks,
>> Arnaud
>>
>> On Oct 21, 2011, at 7:36 AM, Bill Lorensen wrote:
>>
>> Folks,
>>
>> Valgrind issues:
>> http://www.cdash.org/CDash/viewDynamicAnalysis.php?buildid=1646825
>>
>> The following tests have valgrind defects:
>>
>> TemporalProcessObjectTest
>> VideoSourceTest
>> VideoToVideoFilterTest
>> itkANTSNeighborhoodCorrelationImageToImageObjectMetricTest
>> itkANTSNeighborhoodCorrelationImageToImageObjectRegistrationTest
>> itkBinaryClosingByReconstructionImageFilterTest
>> itkBinaryDilateImageFilterTest
>> itkImageToImageObjectMetricTest
>> itkMultiLevelSetsv4ChanAndVeseInternalTermTest
>> itkMultiLevelSetsv4SetEvolutionTest
>> itkObjectToObjectOptimizerBaseTest
>> itkSingleLevelSetsv4DenseImage2DTest
>> itkSingleLevelSetsv4MalcolmImage2DTest
>> itkSingleLevelSetsv4ShiImage2DTest
>> itkSingleLevelSetsv4WhitakerImage2DTest
>> itkSingleLevelSetsv4WhitakerImage2DWithCurvatureTest
>> itkSingleLevelSetsv4WhitakerImage2DWithLaplacianTest
>> itkSingleLevelSetsv4WhitakerImage2DWithPropagationTest
>> itkTwoLevelSetsv4DenseImage2DTest
>> itkTwoLevelSetsv4MalcolmImage2DTest
>> itkTwoLevelSetsv4ShiImage2DTest
>> itkTwoLevelSetsv4WhitakerImage2DTest
>>
>> Bill
>>
>> <ATT00001.txt>
>>
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20111023/366517ab/attachment.htm>