REFAC: don't duplicate code in Application::GetParameterImage
Why are there two methods called Application::GetParameterImage
, one templated on its return type and one not templated? The code between the two is partially duplicated. The only practical difference is that they call a different GetImage
, one templated on its return type and the other not. But it's equivalent in the end because the non template GetImage
calls GetImage<FloatVectorImageType>
.
This diff is a proposal to fix but I wonder if there is some good reason to have a template method and a non-template one, where a default template argument would do the job? (@lhermitte)
I would also propose the same fix for InputImageParameter::GetImage
, but there the situation is even less clear with 3 implementations: a non template, a specialization for ImageBaseType
, and a template.