Automatic conversions and casts

Once variables have been declared, they can be manipulated and combined in various ways. Exactly which operators may be applied to a variable depends on its type: for example all the numerical types support the usual arithmetic operations (addition, multiplication etc.) but only integer types support the modulus operator, %.

Sometimes you need to convert from one type to another, for example when assigning an int to a floating-point type. It is clear what the result of such a conversion should be, and C is prepared to convert between most pairs of related types automatically:

int a = 6; float b = 3.5; int c; c = b; /* ok, b gets rounded towards zero, so c=3 */ b = a; /* ok, b becomes the real number 6 */

To convert between types for which an automatic conversion does not exist, or to force a conversion at a convenient moment, use a cast:

int n = 4; int m = 6; float n_over_m; n_over_m = n/m; /* yields 0 */ n_over_m = n/ (float)m; /* (cast m to a float) yields 0.66666 etc. */

Without the cast, n/m will be evaluated using integer division (which returns the quotient, discarding the remainder). The cast forces m to be converted to a float, after which n is automatically converted to a floating-point type and the division is done floating-point.