On Wed, 16 May 2018, Mehdi Saada wrote:

> Is it likely or not that we see some improvements over multidimensional arrays ?

Why?

> But it should be allowed to provide a proper initial value at initialization.
> Imagine a matrix whose first row and first column's elements are all at zero.

What is wrong with the code below?

    type Matrix is array (Natural range <>, Natural range <>) of Natural;

    M1: Matrix(1 .. 10, 1 .. 5) :=
      (1      => (others => 0),        -- first row 0
       others => (1 => 0, others => 1) -- first column in other rows 0,
 	                              -- all other entries 1
      );

    M2: Matrix(1 .. 10, 1 .. 5) :=
      (1 =>      (others => 0),         -- first row 0
       others => (1 => 0, others => <>) -- first column in other rows 0,
 	                               -- all other entries left undefined
      );

--------  I  love  the  taste  of  Cryptanalysis  in  the morning!  --------
www.uni-weimar.de/de/medien/professuren/mediensicherheit/people/stefan-lucks
----Stefan.Lucks (at) uni-weimar.de, Bauhaus-Universität Weimar, Germany----