Node:Mathematical functions, Next:Questions for Chapter 13, Previous:Common library functions, Up:Libraries

---

Mathematical functions

Let us now examine some simple math library functions. (This section presupposes some familiarity on your part with trigonometry. If you have none, you might want to skip this section for now - but reading it won't hurt you!)

The following mathematical functions, among others, are available on GNU systems. Many of the functions described below are macros, so the usual caveats about macro parameters apply. (See Macro functions.) All of these functions require parameters of type double or long float. Constants used must be written in floating point form: for instance, write 7.0 instead of just 7.

Here is a list of some functions you can expect to find in the headers math.h, tgmath.h, and limits.h.

abs

Returns the unsigned value of the parameter in brackets. This function is a macro; see fabs for a proper function version.

acos

Returns the arccosine (or inverse cosine) of the parameter, which must lie between -1.0 and +1.0 inclusive. (The result is always in radians.)

asin

Returns the arcsine (or inverse sine) of the parameter, which must lie between -1.0 and +1.0 inclusive. (The result is always in radians.)

atan

Returns the arctangent (or inverse tangent) of the parameter. (The result is always in radians.)

atan2

This is a special function for calculating the inverse tangent of the second parameter divided by the first. atan2 will find the result more accurately than atan will.

result = atan2 (x, y);
result = atan2 (x, 3.14);

ceil

Returns the ceiling for the parameter: that is, the integer just above it. In effect, rounds the parameter up.

cos

Returns the cosine of the parameter in radians. (The parameter is also assumed to be specified in radians.)

cosh

Returns the hyperbolic cosine of the parameter.

exp

Returns the exponential function of the parameter (i.e. e to the power of the parameter).

fabs

Returns the absolute or unsigned value of the parameter in brackets. This is the version that is a proper function; see abs if you want one that is a macro.

floor

Returns the floor for the parameter: that is, the integer just below it. In effect, rounds the parameter down to the nearest integer value, i.e. truncates it.

log

Returns the natural logarithm of the parameter. The parameter used must be greater than zero, but does not have to be declared as unsigned.

log10

Returns the base 10 logarithm of the parameter. The parameter used must be greater than zero, but does not have to be declared as unsigned.

pow

Returns the first parameter raised to the power of the second.

result = pow (x,y); /*raise x to the power y */
result = pow (x,2); /* square x */

sin

Returns the sine of the parameter in radians. (The parameter is also assumed to be specified in radians.)

sinh

Returns the hyperbolic sine of the parameter. (Pronounced "shine" or "sinch".)

sqrt

Returns the positive square root of the parameter.

tan

Returns the tangent of the parameter in radians. (The parameter is also assumed to be specified in radians.)

tanh

Returns the hyperbolic tangent of the parameter.

Here is a code example that uses a few of the math library routines listed above.

#include <stdio.h>
#include <math.h>

int main()
{
  double my_pi;

  my_pi = 4 * atan(1.0);

  /* Print the value of pi we just calculated, to 32 digits. */
  printf ("my_pi = %.32f\n", my_pi);

  /* Print value of pi from math library, to 32 digits. */
  printf ("M_PI  = %.32f\n", M_PI);

  return 0;
}

If you save the above example as pi.c, you will have to enter a command such as the one below to compile it.

gcc pi.c -o pi -lm

When you compile and run the code example, it should print the following results:

my_pi = 3.14159265358979311599796346854419
M_PI  = 3.14159265358979311599796346854419