HP-UX Reference (11i v2 04/09) - 3 Library Functions A-M (vol 6)

ManualsBrandsHP ManualsSoftwareHP-UX Reference Manuals

141

142

143

144

145

146

147

148

149

150

catan(3M)

Itanium(R)-based Systems Only

catan(3M)

NAME

catan(), catanf(), catanl(), catanw(), catanq() - complex arctangent functions

SYNOPSIS

#include <complex.h>

double complex catan(double complex z );

float complex catanf(float complex z );

long double complex catanl(long double complex z );

extended complex catanw(extended complex z );

quad complex catanq(quad complex z );

DESCRIPTION

catan() returns the complex arctangent of z, in the range of a strip mathematically unbounded along

the imaginary axis and in the interval [-Pi/2,+Pi/2] along the real axis. Branch cuts are outside the inter-

val [-1,+1] along the imaginary axis.

catanf() is a float complex version of catan()

; it takes a float complex argument and

returns a

float complex result.

catanl() is a long double complex version of catan(); it takes a long double complex argu-

ment and returns a long double complex result.

catanw() is an extended complex version of catan(); it takes an extended complex argument

and returns an extended complex result.

catanq() is equivalent to catanl() on HP-UX systems.

USAGE

To use these functions, compile with the default

-Ae option. To use catanw()

or catanq(), compile

with the

-fpwidetypes option. Make sure your program includes <complex.h>. Link in the math

library by specifying -lm on the compiler or linker command line.

RETURN VALUE

catan(conj(z)) = conj(catan(z)) and catan is odd.

catan(+0+i0) returns +0+i0.

catan(+NaN+i0) returns +NaN+i0.

catan(+0+i1) returns +0+iInf and raises the divide-by-zero ﬂoating-point exception.

catan(Inf+iy) returns Pi/2+i0, for ﬁnite positive-signed y.

catan(NaN+iy) returns NaN+iNaN and optionally raises the invalid ﬂoating-point exception, for

nonzero ﬁnite y.

catan(+x+iInf) returns Pi/2+i0, for ﬁnite positive-signed x.

catan(+Inf+iInf) returns Pi/2+i0.

catan(+NaN+iInf) returns +NaN+i0.

catan(x+iNaN) returns NaN+iNaN and optionally raises the invalid ﬂoating-point exception, for ﬁnite

catan(Inf+iNaN) returns Pi/2+-i0 (where the sign of the imaginary part of the result is unspeciﬁed).

catan(NaN+iNaN) returns NaN+iNaN.

ERRORS

No errors are deﬁned.