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
|
/* Copyright (C) 2001-2012 Artifex Software, Inc.
All Rights Reserved.
This software is provided AS-IS with no warranty, either express or
implied.
This software is distributed under license and may not be copied,
modified or distributed except as expressly authorized under the terms
of the license contained in the file LICENSE in this distribution.
Refer to licensing information at http://www.artifex.com or contact
Artifex Software, Inc., 7 Mt. Lassen Drive - Suite A-134, San Rafael,
CA 94903, U.S.A., +1(415)492-9861, for further information.
*/
/* RasterOp / transparency type definitions */
#ifndef gsropt_INCLUDED
# define gsropt_INCLUDED
/*
* This file defines the types for some library extensions that are
* motivated by PCL5 and also made available for PostScript:
* RasterOp, source and pattern white-pixel transparency, and
* per-pixel "render algorithm" information.
*/
/*
* Define whether we implement transparency correctly, or whether we
* implement it as documented in the H-P manuals.
*/
#define TRANSPARENCY_PER_H_P
/* Raster ops are explained in Chapter 5 of "The PCL 5 Color Technical
* Reference Manual", but essentially, they provide a way of mixing
* various bitmaps together. The most general form is where a texture
* bitmap, a source bitmap, and a destination bitmap are added together to
* give a new destination bitmap value.
*
* Each bitmap value is either 0 or 1, so essentially each raster op is
* a function of the form f(D,S,T). (Confusingly, the HP documentation uses
* the name 'pattern' in the english descriptive text in an inconsistent way
* to sometimes mean 'texture' and sometimes mean something that is used to
* help generate the texture. We will use texture exclusively here.)
*
* There are therefore 8 possible values on input to such functions, and 2
* possible output values for each input. Hence 2^8 = 256 possible raster
* op functions.
*
* For any given function, f(D,S,T), we can form an 8 bit number as follows:
* Bit 0 = f(0,0,0)
* Bit 1 = f(1,0,0)
* Bit 2 = f(0,1,0)
* Bit 3 = f(1,1,0)
* Bit 4 = f(0,0,1)
* Bit 5 = f(1,0,1)
* Bit 6 = f(0,1,1)
* Bit 7 = f(1,1,1)
* Running this in reverse for each number from 0 to 255 we have an
* enumeratation of the possible 3 input raster op functions.
*
* We could therefore define F(op,D,S,T) to be a 'master' function that
* encapsulates all these raster op functions as follows:
* F(op,D,S,T) = !!(op & ((D?0xAA:0x55) & (S?0xCC:0x33) & (T?0xF0:0xF)))
* We don't actually use that technique here (as it only works for 1bpp
* inputs, and we actually use the rasterops for doing multiple bits at
* once - it is mentioned for interests sake only.
*/
/*
* By the magic of Boolean algebra, we can operate on the rop codes using
* Boolean operators and get the right result. E.g., the value of
* (rop3_S & rop3_D) is the rop3 code for S & D. We just have to remember
* to mask results with rop2_1 or rop3_1 if necessary.
*/
/* 2-input RasterOp */
typedef enum {
rop2_0 = 0,
rop2_S = 0xc, /* source */
#define rop2_S_shift 2
rop2_D = 0xa, /* destination */
#define rop2_D_shift 1
rop2_1 = 0xf,
#define rop2_operand(shift, d, s)\
((shift) == 2 ? (s) : (d))
rop2_default = rop2_S
} gs_rop2_t;
/*
* For the 3-input case, we follow H-P's inconsistent terminology:
* the transparency mode is called pattern transparency, but the third
* RasterOp operand is called texture, not pattern.
*/
/* 3-input RasterOp */
typedef enum {
rop3_0 = 0,
#define rop3_T 0xf0 /* texture */
#define rop3_T_shift 4
#define rop3_S 0xcc /* source */
#define rop3_S_shift 2
rop3_D = 0xaa, /* destination */
#define rop3_D_shift 1
rop3_1 = 0xff,
rop3_default = rop3_T | rop3_S
} gs_rop3_t;
/* All the transformations on rop3s are designed so that */
/* they can also be used on lops. The only place this costs anything */
/* is in rop3_invert. */
/*
* Invert an operand.
*/
#define rop3_invert_(op, mask, shift)\
( (((op) & mask) >> shift) | (((op) & (rop3_1 - mask)) << shift) |\
((op) & ~rop3_1) )
#define rop3_invert_D(op) rop3_invert_(op, rop3_D, rop3_D_shift)
#define rop3_invert_S(op) rop3_invert_(op, rop3_S, rop3_S_shift)
#define rop3_invert_T(op) rop3_invert_(op, rop3_T, rop3_T_shift)
/*
* Pin an operand to 0.
*/
#define rop3_know_0_(op, mask, shift)\
( (((op) & (rop3_1 - mask)) << shift) | ((op) & ~mask) )
#define rop3_know_D_0(op) rop3_know_0_(op, rop3_D, rop3_D_shift)
#define rop3_know_S_0(op) rop3_know_0_(op, rop3_S, rop3_S_shift)
#define rop3_know_T_0(op) rop3_know_0_(op, rop3_T, rop3_T_shift)
/*
* Pin an operand to 1.
*/
#define rop3_know_1_(op, mask, shift)\
( (((op) & mask) >> shift) | ((op) & ~(rop3_1 - mask)) )
#define rop3_know_D_1(op) rop3_know_1_(op, rop3_D, rop3_D_shift)
#define rop3_know_S_1(op) rop3_know_1_(op, rop3_S, rop3_S_shift)
#define rop3_know_T_1(op) rop3_know_1_(op, rop3_T, rop3_T_shift)
/*
* Swap S and T.
*/
#define rop3_swap_S_T(op)\
( (((op) & rop3_S & ~rop3_T) << (rop3_T_shift - rop3_S_shift)) |\
(((op) & ~rop3_S & rop3_T) >> (rop3_T_shift - rop3_S_shift)) |\
((op) & ~(rop3_S ^ rop3_T)) )
#define lop_swap_S_T(op)\
( ((rop3_swap_S_T(op)) & ~(lop_S_transparent | lop_T_transparent)) |\
(((op) & lop_S_transparent) ? lop_T_transparent : 0) |\
(((op) & lop_T_transparent) ? lop_S_transparent : 0) )
/*
* Account for transparency.
*/
#define rop3_use_D_when_0_(op, mask)\
(((op) & ~(rop3_1 - mask)) | (rop3_D & ~mask))
#define rop3_use_D_when_1_(op, mask)\
(((op) & ~mask) | (rop3_D & mask))
#define rop3_use_D_when_S_0(op) rop3_use_D_when_0_(op, rop3_S)
#define rop3_use_D_when_S_1(op) rop3_use_D_when_1_(op, rop3_S)
#define rop3_use_D_when_T_0(op) rop3_use_D_when_0_(op, rop3_T)
#define rop3_use_D_when_T_1(op) rop3_use_D_when_1_(op, rop3_T)
/*
* Invert the result.
*/
#define rop3_not(op) ((op) ^ rop3_1)
/*
* Test whether an operand is used.
*/
#define rop3_uses_(op, mask, shift)\
( ((((op) << shift) ^ (op)) & mask) != 0 )
#define rop3_uses_D(op) rop3_uses_(op, rop3_D, rop3_D_shift)
#define rop3_uses_S(op) rop3_uses_(op, rop3_S, rop3_S_shift)
#define rop3_uses_T(op) rop3_uses_(op, rop3_T, rop3_T_shift)
/*
* Test whether an operation is idempotent, i.e., whether
* f(D, S, T) = f(f(D, S, T), S, T) (for all D,S,T)
* f has a range of 0 or 1, so this is equivalent to the condition that:
* There exists no s,t, s.t. (f(0,s,t) == 1 && f(1,s,t) == 0)
* or equivalently:
* For all s, t, ~(f(0,s,t) == 1 && f(1,s,t) == 0 )
* or For all s, t, ~(f(0,s,t) == 1 && ~(f(1,s,t) == 1))
* or For all s, t, ~((f(0,s,t) & ~f(1,s,t)) == 1)
* or For all s, t, ((f(0,s,t) & ~f(1,s,t)) == 0)
*
* We can code this as a logical operation by observing that:
* ((~op) & rop3_D) gives us a bitfield of ~f(1,s,t)
* (((op)<<rop_3_D_shift) & rop3_D) gives us a bitfield of f(0,s,t)
* So
* ((~op) & rop3_D) & ((op<<rop3_D_shift) & rop3_D) == 0 iff idempotent
* === ((~op) & (op<<rop3_D_shift) & rop3_D) == 0 iff idempotent
*/
#define rop3_is_idempotent(op)\
!( (~op) & ((op) << rop3_D_shift) & rop3_D )
/* Transparency */
#define source_transparent_default false
#define pattern_transparent_default false
/*
* We define a logical operation as a RasterOp, transparency flags,
* and render algorithm all packed into a single integer.
* In principle, we should use a structure, but most C implementations
* implement structure values very inefficiently.
*
* In addition, we define a "pdf14" flag which indicates that PDF
* transparency is in effect. This doesn't change rendering in any way,
* but does force the lop to be considered non-idempotent.
*/
#define lop_rop(lop) ((gs_rop3_t)((lop) & 0xff)) /* must be low-order bits */
#define lop_S_transparent 0x100
#define lop_T_transparent 0x200
#define lop_pdf14 0x400
/* Also, we abuse the lop even further, by allowing it to specify a specific
* plane for an operation to work on (in a planar device context). To specify
* a particular plane, set lop_planar, and then or in the plane number
* shifted up by lop_planar_shift.
*/
#define lop_planar 0x800
#define lop_planar_shift 12
typedef uint gs_logical_operation_t;
#define lop_default\
(rop3_default |\
(source_transparent_default ? lop_S_transparent : 0) |\
(pattern_transparent_default ? lop_T_transparent : 0))
/* Test whether a logical operation uses S or T. */
#ifdef TRANSPARENCY_PER_H_P /* bizarre but necessary definition */
#define lop_uses_S(lop)\
(rop3_uses_S(lop) || ((lop) & (lop_S_transparent | lop_T_transparent)))
#else /* reasonable definition */
#define lop_uses_S(lop)\
(rop3_uses_S(lop) || ((lop) & lop_S_transparent))
#endif
#define lop_uses_T(lop)\
(rop3_uses_T(lop) || ((lop) & lop_T_transparent))
/* Test whether a logical operation just sets D = x if y = 0. */
#define lop_no_T_is_S(lop)\
(((lop) & (lop_S_transparent | (rop3_1 - rop3_T))) == (rop3_S & ~rop3_T))
#define lop_no_S_is_T(lop)\
(((lop) & (lop_T_transparent | (rop3_1 - rop3_S))) == (rop3_T & ~rop3_S))
/* Test whether a logical operation is idempotent. */
#define lop_is_idempotent(lop) (rop3_is_idempotent(lop) && !(lop & lop_pdf14))
/*
* Define the logical operation versions of some RasterOp transformations.
* Note that these also affect the transparency flags.
*/
#define lop_know_S_0(lop)\
(rop3_know_S_0(lop) & ~lop_S_transparent)
#define lop_know_T_0(lop)\
(rop3_know_T_0(lop) & ~lop_T_transparent)
#define lop_know_S_1(lop)\
(lop & lop_S_transparent ? rop3_D : rop3_know_S_1(lop))
#define lop_know_T_1(lop)\
(lop & lop_T_transparent ? rop3_D : rop3_know_T_1(lop))
/* Define the interface to the table of 256 RasterOp procedures. */
typedef unsigned long rop_operand;
typedef rop_operand (*rop_proc)(rop_operand D, rop_operand S, rop_operand T);
/* Define the table of operand usage by the 256 RasterOp operations. */
typedef enum {
rop_usage_none = 0,
rop_usage_D = 1,
rop_usage_S = 2,
rop_usage_DS = 3,
rop_usage_T = 4,
rop_usage_DT = 5,
rop_usage_ST = 6,
rop_usage_DST = 7
} rop_usage_t;
/* Define the table of RasterOp implementation procedures. */
extern const rop_proc rop_proc_table[256];
/* Define the table of RasterOp operand usage. */
extern const byte /*rop_usage_t*/ rop_usage_table[256];
/* Rather than just doing one raster operation at a time (which costs us
* a significant amount of time due to function call overheads), we attempt
* to ameliorate this by doing runs of raster operations at a time.
*/
/* The raster op run operator type */
typedef struct rop_run_op_s rop_run_op;
/* The contents of these structs should really be private, but we define them
* in the open here to allow us to easily allocate the space on the stack
* and thus avoid costly malloc/free overheads. */
typedef union rop_source_s {
struct {
const byte *ptr;
int pos;
} b;
rop_operand c;
} rop_source;
struct rop_run_op_s {
void (*run)(rop_run_op *, byte *dest, int len);
void (*runswap)(rop_run_op *, byte *dest, int len);
rop_source s;
rop_source t;
int rop;
byte depth;
byte flags;
byte dpos;
const byte *scolors;
const byte *tcolors;
void (*release)(rop_run_op *);
void *opaque;
};
/* Flags for passing into rop_get_run_op */
enum {
rop_s_constant = 1,
rop_t_constant = 2,
rop_s_1bit = 4,
rop_t_1bit = 8
};
/* To use a rop_run_op, allocate it on the stack, then call
* rop_get_run_op with a pointer to it to fill it in with details of an
* implementer. If you're lucky (doing a popular rop) you'll get an optimised
* implementation. If you're not, you'll get a general purpose slow rop. You
* will always get an implementation of some kind though.
*
* You should logical or together the flags - this tells the routine whether
* s and t are constant, or will be varying across the run.
*/
void rop_get_run_op(rop_run_op *op, int rop, int depth, int flags);
/* Next, you should set the values of S and T. Each of these can either be
* a constant value, or a pointer to a run of bytes. It is the callers
* responsibility to set these in the correct way (corresponding to the flags
* passed into the call to rop_get_run_op.
*
* For cases where depth < 8, and a bitmap is used, we have to specify the
* start bit position within the byte. (All data in rop bitmaps is considered
* to be bigendian). */
void rop_set_s_constant(rop_run_op *op, int s);
void rop_set_s_bitmap(rop_run_op *op, const byte *s);
void rop_set_s_bitmap_subbyte(rop_run_op *op, const byte *s, int startbitpos);
void rop_set_s_colors(rop_run_op *op, const byte *scolors);
void rop_set_t_constant(rop_run_op *op, int t);
void rop_set_t_bitmap(rop_run_op *op, const byte *t);
void rop_set_t_bitmap_subbyte(rop_run_op *op, const byte *s, int startbitpos);
void rop_set_t_colors(rop_run_op *op, const byte *scolors);
/* At last we call the rop_run function itself. (This can happen multiple
* times, perhaps once per scanline, with any required calls to the
* rop_set_{s,t} functions.) The length field is given in terms of 'number
* of depth bits'.
*
* This version is for the depth >= 8 case. */
void rop_run(rop_run_op *op, byte *d, int len);
/* This version tells us how many bits to skip over in the first byte, and
* is therefore only useful in the depth < 8 case. */
void rop_run_subbyte(rop_run_op *op, byte *d, int startbitpos, int len);
/* And finally we release our rop (in case any allocation was done). */
void rop_release_run_op(rop_run_op *op);
/* We offer some of these functions implemented as macros for speed */
#define rop_set_s_constant(OP,S) ((OP)->s.c = (S))
#define rop_set_s_bitmap(OP,S) ((OP)->s.b.ptr = (S))
#define rop_set_s_bitmap_subbyte(OP,S,B) (((OP)->s.b.ptr = (S)),((OP)->s.b.pos = (B)))
#define rop_set_s_colors(OP,S) ((OP)->scolors = (const byte *)(S))
#define rop_set_t_constant(OP,T) ((OP)->t.c = (T))
#define rop_set_t_bitmap(OP,T) ((OP)->t.b.ptr = (T))
#define rop_set_t_bitmap_subbyte(OP,T,B) (((OP)->t.b.ptr = (T)),((OP)->t.b.pos = (B)))
#define rop_set_t_colors(OP,T) ((OP)->tcolors = (const byte *)(T))
#define rop_run(OP,D,LEN) (((OP)->run)(OP,D,LEN))
#define rop_run_subbyte(OP,D,S,L) (((OP)->dpos=(byte)S),((OP)->run)(OP,D,L))
#define rop_release_run_op(OP) do { rop_run_op *OP2 = (OP); \
if (OP2->release) OP2->release(OP2); \
} while (0==1)
#ifdef _MSC_VER /* Are we using MSVC? */
# if defined(_M_IX86) || defined(_M_AMD64) /* Are we on an x86? */
# if MSC_VER >= 1400 /* old versions have _byteswap_ulong() in stdlib.h */
# include "intrin.h"
# endif
# define ENDIAN_SWAP_INT _byteswap_ulong
#endif
#elif defined(HAVE_BSWAP32)
#define ENDIAN_SWAP_INT __builtin_bswap32
#elif defined(HAVE_BYTESWAP_H)
#include <byteswap.h>
#define ENDIAN_SWAP_INT bswap_32
#endif
#endif /* gsropt_INCLUDED */
|