1 | (*
|
---|
2 | (c) copyright 1988 by the Vrije Universiteit, Amsterdam, The Netherlands.
|
---|
3 | See the copyright notice in the ACK home directory, in the file "Copyright".
|
---|
4 | *)
|
---|
5 |
|
---|
6 | (*$R-*)
|
---|
7 | IMPLEMENTATION MODULE Mathlib;
|
---|
8 | (*
|
---|
9 | Module: Mathematical functions
|
---|
10 | Author: Ceriel J.H. Jacobs
|
---|
11 | Version: $Header: /cvsup/minix/src/lib/ack/libm2/Mathlib.mod,v 1.1 2005/10/10 15:27:46 beng Exp $
|
---|
12 | *)
|
---|
13 |
|
---|
14 | FROM EM IMPORT FIF, FEF;
|
---|
15 | FROM Traps IMPORT Message;
|
---|
16 |
|
---|
17 | CONST
|
---|
18 | OneRadianInDegrees = 57.295779513082320876798155D;
|
---|
19 | OneDegreeInRadians = 0.017453292519943295769237D;
|
---|
20 | OneOverSqrt2 = 0.70710678118654752440084436210484904D;
|
---|
21 |
|
---|
22 | (* basic functions *)
|
---|
23 |
|
---|
24 | PROCEDURE pow(x: REAL; i: INTEGER): REAL;
|
---|
25 | BEGIN
|
---|
26 | RETURN SHORT(longpow(LONG(x), i));
|
---|
27 | END pow;
|
---|
28 |
|
---|
29 | PROCEDURE longpow(x: LONGREAL; i: INTEGER): LONGREAL;
|
---|
30 | VAR val: LONGREAL;
|
---|
31 | ri: LONGREAL;
|
---|
32 | BEGIN
|
---|
33 | ri := FLOATD(i);
|
---|
34 | IF x < 0.0D THEN
|
---|
35 | val := longexp(longln(-x) * ri);
|
---|
36 | IF ODD(i) THEN RETURN -val;
|
---|
37 | ELSE RETURN val;
|
---|
38 | END;
|
---|
39 | ELSIF x = 0.0D THEN
|
---|
40 | RETURN 0.0D;
|
---|
41 | ELSE
|
---|
42 | RETURN longexp(longln(x) * ri);
|
---|
43 | END;
|
---|
44 | END longpow;
|
---|
45 |
|
---|
46 | PROCEDURE sqrt(x: REAL): REAL;
|
---|
47 | BEGIN
|
---|
48 | RETURN SHORT(longsqrt(LONG(x)));
|
---|
49 | END sqrt;
|
---|
50 |
|
---|
51 | PROCEDURE longsqrt(x: LONGREAL): LONGREAL;
|
---|
52 | VAR
|
---|
53 | temp: LONGREAL;
|
---|
54 | exp, i: INTEGER;
|
---|
55 | BEGIN
|
---|
56 | IF x <= 0.0D THEN
|
---|
57 | IF x < 0.0D THEN
|
---|
58 | Message("sqrt: negative argument");
|
---|
59 | HALT
|
---|
60 | END;
|
---|
61 | RETURN 0.0D;
|
---|
62 | END;
|
---|
63 | temp := FEF(x,exp);
|
---|
64 | (*
|
---|
65 | * NOTE
|
---|
66 | * this wont work on 1's comp
|
---|
67 | *)
|
---|
68 | IF ODD(exp) THEN
|
---|
69 | temp := 2.0D * temp;
|
---|
70 | DEC(exp);
|
---|
71 | END;
|
---|
72 | temp := 0.5D*(1.0D + temp);
|
---|
73 |
|
---|
74 | WHILE exp > 28 DO
|
---|
75 | temp := temp * 16384.0D;
|
---|
76 | exp := exp - 28;
|
---|
77 | END;
|
---|
78 | WHILE exp < -28 DO
|
---|
79 | temp := temp / 16384.0D;
|
---|
80 | exp := exp + 28;
|
---|
81 | END;
|
---|
82 | WHILE exp >= 2 DO
|
---|
83 | temp := temp * 2.0D;
|
---|
84 | exp := exp - 2;
|
---|
85 | END;
|
---|
86 | WHILE exp <= -2 DO
|
---|
87 | temp := temp / 2.0D;
|
---|
88 | exp := exp + 2;
|
---|
89 | END;
|
---|
90 | FOR i := 0 TO 5 DO
|
---|
91 | temp := 0.5D*(temp + x/temp);
|
---|
92 | END;
|
---|
93 | RETURN temp;
|
---|
94 | END longsqrt;
|
---|
95 |
|
---|
96 | PROCEDURE ldexp(x:LONGREAL; n: INTEGER): LONGREAL;
|
---|
97 | BEGIN
|
---|
98 | WHILE n >= 16 DO
|
---|
99 | x := x * 65536.0D;
|
---|
100 | n := n - 16;
|
---|
101 | END;
|
---|
102 | WHILE n > 0 DO
|
---|
103 | x := x * 2.0D;
|
---|
104 | DEC(n);
|
---|
105 | END;
|
---|
106 | WHILE n <= -16 DO
|
---|
107 | x := x / 65536.0D;
|
---|
108 | n := n + 16;
|
---|
109 | END;
|
---|
110 | WHILE n < 0 DO
|
---|
111 | x := x / 2.0D;
|
---|
112 | INC(n);
|
---|
113 | END;
|
---|
114 | RETURN x;
|
---|
115 | END ldexp;
|
---|
116 |
|
---|
117 | PROCEDURE exp(x: REAL): REAL;
|
---|
118 | BEGIN
|
---|
119 | RETURN SHORT(longexp(LONG(x)));
|
---|
120 | END exp;
|
---|
121 |
|
---|
122 | PROCEDURE longexp(x: LONGREAL): LONGREAL;
|
---|
123 | (* Algorithm and coefficients from:
|
---|
124 | "Software manual for the elementary functions"
|
---|
125 | by W.J. Cody and W. Waite, Prentice-Hall, 1980
|
---|
126 | *)
|
---|
127 | CONST
|
---|
128 | p0 = 0.25000000000000000000D+00;
|
---|
129 | p1 = 0.75753180159422776666D-02;
|
---|
130 | p2 = 0.31555192765684646356D-04;
|
---|
131 | q0 = 0.50000000000000000000D+00;
|
---|
132 | q1 = 0.56817302698551221787D-01;
|
---|
133 | q2 = 0.63121894374398503557D-03;
|
---|
134 | q3 = 0.75104028399870046114D-06;
|
---|
135 |
|
---|
136 | VAR
|
---|
137 | neg: BOOLEAN;
|
---|
138 | n: INTEGER;
|
---|
139 | xn, g, x1, x2: LONGREAL;
|
---|
140 | BEGIN
|
---|
141 | neg := x < 0.0D;
|
---|
142 | IF neg THEN
|
---|
143 | x := -x;
|
---|
144 | END;
|
---|
145 | n := TRUNC(x/longln2 + 0.5D);
|
---|
146 | xn := FLOATD(n);
|
---|
147 | x1 := FLOATD(TRUNCD(x));
|
---|
148 | x2 := x - x1;
|
---|
149 | g := ((x1 - xn * 0.693359375D)+x2) - xn * (-2.1219444005469058277D-4);
|
---|
150 | IF neg THEN
|
---|
151 | g := -g;
|
---|
152 | n := -n;
|
---|
153 | END;
|
---|
154 | xn := g*g;
|
---|
155 | x := g*((p2*xn+p1)*xn+p0);
|
---|
156 | INC(n);
|
---|
157 | RETURN ldexp(0.5D + x/((((q3*xn+q2)*xn+q1)*xn+q0) - x), n);
|
---|
158 | END longexp;
|
---|
159 |
|
---|
160 | PROCEDURE ln(x: REAL): REAL; (* natural log *)
|
---|
161 | BEGIN
|
---|
162 | RETURN SHORT(longln(LONG(x)));
|
---|
163 | END ln;
|
---|
164 |
|
---|
165 | PROCEDURE longln(x: LONGREAL): LONGREAL; (* natural log *)
|
---|
166 | (* Algorithm and coefficients from:
|
---|
167 | "Software manual for the elementary functions"
|
---|
168 | by W.J. Cody and W. Waite, Prentice-Hall, 1980
|
---|
169 | *)
|
---|
170 | CONST
|
---|
171 | p0 = -0.64124943423745581147D+02;
|
---|
172 | p1 = 0.16383943563021534222D+02;
|
---|
173 | p2 = -0.78956112887491257267D+00;
|
---|
174 | q0 = -0.76949932108494879777D+03;
|
---|
175 | q1 = 0.31203222091924532844D+03;
|
---|
176 | q2 = -0.35667977739034646171D+02;
|
---|
177 | q3 = 1.0D;
|
---|
178 | VAR
|
---|
179 | exp: INTEGER;
|
---|
180 | z, znum, zden, w: LONGREAL;
|
---|
181 |
|
---|
182 | BEGIN
|
---|
183 | IF x <= 0.0D THEN
|
---|
184 | Message("ln: argument <= 0");
|
---|
185 | HALT
|
---|
186 | END;
|
---|
187 | x := FEF(x, exp);
|
---|
188 | IF x > OneOverSqrt2 THEN
|
---|
189 | znum := (x - 0.5D) - 0.5D;
|
---|
190 | zden := x * 0.5D + 0.5D;
|
---|
191 | ELSE
|
---|
192 | znum := x - 0.5D;
|
---|
193 | zden := znum * 0.5D + 0.5D;
|
---|
194 | DEC(exp);
|
---|
195 | END;
|
---|
196 | z := znum / zden;
|
---|
197 | w := z * z;
|
---|
198 | x := z + z * w * (((p2*w+p1)*w+p0)/(((q3*w+q2)*w+q1)*w+q0));
|
---|
199 | z := FLOATD(exp);
|
---|
200 | x := x + z * (-2.121944400546905827679D-4);
|
---|
201 | RETURN x + z * 0.693359375D;
|
---|
202 | END longln;
|
---|
203 |
|
---|
204 | PROCEDURE log(x: REAL): REAL; (* log with base 10 *)
|
---|
205 | BEGIN
|
---|
206 | RETURN SHORT(longlog(LONG(x)));
|
---|
207 | END log;
|
---|
208 |
|
---|
209 | PROCEDURE longlog(x: LONGREAL): LONGREAL; (* log with base 10 *)
|
---|
210 | BEGIN
|
---|
211 | RETURN longln(x)/longln10;
|
---|
212 | END longlog;
|
---|
213 |
|
---|
214 | (* trigonometric functions; arguments in radians *)
|
---|
215 |
|
---|
216 | PROCEDURE sin(x: REAL): REAL;
|
---|
217 | BEGIN
|
---|
218 | RETURN SHORT(longsin(LONG(x)));
|
---|
219 | END sin;
|
---|
220 |
|
---|
221 | PROCEDURE sinus(x: LONGREAL; cosflag: BOOLEAN) : LONGREAL;
|
---|
222 | (* Algorithm and coefficients from:
|
---|
223 | "Software manual for the elementary functions"
|
---|
224 | by W.J. Cody and W. Waite, Prentice-Hall, 1980
|
---|
225 | *)
|
---|
226 | CONST
|
---|
227 | r0 = -0.16666666666666665052D+00;
|
---|
228 | r1 = 0.83333333333331650314D-02;
|
---|
229 | r2 = -0.19841269841201840457D-03;
|
---|
230 | r3 = 0.27557319210152756119D-05;
|
---|
231 | r4 = -0.25052106798274584544D-07;
|
---|
232 | r5 = 0.16058936490371589114D-09;
|
---|
233 | r6 = -0.76429178068910467734D-12;
|
---|
234 | r7 = 0.27204790957888846175D-14;
|
---|
235 | A1 = 3.1416015625D;
|
---|
236 | A2 = -8.908910206761537356617D-6;
|
---|
237 | VAR
|
---|
238 | x1, x2, y : LONGREAL;
|
---|
239 | neg : BOOLEAN;
|
---|
240 | BEGIN
|
---|
241 | IF x < 0.0D THEN
|
---|
242 | neg := TRUE;
|
---|
243 | x := -x
|
---|
244 | ELSE neg := FALSE
|
---|
245 | END;
|
---|
246 | IF cosflag THEN
|
---|
247 | neg := FALSE;
|
---|
248 | y := longhalfpi + x
|
---|
249 | ELSE
|
---|
250 | y := x
|
---|
251 | END;
|
---|
252 | y := y / longpi + 0.5D;
|
---|
253 |
|
---|
254 | IF FIF(y, 1.0D, y) < 0.0D THEN ; END;
|
---|
255 | IF FIF(y, 0.5D, x1) # 0.0D THEN neg := NOT neg END;
|
---|
256 | IF cosflag THEN y := y - 0.5D END;
|
---|
257 | x2 := FIF(x, 1.0, x1);
|
---|
258 | x := x1 - y * A1;
|
---|
259 | x := x + x2;
|
---|
260 | x := x - y * A2;
|
---|
261 |
|
---|
262 | IF x < 0.0D THEN
|
---|
263 | neg := NOT neg;
|
---|
264 | x := -x
|
---|
265 | END;
|
---|
266 | y := x * x;
|
---|
267 | x := x + x * y * (((((((r7*y+r6)*y+r5)*y+r4)*y+r3)*y+r2)*y+r1)*y+r0);
|
---|
268 | IF neg THEN RETURN -x END;
|
---|
269 | RETURN x;
|
---|
270 | END sinus;
|
---|
271 |
|
---|
272 | PROCEDURE longsin(x: LONGREAL): LONGREAL;
|
---|
273 | BEGIN
|
---|
274 | RETURN sinus(x, FALSE);
|
---|
275 | END longsin;
|
---|
276 |
|
---|
277 | PROCEDURE cos(x: REAL): REAL;
|
---|
278 | BEGIN
|
---|
279 | RETURN SHORT(longcos(LONG(x)));
|
---|
280 | END cos;
|
---|
281 |
|
---|
282 | PROCEDURE longcos(x: LONGREAL): LONGREAL;
|
---|
283 | BEGIN
|
---|
284 | IF x < 0.0D THEN x := -x; END;
|
---|
285 | RETURN sinus(x, TRUE);
|
---|
286 | END longcos;
|
---|
287 |
|
---|
288 | PROCEDURE tan(x: REAL): REAL;
|
---|
289 | BEGIN
|
---|
290 | RETURN SHORT(longtan(LONG(x)));
|
---|
291 | END tan;
|
---|
292 |
|
---|
293 | PROCEDURE longtan(x: LONGREAL): LONGREAL;
|
---|
294 | (* Algorithm and coefficients from:
|
---|
295 | "Software manual for the elementary functions"
|
---|
296 | by W.J. Cody and W. Waite, Prentice-Hall, 1980
|
---|
297 | *)
|
---|
298 |
|
---|
299 | CONST
|
---|
300 | p1 = -0.13338350006421960681D+00;
|
---|
301 | p2 = 0.34248878235890589960D-02;
|
---|
302 | p3 = -0.17861707342254426711D-04;
|
---|
303 |
|
---|
304 | q0 = 1.0D;
|
---|
305 | q1 = -0.46671683339755294240D+00;
|
---|
306 | q2 = 0.25663832289440112864D-01;
|
---|
307 | q3 = -0.31181531907010027307D-03;
|
---|
308 | q4 = 0.49819433993786512270D-06;
|
---|
309 |
|
---|
310 | A1 = 1.57080078125D;
|
---|
311 | A2 = -4.454455103380768678308D-06;
|
---|
312 |
|
---|
313 | VAR y, x1, x2: LONGREAL;
|
---|
314 | negative: BOOLEAN;
|
---|
315 | invert: BOOLEAN;
|
---|
316 | BEGIN
|
---|
317 | negative := x < 0.0D;
|
---|
318 | y := x / longhalfpi + 0.5D;
|
---|
319 |
|
---|
320 | (* Use extended precision to calculate reduced argument.
|
---|
321 | Here we used 12 bits of the mantissa for a1.
|
---|
322 | Also split x in integer part x1 and fraction part x2.
|
---|
323 | *)
|
---|
324 | IF FIF(y, 1.0D, y) < 0.0D THEN ; END;
|
---|
325 | invert := FIF(y, 0.5D, x1) # 0.0D;
|
---|
326 | x2 := FIF(x, 1.0D, x1);
|
---|
327 | x := x1 - y * A1;
|
---|
328 | x := x + x2;
|
---|
329 | x := x - y * A2;
|
---|
330 |
|
---|
331 | y := x * x;
|
---|
332 | x := x + x * y * ((p3*y+p2)*y+p1);
|
---|
333 | y := (((q4*y+q3)*y+q2)*y+q1)*y+q0;
|
---|
334 | IF negative THEN x := -x END;
|
---|
335 | IF invert THEN RETURN -y/x END;
|
---|
336 | RETURN x/y;
|
---|
337 | END longtan;
|
---|
338 |
|
---|
339 | PROCEDURE arcsin(x: REAL): REAL;
|
---|
340 | BEGIN
|
---|
341 | RETURN SHORT(longarcsin(LONG(x)));
|
---|
342 | END arcsin;
|
---|
343 |
|
---|
344 | PROCEDURE arcsincos(x: LONGREAL; cosfl: BOOLEAN): LONGREAL;
|
---|
345 | CONST
|
---|
346 | p0 = -0.27368494524164255994D+02;
|
---|
347 | p1 = 0.57208227877891731407D+02;
|
---|
348 | p2 = -0.39688862997540877339D+02;
|
---|
349 | p3 = 0.10152522233806463645D+02;
|
---|
350 | p4 = -0.69674573447350646411D+00;
|
---|
351 |
|
---|
352 | q0 = -0.16421096714498560795D+03;
|
---|
353 | q1 = 0.41714430248260412556D+03;
|
---|
354 | q2 = -0.38186303361750149284D+03;
|
---|
355 | q3 = 0.15095270841030604719D+03;
|
---|
356 | q4 = -0.23823859153670238830D+02;
|
---|
357 | q5 = 1.0D;
|
---|
358 | VAR
|
---|
359 | negative : BOOLEAN;
|
---|
360 | big: BOOLEAN;
|
---|
361 | g: LONGREAL;
|
---|
362 | BEGIN
|
---|
363 | negative := x < 0.0D;
|
---|
364 | IF negative THEN x := -x; END;
|
---|
365 | IF x > 0.5D THEN
|
---|
366 | big := TRUE;
|
---|
367 | IF x > 1.0D THEN
|
---|
368 | Message("arcsin or arccos: argument > 1");
|
---|
369 | HALT
|
---|
370 | END;
|
---|
371 | g := 0.5D - 0.5D * x;
|
---|
372 | x := -longsqrt(g);
|
---|
373 | x := x + x;
|
---|
374 | ELSE
|
---|
375 | big := FALSE;
|
---|
376 | g := x * x;
|
---|
377 | END;
|
---|
378 | x := x + x * g *
|
---|
379 | ((((p4*g+p3)*g+p2)*g+p1)*g+p0)/(((((q5*g+q4)*g+q3)*g+q2)*g+q1)*g+q0);
|
---|
380 | IF cosfl AND NOT negative THEN x := -x END;
|
---|
381 | IF cosfl = NOT big THEN
|
---|
382 | x := (x + longquartpi) + longquartpi;
|
---|
383 | ELSIF cosfl AND negative AND big THEN
|
---|
384 | x := (x + longhalfpi) + longhalfpi;
|
---|
385 | END;
|
---|
386 | IF negative AND NOT cosfl THEN x := -x END;
|
---|
387 | RETURN x;
|
---|
388 | END arcsincos;
|
---|
389 |
|
---|
390 | PROCEDURE longarcsin(x: LONGREAL): LONGREAL;
|
---|
391 | BEGIN
|
---|
392 | RETURN arcsincos(x, FALSE);
|
---|
393 | END longarcsin;
|
---|
394 |
|
---|
395 | PROCEDURE arccos(x: REAL): REAL;
|
---|
396 | BEGIN
|
---|
397 | RETURN SHORT(longarccos(LONG(x)));
|
---|
398 | END arccos;
|
---|
399 |
|
---|
400 | PROCEDURE longarccos(x: LONGREAL): LONGREAL;
|
---|
401 | BEGIN
|
---|
402 | RETURN arcsincos(x, TRUE);
|
---|
403 | END longarccos;
|
---|
404 |
|
---|
405 | PROCEDURE arctan(x: REAL): REAL;
|
---|
406 | BEGIN
|
---|
407 | RETURN SHORT(longarctan(LONG(x)));
|
---|
408 | END arctan;
|
---|
409 |
|
---|
410 | VAR A: ARRAY[0..3] OF LONGREAL;
|
---|
411 | arctaninit: BOOLEAN;
|
---|
412 |
|
---|
413 | PROCEDURE longarctan(x: LONGREAL): LONGREAL;
|
---|
414 | (* Algorithm and coefficients from:
|
---|
415 | "Software manual for the elementary functions"
|
---|
416 | by W.J. Cody and W. Waite, Prentice-Hall, 1980
|
---|
417 | *)
|
---|
418 | CONST
|
---|
419 | p0 = -0.13688768894191926929D+02;
|
---|
420 | p1 = -0.20505855195861651981D+02;
|
---|
421 | p2 = -0.84946240351320683534D+01;
|
---|
422 | p3 = -0.83758299368150059274D+00;
|
---|
423 | q0 = 0.41066306682575781263D+02;
|
---|
424 | q1 = 0.86157349597130242515D+02;
|
---|
425 | q2 = 0.59578436142597344465D+02;
|
---|
426 | q3 = 0.15024001160028576121D+02;
|
---|
427 | q4 = 1.0D;
|
---|
428 | VAR
|
---|
429 | g: LONGREAL;
|
---|
430 | neg: BOOLEAN;
|
---|
431 | n: INTEGER;
|
---|
432 | BEGIN
|
---|
433 | IF NOT arctaninit THEN
|
---|
434 | arctaninit := TRUE;
|
---|
435 | A[0] := 0.0D;
|
---|
436 | A[1] := 0.52359877559829887307710723554658381D; (* p1/6 *)
|
---|
437 | A[2] := longhalfpi;
|
---|
438 | A[3] := 1.04719755119659774615421446109316763D; (* pi/3 *)
|
---|
439 | END;
|
---|
440 | neg := FALSE;
|
---|
441 | IF x < 0.0D THEN
|
---|
442 | neg := TRUE;
|
---|
443 | x := -x;
|
---|
444 | END;
|
---|
445 | IF x > 1.0D THEN
|
---|
446 | x := 1.0D/x;
|
---|
447 | n := 2
|
---|
448 | ELSE
|
---|
449 | n := 0
|
---|
450 | END;
|
---|
451 | IF x > 0.26794919243112270647D (* 2-sqrt(3) *) THEN
|
---|
452 | INC(n);
|
---|
453 | x := (((0.73205080756887729353D*x-0.5D)-0.5D)+x)/
|
---|
454 | (1.73205080756887729353D + x);
|
---|
455 | END;
|
---|
456 | g := x*x;
|
---|
457 | x := x + x * g * (((p3*g+p2)*g+p1)*g+p0) / ((((q4*g+q3)*g+q2)*g+q1)*g+q0);
|
---|
458 | IF n > 1 THEN x := -x END;
|
---|
459 | x := x + A[n];
|
---|
460 | IF neg THEN RETURN -x; END;
|
---|
461 | RETURN x;
|
---|
462 | END longarctan;
|
---|
463 |
|
---|
464 | (* hyperbolic functions *)
|
---|
465 | (* The C math library has better implementations for some of these, but
|
---|
466 | they depend on some properties of the floating point implementation,
|
---|
467 | and, for now, we don't want that in the Modula-2 system.
|
---|
468 | *)
|
---|
469 |
|
---|
470 | PROCEDURE sinh(x: REAL): REAL;
|
---|
471 | BEGIN
|
---|
472 | RETURN SHORT(longsinh(LONG(x)));
|
---|
473 | END sinh;
|
---|
474 |
|
---|
475 | PROCEDURE longsinh(x: LONGREAL): LONGREAL;
|
---|
476 | VAR expx: LONGREAL;
|
---|
477 | BEGIN
|
---|
478 | expx := longexp(x);
|
---|
479 | RETURN (expx - 1.0D/expx)/2.0D;
|
---|
480 | END longsinh;
|
---|
481 |
|
---|
482 | PROCEDURE cosh(x: REAL): REAL;
|
---|
483 | BEGIN
|
---|
484 | RETURN SHORT(longcosh(LONG(x)));
|
---|
485 | END cosh;
|
---|
486 |
|
---|
487 | PROCEDURE longcosh(x: LONGREAL): LONGREAL;
|
---|
488 | VAR expx: LONGREAL;
|
---|
489 | BEGIN
|
---|
490 | expx := longexp(x);
|
---|
491 | RETURN (expx + 1.0D/expx)/2.0D;
|
---|
492 | END longcosh;
|
---|
493 |
|
---|
494 | PROCEDURE tanh(x: REAL): REAL;
|
---|
495 | BEGIN
|
---|
496 | RETURN SHORT(longtanh(LONG(x)));
|
---|
497 | END tanh;
|
---|
498 |
|
---|
499 | PROCEDURE longtanh(x: LONGREAL): LONGREAL;
|
---|
500 | VAR expx: LONGREAL;
|
---|
501 | BEGIN
|
---|
502 | expx := longexp(x);
|
---|
503 | RETURN (expx - 1.0D/expx) / (expx + 1.0D/expx);
|
---|
504 | END longtanh;
|
---|
505 |
|
---|
506 | PROCEDURE arcsinh(x: REAL): REAL;
|
---|
507 | BEGIN
|
---|
508 | RETURN SHORT(longarcsinh(LONG(x)));
|
---|
509 | END arcsinh;
|
---|
510 |
|
---|
511 | PROCEDURE longarcsinh(x: LONGREAL): LONGREAL;
|
---|
512 | VAR neg: BOOLEAN;
|
---|
513 | BEGIN
|
---|
514 | neg := FALSE;
|
---|
515 | IF x < 0.0D THEN
|
---|
516 | neg := TRUE;
|
---|
517 | x := -x;
|
---|
518 | END;
|
---|
519 | x := longln(x + longsqrt(x*x+1.0D));
|
---|
520 | IF neg THEN RETURN -x; END;
|
---|
521 | RETURN x;
|
---|
522 | END longarcsinh;
|
---|
523 |
|
---|
524 | PROCEDURE arccosh(x: REAL): REAL;
|
---|
525 | BEGIN
|
---|
526 | RETURN SHORT(longarccosh(LONG(x)));
|
---|
527 | END arccosh;
|
---|
528 |
|
---|
529 | PROCEDURE longarccosh(x: LONGREAL): LONGREAL;
|
---|
530 | BEGIN
|
---|
531 | IF x < 1.0D THEN
|
---|
532 | Message("arccosh: argument < 1");
|
---|
533 | HALT
|
---|
534 | END;
|
---|
535 | RETURN longln(x + longsqrt(x*x - 1.0D));
|
---|
536 | END longarccosh;
|
---|
537 |
|
---|
538 | PROCEDURE arctanh(x: REAL): REAL;
|
---|
539 | BEGIN
|
---|
540 | RETURN SHORT(longarctanh(LONG(x)));
|
---|
541 | END arctanh;
|
---|
542 |
|
---|
543 | PROCEDURE longarctanh(x: LONGREAL): LONGREAL;
|
---|
544 | BEGIN
|
---|
545 | IF (x <= -1.0D) OR (x >= 1.0D) THEN
|
---|
546 | Message("arctanh: ABS(argument) >= 1");
|
---|
547 | HALT
|
---|
548 | END;
|
---|
549 | RETURN longln((1.0D + x)/(1.0D - x)) / 2.0D;
|
---|
550 | END longarctanh;
|
---|
551 |
|
---|
552 | (* conversions *)
|
---|
553 |
|
---|
554 | PROCEDURE RadianToDegree(x: REAL): REAL;
|
---|
555 | BEGIN
|
---|
556 | RETURN SHORT(longRadianToDegree(LONG(x)));
|
---|
557 | END RadianToDegree;
|
---|
558 |
|
---|
559 | PROCEDURE longRadianToDegree(x: LONGREAL): LONGREAL;
|
---|
560 | BEGIN
|
---|
561 | RETURN x * OneRadianInDegrees;
|
---|
562 | END longRadianToDegree;
|
---|
563 |
|
---|
564 | PROCEDURE DegreeToRadian(x: REAL): REAL;
|
---|
565 | BEGIN
|
---|
566 | RETURN SHORT(longDegreeToRadian(LONG(x)));
|
---|
567 | END DegreeToRadian;
|
---|
568 |
|
---|
569 | PROCEDURE longDegreeToRadian(x: LONGREAL): LONGREAL;
|
---|
570 | BEGIN
|
---|
571 | RETURN x * OneDegreeInRadians;
|
---|
572 | END longDegreeToRadian;
|
---|
573 |
|
---|
574 | BEGIN
|
---|
575 | arctaninit := FALSE;
|
---|
576 | END Mathlib.
|
---|