mustBeUnderlyingType

Create a distributed array (requires Parallel Computing Toolbox™) and then validate that the underlying data type is double.

x = distributed(single(1:10));
mustBeUnderlyingType(x,"double")

Starting parallel pool (parpool) using the 'local' profile ...
Connected to the parallel pool (number of workers: 6).

Value must have one of these underlying data types: 'double'.

mustBeUnderlyingType throws an error because the underlying type of the distributed array is single.

You can specify more than one data type to check against. (since R2024a) Define a second distributed array of type double.

y = distributed(double(1:10));

mustBeUnderlyingType does not throw an error for x or y when both double and single are specified.

mustBeUnderlyingType(x,["double","single"])
mustBeUnderlyingType(y,["double","single"])

This function declares one input argument. In the arguments block, the input is required to have an underlying data type of single.

function mbSingle(input)
    arguments
        input {mustBeUnderlyingType(input,"single")}
    end

    disp("Input is class " + class(input) + ...
        " with underlying type " + underlyingType(input) + ".")
end

Create a distributed vector (requires Parallel Computing Toolbox) that has underlying data of type single.

x = distributed(single(1:10));

Because the input passes the argument validation, the mbSingle function prints information about the class and underlying type.

mbSingle(x)

Input is class distributed with underlying type single.

This function declares one input argument. In the arguments block, the input is required to have an underlying data type of double.

function mbDouble(input)
    arguments
        input {mustBeUnderlyingType(input,"double")}
    end

    disp("Input is class " + class(input) + ...
        " with underlying type " + underlyingType(input) + ".")
end

Call the function with a cell array that contains only double values. Because mustBeUnderlyingType identifies the data type as a cell array, the validation fails even though the contents are double.

x = {1,2,3};
mbDouble(x)

Error using mbDouble (line 3)
 mbDouble(x)
          ^
Invalid argument at position 1. Value must have one of these underlying types: double.

This function, which requires that the input be a cell array, accepts x without error.

function mbCell(input)
    arguments
        input {mustBeUnderlyingType(input,"cell")}
    end
    
    disp("Input is class " + class(input) + ...
        " with underlying type " + underlyingType(input) + ".")
end

mbCell(x)

Input is class cell with underlying type cell.

For more examples, see underlyingType.