Any approaches to slice an array with &[[usize]; ndim] ?

[Ionizing]

Hi, I just came across a problem with the slicing:

Suppose we have an array A of ndim dimensions and &[usize]s containing indices of each dimension, how to operate A with the indices ?

For example, in numpy, we can

A = np.arange(27).reshape(3, 3, 3)
id_dim1 = [0, 2]
id_dim2 = [0, 1]
id_dim3 = [1, 2]

A[id_dim1, id_dim2, id_dim3] *= 2

# produces
A = array([[[ 0,  2,  2],
        [ 3,  4,  5],
        [ 6,  7,  8]],

       [[ 9, 10, 11],
        [12, 13, 14],
        [15, 16, 17]],

       [[18, 19, 20],
        [21, 22, 46],
        [24, 25, 26]]])

Is it possible to do the same thing within ndarray in an elegant way ?

[Ionizing]

It seems this question is duplicated with #526 ... but can we do it now?

[Ionizing]

I've tested via julia, this approach is a bit slower than direct loop sum due to the extra allocation. Maybe I should close this issue.