Re: Initialization of dynamic multidimensional array

[Ali Çehreli via Digitalmars-d-learn]

On 02/05/2017 12:33 PM, berni wrote:
With X not known at compile time:
auto arr = new int[][](X,X);
for (int i=0;i<X;i++)
  for (int j=0;j<X;j++)
    arr[i][j] = -1;

Is there anything better for this? I mean, the program will fill the
array with zeroes, just to overwrite all of them with -1. That's wasted
execution time and doesn't feel D-ish to me.

Here is something that returns the tail as dynamic array. (Otherwise, 
being a value type, the whole static array would be copied out of the 
function):

template makeMulDim(dims...) {
    template MultDimType(T, dims...) {
        static if (dims.length == 0) {
            alias MultDimType = T;
        } else static if (dims.length == 1) {
            alias MultDimType = T[];
        } else {
            alias MultDimType = MultDimType!(T[dims[0]], dims[1..$]);
        }
    }

    auto makeMulDim(T)(T initValue) {
        import std.algorithm : fold;

        const elementCount = [ dims ].fold!((a, b) => a * b);
        auto storage = new T[](elementCount);
        storage[] = initValue;
        return cast(MultDimType!(T, dims))storage;
    }
}

unittest {
    auto arr = makeMulDim!(2, 3, 4)(1.1);
    static assert (is (typeof(arr) == double[2][3][]));
}

void main() {
    import std.stdio;

    auto arr = makeMulDim!(2, 3, 4)(-1);
    pragma(msg, typeof(arr));
    arr[3][2][1] = 42;
    writefln("%(%s\n%)", arr);
}

So, the returned type in main is int[2][3][].

Ali