Although JScript includes many built-in functions, you can create your own functions. A function definition consists of a function statement and a block of JScript statements.

# Defining Your Own Function

The `checkTriplet`

function in the following example takes the lengths of the sides of a triangle as its arguments. It calculates whether the triangle is a right triangle by checking whether the three numbers constitute a Pythagorean triplet (the square of the length of the hypotenuse of a right triangle is equal to the sum of the squares of the lengths of the other two sides). The `checkTriplet`

function calls one of two other functions to make the actual test.

Notice the use of a very small number (`epsilon`

) as a testing variable in the floating-point version of the test. Because of uncertainties and round-off errors in floating-point calculations, it is not practical to make a direct test of whether the three numbers constitute a Pythagorean triplet unless all three values in question are known to be integers. Because a direct test is more accurate, the code in this example determines whether it is appropriate and, if it is, uses it.

Type annotation is not used when defining these functions. For this application, it is useful for the `checkTriplet`

function to take both integer and floating-point data types.

В | Copy Code |
---|---|

const epsilon = 0.00000000001; // Some very small number to test against. // Type annotate the function parameters and return type. function integerCheck(a : int, b : int, c : int) : boolean { // The test function for integers. // Return true if a Pythagorean triplet. return ( ((a*a) + (b*b)) == (c*c) ); } // End of the integer checking function. function floatCheck(a : double, b : double, c : double) : boolean { // The test function for floating-point numbers. // delta should be zero for a Pythagorean triplet. var delta = Math.abs( ((a*a) + (b*b) - (c*c)) * 100 / (c*c)); // Return true if a Pythagorean triplet (if delta is small enough). return (delta < epsilon); } // End of the floating-poing check function. // Type annotation is not used for parameters here. This allows // the function to accept both integer and floating-point values // without coercing either type. function checkTriplet(a, b, c) : boolean { // The main triplet checker function. // First, move the longest side to position c. var d = 0; // Create a temporary variable for swapping values if (b > c) { // Swap b and c. d = c; c = b; b = d; } if (a > c) { // Swap a and c. d = c; c = a; a = d; } // Test all 3 values. Are they integers? if ((int(a) == a) && (int(b) == b) && (int(c) == c)) { // If so, use the precise check. return integerCheck(a, b, c); } else { // If not, get as close as is reasonably possible. return floatCheck(a, b, c); } } // End of the triplet check function. // Test the function with several triplets and print the results. // Call with a Pythagorean triplet of integers. print(checkTriplet(3,4,5)); // Call with a Pythagorean triplet of floating-point numbers. print(checkTriplet(5.0,Math.sqrt(50.0),5.0)); // Call with three integers that do not form a Pythagorean triplet. print(checkTriplet(5,5,5)); |

The output of this program is:

В | Copy Code |
---|---|

true true false |