Block transposition, permutation, and reshaping of arrays

Analogues for transpose(), reshape(), and permute() but where sub-blocks of the array are treated as scalar elements.

MATLAB provides various functions like permute(), transpose(), and reshape() allowing one to change the dimensions of an array and reorder its elements. At times, however, I find I wish to do things similar to these, but where instead of reordering the individual scalar entries of an array, certain sub-blocks are treated as the "elements". Thus, for instance, if A, B, C, and D are matrices of the same size and M=[A B;C D] is a block matrix, I would like to transpose the block positions to obtain M=[A C; B D] but without reordering the contents of the sub-matrices A,B,C, and D themselves.

This FEX submission contains a set of files for doing just this kind of thing. The assumption throughout is that the sub-blocks are equal-sized sub-tiles of the given array. The tools work on any MATLAB array that can be reshaped and permuted in N-dimensions. They will also work for sparse matrices if you download my ndSparse class.

The usage of these tools is demonstrated in the Examples tab.