I was looking at this interesting thread: https://stackoverflow.com/a/16596463/2436175
My specific case concerns declaring a templated function using a std container of cv::Point_ and cv::Rect_ from opencv. I want to template against:
- the type of std container I will use
- the basic data types to complete the definition of cv::Point_ and cv::Rect_
I ended up with the following declaration:
template <typename T, template <typename, typename> class Container_t>
void CreateRects(const Container_t<cv::Point_<T>,std::allocator<cv::Point_<T> > >& points,
const T value,
Container_t<cv::Rect_<T>,std::allocator<cv::Rect_<T> > >& rects) {
}
which compiles fine with this:
void dummy() {
const std::vector<cv::Point_<double> > points;
std::vector<cv::Rect_<double> > rects;
CreateRects(points,5.0,rects);
}
(I have also seen that I can also use, for example, CreateRects<double>(points,5,rects))
I was wondering if there existed any way to make my declaration more compact, e.g. without the need to specify 2 times the default allocator.