1 /* cairo - a vector graphics library with display and print output
2 *
3 * Copyright © 2002 University of Southern California
4 *
5 * This library is free software; you can redistribute it and/or
6 * modify it either under the terms of the GNU Lesser General Public
7 * License version 2.1 as published by the Free Software Foundation
8 * (the "LGPL") or, at your option, under the terms of the Mozilla
9 * Public License Version 1.1 (the "MPL"). If you do not alter this
10 * notice, a recipient may use your version of this file under either
11 * the MPL or the LGPL.
12 *
13 * You should have received a copy of the LGPL along with this library
14 * in the file COPYING-LGPL-2.1; if not, write to the Free Software
15 * Foundation, Inc., 51 Franklin Street, Suite 500, Boston, MA 02110-1335, USA
16 * You should have received a copy of the MPL along with this library
17 * in the file COPYING-MPL-1.1
18 *
19 * The contents of this file are subject to the Mozilla Public License
20 * Version 1.1 (the "License"); you may not use this file except in
21 * compliance with the License. You may obtain a copy of the License at
22 * http://www.mozilla.org/MPL/
23 *
24 * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY
25 * OF ANY KIND, either express or implied. See the LGPL or the MPL for
26 * the specific language governing rights and limitations.
27 *
28 * The Original Code is the cairo graphics library.
29 *
30 * The Initial Developer of the Original Code is University of Southern
31 * California.
32 *
33 * Contributor(s):
34 * Carl D. Worth <cworth@cworth.org>
35 */
36
37 #include "cairoint.h"
38 #include "cairo-error-private.h"
39
40 #if _XOPEN_SOURCE >= 600 || defined (_ISOC99_SOURCE)
41 #define ISFINITE(x) isfinite (x)
42 #else
43 #define ISFINITE(x) ((x) * (x) >= 0.) /* check for NaNs */
44 #endif
45
46 /**
47 * SECTION:cairo-matrix
48 * @Title: cairo_matrix_t
49 * @Short_Description: Generic matrix operations
50 * @See_Also: #cairo_t
51 *
52 * #cairo_matrix_t is used throughout cairo to convert between different
53 * coordinate spaces. A #cairo_matrix_t holds an affine transformation,
54 * such as a scale, rotation, shear, or a combination of these.
55 * The transformation of a point (<literal>x</literal>,<literal>y</literal>)
56 * is given by:
57 *
58 * <programlisting>
59 * x_new = xx * x + xy * y + x0;
60 * y_new = yx * x + yy * y + y0;
61 * </programlisting>
62 *
63 * The current transformation matrix of a #cairo_t, represented as a
64 * #cairo_matrix_t, defines the transformation from user-space
65 * coordinates to device-space coordinates. See cairo_get_matrix() and
66 * cairo_set_matrix().
67 */
68
69 static void
70 _cairo_matrix_scalar_multiply (cairo_matrix_t *matrix, double scalar);
71
72 static void
73 _cairo_matrix_compute_adjoint (cairo_matrix_t *matrix);
74
75 /**
76 * cairo_matrix_init_identity:
77 * @matrix: a #cairo_matrix_t
78 *
79 * Modifies @matrix to be an identity transformation.
80 **/
81 void
cairo_matrix_init_identity(cairo_matrix_t * matrix)82 cairo_matrix_init_identity (cairo_matrix_t *matrix)
83 {
84 cairo_matrix_init (matrix,
85 1, 0,
86 0, 1,
87 0, 0);
88 }
89 slim_hidden_def(cairo_matrix_init_identity);
90
91 /**
92 * cairo_matrix_init:
93 * @matrix: a #cairo_matrix_t
94 * @xx: xx component of the affine transformation
95 * @yx: yx component of the affine transformation
96 * @xy: xy component of the affine transformation
97 * @yy: yy component of the affine transformation
98 * @x0: X translation component of the affine transformation
99 * @y0: Y translation component of the affine transformation
100 *
101 * Sets @matrix to be the affine transformation given by
102 * @xx, @yx, @xy, @yy, @x0, @y0. The transformation is given
103 * by:
104 * <programlisting>
105 * x_new = xx * x + xy * y + x0;
106 * y_new = yx * x + yy * y + y0;
107 * </programlisting>
108 **/
109 void
cairo_matrix_init(cairo_matrix_t * matrix,double xx,double yx,double xy,double yy,double x0,double y0)110 cairo_matrix_init (cairo_matrix_t *matrix,
111 double xx, double yx,
112
113 double xy, double yy,
114 double x0, double y0)
115 {
116 matrix->xx = xx; matrix->yx = yx;
117 matrix->xy = xy; matrix->yy = yy;
118 matrix->x0 = x0; matrix->y0 = y0;
119 }
120 slim_hidden_def(cairo_matrix_init);
121
122 /**
123 * _cairo_matrix_get_affine:
124 * @matrix: a #cairo_matrix_t
125 * @xx: location to store xx component of matrix
126 * @yx: location to store yx component of matrix
127 * @xy: location to store xy component of matrix
128 * @yy: location to store yy component of matrix
129 * @x0: location to store x0 (X-translation component) of matrix, or %NULL
130 * @y0: location to store y0 (Y-translation component) of matrix, or %NULL
131 *
132 * Gets the matrix values for the affine transformation that @matrix represents.
133 * See cairo_matrix_init().
134 *
135 *
136 * This function is a leftover from the old public API, but is still
137 * mildly useful as an internal means for getting at the matrix
138 * members in a positional way. For example, when reassigning to some
139 * external matrix type, or when renaming members to more meaningful
140 * names (such as a,b,c,d,e,f) for particular manipulations.
141 **/
142 void
_cairo_matrix_get_affine(const cairo_matrix_t * matrix,double * xx,double * yx,double * xy,double * yy,double * x0,double * y0)143 _cairo_matrix_get_affine (const cairo_matrix_t *matrix,
144 double *xx, double *yx,
145 double *xy, double *yy,
146 double *x0, double *y0)
147 {
148 *xx = matrix->xx;
149 *yx = matrix->yx;
150
151 *xy = matrix->xy;
152 *yy = matrix->yy;
153
154 if (x0)
155 *x0 = matrix->x0;
156 if (y0)
157 *y0 = matrix->y0;
158 }
159
160 /**
161 * cairo_matrix_init_translate:
162 * @matrix: a #cairo_matrix_t
163 * @tx: amount to translate in the X direction
164 * @ty: amount to translate in the Y direction
165 *
166 * Initializes @matrix to a transformation that translates by @tx and
167 * @ty in the X and Y dimensions, respectively.
168 **/
169 void
cairo_matrix_init_translate(cairo_matrix_t * matrix,double tx,double ty)170 cairo_matrix_init_translate (cairo_matrix_t *matrix,
171 double tx, double ty)
172 {
173 cairo_matrix_init (matrix,
174 1, 0,
175 0, 1,
176 tx, ty);
177 }
178 slim_hidden_def(cairo_matrix_init_translate);
179
180 /**
181 * cairo_matrix_translate:
182 * @matrix: a #cairo_matrix_t
183 * @tx: amount to translate in the X direction
184 * @ty: amount to translate in the Y direction
185 *
186 * Applies a translation by @tx, @ty to the transformation in
187 * @matrix. The effect of the new transformation is to first translate
188 * the coordinates by @tx and @ty, then apply the original transformation
189 * to the coordinates.
190 **/
191 void
cairo_matrix_translate(cairo_matrix_t * matrix,double tx,double ty)192 cairo_matrix_translate (cairo_matrix_t *matrix, double tx, double ty)
193 {
194 cairo_matrix_t tmp;
195
196 cairo_matrix_init_translate (&tmp, tx, ty);
197
198 cairo_matrix_multiply (matrix, &tmp, matrix);
199 }
200 slim_hidden_def (cairo_matrix_translate);
201
202 /**
203 * cairo_matrix_init_scale:
204 * @matrix: a #cairo_matrix_t
205 * @sx: scale factor in the X direction
206 * @sy: scale factor in the Y direction
207 *
208 * Initializes @matrix to a transformation that scales by @sx and @sy
209 * in the X and Y dimensions, respectively.
210 **/
211 void
cairo_matrix_init_scale(cairo_matrix_t * matrix,double sx,double sy)212 cairo_matrix_init_scale (cairo_matrix_t *matrix,
213 double sx, double sy)
214 {
215 cairo_matrix_init (matrix,
216 sx, 0,
217 0, sy,
218 0, 0);
219 }
220 slim_hidden_def(cairo_matrix_init_scale);
221
222 /**
223 * cairo_matrix_scale:
224 * @matrix: a #cairo_matrix_t
225 * @sx: scale factor in the X direction
226 * @sy: scale factor in the Y direction
227 *
228 * Applies scaling by @sx, @sy to the transformation in @matrix. The
229 * effect of the new transformation is to first scale the coordinates
230 * by @sx and @sy, then apply the original transformation to the coordinates.
231 **/
232 void
cairo_matrix_scale(cairo_matrix_t * matrix,double sx,double sy)233 cairo_matrix_scale (cairo_matrix_t *matrix, double sx, double sy)
234 {
235 cairo_matrix_t tmp;
236
237 cairo_matrix_init_scale (&tmp, sx, sy);
238
239 cairo_matrix_multiply (matrix, &tmp, matrix);
240 }
241 slim_hidden_def(cairo_matrix_scale);
242
243 /**
244 * cairo_matrix_init_rotate:
245 * @matrix: a #cairo_matrix_t
246 * @radians: angle of rotation, in radians. The direction of rotation
247 * is defined such that positive angles rotate in the direction from
248 * the positive X axis toward the positive Y axis. With the default
249 * axis orientation of cairo, positive angles rotate in a clockwise
250 * direction.
251 *
252 * Initialized @matrix to a transformation that rotates by @radians.
253 **/
254 void
cairo_matrix_init_rotate(cairo_matrix_t * matrix,double radians)255 cairo_matrix_init_rotate (cairo_matrix_t *matrix,
256 double radians)
257 {
258 double s;
259 double c;
260
261 s = sin (radians);
262 c = cos (radians);
263
264 cairo_matrix_init (matrix,
265 c, s,
266 -s, c,
267 0, 0);
268 }
269 slim_hidden_def(cairo_matrix_init_rotate);
270
271 /**
272 * cairo_matrix_rotate:
273 * @matrix: a #cairo_matrix_t
274 * @radians: angle of rotation, in radians. The direction of rotation
275 * is defined such that positive angles rotate in the direction from
276 * the positive X axis toward the positive Y axis. With the default
277 * axis orientation of cairo, positive angles rotate in a clockwise
278 * direction.
279 *
280 * Applies rotation by @radians to the transformation in
281 * @matrix. The effect of the new transformation is to first rotate the
282 * coordinates by @radians, then apply the original transformation
283 * to the coordinates.
284 **/
285 void
cairo_matrix_rotate(cairo_matrix_t * matrix,double radians)286 cairo_matrix_rotate (cairo_matrix_t *matrix, double radians)
287 {
288 cairo_matrix_t tmp;
289
290 cairo_matrix_init_rotate (&tmp, radians);
291
292 cairo_matrix_multiply (matrix, &tmp, matrix);
293 }
294
295 /**
296 * cairo_matrix_multiply:
297 * @result: a #cairo_matrix_t in which to store the result
298 * @a: a #cairo_matrix_t
299 * @b: a #cairo_matrix_t
300 *
301 * Multiplies the affine transformations in @a and @b together
302 * and stores the result in @result. The effect of the resulting
303 * transformation is to first apply the transformation in @a to the
304 * coordinates and then apply the transformation in @b to the
305 * coordinates.
306 *
307 * It is allowable for @result to be identical to either @a or @b.
308 **/
309 /*
310 * XXX: The ordering of the arguments to this function corresponds
311 * to [row_vector]*A*B. If we want to use column vectors instead,
312 * then we need to switch the two arguments and fix up all
313 * uses.
314 */
315 void
cairo_matrix_multiply(cairo_matrix_t * result,const cairo_matrix_t * a,const cairo_matrix_t * b)316 cairo_matrix_multiply (cairo_matrix_t *result, const cairo_matrix_t *a, const cairo_matrix_t *b)
317 {
318 cairo_matrix_t r;
319
320 r.xx = a->xx * b->xx + a->yx * b->xy;
321 r.yx = a->xx * b->yx + a->yx * b->yy;
322
323 r.xy = a->xy * b->xx + a->yy * b->xy;
324 r.yy = a->xy * b->yx + a->yy * b->yy;
325
326 r.x0 = a->x0 * b->xx + a->y0 * b->xy + b->x0;
327 r.y0 = a->x0 * b->yx + a->y0 * b->yy + b->y0;
328
329 *result = r;
330 }
331 slim_hidden_def(cairo_matrix_multiply);
332
333 /**
334 * cairo_matrix_transform_distance:
335 * @matrix: a #cairo_matrix_t
336 * @dx: X component of a distance vector. An in/out parameter
337 * @dy: Y component of a distance vector. An in/out parameter
338 *
339 * Transforms the distance vector (@dx,@dy) by @matrix. This is
340 * similar to cairo_matrix_transform_point() except that the translation
341 * components of the transformation are ignored. The calculation of
342 * the returned vector is as follows:
343 *
344 * <programlisting>
345 * dx2 = dx1 * a + dy1 * c;
346 * dy2 = dx1 * b + dy1 * d;
347 * </programlisting>
348 *
349 * Affine transformations are position invariant, so the same vector
350 * always transforms to the same vector. If (@x1,@y1) transforms
351 * to (@x2,@y2) then (@x1+@dx1,@y1+@dy1) will transform to
352 * (@x1+@dx2,@y1+@dy2) for all values of @x1 and @x2.
353 **/
354 void
cairo_matrix_transform_distance(const cairo_matrix_t * matrix,double * dx,double * dy)355 cairo_matrix_transform_distance (const cairo_matrix_t *matrix, double *dx, double *dy)
356 {
357 double new_x, new_y;
358
359 new_x = (matrix->xx * *dx + matrix->xy * *dy);
360 new_y = (matrix->yx * *dx + matrix->yy * *dy);
361
362 *dx = new_x;
363 *dy = new_y;
364 }
365 slim_hidden_def(cairo_matrix_transform_distance);
366
367 /**
368 * cairo_matrix_transform_point:
369 * @matrix: a #cairo_matrix_t
370 * @x: X position. An in/out parameter
371 * @y: Y position. An in/out parameter
372 *
373 * Transforms the point (@x, @y) by @matrix.
374 **/
375 void
cairo_matrix_transform_point(const cairo_matrix_t * matrix,double * x,double * y)376 cairo_matrix_transform_point (const cairo_matrix_t *matrix, double *x, double *y)
377 {
378 cairo_matrix_transform_distance (matrix, x, y);
379
380 *x += matrix->x0;
381 *y += matrix->y0;
382 }
383 slim_hidden_def(cairo_matrix_transform_point);
384
385 void
_cairo_matrix_transform_bounding_box(const cairo_matrix_t * matrix,double * x1,double * y1,double * x2,double * y2,cairo_bool_t * is_tight)386 _cairo_matrix_transform_bounding_box (const cairo_matrix_t *matrix,
387 double *x1, double *y1,
388 double *x2, double *y2,
389 cairo_bool_t *is_tight)
390 {
391 int i;
392 double quad_x[4], quad_y[4];
393 double min_x, max_x;
394 double min_y, max_y;
395
396 if (matrix->xy == 0. && matrix->yx == 0.) {
397 /* non-rotation/skew matrix, just map the two extreme points */
398
399 if (matrix->xx != 1.) {
400 quad_x[0] = *x1 * matrix->xx;
401 quad_x[1] = *x2 * matrix->xx;
402 if (quad_x[0] < quad_x[1]) {
403 *x1 = quad_x[0];
404 *x2 = quad_x[1];
405 } else {
406 *x1 = quad_x[1];
407 *x2 = quad_x[0];
408 }
409 }
410 if (matrix->x0 != 0.) {
411 *x1 += matrix->x0;
412 *x2 += matrix->x0;
413 }
414
415 if (matrix->yy != 1.) {
416 quad_y[0] = *y1 * matrix->yy;
417 quad_y[1] = *y2 * matrix->yy;
418 if (quad_y[0] < quad_y[1]) {
419 *y1 = quad_y[0];
420 *y2 = quad_y[1];
421 } else {
422 *y1 = quad_y[1];
423 *y2 = quad_y[0];
424 }
425 }
426 if (matrix->y0 != 0.) {
427 *y1 += matrix->y0;
428 *y2 += matrix->y0;
429 }
430
431 if (is_tight)
432 *is_tight = TRUE;
433
434 return;
435 }
436
437 /* general matrix */
438 quad_x[0] = *x1;
439 quad_y[0] = *y1;
440 cairo_matrix_transform_point (matrix, &quad_x[0], &quad_y[0]);
441
442 quad_x[1] = *x2;
443 quad_y[1] = *y1;
444 cairo_matrix_transform_point (matrix, &quad_x[1], &quad_y[1]);
445
446 quad_x[2] = *x1;
447 quad_y[2] = *y2;
448 cairo_matrix_transform_point (matrix, &quad_x[2], &quad_y[2]);
449
450 quad_x[3] = *x2;
451 quad_y[3] = *y2;
452 cairo_matrix_transform_point (matrix, &quad_x[3], &quad_y[3]);
453
454 min_x = max_x = quad_x[0];
455 min_y = max_y = quad_y[0];
456
457 for (i=1; i < 4; i++) {
458 if (quad_x[i] < min_x)
459 min_x = quad_x[i];
460 if (quad_x[i] > max_x)
461 max_x = quad_x[i];
462
463 if (quad_y[i] < min_y)
464 min_y = quad_y[i];
465 if (quad_y[i] > max_y)
466 max_y = quad_y[i];
467 }
468
469 *x1 = min_x;
470 *y1 = min_y;
471 *x2 = max_x;
472 *y2 = max_y;
473
474 if (is_tight) {
475 /* it's tight if and only if the four corner points form an axis-aligned
476 rectangle.
477 And that's true if and only if we can derive corners 0 and 3 from
478 corners 1 and 2 in one of two straightforward ways...
479 We could use a tolerance here but for now we'll fall back to FALSE in the case
480 of floating point error.
481 */
482 *is_tight =
483 (quad_x[1] == quad_x[0] && quad_y[1] == quad_y[3] &&
484 quad_x[2] == quad_x[3] && quad_y[2] == quad_y[0]) ||
485 (quad_x[1] == quad_x[3] && quad_y[1] == quad_y[0] &&
486 quad_x[2] == quad_x[0] && quad_y[2] == quad_y[3]);
487 }
488 }
489
490 cairo_private void
_cairo_matrix_transform_bounding_box_fixed(const cairo_matrix_t * matrix,cairo_box_t * bbox,cairo_bool_t * is_tight)491 _cairo_matrix_transform_bounding_box_fixed (const cairo_matrix_t *matrix,
492 cairo_box_t *bbox,
493 cairo_bool_t *is_tight)
494 {
495 double x1, y1, x2, y2;
496
497 _cairo_box_to_doubles (bbox, &x1, &y1, &x2, &y2);
498 _cairo_matrix_transform_bounding_box (matrix, &x1, &y1, &x2, &y2, is_tight);
499 _cairo_box_from_doubles (bbox, &x1, &y1, &x2, &y2);
500 }
501
502 static void
_cairo_matrix_scalar_multiply(cairo_matrix_t * matrix,double scalar)503 _cairo_matrix_scalar_multiply (cairo_matrix_t *matrix, double scalar)
504 {
505 matrix->xx *= scalar;
506 matrix->yx *= scalar;
507
508 matrix->xy *= scalar;
509 matrix->yy *= scalar;
510
511 matrix->x0 *= scalar;
512 matrix->y0 *= scalar;
513 }
514
515 /* This function isn't a correct adjoint in that the implicit 1 in the
516 homogeneous result should actually be ad-bc instead. But, since this
517 adjoint is only used in the computation of the inverse, which
518 divides by det (A)=ad-bc anyway, everything works out in the end. */
519 static void
_cairo_matrix_compute_adjoint(cairo_matrix_t * matrix)520 _cairo_matrix_compute_adjoint (cairo_matrix_t *matrix)
521 {
522 /* adj (A) = transpose (C:cofactor (A,i,j)) */
523 double a, b, c, d, tx, ty;
524
525 _cairo_matrix_get_affine (matrix,
526 &a, &b,
527 &c, &d,
528 &tx, &ty);
529
530 cairo_matrix_init (matrix,
531 d, -b,
532 -c, a,
533 c*ty - d*tx, b*tx - a*ty);
534 }
535
536 /**
537 * cairo_matrix_invert:
538 * @matrix: a #cairo_matrix_t
539 *
540 * Changes @matrix to be the inverse of its original value. Not
541 * all transformation matrices have inverses; if the matrix
542 * collapses points together (it is <firstterm>degenerate</firstterm>),
543 * then it has no inverse and this function will fail.
544 *
545 * Returns: If @matrix has an inverse, modifies @matrix to
546 * be the inverse matrix and returns %CAIRO_STATUS_SUCCESS. Otherwise,
547 * returns %CAIRO_STATUS_INVALID_MATRIX.
548 **/
549 cairo_status_t
cairo_matrix_invert(cairo_matrix_t * matrix)550 cairo_matrix_invert (cairo_matrix_t *matrix)
551 {
552 double det;
553
554 /* Simple scaling|translation matrices are quite common... */
555 if (matrix->xy == 0. && matrix->yx == 0.) {
556 matrix->x0 = -matrix->x0;
557 matrix->y0 = -matrix->y0;
558
559 if (matrix->xx != 1.) {
560 if (matrix->xx == 0.)
561 return _cairo_error (CAIRO_STATUS_INVALID_MATRIX);
562
563 matrix->xx = 1. / matrix->xx;
564 matrix->x0 *= matrix->xx;
565 }
566
567 if (matrix->yy != 1.) {
568 if (matrix->yy == 0.)
569 return _cairo_error (CAIRO_STATUS_INVALID_MATRIX);
570
571 matrix->yy = 1. / matrix->yy;
572 matrix->y0 *= matrix->yy;
573 }
574
575 return CAIRO_STATUS_SUCCESS;
576 }
577
578 /* inv (A) = 1/det (A) * adj (A) */
579 det = _cairo_matrix_compute_determinant (matrix);
580
581 if (! ISFINITE (det))
582 return _cairo_error (CAIRO_STATUS_INVALID_MATRIX);
583
584 if (det == 0)
585 return _cairo_error (CAIRO_STATUS_INVALID_MATRIX);
586
587 _cairo_matrix_compute_adjoint (matrix);
588 _cairo_matrix_scalar_multiply (matrix, 1 / det);
589
590 return CAIRO_STATUS_SUCCESS;
591 }
592 slim_hidden_def(cairo_matrix_invert);
593
594 cairo_bool_t
_cairo_matrix_is_invertible(const cairo_matrix_t * matrix)595 _cairo_matrix_is_invertible (const cairo_matrix_t *matrix)
596 {
597 double det;
598
599 det = _cairo_matrix_compute_determinant (matrix);
600
601 return ISFINITE (det) && det != 0.;
602 }
603
604 cairo_bool_t
_cairo_matrix_is_scale_0(const cairo_matrix_t * matrix)605 _cairo_matrix_is_scale_0 (const cairo_matrix_t *matrix)
606 {
607 return matrix->xx == 0. &&
608 matrix->xy == 0. &&
609 matrix->yx == 0. &&
610 matrix->yy == 0.;
611 }
612
613 double
_cairo_matrix_compute_determinant(const cairo_matrix_t * matrix)614 _cairo_matrix_compute_determinant (const cairo_matrix_t *matrix)
615 {
616 double a, b, c, d;
617
618 a = matrix->xx; b = matrix->yx;
619 c = matrix->xy; d = matrix->yy;
620
621 return a*d - b*c;
622 }
623
624 /**
625 * _cairo_matrix_compute_basis_scale_factors:
626 * @matrix: a matrix
627 * @basis_scale: the scale factor in the direction of basis
628 * @normal_scale: the scale factor in the direction normal to the basis
629 * @x_basis: basis to use. X basis if true, Y basis otherwise.
630 *
631 * Computes |Mv| and det(M)/|Mv| for v=[1,0] if x_basis is true, and v=[0,1]
632 * otherwise, and M is @matrix.
633 *
634 * Return value: the scale factor of @matrix on the height of the font,
635 * or 1.0 if @matrix is %NULL.
636 **/
637 cairo_status_t
_cairo_matrix_compute_basis_scale_factors(const cairo_matrix_t * matrix,double * basis_scale,double * normal_scale,cairo_bool_t x_basis)638 _cairo_matrix_compute_basis_scale_factors (const cairo_matrix_t *matrix,
639 double *basis_scale, double *normal_scale,
640 cairo_bool_t x_basis)
641 {
642 double det;
643
644 det = _cairo_matrix_compute_determinant (matrix);
645
646 if (! ISFINITE (det))
647 return _cairo_error (CAIRO_STATUS_INVALID_MATRIX);
648
649 if (det == 0)
650 {
651 *basis_scale = *normal_scale = 0;
652 }
653 else
654 {
655 double x = x_basis != 0;
656 double y = x == 0;
657 double major, minor;
658
659 cairo_matrix_transform_distance (matrix, &x, &y);
660 major = hypot (x, y);
661 /*
662 * ignore mirroring
663 */
664 if (det < 0)
665 det = -det;
666 if (major)
667 minor = det / major;
668 else
669 minor = 0.0;
670 if (x_basis)
671 {
672 *basis_scale = major;
673 *normal_scale = minor;
674 }
675 else
676 {
677 *basis_scale = minor;
678 *normal_scale = major;
679 }
680 }
681
682 return CAIRO_STATUS_SUCCESS;
683 }
684
685 cairo_bool_t
_cairo_matrix_is_identity(const cairo_matrix_t * matrix)686 _cairo_matrix_is_identity (const cairo_matrix_t *matrix)
687 {
688 return (matrix->xx == 1.0 && matrix->yx == 0.0 &&
689 matrix->xy == 0.0 && matrix->yy == 1.0 &&
690 matrix->x0 == 0.0 && matrix->y0 == 0.0);
691 }
692
693 cairo_bool_t
_cairo_matrix_is_translation(const cairo_matrix_t * matrix)694 _cairo_matrix_is_translation (const cairo_matrix_t *matrix)
695 {
696 return (matrix->xx == 1.0 && matrix->yx == 0.0 &&
697 matrix->xy == 0.0 && matrix->yy == 1.0);
698 }
699
700 cairo_bool_t
_cairo_matrix_is_integer_translation(const cairo_matrix_t * matrix,int * itx,int * ity)701 _cairo_matrix_is_integer_translation (const cairo_matrix_t *matrix,
702 int *itx, int *ity)
703 {
704 if (_cairo_matrix_is_translation (matrix))
705 {
706 cairo_fixed_t x0_fixed = _cairo_fixed_from_double (matrix->x0);
707 cairo_fixed_t y0_fixed = _cairo_fixed_from_double (matrix->y0);
708
709 if (_cairo_fixed_is_integer (x0_fixed) &&
710 _cairo_fixed_is_integer (y0_fixed))
711 {
712 if (itx)
713 *itx = _cairo_fixed_integer_part (x0_fixed);
714 if (ity)
715 *ity = _cairo_fixed_integer_part (y0_fixed);
716
717 return TRUE;
718 }
719 }
720
721 return FALSE;
722 }
723
724 cairo_bool_t
_cairo_matrix_has_unity_scale(const cairo_matrix_t * matrix)725 _cairo_matrix_has_unity_scale (const cairo_matrix_t *matrix)
726 {
727 if (matrix->xy == 0.0 && matrix->yx == 0.0) {
728 if (! (matrix->xx == 1.0 || matrix->xx == -1.0))
729 return FALSE;
730 if (! (matrix->yy == 1.0 || matrix->yy == -1.0))
731 return FALSE;
732 } else if (matrix->xx == 0.0 && matrix->yy == 0.0) {
733 if (! (matrix->xy == 1.0 || matrix->xy == -1.0))
734 return FALSE;
735 if (! (matrix->yx == 1.0 || matrix->yx == -1.0))
736 return FALSE;
737 } else
738 return FALSE;
739
740 return TRUE;
741 }
742
743 /* By pixel exact here, we mean a matrix that is composed only of
744 * 90 degree rotations, flips, and integer translations and produces a 1:1
745 * mapping between source and destination pixels. If we transform an image
746 * with a pixel-exact matrix, filtering is not useful.
747 */
748 cairo_bool_t
_cairo_matrix_is_pixel_exact(const cairo_matrix_t * matrix)749 _cairo_matrix_is_pixel_exact (const cairo_matrix_t *matrix)
750 {
751 cairo_fixed_t x0_fixed, y0_fixed;
752
753 if (! _cairo_matrix_has_unity_scale (matrix))
754 return FALSE;
755
756 x0_fixed = _cairo_fixed_from_double (matrix->x0);
757 y0_fixed = _cairo_fixed_from_double (matrix->y0);
758
759 return _cairo_fixed_is_integer (x0_fixed) && _cairo_fixed_is_integer (y0_fixed);
760 }
761
762 /*
763 A circle in user space is transformed into an ellipse in device space.
764
765 The following is a derivation of a formula to calculate the length of the
766 major axis for this ellipse; this is useful for error bounds calculations.
767
768 Thanks to Walter Brisken <wbrisken@aoc.nrao.edu> for this derivation:
769
770 1. First some notation:
771
772 All capital letters represent vectors in two dimensions. A prime '
773 represents a transformed coordinate. Matrices are written in underlined
774 form, ie _R_. Lowercase letters represent scalar real values.
775
776 2. The question has been posed: What is the maximum expansion factor
777 achieved by the linear transformation
778
779 X' = X _R_
780
781 where _R_ is a real-valued 2x2 matrix with entries:
782
783 _R_ = [a b]
784 [c d] .
785
786 In other words, what is the maximum radius, MAX[ |X'| ], reached for any
787 X on the unit circle ( |X| = 1 ) ?
788
789 3. Some useful formulae
790
791 (A) through (C) below are standard double-angle formulae. (D) is a lesser
792 known result and is derived below:
793
794 (A) sin²(θ) = (1 - cos(2*θ))/2
795 (B) cos²(θ) = (1 + cos(2*θ))/2
796 (C) sin(θ)*cos(θ) = sin(2*θ)/2
797 (D) MAX[a*cos(θ) + b*sin(θ)] = sqrt(a² + b²)
798
799 Proof of (D):
800
801 find the maximum of the function by setting the derivative to zero:
802
803 -a*sin(θ)+b*cos(θ) = 0
804
805 From this it follows that
806
807 tan(θ) = b/a
808
809 and hence
810
811 sin(θ) = b/sqrt(a² + b²)
812
813 and
814
815 cos(θ) = a/sqrt(a² + b²)
816
817 Thus the maximum value is
818
819 MAX[a*cos(θ) + b*sin(θ)] = (a² + b²)/sqrt(a² + b²)
820 = sqrt(a² + b²)
821
822 4. Derivation of maximum expansion
823
824 To find MAX[ |X'| ] we search brute force method using calculus. The unit
825 circle on which X is constrained is to be parameterized by t:
826
827 X(θ) = (cos(θ), sin(θ))
828
829 Thus
830
831 X'(θ) = X(θ) * _R_ = (cos(θ), sin(θ)) * [a b]
832 [c d]
833 = (a*cos(θ) + c*sin(θ), b*cos(θ) + d*sin(θ)).
834
835 Define
836
837 r(θ) = |X'(θ)|
838
839 Thus
840
841 r²(θ) = (a*cos(θ) + c*sin(θ))² + (b*cos(θ) + d*sin(θ))²
842 = (a² + b²)*cos²(θ) + (c² + d²)*sin²(θ)
843 + 2*(a*c + b*d)*cos(θ)*sin(θ)
844
845 Now apply the double angle formulae (A) to (C) from above:
846
847 r²(θ) = (a² + b² + c² + d²)/2
848 + (a² + b² - c² - d²)*cos(2*θ)/2
849 + (a*c + b*d)*sin(2*θ)
850 = f + g*cos(φ) + h*sin(φ)
851
852 Where
853
854 f = (a² + b² + c² + d²)/2
855 g = (a² + b² - c² - d²)/2
856 h = (a*c + d*d)
857 φ = 2*θ
858
859 It is clear that MAX[ |X'| ] = sqrt(MAX[ r² ]). Here we determine MAX[ r² ]
860 using (D) from above:
861
862 MAX[ r² ] = f + sqrt(g² + h²)
863
864 And finally
865
866 MAX[ |X'| ] = sqrt( f + sqrt(g² + h²) )
867
868 Which is the solution to this problem.
869
870 Walter Brisken
871 2004/10/08
872
873 (Note that the minor axis length is at the minimum of the above solution,
874 which is just sqrt ( f - sqrt(g² + h²) ) given the symmetry of (D)).
875
876
877 For another derivation of the same result, using Singular Value Decomposition,
878 see doc/tutorial/src/singular.c.
879 */
880
881 /* determine the length of the major axis of a circle of the given radius
882 after applying the transformation matrix. */
883 double
_cairo_matrix_transformed_circle_major_axis(const cairo_matrix_t * matrix,double radius)884 _cairo_matrix_transformed_circle_major_axis (const cairo_matrix_t *matrix,
885 double radius)
886 {
887 double a, b, c, d, f, g, h, i, j;
888
889 _cairo_matrix_get_affine (matrix,
890 &a, &b,
891 &c, &d,
892 NULL, NULL);
893
894 i = a*a + b*b;
895 j = c*c + d*d;
896
897 f = 0.5 * (i + j);
898 g = 0.5 * (i - j);
899 h = a*c + b*d;
900
901 return radius * sqrt (f + hypot (g, h));
902
903 /*
904 * we don't need the minor axis length, which is
905 * double min = radius * sqrt (f - sqrt (g*g+h*h));
906 */
907 }
908
909 void
_cairo_matrix_to_pixman_matrix(const cairo_matrix_t * matrix,pixman_transform_t * pixman_transform,double xc,double yc)910 _cairo_matrix_to_pixman_matrix (const cairo_matrix_t *matrix,
911 pixman_transform_t *pixman_transform,
912 double xc,
913 double yc)
914 {
915 static const pixman_transform_t pixman_identity_transform = {{
916 {1 << 16, 0, 0},
917 { 0, 1 << 16, 0},
918 { 0, 0, 1 << 16}
919 }};
920
921 if (_cairo_matrix_is_identity (matrix)) {
922 *pixman_transform = pixman_identity_transform;
923 } else {
924 cairo_matrix_t inv;
925 unsigned max_iterations;
926
927 pixman_transform->matrix[0][0] = _cairo_fixed_16_16_from_double (matrix->xx);
928 pixman_transform->matrix[0][1] = _cairo_fixed_16_16_from_double (matrix->xy);
929 pixman_transform->matrix[0][2] = _cairo_fixed_16_16_from_double (matrix->x0);
930
931 pixman_transform->matrix[1][0] = _cairo_fixed_16_16_from_double (matrix->yx);
932 pixman_transform->matrix[1][1] = _cairo_fixed_16_16_from_double (matrix->yy);
933 pixman_transform->matrix[1][2] = _cairo_fixed_16_16_from_double (matrix->y0);
934
935 pixman_transform->matrix[2][0] = 0;
936 pixman_transform->matrix[2][1] = 0;
937 pixman_transform->matrix[2][2] = 1 << 16;
938
939 /* The conversion above breaks cairo's translation invariance:
940 * a translation of (a, b) in device space translates to
941 * a translation of (xx * a + xy * b, yx * a + yy * b)
942 * for cairo, while pixman uses rounded versions of xx ... yy.
943 * This error increases as a and b get larger.
944 *
945 * To compensate for this, we fix the point (xc, yc) in pattern
946 * space and adjust pixman's transform to agree with cairo's at
947 * that point.
948 */
949
950 if (_cairo_matrix_has_unity_scale (matrix))
951 return;
952
953 /* Note: If we can't invert the transformation, skip the adjustment. */
954 inv = *matrix;
955 if (cairo_matrix_invert (&inv) != CAIRO_STATUS_SUCCESS)
956 return;
957
958 /* find the pattern space coordinate that maps to (xc, yc) */
959 xc += .5; yc += .5; /* offset for the pixel centre */
960 max_iterations = 5;
961 do {
962 double x,y;
963 pixman_vector_t vector;
964 cairo_fixed_16_16_t dx, dy;
965
966 vector.vector[0] = _cairo_fixed_16_16_from_double (xc);
967 vector.vector[1] = _cairo_fixed_16_16_from_double (yc);
968 vector.vector[2] = 1 << 16;
969
970 if (! pixman_transform_point_3d (pixman_transform, &vector))
971 return;
972
973 x = pixman_fixed_to_double (vector.vector[0]);
974 y = pixman_fixed_to_double (vector.vector[1]);
975 cairo_matrix_transform_point (&inv, &x, &y);
976
977 /* Ideally, the vector should now be (xc, yc).
978 * We can now compensate for the resulting error.
979 */
980 x -= xc;
981 y -= yc;
982 cairo_matrix_transform_distance (matrix, &x, &y);
983 dx = _cairo_fixed_16_16_from_double (x);
984 dy = _cairo_fixed_16_16_from_double (y);
985 pixman_transform->matrix[0][2] -= dx;
986 pixman_transform->matrix[1][2] -= dy;
987
988 if (dx == 0 && dy == 0)
989 break;
990 } while (--max_iterations);
991 }
992 }
993