Specify Input Types Using assert Statements in MATLAB Code

Specify Input Types Using `assert` Statements in MATLAB Code

For code generation, you can use assert statements to define the types of entry-point function inputs within your MATLAB^® code. This approach is called preconditioning. Alternatively, you can specify input types for code generation by using arguments blocks in your MATLAB code, by using the MATLAB Coder™ app, or by using the -args argument with codegen (MATLAB Coder) at the command line. See Specify Types of Entry-Point Function Inputs (MATLAB Coder).

How to Use assert with MATLAB Coder

Use the assert function to invoke standard MATLAB functions for specifying the class, size, and complexity of primary function inputs.

If your MATLAB function uses assert statements to define input types, the app uses these definitions by default. In the Entry Points pane, the app selects the Source input types from file check box to the right of the function signature.

Entry-point pand showing function bar with the type of input argument d sourced from MATLAB code.

When specifying input properties using the assert function, use one of the following methods. Use the exact syntax that is provided; do not modify it.

Specify Any Class

assert(isa(param,'class_name'))

Sets the input parameter param to the MATLAB class class_name. For example, to set the class of input U to a 32-bit signed integer, call:

... 
assert(isa(U,'int32'));
...

Specify fi Class

assert(isfi(param))
assert(isa(param,'embedded.fi'))

Sets the input parameter param to the MATLAB class fi (fixed-point numeric object). For example, to set the class of input U to fi, call:

... 
assert(isfi(U));
...

...
assert(isa(U,'embedded.fi'));
...

You must specify both the fi class and the numerictype. See Specify numerictype of Fixed-Point Input. You can also set the fimath properties, see Specify fimath of Fixed-Point Input. If you do not set the fimath properties, codegen uses the MATLAB default fimath value.

Specify Structure Class

assert(isstruct(param))
assert(isa(param,'struct'))

Sets the input parameter param to the MATLAB class struct (structure). For example, to set the class of input U to a struct, call:

...
assert(isstruct(U));
...

...
assert(isa(U,'struct'));
...

If you set the class of an input parameter to struct, you must specify the properties of all fields in the order that they appear in the structure definition.

Specify Cell Array Class

assert(iscell( param))
assert(isa(param,'cell'))

Sets the input parameter param to the MATLAB class cell (cell array). For example, to set the class of input C to a cell, call:

...
assert(iscell(C));
...

...
assert(isa(C,'cell'));
...

To specify the properties of cell array elements, see Specifying Properties of Cell Arrays.

Specify Fixed Size

assert(all(size (param) == [dims ]))

Sets the input parameter param to the size that dimensions dims specifies. For example, to set the size of input U to a 3-by-2 matrix, call:

...
assert(all(size(U)== [3 2]));
...

Specify Scalar Size

assert(isscalar (param))
assert(all(size (param) == [ 1 ]))

Sets the size of input parameter param to scalar. To set the size of input U to scalar, call:

...
assert(isscalar(U));
...

...
assert(all(size(U)== [1]));
...

Specify Upper Bounds for Variable-Size Inputs

assert(all(size(param)<=[N0 N1 ...]));
assert(all(size(param)<[N0 N1 ...]));

Sets the upper-bound size of each dimension of input parameter param. To set the upper-bound size of input U to be less than or equal to a 3-by-2 matrix, call:

assert(all(size(U)<=[3 2]));

Note

You can also specify upper bounds for variable-size inputs using coder.varsize.

Specify Inputs with Fixed- and Variable-Size Dimensions

assert(all(size(param)>=[M0 M1 ...]));
assert(all(size(param)<=[N0 N1 ...]));

When you use assert(all(size(param)>=[M0 M1 ...])) to specify the lower-bound size of each dimension of an input parameter:

You must also specify an upper-bound size for each dimension of the input parameter.
For each dimension, k, the lower-bound Mk must be less than or equal to the upper-bound Nk.
To specify a fixed-size dimension, set the lower and upper bound of a dimension to the same value.
Bounds must be nonnegative.

To fix the size of the first dimension of input U to 3 and set the second dimension as variable size with upper bound of 2, call:

assert(all(size(U)>=[3 0]));
assert(all(size(U)<=[3 2]));

Specify Size of Individual Dimensions

assert (size(param, k)==Nk);
assert (size(param, k)<=Nk);
assert (size(param, k)<Nk);

You can specify individual dimensions and all dimensions simultaneously. You can also specify individual dimensions instead of specifying all dimensions simultaneously. The following rules apply:

You must specify the size of each dimension at least once.
The last dimension specification takes precedence over earlier specifications.

Sets the upper-bound size of dimension k of input parameter param. To set the upper-bound size of the first dimension of input U to 3, call:

assert(size(U,1)<=3)

To fix the size of the second dimension of input U to 2, call:

assert(size(U,2)==2)

Specify Real Input

assert(isreal (param))

Specifies that the input parameter param is real. To specify that input U is real, call:

...
assert(isreal(U));
...

Specify Complex Input

assert(~isreal (param))

Specifies that the input parameter param is complex. To specify that input U is complex, call:

...
assert(~isreal(U));
...

Specify `numerictype` of Fixed-Point Input

assert(isequal(numerictype(fiparam ), T))

Sets the numerictype properties of fi input parameter fiparam to the numerictype object T. For example, to specify the numerictype property of fixed-point input U as a signed numerictype object T with 32-bit word length and 30-bit fraction length, use the following code:

%#codegen
...
% Define the numerictype object.
T = numerictype(1, 32, 30);

% Set the numerictype property of input U to T.
assert(isequal(numerictype(U),T));
...

Specifying the numerictype for a variable does not automatically specify that the variable is fixed point. You must specify both the fi class and the numerictype.

Specify fimath of Fixed-Point Input

assert(isequal(fimath(fiparam ), F))

Sets the fimath properties of fi input parameter fiparam to the fimath object F. For example, to specify the fimath property of fixed-point input U so that it saturates on integer overflow, use the following code:

%#codegen
...
% Define the fimath object.
F = fimath('OverflowMode','saturate');

% Set the fimath property of input U to F.
assert(isequal(fimath(U),F));
...

If you do not specify the fimath properties using assert, codegen uses the MATLAB default fimath value.

Specify Multiple Properties of Input

assert(function1(params)&& 
         function2(params)&& 
         function3(params(&&...)

Specifies the class, size, and complexity of one or more inputs using a single assert function call. For example, the following code specifies that input U is a double, complex, 3-by-3 matrix, and input V is a 16-bit unsigned integer:

%#codegen
...
assert(isa(U,'double') && 
       ~isreal(U) && 
       all(size(U) == [3 3]) && 
       isa(V,'uint16'));
...

Rules for Using assert Function

When using the assert function to specify the properties of primary function inputs, follow these rules:

Call assert functions at the beginning of the primary function, before control-flow operations such as if statements or subroutine calls.
Do not call assert functions inside conditional constructs, such as if, for, while, and switch statements.
For a fixed-point input, you must specify both the fi class and the numerictype. See Specify numerictype of Fixed-Point Input. You can also set the fimath properties. See Specify fimath of Fixed-Point Input. If you do not set the fimath properties, codegen uses the MATLAB default fimath value.
If you set the class of an input parameter to struct, you must specify the class, size, and complexity of all fields in the order that they appear in the structure definition.
When you use assert(all(size(param)>=[M0 M1 ...])) to specify the lower-bound size of each dimension of an input parameter:
- You must also specify an upper-bound size for each dimension of the input parameter.
- For each dimension, k, the lower-bound Mk must be less than or equal to the upper-bound Nk.
- To specify a fixed-size dimension, set the lower and upper bound of a dimension to the same value.
- Bounds must be nonnegative.
If you specify individual dimensions, the following rules apply:
- You must specify the size of each dimension at least once.
- The last dimension specification takes precedence over earlier specifications.

Note

The generation of single precision C/C++ code is not supported when specifying input types using preconditioning. To generate single precision C/C++ code, use codegen -singleC and specify input types using -args.

Specifying General Properties of Primary Inputs

In the following code excerpt, a primary MATLAB function mcspecgram takes two inputs: pennywhistle and win. The code specifies the following properties for these inputs.

Input	Property	Value
`pennywhistle`	class	`int16`
	size	220500-by-1 vector
	complexity	`real` (by default)
`win`	class	`double`
	size	1024-by-1 vector
	complexity	`real` (by default)

%#codegen
function y = mcspecgram(pennywhistle,win)
nx = 220500;
nfft = 1024;
assert(isa(pennywhistle,'int16'));
assert(all(size(pennywhistle) == [nx 1]));
assert(isa(win,'double'));
assert(all(size(win) == [nfft 1]));
...

Alternatively, you can combine property specifications for one or more inputs inside assert commands:

%#codegen
function y = mcspecgram(pennywhistle,win)
nx = 220500;
nfft = 1024;
assert(isa(pennywhistle,'int16') && all(size(pennywhistle) == [nx 1]));
assert(isa(win,'double') && all(size(win) == [nfft 1]));
...

Specifying Properties of Primary Fixed-Point Inputs

To specify fixed-point inputs, you must install Fixed-Point Designer™ software.

In the following example, the primary MATLAB function mcsqrtfi takes one fixed-point input x. The code specifies the following properties for this input.

Property	Value
class	`fi`
`numerictype`	`numerictype` object `T`, as specified in the primary function
`fimath`	`fimath` object `F`, as specified in the primary function
size	`scalar`
complexity	`real` (by default)

function y = mcsqrtfi(x) %#codegen
T = numerictype('WordLength',32,'FractionLength',23,...
                'Signed',true);
F = fimath('SumMode','SpecifyPrecision',...
           'SumWordLength',32,'SumFractionLength',23,...
           'ProductMode','SpecifyPrecision',...
           'ProductWordLength',32,'ProductFractionLength',23);
assert(isfi(x));
assert(isequal(numerictype(x),T));
assert(isequal(fimath(x),F));

y = sqrt(x);

You must specify both the fi class and the numerictype.

Specifying Properties of Cell Arrays

To specify the class cell (cell array), use one of the following syntaxes:

assert(iscell(param))
assert(isa( param,'cell'))

For example, to set the class of input C to cell, use:

...
assert(iscell(C));
...

...
assert(isa(C,'cell'));
...

You can also specify the size of the cell array and the properties of the cell array elements. The number of elements that you specify determines whether the cell array is homogeneous or heterogeneous. See Code Generation for Cell Arrays (MATLAB Coder).

If you specify the properties of the first element only, the cell array is homogeneous. For example, the following code specifies that C is a 1x3 homogeneous cell array whose elements are 1x1 double.

...
assert(isa(C,'cell'));
assert(all(size(C) == [1  3]));
assert(isa(C{1},'double'));
...

If you specify the properties of the first element only, but also assign a structure type name to the cell array, the cell array is heterogeneous. Each element has the properties of the first element. For example, the following code specifies that C is a 1x3 heterogeneous cell array. Each element is a 1x1 double.

...
assert(isa(C,'cell'));
assert(all(size(C) == [1  3]));
assert(isa(C{1},'double'));
coder.cstructname(C,'myname');
...

If you specify the properties of each element, the cell array is heterogeneous. For example, the following code specifies a 1x2 heterogeneous cell array whose first element is 1x1 char and whose second element is 1x3 double.

...
assert(isa(C,'cell'));
assert(all(size(C) == [1  2]));
assert(isa(C{1},'char'));
assert(all(size(C{2}) == [1 3]));
assert(isa(C{2},'double'));
...

If you specify more than one element, you cannot specify that the cell array is variable size, even if all elements have the same properties. For example, the following code specifies a variable-size cell array. Because the code specifies the properties of the first and second elements, code generation fails.

...
assert(isa(C,'cell'));
assert(all(size(C) <= [1  2]));
assert(isa(C{1},'double'));
assert(isa(C{2},'double'));
...

In the previous example, if you specify the first element only, you can specify that the cell array is variable-size. For example:

...
assert(isa(C,'cell'));
assert(all(size(C) <= [1  2]));
assert(isa(C{1},'double'));
...

Specifying Class and Size of Scalar Structure

Suppose that you define S as the following scalar MATLAB structure:

S = struct('r',double(1),'i',int8(4));

The following code specifies the properties of the function input S and its fields:

function y = fcn(S)  %#codegen


% Specify the class of the input as struct.
assert(isstruct(S));

% Specify the class and size of the fields r and i
% in the order in which you defined them.
assert(isa(S.r,'double'));
assert(isa(S.i,'int8');
...

In most cases, when you do not explicitly specify values for properties, MATLAB Coder uses defaults—except for structure fields. The only way to name a field in a structure is to set at least one of its properties. At a minimum, you must specify the class of a structure field.

Specifying Class and Size of Structure Array

For structure arrays, you must choose a representative element of the array for specifying the properties of each field. For example, assume that you have defined S as the following 1-by-2 array of MATLAB structures:

S = struct('r',{double(1),double(2)},'i',{int8(4), int8(5)});

The following code specifies the class and size of each field of structure input S by using the first element of the array:

%#codegen
function y = fcn(S)

% Specify the class of the input S as struct.
assert(isstruct(S));

% Specify the size of the fields r and i
% based on the first element of the array.
assert(all(size(S) == [1 2]));
assert(isa(S(1).r,'double'));
assert(isa(S(1).i,'int8'));

The only way to name a field in a structure is to set at least one of its properties. At a minimum, you must specify the class of all fields.