Repository navigation
Expand file tree
/
Copy pathSDL_RLEaccel.c
More file actions
1602 lines (1475 loc) · 71.7 KB
/
Copy pathSDL_RLEaccel.c
File metadata and controls
1602 lines (1475 loc) · 71.7 KB
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
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
Simple DirectMedia Layer
Copyright (C) 1997-2024 Sam Lantinga <slouken@libsdl.org>
This software is provided 'as-is', without any express or implied
warranty. In no event will the authors be held liable for any damages
arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented; you must not
claim that you wrote the original software. If you use this software
in a product, an acknowledgment in the product documentation would be
appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
3. This notice may not be removed or altered from any source distribution.
*/
#include "SDL_internal.h"
#if SDL_HAVE_RLE
/*
* RLE encoding for software colorkey and alpha-channel acceleration
*
* Original version by Sam Lantinga
*
* Mattias Engdegård (Yorick): Rewrite. New encoding format, encoder and
* decoder. Added per-surface alpha blitter. Added per-pixel alpha
* format, encoder and blitter.
*
* Many thanks to Xark and johns for hints, benchmarks and useful comments
* leading to this code.
*
* Welcome to Macro Mayhem.
*/
/*
* The encoding translates the image data to a stream of segments of the form
*
* <skip> <run> <data>
*
* where <skip> is the number of transparent pixels to skip,
* <run> is the number of opaque pixels to blit,
* and <data> are the pixels themselves.
*
* This basic structure is used both for colorkeyed surfaces, used for simple
* binary transparency and for per-surface alpha blending, and for surfaces
* with per-pixel alpha. The details differ, however:
*
* Encoding of colorkeyed surfaces:
*
* Encoded pixels always have the same format as the target surface.
* <skip> and <run> are unsigned 8 bit integers, except for 32 bit depth
* where they are 16 bit. This makes the pixel data aligned at all times.
* Segments never wrap around from one scan line to the next.
*
* The end of the sequence is marked by a zero <skip>,<run> pair at the *
* beginning of a line.
*
* Encoding of surfaces with per-pixel alpha:
*
* The sequence begins with a struct RLEDestFormat describing the target
* pixel format, to provide reliable un-encoding.
*
* Each scan line is encoded twice: First all completely opaque pixels,
* encoded in the target format as described above, and then all
* partially transparent (translucent) pixels (where 1 <= alpha <= 254),
* in the following 32-bit format:
*
* For 32-bit targets, each pixel has
10BC8
the target RGB format but with
* the alpha value occupying the highest 8 bits. The <skip> and <run>
* counts are 16 bit.
*
* For 16-bit targets, each pixel has the target RGB format, but with
* the middle component (usually green) shifted 16 steps to the left,
* and the hole filled with the 5 most significant bits of the alpha value.
* i.e. if the target has the format rrrrrggggggbbbbb,
* the encoded pixel will be 00000gggggg00000rrrrr0aaaaabbbbb.
* The <skip> and <run> counts are 8 bit for the opaque lines, 16 bit
* for the translucent lines. Two padding bytes may be inserted
* before each translucent line to keep them 32-bit aligned.
*
* The end of the sequence is marked by a zero <skip>,<run> pair at the
* beginning of an opaque line.
*/
#include "SDL_sysvideo.h"
#include "SDL_blit.h"
#include "SDL_RLEaccel_c.h"
#define PIXEL_COPY(to, from, len, bpp) \
SDL_memcpy(to, from, (size_t)(len) * (bpp))
/*
* Various colorkey blit methods, for opaque and per-surface alpha
*/
#define OPAQUE_BLIT(to, from, length, bpp, alpha) \
PIXEL_COPY(to, from, length, bpp)
/*
* For 32bpp pixels on the form 0x00rrggbb:
* If we treat the middle component separately, we can process the two
* remaining in parallel. This is safe to do because of the gap to the left
* of each component, so the bits from the multiplication don't collide.
* This can be used for any RGB permutation of course.
*/
#define ALPHA_BLIT32_888(to, from, length, bpp, alpha) \
do { \
int i; \
Uint32 *src = (Uint32 *)(from); \
Uint32 *dst = (Uint32 *)(to); \
for (i = 0; i < (int)(length); i++) { \
Uint32 s = *src++; \
Uint32 d = *dst; \
Uint32 s1 = s & 0xff00ff; \
Uint32 d1 = d & 0xff00ff; \
d1 = (d1 + ((s1 - d1) * alpha >> 8)) & 0xff00ff; \
s &= 0xff00; \
d &= 0xff00; \
d = (d + ((s - d) * alpha >> 8)) & 0xff00; \
*dst++ = d1 | d; \
} \
} while (0)
/*
* For 16bpp pixels we can go a step further: put the middle component
* in the high 16 bits of a 32 bit word, and process all three RGB
* components at the same time. Since the smallest gap is here just
* 5 bits, we have to scale alpha down to 5 bits as well.
*/
#define ALPHA_BLIT16_565(to, from, length, bpp, alpha) \
do { \
int i; \
Uint16 *src = (Uint16 *)(from); \
Uint16 *dst = (Uint16 *)(to); \
Uint32 ALPHA = alpha >> 3; \
for (i = 0; i < (int)(length); i++) { \
Uint32 s = *src++; \
Uint32 d = *dst; \
s = (s | s << 16) & 0x07e0f81f; \
d = (d | d << 16) & 0x07e0f81f; \
d += (s - d) * ALPHA >> 5; \
d &= 0x07e0f81f; \
*dst++ = (Uint16)(d | d >> 16); \
} \
} while (0)
#define ALPHA_BLIT16_555(to, from, length, bpp, alpha) \
do { \
int i; \
Uint16 *src = (Uint16 *)(from); \
Uint16 *dst = (Uint16 *)(to); \
Uint32 ALPHA = alpha >> 3; \
for (i = 0; i < (int)(length); i++) { \
Uint32 s = *src++; \
Uint32 d = *dst; \
s = (s | s << 16) & 0x03e07c1f; \
d = (d | d << 16) & 0x03e07c1f; \
d += (s - d) * ALPHA >> 5; \
d &= 0x03e07c1f; \
*dst++ = (Uint16)(d | d >> 16); \
} \
} while (0)
/*
* The general slow catch-all function, for remaining depths and formats
*/
#define ALPHA_BLIT_ANY(to, from, length, bpp, alpha) \
do { \
int i; \
Uint8 *src = from; \
Uint8 *dst = to; \
for (i = 0; i < (int)(length); i++) { \
Uint32 s = 0, d = 0; \
unsigned rs, gs, bs, rd, gd, bd; \
switch (bpp) { \
case 2: \
s = *(Uint16 *)src; \
d = *(Uint16 *)dst; \
break; \
case 3: \
if (SDL_BYTEORDER == SDL_BIG_ENDIAN) { \
s = (src[0] << 16) | (src[1] << 8) | src[2]; \
d = (dst[0] << 16) | (dst[1] << 8) | dst[2]; \
} else { \
s = (src[2] << 16) | (src[1] << 8) | src[0]; \
d = (dst[2] << 16) | (dst[1] << 8) | dst[0]; \
} \
break; \
case 4: \
s = *(Uint32 *)src; \
d = *(Uint32 *)dst; \
break; \
} \
RGB_FROM_PIXEL(s, fmt, rs, gs, bs); \
RGB_FROM_PIXEL(d, fmt, rd, gd, bd); \
rd += (rs - rd) * alpha >> 8; \
gd += (gs - gd) * alpha >> 8; \
bd += (bs - bd) * alpha >> 8; \
PIXEL_FROM_RGB(d, fmt, rd, gd, bd); \
switch (bpp) { \
case 2: \
*(Uint16 *)dst = (Uint16)d; \
break; \
case 3: \
if (SDL_BYTEORDER == SDL_BIG_ENDIAN) { \
dst[0] = (Uint8)(d >> 16); \
dst[1] = (Uint8)(d >> 8); \
dst[2] = (Uint8)(d); \
} else { \
dst[0] = (Uint8)d; \
dst[1] = (Uint8)(d >> 8); \
dst[2] = (Uint8)(d >> 16); \
} \
break; \
case 4: \
*(Uint32 *)dst = d; \
break; \
} \
src += bpp; \
dst += bpp; \
} \
} while (0)
/*
* Special case: 50% alpha (alpha=128)
* This is treated specially because it can be optimized very well, and
* since it is good for many cases of semi-translucency.
* The theory is to do all three components at the same time:
* First zero the lowest bit of each component, which gives us room to
* add them. Then shift right and add the sum of the lowest bits.
*/
#define ALPHA_BLIT32_888_50(to, from, length, bpp, alpha) \
do { \
int i; \
Uint32 *src = (Uint32 *)(from); \
Uint32 *dst = (Uint32 *)(to); \
for (i = 0; i < (int)(length); i++) { \
Uint32 s = *src++; \
Uint32 d = *dst; \
*dst++ = (((s & 0x00fefefe) + (d & 0x00fefefe)) >> 1) + (s & d & 0x00010101); \
} \
} while (0)
/*
* For 16bpp, we can actually blend two pixels in parallel, if we take
* care to shift before we add, not after.
*/
/* helper: blend a single 16 bit pixel at 50% */
#define BLEND16_50(dst, src, mask) \
do { \
Uint32 s = *src++; \
Uint32 d = *dst; \
*dst++ = (Uint16)((((s & mask) + (d & mask)) >> 1) + \
(s & d & (~mask & 0xffff))); \
} while (0)
/* basic 16bpp blender. mask is the pixels to keep when adding. */
#define ALPHA_BLIT16_50(to, from, length, bpp, alpha, mask) \
do { \
unsigned n = (length); \
Uint16 *src = (Uint16 *)(from); \
Uint16 *dst = (Uint16 *)(to); \
if (((uintptr_t)src ^ (uintptr_t)dst) & 3) { \
/* source and destination not in phase, blit one by one */ \
while (n--) \
BLEND16_50(dst, src, mask); \
} else { \
if ((uintptr_t)src & 3) { \
/* first odd pixel */ \
BLEND16_50(dst, src, mask); \
n--; \
} \
for (; n > 1; n -= 2) { \
Uint32 s = *(Uint32 *)src; \
Uint32 d = *(Uint32 *)dst; \
*(Uint32 *)dst = ((s & (mask | mask << 16)) >> 1) + ((d & (mask | mask << 16)) >> 1) + (s & d & (~(mask | mask << 16))); \
src += 2; \
dst += 2; \
} \
if (n) \
BLEND16_50(dst, src, mask); /* last odd pixel */ \
} \
} while (0)
#define ALPHA_BLIT16_565_50(to, from, length, bpp, alpha) \
ALPHA_BLIT16_50(to, from, length, bpp, alpha, 0xf7deU)
#define ALPHA_BLIT16_555_50(to, from, length, bpp, alpha) \
ALPHA_BLIT16_50(to, from, length, bpp, alpha, 0xfbdeU)
#define CHOOSE_BLIT(blitter, alpha, fmt) \
do { \
if (alpha == 255) { \
switch (fmt->bytes_per_pixel) { \
case 1: \
blitter(1, Uint8, OPAQUE_BLIT); \
break; \
case 2: \
blitter(2, Uint8, OPAQUE_BLIT); \
break; \
case 3: \
blitter(3, Uint8, OPAQUE_BLIT); \
10BC8
break; \
case 4: \
blitter(4, Uint16, OPAQUE_BLIT); \
break; \
} \
} else { \
switch (fmt->bytes_per_pixel) { \
case 1: \
/* No 8bpp alpha blitting */ \
break; \
\
case 2: \
switch (fmt->Rmask | fmt->Gmask | fmt->Bmask) { \
case 0xffff: \
if (fmt->Gmask == 0x07e0 || fmt->Rmask == 0x07e0 || fmt->Bmask == 0x07e0) { \
if (alpha == 128) { \
blitter(2, Uint8, ALPHA_BLIT16_565_50); \
} else { \
blitter(2, Uint8, ALPHA_BLIT16_565); \
} \
} else { \
goto general16; \
} \
break; \
\
case 0x7fff: \
if (fmt->Gmask == 0x03e0 || fmt->Rmask == 0x03e0 || fmt->Bmask == 0x03e0) { \
if (alpha == 128) { \
blitter(2, Uint8, ALPHA_BLIT16_555_50); \
} else { \
blitter(2, Uint8, ALPHA_BLIT16_555); \
} \
break; \
} else { \
goto general16; \
} \
break; \
\
default: \
general16: \
blitter(2, Uint8, ALPHA_BLIT_ANY); \
} \
break; \
\
case 3: \
blitter(3, Uint8, ALPHA_BLIT_ANY); \
break; \
\
case 4: \
if ((fmt->Rmask | fmt->Gmask | fmt->Bmask) == 0x00ffffff && (fmt->Gmask == 0xff00 || fmt->Rmask == 0xff00 || fmt->Bmask == 0xff00)) { \
if (alpha == 128) { \
blitter(4, Uint16, ALPHA_BLIT32_888_50); \
} else { \
blitter(4, Uint16, ALPHA_BLIT32_888); \
} \
} else { \
blitter(4, Uint16, ALPHA_BLIT_ANY); \
} \
break; \
} \
} \
} while (0)
/*
* Set a pixel value using the given format, except that the alpha value is
* placed in the top byte. This is the format used for RLE with alpha.
*/
#define RLEPIXEL_FROM_RGBA(Pixel, fmt, r, g, b, a) \
{ \
Pixel = ((r >> fmt->Rloss) << fmt->Rshift) | \
((g >> fmt->Gloss) << fmt->Gshift) | \
((b >> fmt->Bloss) << fmt->Bshift) | \
(a << 24); \
}
/*
* This takes care of the case when the surface is clipped on the left and/or
* right. Top clipping has already been taken care of.
*/
#define RLECLIPBLIT(bpp, Type, do_blit) \
do { \
int linecount = srcrect->h; \
int ofs = 0; \
int left = srcrect->x; \
int right = left + srcrect->w; \
dstbuf -= left * bpp; \
for (;;) { \
int run; \
ofs += *(Type *)srcbuf; \
run = ((Type *)srcbuf)[1]; \
srcbuf += 2 * sizeof(Type); \
if (run) { \
/* clip to left and right borders */ \
if (ofs < right) { \
int start = 0; \
int len = run; \
int startcol; \
if (left - ofs > 0) { \
start = left - ofs; \
len -= start; \
if (len <= 0) \
goto nocopy##bpp##do_blit; \
} \
startcol = ofs + start; \
if (len > right - startcol) \
len = right - startcol; \
do_blit(dstbuf + startcol * bpp, srcbuf + start * bpp, \
len, bpp, alpha); \
} \
nocopy##bpp##do_blit : srcbuf += run * bpp; \
ofs += run; \
} else if (!ofs) { \
break; \
} \
\
if (ofs == w) { \
ofs = 0; \
dstbuf += surf_dst->pitch; \
if (!--linecount) { \
break; \
} \
} \
} \
} while (0)
static void RLEClipBlit(int w, Uint8 *srcbuf, SDL_Surface *surf_dst,
Uint8 *dstbuf, const SDL_Rect *srcrect, unsigned alpha)
{
SDL_PixelFormat *fmt = surf_dst->format;
CHOOSE_BLIT(RLECLIPBLIT, alpha, fmt);
}
#undef RLECLIPBLIT
/* blit a colorkeyed RLE surface */
static int SDLCALL SDL_RLEBlit(SDL_Surface *surf_src, const SDL_Rect *srcrect,
SDL_Surface *surf_dst, const SDL_Rect *dstrect)
{
Uint8 *dstbuf;
Uint8 *srcbuf;
int x, y;
int w = surf_src->w;
unsigned alpha;
/* Lock the destination if necessary */
if (SDL_MUSTLOCK(surf_dst)) {
if (SDL_LockSurface(surf_dst) < 0) {
return -1;
}
}
/* Set up the source and destination pointers */
x = dstrect->x;
y = dstrect->y;
dstbuf = (Uint8 *)surf_dst->pixels + y * surf_dst->pitch + x * surf_src->format->bytes_per_pixel;
srcbuf = (Uint8 *)surf_src->map->data;
{
/* skip lines at the top if necessary */
int vskip = srcrect->y;
int ofs = 0;
if (vskip) {
#define RLESKIP(bpp, Type) \
for (;;) { \
int run; \
ofs += *(Type *)srcbuf; \
run = ((Type *)srcbuf)[1]; \
srcbuf += sizeof(Type) * 2; \
if (run) { \
srcbuf += run * bpp; \
ofs += run; \
} else if (!ofs) \
goto done; \
if (ofs == w) { \
ofs = 0; \
if (!--vskip) \
break; \
} \
}
switch (surf_src->format->bytes_per_pixel) {
case 1:
RLESKIP(1, Uint8);
break;
case 2:
RLESKIP(2, Uint8);
break;
case 3:
RLESKIP(3, Uint8);
break;
case 4:
RLESKIP(4, Uint16);
break;
}
#undef RLESKIP
}
}
alpha = surf_src->map->info.a;
/* if left or right edge clipping needed, call clip blit */
if (srcrect->x || srcrect->w != surf_src->w) {
RLEClipBlit(w, srcbuf, surf_dst, dstbuf, srcrect, alpha);
} else {
SDL_PixelFormat *fmt = surf_src->format;
#define RLEBLIT(bpp, Type, do_blit) \
do { \
int linecount = srcrect->h; \
int ofs = 0; \
for (;;) { \
unsigned run; \
ofs += *(Type *)srcbuf; \
run = ((Type *)srcbuf)[1]; \
srcbuf += 2 * sizeof(Type); \
if (run) { \
do_blit(dstbuf + ofs * bpp, srcbuf, run, bpp, alpha); \
srcbuf += run * bpp; \
ofs += run; \
} else if (!ofs) \
break; \
if (ofs == w) { \
ofs = 0; \
dstbuf += surf_dst->pitch; \
if (!--linecount) \
break; \
} \
} \
} while (0)
CHOOSE_BLIT(RLEBLIT, alpha, fmt);
#undef RLEBLIT
}
done:
/* Unlock the destination if necessary */
if (SDL_MUSTLOCK(surf_dst)) {
SDL_UnlockSurface(surf_dst);
}
return 0;
}
#undef OPAQUE_BLIT
/*
* Per-pixel blitting macros for translucent pixels:
* These use the same techniques as the per-surface blitting macros
*/
/*
* For 32bpp pixels, we have made sure the alpha is stored in the top
* 8 bits, so proceed as usual
*/
#define BLIT_TRANSL_888(src, dst) \
do { \
Uint32 s = src; \
Uint32 d = dst; \
unsigned alpha = s >> 24; \
Uint32 s1 = s & 0xff00ff; \
Uint32 d1 = d & 0xff00ff; \
d1 = (d1 + ((s1 - d1) * alpha >> 8)) & 0xff00ff; \
s &= 0xff00; \
d &= 0xff00; \
d = (d + ((s - d) * alpha >> 8)) & 0xff00; \
dst = d1 | d | 0xff000000; \
} while (0)
/*
* For 16bpp pixels, we have stored the 5 most significant alpha bits in
* bits 5-10. As before, we can process all 3 RGB components at the same time.
*/
#define BLIT_TRANSL_565(src, dst) \
do { \
Uint32 s = src; \
Uint32 d = dst; \
unsigned alpha = (s & 0x3e0) >> 5; \
s &= 0x07e0f81f; \
d = (d | d << 16) & 0x07e0f81f; \
d += (s - d) * alpha >> 5; \
d &= 0x07e0f81f; \
dst = (Uint16)(d | d >> 16); \
} while (0)
#define BLIT_TRANSL_555(src, dst) \
do { \
Uint32 s = src; \
Uint32 d = dst; \
unsigned alpha = (s & 0x3e0) >> 5; \
s &= 0x03e07c1f; \
d = (d | d << 16) & 0x03e07c1f; \
d += (s - d) * alpha >> 5; \
d &= 0x03e07c1f; \
dst = (Uint16)(d | d >> 16); \
} while (0)
/* used to save the destination format in the encoding. Designed to be
macro-compatible with SDL_PixelFormat but without the unneeded fields */
typedef struct
{
Uint8 bytes_per_pixel;
Uint8 padding[3];
Uint32 Rmask;
Uint32 Gmask;
Uint32 Bmask;
Uint32 Amask;
Uint8 Rloss;
Uint8 Gloss;
Uint8 Bloss;
Uint8 Aloss;
Uint8 Rshift;
Uint8 Gshift;
Uint8 Bshift;
Uint8 Ashift;
} RLEDestFormat;
/* blit a pixel-alpha RLE surface clipped at the right and/or left edges */
static void RLEAlphaClipBlit(int w, Uint8 *srcbuf, SDL_Surface *surf_dst,
Uint8 *dstbuf, const SDL_Rect *srcrect)
{
SDL_PixelFormat *df = surf_dst->format;
/*
* clipped blitter: Ptype is the destination pixel type,
* Ctype the translucent count type, and do_blend the macro
* to blend one pixel.
*/
#define RLEALPHACLIPBLIT(Ptype, Ctype, do_blend) \
do { \
int linecount = srcrect->h; \
int left = srcrect->x; \
int right = left + srcrect->w; \
dstbuf -= left * sizeof(Ptype); \
do { \
int ofs = 0; \
/* blit opaque pixels on one line */ \
do { \
unsigned run; \
ofs += ((Ctype *)srcbuf)[0]; \
run = ((Ctype *)srcbuf)[1]; \
srcbuf += 2 * sizeof(Ctype); \
if (run) { \
/* clip to left and right borders */ \
int cofs = ofs; \
int crun = run; \
if (left - cofs > 0) { \
crun -= left - cofs; \
cofs = left; \
} \
if (crun > right - cofs) \
crun = right - cofs; \
if (crun > 0) \
PIXEL_COPY(dstbuf + cofs * sizeof(Ptype), \
srcbuf + (cofs - ofs) * sizeof(Ptype), \
(unsigned)crun, sizeof(Ptype)); \
srcbuf += run * sizeof(Ptype); \
ofs += run; \
} else if (!ofs) \
return; \
} while (ofs < w); \
/* skip padding if necessary */ \
if (sizeof(Ptype) == 2) \
srcbuf += (uintptr_t)srcbuf & 2; \
/* blit translucent pixels on the same line */ \
ofs = 0; \
do { \
unsigned run; \
ofs += ((Uint16 *)srcbuf)[0]; \
run = ((Uint16 *)srcbuf)[1]; \
srcbuf += 4; \
if (run) { \
/* clip to left and right borders */ \
int cofs = ofs; \
int crun = run; \
if (left - cofs > 0) { \
crun -= left - cofs; \
cofs = left; \
} \
if (crun > right - cofs) \
crun = right - cofs; \
if (crun > 0) { \
Ptype *dst = (Ptype *)dstbuf + cofs; \
Uint32 *src = (Uint32 *)srcbuf + (cofs - ofs); \
int i; \
for (i = 0; i < crun; i++) \
do_blend(src[i], dst[i]); \
} \
srcbuf += run * 4; \
ofs += run; \
} \
} while (ofs < w); \
dstbuf += surf_dst->pitch; \
} while (--linecount); \
} while (0)
switch (df->bytes_per_pixel) {
case 2:
if (df->Gmask == 0x07e0 || df->Rmask == 0x07e0 || df->Bmask == 0x07e0) {
RLEALPHACLIPBLIT(Uint16, Uint8, BLIT_TRANSL_565);
} else {
RLEALPHACLIPBLIT(Uint16, Uint8, BLIT_TRANSL_555);
}
break;
case 4:
RLEALPHACLIPBLIT(Uint32, Uint16, BLIT_TRANSL_888);
break;
}
}
/* blit a pixel-alpha RLE surface */
static int SDLCALL SDL_RLEAlphaBlit(SDL_Surface *surf_src, const SDL_Rect *srcrect,
SDL_Surface *surf_dst, const SDL_Rect *dstrect)
{
int x, y;
int w = surf_src->w;
Uint8 *srcbuf, *dstbuf;
SDL_PixelFormat *df = surf_dst->format;
/* Lock the destination if necessary */
if (SDL_MUSTLOCK(surf_dst)) {
if (SDL_LockSurface(surf_dst) < 0) {
return -1;
}
}
x = dstrect->x;
y = dstrect->y;
dstbuf = (Uint8 *)surf_dst->pixels + y * surf_dst->pitch + x * df->bytes_per_pixel;
srcbuf = (Uint8 *)surf_src->map->data + sizeof(RLEDestFormat);
{
/* skip lines at the top if necessary */
int vskip = srcrect->y;
if (vskip) {
int ofs;
if (df->bytes_per_pixel == 2) {
/* the 16/32 interleaved format */
do {
/* skip opaque line */
ofs = 0;
do {
int run;
ofs += srcbuf[0];
run = srcbuf[1];
srcbuf += 2;
if (run) {
srcbuf += 2 * run;
ofs += run;
} else if (ofs == 0) {
goto done;
}
} while (ofs < w);
/* skip padding */
srcbuf += (uintptr_t)srcbuf & 2;
/* skip translucent line */
ofs = 0;
do {
int run;
ofs += ((Uint16 *)srcbuf)[0];
run = ((Uint16 *)srcbuf)[1];
srcbuf += 4 * (run + 1);
ofs += run;
} while (ofs < w);
} while (--vskip);
} else {
/* the 32/32 interleaved format */
vskip <<= 1; /* opaque and translucent have same format */
do {
ofs = 0;
do {
int run;
ofs += ((Uint16 *)srcbuf)[0];
run = ((Uint16 *)srcbuf)[1];
srcbuf += 4;
if (run) {
srcbuf += 4 * run;
ofs += run;
} else if (ofs == 0) {
goto done;
}
} while (ofs < w);
} while (--vskip);
}
}
}
/* if left or right edge clipping needed, call clip blit */
if (srcrect->x || srcrect->w != surf_src->w) {
RLEAlphaClipBlit(w, srcbuf, surf_dst, dstbuf, srcrect);
} else {
/*
* non-clipped blitter. Ptype is the destination pixel type,
* Ctype the translucent count type, and do_blend the
* macro to blend one pixel.
*/
#define RLEALPHABLIT(Ptype, Ctype, do_blend) \
do { \
int linecount = srcrect->h; \
do { \
int ofs = 0; \
/* blit opaque pixels on one line */ \
do { \
unsigned run; \
ofs += ((Ctype *)srcbuf)[0]; \
run = ((Ctype *)srcbuf)[1]; \
srcbuf += 2 * sizeof(Ctype); \
if (run) { \
PIXEL_COPY(dstbuf + ofs * sizeof(Ptype), srcbuf, \
run, sizeof(Ptype)); \
srcbuf += run * sizeof(Ptype); \
ofs += run; \
} else if (!ofs) \
goto done; \
} while (ofs < w); \
/* skip padding if necessary */ \
if (sizeof(Ptype) == 2) \
srcbuf += (uintptr_t)srcbuf & 2; \
/* blit translucent pixels on the same line */ \
ofs = 0; \
do { \
unsigned run; \
ofs += ((Uint16 *)srcbuf)[0]; \
run = ((Uint16 *)srcbuf)[1]; \
srcbuf += 4; \
if (run) { \
Ptype *dst = (Ptype *)dstbuf + ofs; \
unsigned i; \
for (i = 0; i < run; i++) { \
Uint32 src = *(Uint32 *)srcbuf; \
do_blend(src, *dst); \
srcbuf += 4; \
dst++; \
} \
ofs += run; \
} \
} while (ofs < w); \
dstbuf += surf_dst->pitch; \
} while (--linecount); \
} while (0)
switch (df->bytes_per_pixel) {
case 2:
if (df->Gmask == 0x07e0 || df->Rmask == 0x07e0 || df->Bmask == 0x07e0) {
RLEALPHABLIT(Uint16, Uint8, BLIT_TRANSL_565);
} else {
RLEALPHABLIT(Uint16, Uint8, BLIT_TRANSL_555);
}
break;
case 4:
RLEALPHABLIT(Uint32, Uint16, BLIT_TRANSL_888);
break;
}
}
done:
/* Unlock the destination if necessary */
if (SDL_MUSTLOCK(surf_dst)) {
SDL_UnlockSurface(surf_dst);
}
return 0;
}
/*
* Auxiliary functions:
* The encoding functions take 32bpp rgb + a, and
* return the number of bytes copied to the destination.
* The decoding functions copy to 32bpp rgb + a, and
* return the number of bytes copied from the source.
* These are only used in the encoder and un-RLE code and are therefore not
* highly optimised.
*/
/* encode 32bpp rgb + a into 16bpp rgb, losing alpha */
static int copy_opaque_16(void *dst, const Uint32 *src, int n,
SDL_PixelFormat *sfmt, SDL_PixelFormat *dfmt)
{
int i;
Uint16 *d = (Uint16 *)dst;
for (i = 0; i < n; i++) {
unsigned r, g, b;
RGB_FROM_PIXEL(*src, sfmt, r, g, b);
PIXEL_FROM_RGB(*d, dfmt, r, g, b);
src++;
d++;
}
return n * 2;
}
/* decode opaque pixels from 16bpp to 32bpp rgb + a */
static int uncopy_opaque_16(Uint32 *dst, const void *src, int n,
RLEDestFormat *sfmt, SDL_PixelFormat *dfmt)
{
int i;
const Uint16 *s = (const Uint16 *)src;
unsigned alpha = dfmt->Amask ? 255 : 0;
for (i = 0; i < n; i++) {
unsigned r, g, b;
RGB_FROM_PIXEL(*s, sfmt, r, g, b);
PIXEL_FROM_RGBA(*dst, dfmt, r, g, b, alpha);
s++;
dst++;
}
return n * 2;
}
/* encode 32bpp rgb + a into 32bpp G0RAB format for blitting into 565 */
static int copy_transl_565(void *dst, const Uint32 *src, int n,
SDL_PixelFormat *sfmt, SDL_PixelFormat *dfmt)
{
int i;
Uint32 *d = (Uint32 *)dst;
for (i = 0; i < n; i++) {
unsigned r, g, b, a;
Uint16 pix;
RGBA_FROM_8888(*src, sfmt, r, g, b, a);
PIXEL_FROM_RGB(pix, dfmt, r, g, b);
*d = ((pix & 0x7e0) << 16) | (pix & 0xf81f) | ((a << 2) & 0x7e0);
src++;
d++;
}
return n * 4;
}
/* encode 32bpp rgb + a into 32bpp G0RAB format for blitting into 555 */
static int copy_transl_555(void *dst, const Uint32 *src, int n,
SDL_PixelFormat *sfmt, SDL_PixelFormat *dfmt)
{
int i;
Uint32 *d = (Uint32 *)dst;
for (i = 0; i < n; i++) {
unsigned r, g, b, a;
Uint16 pix;
RGBA_FROM_8888(*src, sfmt, r, g, b, a);
PIXEL_FROM_RGB(pix, dfmt, r, g, b);
*d = ((pix & 0x3e0) << 16) | (pix & 0xfc1f) | ((a << 2) & 0x3e0);
src++;
d++;
}
return n * 4;
}
/* decode translucent pixels from 32bpp GORAB to 32bpp rgb + a */
static int uncopy_transl_16(Uint32 *dst, const void *src, int n,
RLEDestFormat *sfmt, SDL_PixelFormat *dfmt)
{
int i;
const Uint32 *s = (const Uint32 *)src;
for (i = 0; i < n; i++) {
unsigned r, g, b, a;
Uint32 pix = *s++;
a = (pix & 0x3e0) >> 2;
pix = (pix & ~0x3e0) | pix >> 16;
RGB_FROM_PIXEL(pix, sfmt, r, g, b);
PIXEL_FROM_RGBA(*dst, dfmt, r, g, b, a);
dst++;
}
return n * 4;
}
/* encode 32bpp rgba into 32bpp rgba, keeping alpha (dual purpose) */
static int copy_32(void *dst, const Uint32 *src, int n,
SDL_PixelFormat *sfmt, SDL_PixelFormat *dfmt)
{
int i;
Uint32 *d = (Uint32 *)dst;
for (i = 0; i < n; i++) {
unsigned r, g, b, a;
RGBA_FROM_8888(*src, sfmt, r, g, b, a);
RLEPIXEL_FROM_RGBA(*d, dfmt, r, g, b, a);
d++;
src++;
}
return n * 4;
}
/* decode 32bpp rgba into 32bpp rgba, keeping alpha (dual purpose) */
static int uncopy_32(Uint32 *dst, const void *src, int n,
RLEDestFormat *sfmt, SDL_PixelFormat *dfmt)
{
int i;
const Uint32 *s = (const Uint32 *)src;
for (i = 0; i < n; i++) {
unsigned r, g, b, a;
Uint32 pixel = *s++;
RGB_FROM_PIXEL(pixel, sfmt, r, g, b);
a = pixel >> 24;
PIXEL_FROM_RGBA(*dst, dfmt, r, g, b, a);
dst++;