summaryrefslogtreecommitdiff
path: root/Master/texmf-dist/source/latex/l3kernel/l3fp-trig.dtx
blob: d6682a661c36aad1195c430cf08f6358ebdd4593 (plain)
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
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
% \iffalse meta-comment
%
%% File: l3fp-trig.dtx Copyright (C) 2011-2012 The LaTeX3 Project
%%
%% It may be distributed and/or modified under the conditions of the
%% LaTeX Project Public License (LPPL), either version 1.3c of this
%% license or (at your option) any later version.  The latest version
%% of this license is in the file
%%
%%    http://www.latex-project.org/lppl.txt
%%
%% This file is part of the "l3kernel bundle" (The Work in LPPL)
%% and all files in that bundle must be distributed together.
%%
%% The released version of this bundle is available from CTAN.
%%
%% -----------------------------------------------------------------------
%%
%% The development version of the bundle can be found at
%%
%%    http://www.latex-project.org/svnroot/experimental/trunk/
%%
%% for those people who are interested.
%%
%%%%%%%%%%%
%% NOTE: %%
%%%%%%%%%%%
%%
%%   Snapshots taken from the repository represent work in progress and may
%%   not work or may contain conflicting material!  We therefore ask
%%   people _not_ to put them into distributions, archives, etc. without
%%   prior consultation with the LaTeX Project Team.
%%
%% -----------------------------------------------------------------------
%%
%
%<*driver>
\RequirePackage{l3bootstrap}
\GetIdInfo$Id: l3fp-trig.dtx 3514 2012-03-08 06:14:48Z bruno $
  {L3 Floating-point trigonometric functions}
\documentclass[full]{l3doc}
\begin{document}
  \DocInput{\jobname.dtx}
\end{document}
%</driver>
% \fi
%
% \title{The \textsf{l3fp-trig} package\thanks{This file
%         has version number \ExplFileVersion, last
%         revised \ExplFileDate.}\\
% Floating point trigonometric functions}
% \author{^^A
%  The \LaTeX3 Project\thanks
%    {^^A
%      E-mail:
%        \href{mailto:latex-team@latex-project.org}
%          {latex-team@latex-project.org}^^A
%    }^^A
% }
% \date{Released \ExplFileDate}
%
% \maketitle
%
% \begin{documentation}
%
% \end{documentation}
%
% \begin{implementation}
%
% \section{Implementation}
%
%    \begin{macrocode}
%<*initex|package>
%    \end{macrocode}
%
%    \begin{macrocode}
%<@@=fp>
%    \end{macrocode}
%
%^^A todo: check EXP/rEXP everywhere.
%
% \subsection{Inverting a floating point number}
%
% \begin{macro}[int, EXP]{\@@_one_over:w}
%   Expects a floating point of the form \cs{s_@@} \ldots{} |;| and
%   computes its multiplicative inverse.  This is used to compute the
%   cotangent function very near $0$.
%    \begin{macrocode}
\cs_new_nopar:Npx \@@_one_over:w
  {
    \exp_not:N \exp_after:wN
    \exp_not:c { @@_/_o:ww }
    \exp_not:N \c_one_fp
  }
%    \end{macrocode}
% \end{macro}
%
% \subsection{Direct trigonometric functions}
%
% The approach for all trigonometric functions (sine, cosine, tangent,
% and cotangent) is the same.
% \begin{itemize}
% \item Filter out special cases ($\pm 0$, $\pm\inf$ and \nan{}).
% \item Keep the sign for later, and work with the absolute value $|x|$
%   of the argument.
% \item For numbers less than $1$, shift the mantissa to convert them to
%   fixed point numbers.  Very small numbers take a slightly different
%   route.
% \item For numbers $\geq 1$, subtract a multiple of $\pi/2$ to bring
%   them to the range to $[0, \pi/2]$.
% \item Reduce further to $[0, \pi/4]$ using $\sin x = \cos (\pi/2-x)$.
% \item Use the appropriate power series depending on the octant
%   $\lfloor\frac{|x|}{\pi/4}\rfloor \mod 8$, the sign, and the function
%   to compute.
% \end{itemize}
%
% \subsubsection{Sign and special numbers}
%
% \begin{macro}[int, EXP]{\@@_sin_o:w}
%   The sine of $\pm 0$ or \nan{} is the same floating point number.
%   The sine of $\pm\infty$ raises an invalid operation exception.
%   Otherwise, check the exponent, preparing to use
%   \cs{@@_sin_series:NNwww} for the calculation, with a sign |#2|, and
%   an initial octant of $0$.  The question mark is an argument which is
%   not used in this case.
%    \begin{macrocode}
\cs_new:Npn \@@_sin_o:w \s_@@ \@@_chk:w #1#2
  {
    \if_case:w #1 \exp_stop_f:
           \@@_case_return_same_o:w
    \or:
      \exp_after:wN \@@_trig_exponent:NNNNwn
      \exp_after:wN \@@_sin_series:NNwww
      \exp_after:wN ?
      \exp_after:wN #2
      \exp_after:wN \c_zero
    \or:
      \@@_case_use:nw { \@@_invalid_operation_o:nw { sin } }
    \else: \@@_case_return_same_o:w
    \fi:
    \s_@@ \@@_chk:w #1#2
  }
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}[int, EXP]{\@@_cos_o:w}
%   The cosine of $\pm 0$ is $1$.  The cosine of $\pm\infty$ raises an
%   invalid operation exception.  The cosine of \nan{} is itself.
%   Otherwise, check the exponent, preparing to use
%   \cs{@@_sin_series:NNwww} for the calculation, with a positive sign
%   ($0$), and an initial octant of $2$, because $\cos x = \sin ( \pi/2
%   + |x|)$.  The question mark is an argument which is not used in this
%   case.
%    \begin{macrocode}
\cs_new:Npn \@@_cos_o:w \s_@@ \@@_chk:w #1#2
  {
    \if_case:w #1 \exp_stop_f:
           \@@_case_return_o:Nw \c_one_fp
    \or:
      \@@_case_use:nw %^^A todo: is that faster than the exp_after route?
        {
          \@@_trig_exponent:NNNNwn
          \@@_sin_series:NNwww
          ?
          0
          \c_two
        }
    \or:
      \@@_case_use:nw { \@@_invalid_operation_o:nw { cos } }
    \else: \@@_case_return_same_o:w
    \fi:
    \s_@@ \@@_chk:w #1#2
  }
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}[int, EXP]{\@@_tan_o:w}
%   The tangent of $\pm 0$ or \nan{} is the same floating point number.
%   The tangent of $\pm\infty$ raises an invalid operation exception.
%   Otherwise, check the exponent, preparing to use
%   \cs{@@_tan_series:NNwww} for the calculation, with a positive sign
%   ($0$), and an initial octant of $1$, chosen to be distinct from the
%   octants for sine and cosine.  See \cs{@@_cot_o:w} for an
%   explanation of the $0$ argument.
%    \begin{macrocode}
\cs_new:Npn \@@_tan_o:w \s_@@ \@@_chk:w #1#2
  {
    \if_case:w #1 \exp_stop_f:
           \@@_case_return_same_o:w
    \or:
      \exp_after:wN \@@_trig_exponent:NNNNwn
      \exp_after:wN \@@_tan_series:NNwww
      \exp_after:wN 0
      \exp_after:wN #2
      \exp_after:wN \c_one
    \or:
      \@@_case_use:nw { \@@_invalid_operation_o:nw { tan } }
    \else: \@@_case_return_same_o:w
    \fi:
    \s_@@ \@@_chk:w #1#2
  }
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}[int, EXP]{\@@_cot_o:w}
%   The cotangent of $\pm 0$ is $\pm \infty$ with the same sign,
%   produced by \cs{@@_one_over:w}.  The cotangent of $\pm\infty$ raises
%   an invalid operation exception.  The cotangent of \nan{} is itself.
%   We use $\cot x = - \tan (\pi/2 + x)$, and the initial octant for the
%   tangent was chosen to be $1$, so the octant here starts at $3$.  The
%   change in sign is obtained by feeding \cs{@@_tan_series:NNwww} two
%   signs rather than just the sign of the argument: the first of those
%   indicates whether we compute tangent or cotangent.  Those signs are
%   eventually combined.
%    \begin{macrocode}
\cs_new:Npn \@@_cot_o:w \s_@@ \@@_chk:w #1#2
  {
    \if_case:w #1 \exp_stop_f:
           \exp_after:wN \@@_one_over:w
    \or:
      \exp_after:wN \@@_trig_exponent:NNNNwn
      \exp_after:wN \@@_tan_series:NNwww
      \exp_after:wN 2
      \exp_after:wN #2
      \exp_after:wN \c_three
    \or:
      \@@_case_use:nw { \@@_invalid_operation_o:nw { cot } }
    \else: \@@_case_return_same_o:w
    \fi:
    \s_@@ \@@_chk:w #1#2
  }
%    \end{macrocode}
% \end{macro}
%
% \subsubsection{Small and tiny arguments}
%
% \begin{macro}[aux, EXP]{\@@_trig_exponent:NNNNwn}
%   The first four arguments control what trigonometric function we
%   compute, then follows a normal floating point number.  If the
%   floating point is smaller than $10^{-8}$, then call the appropriate
%   \texttt{_epsilon} auxiliary.  Otherwise, call the function |#1|,
%   with arguments |#2|, |#3|, the octant, computed in an integer
%   expression starting with |#4|, and a fixed point number obtained
%   from the floating point number by argument reduction.  Numbers less
%   than $1$ are converted using \cs{@@_trig_small:w} which simply
%   shifts the mantissa, while large numbers need argument reduction.
%    \begin{macrocode}
\cs_new:Npn \@@_trig_exponent:NNNNwn #1#2#3#4 \s_@@ \@@_chk:w 1#5#6
  {
    \if_int_compare:w #6 > - \c_eight
      \exp_after:wN #1
      \exp_after:wN #2
      \exp_after:wN #3
      \int_use:N \__int_eval:w #4
        \if_int_compare:w #6 > \c_zero
          \exp_after:wN \@@_trig_large:w \__int_value:w
        \else:
          \exp_after:wN \@@_trig_small:w \__int_value:w
        \fi:
    \else:
      \if_case:w #4
             \@@_sin_epsilon:w
      \or:   \@@_sin_epsilon:w
      \or:   \@@_cos_epsilon:w
      \else: \@@_cot_epsilon:w
      \fi:
      #5
    \fi:
    #6 ;
  }
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}[aux, EXP]
%   {\@@_sin_epsilon:w, \@@_cos_epsilon:w, \@@_cot_epsilon:w}
%   Sine and tangent of tiny numbers give the number itself: the
%   relative error is less than $5 \cdot 10^{-17}$, which is
%   appropriate.  Cosine simply gives $1$.  Cotangent computes the
%   inverse.  This is actually slightly wrong because further terms in
%   the power series could affect the rounding for cotangent.
%    \begin{macrocode}
\cs_new:Npn \@@_sin_epsilon:w #1 \fi: #2 \fi: #3 ;
  { \fi: \fi: \@@_exp_after_o:w \s_@@ \@@_chk:w 1 #2 {#3} }
\cs_new:Npn \@@_cos_epsilon:w #1 \fi: #2 \fi: #3 ; #4 ;
  { \fi: \fi: \exp_after:wN \c_one_fp }
\cs_new:Npn \@@_cot_epsilon:w \fi: #1 \fi: #2 ;
  { \fi: \fi: \@@_one_over:w \s_@@ \@@_chk:w 1 #1 {#2} }
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}[aux, EXP]{\@@_trig_small:w, \@@_trig_small_aux:wwNN}
%   Floating point numbers less than $1$ are converted to fixed point
%   numbers by shifting the mantissa.  Since we have already filtered
%   out numbers less than $10^{-8}$, no digit is lost in converting to
%   a fixed point number.
%    \begin{macrocode}
\cs_new:Npn \@@_trig_small:w #1;
  {
    \exp_after:wN \exp_after:wN \exp_after:wN \@@_trig_small_aux:wwNN
    \prg_replicate:nn { - #1 } { 0 } ;
  }
\cs_new:Npn \@@_trig_small_aux:wwNN #1; #2#3#4#5;
  {
    \@@_pack_twice_four:wNNNNNNNN
    \@@_pack_twice_four:wNNNNNNNN
    \@@_pack_twice_four:wNNNNNNNN
    .
    ;
    #1#2#3#4#5 0000 0000;
  }
%    \end{macrocode}
% \end{macro}
%
% \subsubsection{Reduction of large arguments}
%
% In the case of a floating point argument greater or equal to $1$, we
% need to perform argument reduction.
%
% \begin{macro}[aux, rEXP]
%   {
%     \@@_trig_large:w, \@@_trig_large_i:www,
%     \@@_trig_large_ii:wnnnnnn, \@@_trig_large_break:w
%   }
%   We shift the mantissa by one digit at a time, subtracting a multiple
%   of $2\pi$ at each step.  We use a value of $2\pi$ rounded up,
%   consistent with the choice of \cs{c_pi_fp}.  This is not quite
%   correct from an accuracy perspective, but has the nice property that
%   $\sin(180\mathrm{deg}) = 0$ exactly.  The arguments of
%   \cs{@@_trig_large_i:www} are a leading block of up to $5$ digits,
%   three brace groups of $4$ digits each, and the exponent, decremented
%   at each step.  The multiple of $2\pi$ to subtract is estimated as
%   $\lfloor |#1| / 6283\rfloor$ (the formula chosen always gives a
%   non-negative integer).  The subtraction has a form similar to our
%   usual multiplications (see \pkg{l3fp-basics} or
%   \pkg{l3fp-extended}).  Once the exponent reaches $0$, we are done
%   subtracting $2\pi$, and we call \cs{@@_trig_octant_loop:nw} to do
%   the reduction by $\pi/2$.
%    \begin{macrocode}
\cs_new:Npn \@@_trig_large:w #1; #2#3;
  { \@@_trig_large_i:www #2; #3 ;  #1; }
\cs_new:Npn \@@_trig_large_i:www #1; #2; #3;
  {
    \if_meaning:w 0 #3 \@@_trig_large_break:w \fi:
    \exp_after:wN \@@_trig_large_ii:wnnnnnn
    \int_use:N \__int_eval:w ( #1 - 3141 ) / 6283 ;
    {#1} #2;
    \int_use:N \__int_eval:w \c_minus_one + #3;
  }
\cs_new:Npn \@@_trig_large_ii:wnnnnnn #1; #2#3#4#5;
  {
    \exp_after:wN \@@_trig_large_i:www
    \int_use:N \__int_eval:w -5 0000 + #20 - #1*62831
      \exp_after:wN \@@_fixed_mul_pack:NNNNNw
      \int_use:N \__int_eval:w 4 9995 0000 + #30 - #1*8530
        \exp_after:wN \@@_fixed_mul_pack:NNNNNw
        \int_use:N \__int_eval:w 4 9995 0000 + #40 - #1*7179
          \exp_after:wN \@@_fixed_mul_pack:NNNNNw
          \int_use:N \__int_eval:w 5 0000 0000 + #50 - #1*5880
    \exp_after:wN ;
    \exp_after:wN ;
  }
\cs_new:Npn \@@_trig_large_break:w \fi: #1; #2;
  { \fi: \@@_trig_octant_loop:nw #2 {0000} {0000} ; }
%    \end{macrocode}
% \end{macro}
%
%^^A todo: optimize: we don't need 6x4 digits here, only 4x4.
%
% \begin{macro}[aux, rEXP]
%   {
%     \@@_trig_octant_loop:nw, \@@_trig_octant_break:w,
%     \@@_trig_octant_neg:w
%   }
%   We receive a fixed point number as argument.  As long as it is
%   greater than $1.5707$ (a slight underestimate of $\pi/2$), subtract
%   $\pi/2$, and leave |+ \c_two| in the integer expression for the
%   octant.  Once it becomes smaller, if it is greater than $0.7854$
%   (overestimate of $\pi/4$), then compute $\pi/2 - x$ and increment
%   the octant.  If it is negative, correct this by changing the sign
%   and decrementing the octant (by adding $7$).  The result is in all
%   cases in the range $[0, 0.7854]$, appropriate for a series
%   expansion.
%    \begin{macrocode}
\cs_new:Npn \@@_trig_octant_loop:nw #1#2;
  {
    \if_int_compare:w #1 < 15707 \exp_stop_f:
      \@@_trig_octant_break:w
    \fi:
    + \c_two
    \@@_fixed_sub_back:wwN
      {15707} {9632} {6794} {8970} {0000} {0000} ;
      {#1} #2;
    \@@_trig_octant_loop:nw
  }
\cs_new:Npn \@@_trig_octant_break:w #1 \fi: + #2#3 #4; #5#6; #7;
  {
    \fi:
    \if_int_compare:w #5 < 7854 \exp_stop_f:
      \if_int_compare:w #5 < \c_zero
        \exp_after:wN \@@_trig_octant_neg:w
      \fi:
      \exp_after:wN \@@_use_i_until_s:nw
      \exp_after:wN .
    \fi:
    + \c_one
    \@@_fixed_sub:wwN
    {15707} {9632} {6794} {8970} {0000} {0000} ;
    {#5} #6 ; . ;
  }
\cs_new:Npn \@@_trig_octant_neg:w #1\fi: #2; #3#4#5#6#7#8; #9
  {
    \fi:
    + \c_seven
    \exp_after:wN \@@_fixed_add_after:NNNNNwN
    \int_use:N \__int_eval:w 1 9999 9998 - #30000 - #4
      \exp_after:wN \@@_fixed_add_pack:NNNNNwN
      \int_use:N \__int_eval:w 1 9999 9998 - #5#6
        \exp_after:wN \@@_fixed_add_pack:NNNNNwN
        \int_use:N \__int_eval:w 2 0000 0000 - #7#8 ; {#9} ;
  }
%    \end{macrocode}
% \end{macro}
%
% \subsection{Computing the power series}
%
% \begin{macro}[aux, EXP]{\@@_sin_series:NNwww, \@@_sin_series_aux:Nnww}
%   Here we receive an unused |?|, a \meta{sign} ($0$ or $2$), a
%   (non-negative) \meta{octant} delimited by a dot, a \meta{fixed
%     point} number, and junk delimited by a semicolon.  The auxiliary
%   receives:
%   \begin{itemize}
%   \item The final sign, which depends on the octant |#3| and the
%     original sign |#2|,
%   \item The octant |#3|, which will control the series we use.
%   \item The square |#4 * #4| of the argument, computed with
%     \cs{@@_fixed_mul:wwn}.
%   \item The number itself.
%   \end{itemize}
%   If the octant is in $\{1,2,5,6,\ldots{}\}$, we are near an extremum
%   of the function and we use the series
%   \[
%   \cos(x) = 1 - x^2 \bigg( \frac{1}{2!} - x^2 \bigg( \frac{1}{4!}
%   - x^2 \bigg( \cdots \bigg) \bigg) \bigg) .
%   \]
%   Otherwise, the series
%   \[
%   \sin(x) = x \bigg( 1 - x^2 \bigg( \frac{1}{3!} - x^2 \bigg(
%   \frac{1}{5!} - x^2 \bigg( \cdots \bigg) \bigg) \bigg) \bigg)
%   \]
%   is used.  Finally, the fixed point number is converted to a floating
%   point number with the given sign, and we check for overflow or
%   underflow. %^^A todo: can over/underflow really happen??
%    \begin{macrocode}
\cs_new:Npn \@@_sin_series:NNwww #1#2#3 . #4; #5;
  {
    \@@_fixed_mul:wwn #4; #4;
    {
      \exp_after:wN \@@_sin_series_aux:Nnww
      \__int_value:w
        \if_int_odd:w \__int_eval:w ( #3 + \c_two ) / \c_four \__int_eval_end:
          #2
        \else:
          \if_meaning:w #2 0 2 \else: 0 \fi:
        \fi:
      {#3}
    }
    #4 ;
  }
\cs_new:Npn \@@_sin_series_aux:Nnww #1#2 #3; #4;
  {
    \if_int_odd:w \__int_eval:w #2 / \c_two \__int_eval_end:
      \exp_after:wN \use_i:nn
    \else:
      \exp_after:wN \use_ii:nn
    \fi:
    {
      \@@_fixed_continue:wn {0000}{0000}{0000}{0001}{5619}{2070}; % 1/18!
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0000}{0000}{0477}{9477}{3324};
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0000}{0011}{4707}{4559}{7730};
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0000}{2087}{6756}{9878}{6810};
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0027}{5573}{1922}{3985}{8907};
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{2480}{1587}{3015}{8730}{1587};
      \@@_fixed_mul_sub_back:wwwn #3; {0013}{8888}{8888}{8888}{8888}{8889};
      \@@_fixed_mul_sub_back:wwwn #3; {0416}{6666}{6666}{6666}{6666}{6667};
      \@@_fixed_mul_sub_back:wwwn #3; {5000}{0000}{0000}{0000}{0000}{0000};
      \@@_fixed_mul_sub_back:wwwn #3;{10000}{0000}{0000}{0000}{0000}{0000};
    }
    {
      \@@_fixed_continue:wn {0000}{0000}{0000}{0028}{1145}{7254}; % 1/17!
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0000}{0000}{7647}{1637}{3182};
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0000}{0160}{5904}{3836}{8216};
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0002}{5052}{1083}{8544}{1719};
      \@@_fixed_mul_sub_back:wwwn #3; {0000}{0275}{5731}{9223}{9858}{9065};
      \@@_fixed_mul_sub_back:wwwn #3; {0001}{9841}{2698}{4126}{9841}{2698};
      \@@_fixed_mul_sub_back:wwwn #3; {0083}{3333}{3333}{3333}{3333}{3333};
      \@@_fixed_mul_sub_back:wwwn #3; {1666}{6666}{6666}{6666}{6666}{6667};
      \@@_fixed_mul_sub_back:wwwn #3;{10000}{0000}{0000}{0000}{0000}{0000};
      \@@_fixed_mul:wwn #4;
    }
    {
      \exp_after:wN \@@_sanitize:Nw
      \exp_after:wN #1
      \int_use:N \__int_eval:w \@@_fixed_to_float:wN
    }
    #1
  }
%    \end{macrocode}
% \end{macro}
%
% \begin{macro}[aux, EXP]{\@@_tan_series:NNwww, \@@_tan_series_aux:Nnww}
%   Similar to \cs{@@_sin_series:NNwww}, but with slightly different
%   rules to find the sign.  The result is expressed as a ratio of
%   polynomials, of the form
%   \[
%   \tan(x) \simeq
%   \frac{x (1 - x^2 (a_1 - x^2 (a_2 - x^2 (a_3 - x^2 (a_4 - x^2 a_5)))))}
%     {1 - x^2 (b_1 - x^2 (b_2 - x^2 (b_3 - x^2 (b_4 - x^2 b_5))))} .
%   \]
%   The ratio of the two fixed point numbers is converted to a floating
%   point number directly to avoid rounding issues.  The two fixed
%   points may be exchanged before computing the ratio, depending on the
%   quadrant.
%    \begin{macrocode}
\cs_new:Npn \@@_tan_series:NNwww #1#2#3. #4; #5;
  {
    \@@_fixed_mul:wwn #4; #4;
    {
      \exp_after:wN \@@_tan_series_aux:Nnww
      \__int_value:w
        \if_int_odd:w \__int_eval:w #3 / \c_two \__int_eval_end:
          \exp_after:wN \reverse_if:N
        \fi:
        \if_meaning:w #1#2 2 \else: 0 \fi:
      {#3}
    }
    #4 ;
  }
\cs_new:Npn \@@_tan_series_aux:Nnww #1 #2 #3; #4;
  {
    \@@_fixed_continue:wn {0000}{0000}{1527}{3493}{0856}{7059};
    \@@_fixed_mul_sub_back:wwwn #3; {0000}{0159}{6080}{0274}{5257}{6472};
    \@@_fixed_mul_sub_back:wwwn #3; {0002}{4571}{2320}{0157}{2558}{8481};
    \@@_fixed_mul_sub_back:wwwn #3; {0115}{5830}{7533}{5397}{3168}{2147};
    \@@_fixed_mul_sub_back:wwwn #3; {1929}{8245}{6140}{3508}{7719}{2982};
    \@@_fixed_mul_sub_back:wwwn #3;{10000}{0000}{0000}{0000}{0000}{0000};
    \@@_fixed_mul:wwn #4;
      {
        \@@_fixed_continue:wn {0000}{0007}{0258}{0681}{9408}{4706};
        \@@_fixed_mul_sub_back:wwwn #3; {0000}{2343}{7175}{1399}{6151}{7670};
        \@@_fixed_mul_sub_back:wwwn #3; {0019}{2638}{4588}{9232}{8861}{3691};
        \@@_fixed_mul_sub_back:wwwn #3; {0536}{6357}{0691}{4344}{6852}{4252};
        \@@_fixed_mul_sub_back:wwwn #3; {5263}{1578}{9473}{6842}{1052}{6315};
        \@@_fixed_mul_sub_back:wwwn #3;{10000}{0000}{0000}{0000}{0000}{0000};
          {
            \exp_after:wN \@@_sanitize:Nw
            \exp_after:wN #1
            \int_use:N \__int_eval:w
              \reverse_if:N \if_int_odd:w
                  \__int_eval:w (#2 - \c_one) / \c_two \__int_eval_end:
                \exp_after:wN \@@_reverse_args:Nww
              \fi:
              \@@_fixed_div_to_float:ww
          }
      }
  }
%    \end{macrocode}
% \end{macro}
%
%    \begin{macrocode}
%</initex|package>
%    \end{macrocode}
%
% \end{implementation}
%
% \PrintChanges
%
% \PrintIndex