How to use lowerBound and upperBound effectively?

[A. Bressan]

Hi, I am translating the following C++ code to D and I have some 
trouble to achieve it using ranges.

#include <vector>
#include <algorithm>

typedef std::vector<int>::const_iterator iter;
std::pair<iter,iter> getSubVector(iter beg, iter end, int val, 
int shift)
{
return {std::upper_bound(beg,end,val)-shift,end};
}

My best result is the following convoluted code, is there a 
better way?

auto getSubVector(T)(const ref T vec, int val, int shift)
if (isInstanceOf!(SortedRange, T))
{
auto 
begin=vec.length-assumeSorted!"a>b"(retro(vec)).lowerBound().length-shift;
return vec[begin..$];
}

The difficulty I have is that, contrary to C++, lowerBound and 
upperBound give the same piece of information because they return 
complementary sub-ranges.
To get the elements that are equal to val outside of the 
upperBound I need to reverse the range.

I tried to use lowerBound on a range sorted by "a<=b", but it 
triggers an error that does not seem to be relevant to binary 
search strategies:

std/algorithm/sorting.d(178): Predicate for isSorted is not 
antisymmetric. Both pred(a, b) and pred(b, a) are true for 
certain values.