File: precess2.cpp

package info (click to toggle)
pluto-lunar 0.0~git20180825.e34c1d1-1
  • links: PTS, VCS
  • area: main
  • in suites: bookworm, bullseye, buster, forky, sid, trixie
  • size: 1,584 kB
  • sloc: cpp: 18,100; makefile: 653; ansic: 368
file content (244 lines) | stat: -rw-r--r-- 8,010 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
/* precess2.cpp: (deprecated version of) functions for computing
Earth precession;  see precess.cpp for current version,  and
'changes.txt' for info on why this is deprecated

Copyright (C) 2010, Project Pluto

This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2
of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
02110-1301, USA.    */

#include <math.h>
#include <string.h>
#include <stdio.h>
#include "watdefs.h"
#include "afuncs.h"
#include "lunar.h"         /* for obliquity( ) prototype */

#define PI 3.1415926535897932384626433832795028841971693993751058209749445923


/* setup_precession fills a 3x3 orthonormal matrix for precessing positions FROM */
/* year t1 TO year t2,  where t1 and t2 are Julian YEARS.  */

int DLL_FUNC setup_precession( double DLLPTR *matrix, double t1,
                               double t2)
{
   double zeta, z, theta, czeta, cz, ctheta, szeta, sz, stheta;
   double ka, kb;
   static double t1_old = -PI, t2_old;
   static double curr_matrix[9];
   int going_backward = 0;

   if( fabs( t1 - t2) < 1.e-5)   /* dates sensibly equal;  spare the tedium */
      {                          /* of doing pointless math */
      set_identity_matrix( matrix);
      return( 0);
      }

         /* Ideally,  precessing from t1 to t2 back to t1 should get your  */
         /* original point.  To ensure that this happens,  we handle only  */
         /* the case t2 > t1;  otherwise,  we swap the times and invert    */
         /* the resulting matrix.                                          */
         /*   The reason is that the following precession formula uses     */
         /* cubic polynomials to approximate zeta,  theta,  and z.  If     */
         /* you feed it (t2,  t1),  it does _not_ create a matrix that is  */
         /* the exact inverse of (t1,  t2);  there is some accumulated     */
         /* error.  Doing it this way avoids having that show.  Also,      */
         /* there is a performance advantage:  if you _do_ call (t1, t2),  */
         /* then (t2, t1),  it's faster to invert the previous result      */
         /* than it would be to do all the math.                           */
   if( t1 < t2)
      {
      double temp = t1;

      t1 = t2;
      t2 = temp;
      going_backward = 1;
      }
         /* It's pretty common to precess a few zillion data points.  So   */
         /* it helps to cache the most recently computed precession matrix */
         /* so that repeated calls don't result in repeated computation.   */
   if( t1 == t1_old && t2 == t2_old)
      {
      FMEMCPY( matrix, curr_matrix, 9 * sizeof( double));
      if( going_backward)
         invert_orthonormal_matrix( matrix);
      return( 0);
      }
   t1_old = t1;
   t2_old = t2;

   t2 = (t2 - t1) / 100.;
   t1 = (t1 - 2000.) / 100.;
   ka = 2306.2181 + 1.39656 * t1 - .000139 * t1 * t1;
   kb = 2004.3109 - 0.85330 * t1 - .000217 * t1 * t1;
   zeta  = t2 * (ka + t2 * ( .30188 - .000345 * t1 + .017998 * t2));
   z     = t2 * (ka + t2 * (1.09468 + .000066 * t1 + .018203 * t2));
   theta = t2 * (kb + t2 * (-.42665 - .000217 * t1 - .041833 * t2));
   theta *= (PI / 180.) / 3600.;
   z     *= (PI / 180.) / 3600.;
   zeta  *= (PI / 180.) / 3600.;
   czeta = cos( zeta);
   szeta = sin( zeta);
   cz = cos( z);
   sz = sin( z);
   ctheta = cos( theta);
   stheta = sin( theta);

   *matrix++ = czeta * ctheta * cz - szeta * sz;
   *matrix++ = -szeta * ctheta * cz - czeta * sz;
   *matrix++ = -stheta * cz;

   *matrix++ = czeta * ctheta * sz + szeta * cz;
   *matrix++ = -szeta * ctheta * sz + czeta * cz;
   *matrix++ = -stheta * sz;

   *matrix++ = czeta * stheta;
   *matrix++ = -szeta * stheta;
   *matrix++ = ctheta;

   matrix -= 9;
   FMEMCPY( curr_matrix, matrix, 9 * sizeof( double));
   if( going_backward)
      invert_orthonormal_matrix( matrix);
   return( 0);
}

static const double sin_obliq_2000 = 0.397777155931913701597179975942380896684;
static const double cos_obliq_2000 = 0.917482062069181825744000384639406458043;

void DLL_FUNC equatorial_to_ecliptic( double *vect)
{
   double temp;

   temp    = vect[2] * cos_obliq_2000 - vect[1] * sin_obliq_2000;
   vect[1] = vect[1] * cos_obliq_2000 + vect[2] * sin_obliq_2000;
   vect[2] = temp;
}

void DLL_FUNC ecliptic_to_equatorial( double *vect)
{
   double temp;

   temp    = vect[2] * cos_obliq_2000 + vect[1] * sin_obliq_2000;
   vect[1] = vect[1] * cos_obliq_2000 - vect[2] * sin_obliq_2000;
   vect[2] = temp;
}

int DLL_FUNC precess_vector( const double DLLPTR *matrix,
                                      const double DLLPTR *v1,
                                      double DLLPTR *v2)
{
   int i = 3;

   while( i--)
      {
      *v2++ = matrix[0] * v1[0] + matrix[1] * v1[1] + matrix[2] * v1[2];
      matrix += 3;
      }
   return( 0);
}

int DLL_FUNC deprecess_vector( const double DLLPTR *matrix,
                                      const double DLLPTR *v1,
                                      double DLLPTR *v2)
{
   int i = 3;

   while( i--)
      {
      *v2++ = matrix[0] * v1[0] + matrix[3] * v1[1] + matrix[6] * v1[2];
      matrix++;
      }
   return( 0);
}

int DLL_FUNC precess_ra_dec( const double DLLPTR *matrix,
                        double DLLPTR *p_out,
                        const double DLLPTR *p_in, int backward)
{
   double v1[3], v2[3];
   const double old_ra = p_in[0];

   v1[0] = cos( p_in[0]) * cos( p_in[1]);
   v1[1] = sin( p_in[0]) * cos( p_in[1]);
   v1[2] =                 sin( p_in[1]);
   if( backward)
      deprecess_vector( matrix, v1, v2);
   else
      precess_vector( matrix, v1, v2);
   if( v2[1] || v2[0])
      p_out[0] = atan2( v2[1], v2[0]);
   else
      p_out[0] = 0.;
   p_out[1] = asine( v2[2]);
   while( p_out[0] - old_ra > PI)
      p_out[0] -= PI * 2.;
   while( p_out[0] - old_ra <-PI)
      p_out[0] += PI * 2.;
   return( 0);
}

/* setup_ecliptic_precession fills a 3x3 orthonormal matrix for precessing */
/* positions _in ecliptic coordinates_ FROM year t1 TO year t2,  where t1  */
/* and t2 are Julian YEARS... much as setup_precession( ) does for RA/dec  */

/* 30 May 2002:  change 'obliquity#' to '-obliquity#' to fix a bug reported */
/* by Jordi Mas,  probably in place since the code was written.             */

int DLL_FUNC setup_ecliptic_precession( double DLLPTR *matrix, const double t1,
                              const double t2)
{
   const double obliquity1 = mean_obliquity( (t1 - 2000.) / 100.);
   const double obliquity2 = mean_obliquity( (t2 - 2000.) / 100.);

   setup_precession( matrix, t1, t2);
   pre_spin_matrix( matrix + 1, matrix + 2, -obliquity1);
   spin_matrix( matrix + 3, matrix + 6, -obliquity2);
   return( 0);
}

#ifdef TEST_MAIN
#include <stdio.h>
#include <stdlib.h>

int main( const int argc, const char **argv)
{
   double t1, t2, matrix[9];
   double p[2];
   int i;

   t1 = atof( argv[1]);
   t2 = atof( argv[2]);
   if( argc > 3)
      {
      p[0] = atof( argv[3]) * PI / 180.;
      p[1] = atof( argv[4]) * PI / 180.;
      }
   if( argc < 6)
      setup_precession( matrix, t1, t2);
   else
      setup_ecliptic_precession( matrix, t1, t2);
   for( i = 0; i < 9; i++)
      printf( "%15.11lf%s", matrix[i], (i % 3 == 2) ? "\n" : " ");
   if( argc > 3)
      {
      precess_ra_dec( matrix, p, p, 0);
      printf( "%lf %lf\n", p[0] * 180. / PI, p[1] * 180. / PI);
      precess_ra_dec( matrix, p, p, 1);
      printf( "%lf %lf\n", p[0] * 180. / PI, p[1] * 180. / PI);
      }
}
#endif