Throw error if condition false


  • assert(cond,msg)
  • assert(cond,msg,A1,...,An)
  • assert(cond,msgID,msg)
  • assert(cond,msgID,msg,A1,...,An)



assert(cond) throws an error if cond is false.


assert(cond,msg) throws an error and displays the error message, msg, if cond is false.

assert(cond,msg,A1,...,An) displays an error message that contains formatting conversion characters, such as those used with the MATLAB® sprintf function, if cond is false. Each conversion character in msg is converted to one of the values A1,...,An.


assert(cond,msgID,msg) throws an error, displays the error message, msg, and includes an error identifier on the exception, if cond is false. The identifier enables you to distinguish errors and to control what happens when MATLAB encounters the errors.

assert(cond,msgID,msg,A1,...,An) includes an error identifier on the exception and displays a formatted error message.


collapse all

Value in Expected Range

Assert that the value, x, is greater than a specified minimum value.

minVal = 7;
x = 26;

assert(minVal < x)

The expression evaluates as true, and the assertion passes.

Assert that the value of x is between the specified minimum and maximum values.

maxVal = 13;

assert((minVal < x) && (x < maxVal))
Assertion failed.

The expression evaluates as false. The assertion fails and MATLAB throws an error.

Expected Data Type

Assert that the product of two numbers is a double-precision number.

a = 13;
b = single(42);
c = a*b;

assert(isa(c,'double'),'Product is not type double.')
Product is not type double.

Enhance the error message to display the data type of c.

assert(isa(c,'double'),'Product is type %s, not double.',class(c))
Product is type single, not double.

Expected Code Conditions

Use the assert function to test for conditions that should not happen in normal code execution. If the coefficients are numeric, the computed roots should be numeric. A quadratic equation using the specified coefficients and computed roots should be zero.

function x = quadraticSolver(C)

validateattributes(C,{'numeric'},{'size',[1 3]})

a = C(1);
b = C(2);
c = C(3);

x(1) = (-b+sqrt(b^2-4*a*c))/(2*a);
x(2) = (-b-sqrt(b^2-4*a*c))/(2*a);
    'Computed roots are not numeric')

y1 = a*x(1)^2+b*x(1)+c;
y2 = a*x(2)^2+b*x(2)+c;
assert(y1 == 0,'quadraticSolver:root1Error','Error in first root')
assert(isequal(y2,0),'quadraticSolver:root2Error','Error in second root')


Related Examples

Input Arguments

collapse all

cond — Condition to assertMATLAB expression

Condition to assert, specified as a valid MATLAB expression. If cond is false, the assert function throws an error. cond can include relational operators (such as < or ==) and logical operators (such as &&, ||, or ~). Use the logical operators and and or to create compound expressions. MATLAB evaluates compound expressions from left to right, adhering to operator precedence rules.

Example: a<0

Example: exist('myfunction.m','file')

msg — Information about assertion failurestring

Information about the assertion failure, specified as a string. This message displays as the error message. To format the string, use escape sequences, such as \t or \n. You also can use any format specifiers supported by the sprintf function, such as %s or %d. Specify values for the conversion specifiers via the A1,...,An input arguments. For more information, see Formatting Strings.

    Note:   You must specify more than one input argument with assert if you want MATLAB to convert special characters (such as \t, \n, %s, and %d) in the error message string.

Example: 'Assertion condition failed.'

A1,...,An — Numeric or character arraysscalar | vector | matrix | multidimensional array

Numeric or character arrays, specified as a scalar, vector, matrix, or multidimensional array. This input argument provides the values that correspond to and replace the conversion specifiers in msg.

msgID — Identifier for assertion failurestring

Identifier for the assertion failure, specified as a string. Use the message identifier to help identify the source of the error or to control a selected subset of the errors in your program.

The message identifier includes a component and mneumonic. The identifier must always contain a colon and follows this simple format: component:mnemonic. The component and mneumonic fields must each begin with a letter. The remaining characters can be alphanumerics (A–Z, a–z, 0–9) and underscores. No whitespace characters can appear anywhere in msgID. For more information, see Message Identifiers.

Example: 'MATLAB:singularMatrix'

Example: 'MATLAB:narginchk:notEnoughInputs'

More About

expand all


  • When you issue an error, MATLAB captures information about it and stores it in a data structure that is an object of the MException class. You can access information in the exception object by using try/catch. Or, if your program terminates because of an exception and returns control to the Command Prompt, you can use MException.last.

  • If an assertion failure occurs within a try block, MATLAB does not cease execution of the program. In this case, MATLAB passes control to the catch block.

See Also


Introduced in R2007a

Was this topic helpful?