| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720 | /*							clogl.c * *	Complex natural logarithm * * * * SYNOPSIS: * * void clogl(); * cmplxl z, w; * * clogl( &z, &w ); * * * * DESCRIPTION: * * Returns complex logarithm to the base e (2.718...) of * the complex argument x. * * If z = x + iy, r = sqrt( x**2 + y**2 ), * then *       w = log(r) + i arctan(y/x). *  * The arctangent ranges from -PI to +PI. * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      7000       8.5e-17     1.9e-17 *    IEEE      -10,+10     30000       5.0e-15     1.1e-16 * * Larger relative error can be observed for z near 1 +i0. * In IEEE arithmetic the peak absolute error is 5.2e-16, rms * absolute error 1.0e-16. */#include <math.h>#ifdef ANSIPROTstatic void cchshl ( long double x, long double *c, long double *s );static long double redupil ( long double x );static long double ctansl ( cmplxl *z );long double cabsl ( cmplxl *x );void csqrtl ( cmplxl *x, cmplxl *y );void caddl ( cmplxl *x, cmplxl *y, cmplxl *z );extern long double fabsl ( long double );extern long double sqrtl ( long double );extern long double logl ( long double );extern long double expl ( long double );extern long double atan2l ( long double, long double );extern long double coshl ( long double );extern long double sinhl ( long double );extern long double asinl ( long double );extern long double sinl ( long double );extern long double cosl ( long double );void clogl ( cmplxl *, cmplxl *);void casinl ( cmplxl *, cmplxl *);#elsestatic void cchshl();static long double redupil();static long double ctansl();long double cabsl(), fabsl(), sqrtl();lnog double logl(), expl(), atan2l(), coshl(), sinhl();long double asinl(), sinl(), cosl();void caddl(), csqrtl(), clogl(), casinl();#endifextern long double MAXNUML, MACHEPL, PIL, PIO2L;void clogl( z, w )register cmplxl *z, *w;{long double p, rr;/*rr = sqrt( z->r * z->r  +  z->i * z->i );*/rr = cabsl(z);p = logl(rr);#if ANSICrr = atan2l( z->i, z->r );#elserr = atan2l( z->r, z->i );if( rr > PIL )	rr -= PIL + PIL;#endifw->i = rr;w->r = p;}/*							cexpl() * *	Complex exponential function * * * * SYNOPSIS: * * void cexpl(); * cmplxl z, w; * * cexpl( &z, &w ); * * * * DESCRIPTION: * * Returns the exponential of the complex argument z * into the complex result w. * * If *     z = x + iy, *     r = exp(x), * * then * *     w = r cos y + i r sin y. * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      8700       3.7e-17     1.1e-17 *    IEEE      -10,+10     30000       3.0e-16     8.7e-17 * */void cexpl( z, w )register cmplxl *z, *w;{long double r;r = expl( z->r );w->r = r * cosl( z->i );w->i = r * sinl( z->i );}/*							csinl() * *	Complex circular sine * * * * SYNOPSIS: * * void csinl(); * cmplxl z, w; * * csinl( &z, &w ); * * * * DESCRIPTION: * * If *     z = x + iy, * * then * *     w = sin x  cosh y  +  i cos x sinh y. * * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      8400       5.3e-17     1.3e-17 *    IEEE      -10,+10     30000       3.8e-16     1.0e-16 * Also tested by csin(casin(z)) = z. * */void csinl( z, w )register cmplxl *z, *w;{long double ch, sh;cchshl( z->i, &ch, &sh );w->r = sinl( z->r ) * ch;w->i = cosl( z->r ) * sh;}/* calculate cosh and sinh */static void cchshl( x, c, s )long double x, *c, *s;{long double e, ei;if( fabsl(x) <= 0.5L )	{	*c = coshl(x);	*s = sinhl(x);	}else	{	e = expl(x);	ei = 0.5L/e;	e = 0.5L * e;	*s = e - ei;	*c = e + ei;	}}/*							ccosl() * *	Complex circular cosine * * * * SYNOPSIS: * * void ccosl(); * cmplxl z, w; * * ccosl( &z, &w ); * * * * DESCRIPTION: * * If *     z = x + iy, * * then * *     w = cos x  cosh y  -  i sin x sinh y. * * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      8400       4.5e-17     1.3e-17 *    IEEE      -10,+10     30000       3.8e-16     1.0e-16 */void ccosl( z, w )register cmplxl *z, *w;{long double ch, sh;cchshl( z->i, &ch, &sh );w->r = cosl( z->r ) * ch;w->i = -sinl( z->r ) * sh;}/*							ctanl() * *	Complex circular tangent * * * * SYNOPSIS: * * void ctanl(); * cmplxl z, w; * * ctanl( &z, &w ); * * * * DESCRIPTION: * * If *     z = x + iy, * * then * *           sin 2x  +  i sinh 2y *     w  =  --------------------. *            cos 2x  +  cosh 2y * * On the real axis the denominator is zero at odd multiples * of PI/2.  The denominator is evaluated by its Taylor * series near these points. * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      5200       7.1e-17     1.6e-17 *    IEEE      -10,+10     30000       7.2e-16     1.2e-16 * Also tested by ctan * ccot = 1 and catan(ctan(z))  =  z. */void ctanl( z, w )register cmplxl *z, *w;{long double d;d = cosl( 2.0L * z->r ) + coshl( 2.0L * z->i );if( fabsl(d) < 0.25L )	d = ctansl(z);if( d == 0.0L )	{	mtherr( "ctan", OVERFLOW );	w->r = MAXNUML;	w->i = MAXNUML;	return;	}w->r = sinl( 2.0L * z->r ) / d;w->i = sinhl( 2.0L * z->i ) / d;}/*							ccotl() * *	Complex circular cotangent * * * * SYNOPSIS: * * void ccotl(); * cmplxl z, w; * * ccotl( &z, &w ); * * * * DESCRIPTION: * * If *     z = x + iy, * * then * *           sin 2x  -  i sinh 2y *     w  =  --------------------. *            cosh 2y  -  cos 2x * * On the real axis, the denominator has zeros at even * multiples of PI/2.  Near these points it is evaluated * by a Taylor series. * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      3000       6.5e-17     1.6e-17 *    IEEE      -10,+10     30000       9.2e-16     1.2e-16 * Also tested by ctan * ccot = 1 + i0. */void ccotl( z, w )register cmplxl *z, *w;{long double d;d = coshl(2.0L * z->i) - cosl(2.0L * z->r);if( fabsl(d) < 0.25L )	d = ctansl(z);if( d == 0.0L )	{	mtherr( "ccot", OVERFLOW );	w->r = MAXNUML;	w->i = MAXNUML;	return;	}w->r = sinl( 2.0L * z->r ) / d;w->i = -sinhl( 2.0L * z->i ) / d;}/* Program to subtract nearest integer multiple of PI *//* extended precision value of PI: */#ifdef UNKstatic double DP1 = 3.14159265160560607910E0;static double DP2 = 1.98418714791870343106E-9;static double DP3 = 1.14423774522196636802E-17;#endif#ifdef DECstatic unsigned short P1[] = {0040511,0007732,0120000,0000000,};static unsigned short P2[] = {0031010,0055060,0100000,0000000,};static unsigned short P3[] = {0022123,0011431,0105056,0001560,};#define DP1 *(double *)P1#define DP2 *(double *)P2#define DP3 *(double *)P3#endif#ifdef IBMPCstatic unsigned short P1[] = {0x0000,0x5400,0x21fb,0x4009};static unsigned short P2[] = {0x0000,0x1000,0x0b46,0x3e21};static unsigned short P3[] = {0xc06e,0x3145,0x6263,0x3c6a};#define DP1 *(double *)P1#define DP2 *(double *)P2#define DP3 *(double *)P3#endif#ifdef MIEEEstatic unsigned short P1[] = {0x4009,0x21fb,0x5400,0x0000};static unsigned short P2[] = {0x3e21,0x0b46,0x1000,0x0000};static unsigned short P3[] = {0x3c6a,0x6263,0x3145,0xc06e};#define DP1 *(double *)P1#define DP2 *(double *)P2#define DP3 *(double *)P3#endifstatic long double redupil(x)long double x;{long double t;long i;t = x/PIL;if( t >= 0.0L )	t += 0.5L;else	t -= 0.5L;i = t;	/* the multiple */t = i;t = ((x - t * DP1) - t * DP2) - t * DP3;return(t);}/*  Taylor series expansion for cosh(2y) - cos(2x)	*/static long double ctansl(z)cmplxl *z;{long double f, x, x2, y, y2, rn, t;long double d;x = fabsl( 2.0L * z->r );y = fabsl( 2.0L * z->i );x = redupil(x);x = x * x;y = y * y;x2 = 1.0L;y2 = 1.0L;f = 1.0L;rn = 0.0;d = 0.0;do	{	rn += 1.0L;	f *= rn;	rn += 1.0L;	f *= rn;	x2 *= x;	y2 *= y;	t = y2 + x2;	t /= f;	d += t;	rn += 1.0L;	f *= rn;	rn += 1.0L;	f *= rn;	x2 *= x;	y2 *= y;	t = y2 - x2;	t /= f;	d += t;	}while( fabsl(t/d) > MACHEPL );return(d);}/*							casinl() * *	Complex circular arc sine * * * * SYNOPSIS: * * void casinl(); * cmplxl z, w; * * casinl( &z, &w ); * * * * DESCRIPTION: * * Inverse complex sine: * *                               2 * w = -i clog( iz + csqrt( 1 - z ) ). * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10     10100       2.1e-15     3.4e-16 *    IEEE      -10,+10     30000       2.2e-14     2.7e-15 * Larger relative error can be observed for z near zero. * Also tested by csin(casin(z)) = z. */void casinl( z, w )cmplxl *z, *w;{static cmplxl ca, ct, zz, z2;long double x, y;x = z->r;y = z->i;if( y == 0.0L )	{	if( fabsl(x) > 1.0L )		{		w->r = PIO2L;		w->i = 0.0L;		mtherr( "casinl", DOMAIN );		}	else		{		w->r = asinl(x);		w->i = 0.0L;		}	return;	}/* Power series expansion *//*b = cabsl(z);if( b < 0.125L ){z2.r = (x - y) * (x + y);z2.i = 2.0L * x * y;cn = 1.0L;n = 1.0L;ca.r = x;ca.i = y;sum.r = x;sum.i = y;do	{	ct.r = z2.r * ca.r  -  z2.i * ca.i;	ct.i = z2.r * ca.i  +  z2.i * ca.r;	ca.r = ct.r;	ca.i = ct.i;	cn *= n;	n += 1.0L;	cn /= n;	n += 1.0L;	b = cn/n;	ct.r *= b;	ct.i *= b;	sum.r += ct.r;	sum.i += ct.i;	b = fabsl(ct.r) + fabs(ct.i);	}while( b > MACHEPL );w->r = sum.r;w->i = sum.i;return;}*/ca.r = x;ca.i = y;ct.r = -ca.i;	/* iz */ct.i = ca.r;	/* sqrt( 1 - z*z) *//* cmul( &ca, &ca, &zz ) */zz.r = (ca.r - ca.i) * (ca.r + ca.i);	/*x * x  -  y * y */zz.i = 2.0L * ca.r * ca.i;zz.r = 1.0L - zz.r;zz.i = -zz.i;csqrtl( &zz, &z2 );caddl( &z2, &ct, &zz );clogl( &zz, &zz );w->r = zz.i;	/* mult by 1/i = -i */w->i = -zz.r;return;}/*							cacosl() * *	Complex circular arc cosine * * * * SYNOPSIS: * * void cacosl(); * cmplxl z, w; * * cacosl( &z, &w ); * * * * DESCRIPTION: * * * w = arccos z  =  PI/2 - arcsin z. * * * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      5200      1.6e-15      2.8e-16 *    IEEE      -10,+10     30000      1.8e-14      2.2e-15 */void cacosl( z, w )cmplxl *z, *w;{casinl( z, w );w->r = PIO2L  -  w->r;w->i = -w->i;}/*							catanl() * *	Complex circular arc tangent * * * * SYNOPSIS: * * void catanl(); * cmplxl z, w; * * catanl( &z, &w ); * * * * DESCRIPTION: * * If *     z = x + iy, * * then *          1       (    2x     ) * Re w  =  - arctan(-----------)  +  k PI *          2       (     2    2) *                  (1 - x  - y ) * *               ( 2         2) *          1    (x  +  (y+1) ) * Im w  =  - log(------------) *          4    ( 2         2) *               (x  +  (y-1) ) * * Where k is an arbitrary integer. * * * * ACCURACY: * *                      Relative error: * arithmetic   domain     # trials      peak         rms *    DEC       -10,+10      5900       1.3e-16     7.8e-18 *    IEEE      -10,+10     30000       2.3e-15     8.5e-17 * The check catan( ctan(z) )  =  z, with |x| and |y| < PI/2, * had peak relative error 1.5e-16, rms relative error * 2.9e-17.  See also clog(). */void catanl( z, w )cmplxl *z, *w;{long double a, t, x, x2, y;x = z->r;y = z->i;if( (x == 0.0L) && (y > 1.0L) )	goto ovrf;x2 = x * x;a = 1.0L - x2 - (y * y);if( a == 0.0L )	goto ovrf;#if ANSICt = atan2l( 2.0L * x, a ) * 0.5L;#elset = atan2l( a, 2.0 * x ) * 0.5L;#endifw->r = redupil( t );t = y - 1.0L;a = x2 + (t * t);if( a == 0.0L )	goto ovrf;t = y + 1.0L;a = (x2 + (t * t))/a;w->i = logl(a)/4.0;return;ovrf:mtherr( "catanl", OVERFLOW );w->r = MAXNUML;w->i = MAXNUML;}
 |