src/character.h

   1 /* Header for multibyte character handler.
   2    Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN.
   3      Licensed to the Free Software Foundation.
   4    Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
   5      National Institute of Advanced Industrial Science and Technology (AIST)
   6      Registration Number H13PRO009
   7
   8 This file is part of GNU Emacs.
   9
  10 GNU Emacs is free software: you can redistribute it and/or modify
  11 it under the terms of the GNU General Public License as published by
  12 the Free Software Foundation, either version 3 of the License, or
  13 (at your option) any later version.
  14
  15 GNU Emacs is distributed in the hope that it will be useful,
  16 but WITHOUT ANY WARRANTY; without even the implied warranty of
  17 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18 GNU General Public License for more details.
  19
  20 You should have received a copy of the GNU General Public License
  21 along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
  22
  23 #ifndef EMACS_CHARACTER_H
  24 #define EMACS_CHARACTER_H
  25
  26 /* character code       1st byte   byte sequence
  27    --------------       --------   -------------
  28         0-7F            00..7F     0xxxxxxx
  29        80-7FF           C2..DF     110xxxxx 10xxxxxx
  30       800-FFFF          E0..EF     1110xxxx 10xxxxxx 10xxxxxx
  31     10000-1FFFFF        F0..F7     11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
  32    200000-3FFF7F        F8         11111000 1000xxxx 10xxxxxx 10xxxxxx 10xxxxxx
  33    3FFF80-3FFFFF        C0..C1     1100000x 10xxxxxx (for eight-bit-char)
  34    400000-...           invalid
  35
  36    invalid 1st byte     80..BF     10xxxxxx
  37                         F9..FF     11111xxx (xxx != 000)
  38 */
  39
  40 /* Maximum character code ((1 << CHARACTERBITS) - 1).  */
  41 #define MAX_CHAR  0x3FFFFF
  42
  43 /* Maximum Unicode character code.  */
  44 #define MAX_UNICODE_CHAR 0x10FFFF
  45
  46 /* Maximum N-byte character codes.  */
  47 #define MAX_1_BYTE_CHAR 0x7F
  48 #define MAX_2_BYTE_CHAR 0x7FF
  49 #define MAX_3_BYTE_CHAR 0xFFFF
  50 #define MAX_4_BYTE_CHAR 0x1FFFFF
  51 #define MAX_5_BYTE_CHAR 0x3FFF7F
  52
  53 /* Minimum leading code of multibyte characters.  */
  54 #define MIN_MULTIBYTE_LEADING_CODE 0xC0
  55 /* Maximum leading code of multibyte characters.  */
  56 #define MAX_MULTIBYTE_LEADING_CODE 0xF8
  57
  58 /* Nonzero iff C is a character that corresponds to a raw 8-bit
  59    byte.  */
  60 #define CHAR_BYTE8_P(c) ((c) > MAX_5_BYTE_CHAR)
  61
  62 /* Return the character code for raw 8-bit byte BYTE.  */
  63 #define BYTE8_TO_CHAR(byte) ((byte) + 0x3FFF00)
  64
  65 #define UNIBYTE_TO_CHAR(byte) \
  66   (ASCII_BYTE_P (byte) ? (byte) : BYTE8_TO_CHAR (byte))
  67
  68 /* Return the raw 8-bit byte for character C.  */
  69 #define CHAR_TO_BYTE8(c)        \
  70   (CHAR_BYTE8_P (c)             \
  71    ? (c) - 0x3FFF00             \
  72    : multibyte_char_to_unibyte (c, Qnil))
  73
  74 /* Return the raw 8-bit byte for character C,
  75    or -1 if C doesn't correspond to a byte.  */
  76 #define CHAR_TO_BYTE_SAFE(c)    \
  77   (CHAR_BYTE8_P (c)             \
  78    ? (c) - 0x3FFF00             \
  79    : multibyte_char_to_unibyte_safe (c))
  80
  81 /* Nonzero iff BYTE is the 1st byte of a multibyte form of a character
  82    that corresponds to a raw 8-bit byte.  */
  83 #define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1)
  84
  85 /* If C is not ASCII, make it unibyte. */
  86 #define MAKE_CHAR_UNIBYTE(c)    \
  87   do {                          \
  88     if (! ASCII_CHAR_P (c))     \
  89       c = CHAR_TO_BYTE8 (c);    \
  90   } while (0)
  91
  92
  93 /* If C is not ASCII, make it multibyte.  Assumes C < 256.  */
  94 #define MAKE_CHAR_MULTIBYTE(c) \
  95   (eassert ((c) >= 0 && (c) < 256), (c) = UNIBYTE_TO_CHAR (c))
  96
  97 /* This is the maximum byte length of multibyte form.  */
  98 #define MAX_MULTIBYTE_LENGTH 5
  99
 100 /* Return a Lisp character whose character code is C.  Assumes C is
 101    a valid character code.  */
 102 #define make_char(c) make_number (c)
 103
 104 /* Nonzero iff C is an ASCII byte.  */
 105 #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
 106
 107 /* Nonzero iff X is a character.  */
 108 #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
 109
 110 /* Nonzero iff C is valid as a character code.  GENERICP is not used.  */
 111 #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
 112
 113 /* Check if Lisp object X is a character or not.  */
 114 #define CHECK_CHARACTER(x) \
 115   CHECK_TYPE (CHARACTERP (x), Qcharacterp, x)
 116
 117 #define CHECK_CHARACTER_CAR(x) \
 118   do {                                  \
 119     Lisp_Object tmp = XCAR (x);         \
 120     CHECK_CHARACTER (tmp);              \
 121     XSETCAR ((x), tmp);                 \
 122   } while (0)
 123
 124 #define CHECK_CHARACTER_CDR(x) \
 125   do {                                  \
 126     Lisp_Object tmp = XCDR (x);         \
 127     CHECK_CHARACTER (tmp);              \
 128     XSETCDR ((x), tmp);                 \
 129   } while (0)
 130
 131 /* Nonzero iff C is an ASCII character.  */
 132 #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
 133
 134 /* Nonzero iff C is a character of code less than 0x100.  */
 135 #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
 136
 137 /* Nonzero if character C has a printable glyph.  */
 138 #define CHAR_PRINTABLE_P(c)     \
 139   (((c) >= 32 && (c) < 127)     \
 140    || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c))))
 141
 142 /* Return byte length of multibyte form for character C.  */
 143 #define CHAR_BYTES(c)                   \
 144   ( (c) <= MAX_1_BYTE_CHAR ? 1          \
 145     : (c) <= MAX_2_BYTE_CHAR ? 2        \
 146     : (c) <= MAX_3_BYTE_CHAR ? 3        \
 147     : (c) <= MAX_4_BYTE_CHAR ? 4        \
 148     : (c) <= MAX_5_BYTE_CHAR ? 5        \
 149     : 2)
 150
 151
 152 /* Return the leading code of multibyte form of C.  */
 153 #define CHAR_LEADING_CODE(c)                            \
 154   ((c) <= MAX_1_BYTE_CHAR ? c                           \
 155    : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6))       \
 156    : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12))      \
 157    : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18))      \
 158    : (c) <= MAX_5_BYTE_CHAR ? 0xF8                      \
 159    : (0xC0 | (((c) >> 6) & 0x01)))
 160
 161
 162 /* Store multibyte form of the character C in P.  The caller should
 163    allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
 164    Returns the length of the multibyte form.  */
 165
 166 #define CHAR_STRING(c, p)                       \
 167   ((unsigned) (c) <= MAX_1_BYTE_CHAR            \
 168    ? ((p)[0] = (c),                             \
 169       1)                                        \
 170    : (unsigned) (c) <= MAX_2_BYTE_CHAR          \
 171    ? ((p)[0] = (0xC0 | ((c) >> 6)),             \
 172       (p)[1] = (0x80 | ((c) & 0x3F)),           \
 173       2)                                        \
 174    : (unsigned) (c) <= MAX_3_BYTE_CHAR          \
 175    ? ((p)[0] = (0xE0 | ((c) >> 12)),            \
 176       (p)[1] = (0x80 | (((c) >> 6) & 0x3F)),    \
 177       (p)[2] = (0x80 | ((c) & 0x3F)),           \
 178       3)                                        \
 179    : char_string ((unsigned) c, p))
 180
 181 /* Store multibyte form of byte B in P.  The caller should allocate at
 182    least MAX_MULTIBYTE_LENGTH bytes area at P in advance.  Returns the
 183    length of the multibyte form.  */
 184
 185 #define BYTE8_STRING(b, p)                      \
 186   ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)),       \
 187    (p)[1] = (0x80 | ((b) & 0x3F)),              \
 188    2)
 189
 190
 191 /* Store multibyte form of the character C in P and advance P to the
 192    end of the multibyte form.  The caller should allocate at least
 193    MAX_MULTIBYTE_LENGTH bytes area at P in advance.  */
 194
 195 #define CHAR_STRING_ADVANCE(c, p)               \
 196   do {                                          \
 197     if ((c) <= MAX_1_BYTE_CHAR)                 \
 198       *(p)++ = (c);                             \
 199     else if ((c) <= MAX_2_BYTE_CHAR)            \
 200       *(p)++ = (0xC0 | ((c) >> 6)),             \
 201         *(p)++ = (0x80 | ((c) & 0x3F));         \
 202     else if ((c) <= MAX_3_BYTE_CHAR)            \
 203       *(p)++ = (0xE0 | ((c) >> 12)),            \
 204         *(p)++ = (0x80 | (((c) >> 6) & 0x3F)),  \
 205         *(p)++ = (0x80 | ((c) & 0x3F));         \
 206     else                                        \
 207       (p) += char_string ((c), (p));            \
 208   } while (0)
 209
 210
 211 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
 212    form.  */
 213 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
 214
 215 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
 216    multibyte form.  */
 217 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
 218
 219 /* Nonzero iff BYTE starts a character in a multibyte form.
 220    This is equivalent to:
 221         (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte))  */
 222 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
 223
 224 /* How many bytes a character that starts with BYTE occupies in a
 225    multibyte form.  */
 226 #define BYTES_BY_CHAR_HEAD(byte)        \
 227   (!((byte) & 0x80) ? 1                 \
 228    : !((byte) & 0x20) ? 2               \
 229    : !((byte) & 0x10) ? 3               \
 230    : !((byte) & 0x08) ? 4               \
 231    : 5)
 232
 233
 234 /* The byte length of multibyte form at unibyte string P ending at
 235    PEND.  If STR doesn't point to a valid multibyte form, return 0.  */
 236
 237 #define MULTIBYTE_LENGTH(p, pend)                               \
 238   (p >= pend ? 0                                                \
 239    : !((p)[0] & 0x80) ? 1                                       \
 240    : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0         \
 241    : ((p)[0] & 0xE0) == 0xC0 ? 2                                \
 242    : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0         \
 243    : ((p)[0] & 0xF0) == 0xE0 ? 3                                \
 244    : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0         \
 245    : ((p)[0] & 0xF8) == 0xF0 ? 4                                \
 246    : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0         \
 247    : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5              \
 248    : 0)
 249
 250
 251 /* Like MULTIBYTE_LENGTH, but don't check the ending address.  */
 252
 253 #define MULTIBYTE_LENGTH_NO_CHECK(p)                    \
 254   (!((p)[0] & 0x80) ? 1                                 \
 255    : ((p)[1] & 0xC0) != 0x80 ? 0                        \
 256    : ((p)[0] & 0xE0) == 0xC0 ? 2                        \
 257    : ((p)[2] & 0xC0) != 0x80 ? 0                        \
 258    : ((p)[0] & 0xF0) == 0xE0 ? 3                        \
 259    : ((p)[3] & 0xC0) != 0x80 ? 0                        \
 260    : ((p)[0] & 0xF8) == 0xF0 ? 4                        \
 261    : ((p)[4] & 0xC0) != 0x80 ? 0                        \
 262    : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5      \
 263    : 0)
 264
 265 /* If P is before LIMIT, advance P to the next character boundary.
 266    Assumes that P is already at a character boundary of the same
 267    multibyte form whose end address is LIMIT.  */
 268
 269 #define NEXT_CHAR_BOUNDARY(p, limit)    \
 270   do {                                  \
 271     if ((p) < (limit))                  \
 272       (p) += BYTES_BY_CHAR_HEAD (*(p)); \
 273   } while (0)
 274
 275
 276 /* If P is after LIMIT, advance P to the previous character boundary.
 277    Assumes that P is already at a character boundary of the same
 278    multibyte form whose beginning address is LIMIT.  */
 279
 280 #define PREV_CHAR_BOUNDARY(p, limit)                                    \
 281   do {                                                                  \
 282     if ((p) > (limit))                                                  \
 283       {                                                                 \
 284         const unsigned char *p0 = (p);                                  \
 285         do {                                                            \
 286           p0--;                                                         \
 287         } while (p0 >= limit && ! CHAR_HEAD_P (*p0));                   \
 288         (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1;    \
 289       }                                                                 \
 290   } while (0)
 291
 292 /* Return the character code of character whose multibyte form is at
 293    P.  */
 294
 295 #define STRING_CHAR(p)                                          \
 296   (!((p)[0] & 0x80)                                             \
 297    ? (p)[0]                                                     \
 298    : ! ((p)[0] & 0x20)                                          \
 299    ? (((((p)[0] & 0x1F) << 6)                                   \
 300        | ((p)[1] & 0x3F))                                       \
 301       + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))       \
 302    : ! ((p)[0] & 0x10)                                          \
 303    ? ((((p)[0] & 0x0F) << 12)                                   \
 304       | (((p)[1] & 0x3F) << 6)                                  \
 305       | ((p)[2] & 0x3F))                                        \
 306    : string_char ((p), NULL, NULL))
 307
 308
 309 /* Like STRING_CHAR, but set ACTUAL_LEN to the length of multibyte
 310    form.  */
 311
 312 #define STRING_CHAR_AND_LENGTH(p, actual_len)                   \
 313   (!((p)[0] & 0x80)                                             \
 314    ? ((actual_len) = 1, (p)[0])                                 \
 315    : ! ((p)[0] & 0x20)                                          \
 316    ? ((actual_len) = 2,                                         \
 317       (((((p)[0] & 0x1F) << 6)                                  \
 318         | ((p)[1] & 0x3F))                                      \
 319        + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)))     \
 320    : ! ((p)[0] & 0x10)                                          \
 321    ? ((actual_len) = 3,                                         \
 322       ((((p)[0] & 0x0F) << 12)                                  \
 323        | (((p)[1] & 0x3F) << 6)                                 \
 324        | ((p)[2] & 0x3F)))                                      \
 325    : string_char ((p), NULL, &actual_len))
 326
 327
 328 /* Like STRING_CHAR, but advance P to the end of multibyte form.  */
 329
 330 #define STRING_CHAR_ADVANCE(p)                                  \
 331   (!((p)[0] & 0x80)                                             \
 332    ? *(p)++                                                     \
 333    : ! ((p)[0] & 0x20)                                          \
 334    ? ((p) += 2,                                                 \
 335       ((((p)[-2] & 0x1F) << 6)                                  \
 336        | ((p)[-1] & 0x3F)                                       \
 337        | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0)))    \
 338    : ! ((p)[0] & 0x10)                                          \
 339    ? ((p) += 3,                                                 \
 340       ((((p)[-3] & 0x0F) << 12)                                 \
 341        | (((p)[-2] & 0x3F) << 6)                                \
 342        | ((p)[-1] & 0x3F)))                                     \
 343    : string_char ((p), &(p), NULL))
 344
 345
 346 /* Fetch the "next" character from Lisp string STRING at byte position
 347    BYTEIDX, character position CHARIDX.  Store it into OUTPUT.
 348
 349    All the args must be side-effect-free.
 350    BYTEIDX and CHARIDX must be lvalues;
 351    we increment them past the character fetched.  */
 352
 353 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX)     \
 354   do                                                                    \
 355     {                                                                   \
 356       CHARIDX++;                                                        \
 357       if (STRING_MULTIBYTE (STRING))                                    \
 358         {                                                               \
 359           unsigned char *ptr = &SDATA (STRING)[BYTEIDX];                \
 360           int len;                                                      \
 361                                                                         \
 362           OUTPUT = STRING_CHAR_AND_LENGTH (ptr, len);                   \
 363           BYTEIDX += len;                                               \
 364         }                                                               \
 365       else                                                              \
 366         {                                                               \
 367           OUTPUT = SREF (STRING, BYTEIDX);                              \
 368           BYTEIDX++;                                                    \
 369         }                                                               \
 370     }                                                                   \
 371   while (0)
 372
 373 /* Like FETCH_STRING_CHAR_ADVANCE, but return a multibyte character
 374    even if STRING is unibyte.  */
 375
 376 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
 377   do                                                                          \
 378     {                                                                         \
 379       CHARIDX++;                                                              \
 380       if (STRING_MULTIBYTE (STRING))                                          \
 381         {                                                                     \
 382           unsigned char *ptr = &SDATA (STRING)[BYTEIDX];                      \
 383           int len;                                                            \
 384                                                                               \
 385           OUTPUT = STRING_CHAR_AND_LENGTH (ptr, len);                         \
 386           BYTEIDX += len;                                                     \
 387         }                                                                     \
 388       else                                                                    \
 389         {                                                                     \
 390           OUTPUT = SREF (STRING, BYTEIDX);                                    \
 391           BYTEIDX++;                                                          \
 392           MAKE_CHAR_MULTIBYTE (OUTPUT);                                       \
 393         }                                                                     \
 394     }                                                                         \
 395   while (0)
 396
 397
 398 /* Like FETCH_STRING_CHAR_ADVANCE, but assumes STRING is multibyte.  */
 399
 400 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
 401   do                                                                         \
 402     {                                                                        \
 403       unsigned char *ptr = &SDATA (STRING)[BYTEIDX];                         \
 404       int len;                                                               \
 405                                                                              \
 406       OUTPUT = STRING_CHAR_AND_LENGTH (ptr, len);                            \
 407       BYTEIDX += len;                                                        \
 408       CHARIDX++;                                                             \
 409     }                                                                        \
 410   while (0)
 411
 412
 413 /* Like FETCH_STRING_CHAR_ADVANCE, but fetch character from the current
 414    buffer.  */
 415
 416 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX)            \
 417   do                                                            \
 418     {                                                           \
 419       CHARIDX++;                                                \
 420       if (!NILP (current_buffer->enable_multibyte_characters))  \
 421         {                                                       \
 422           unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX);         \
 423           int len;                                              \
 424                                                                 \
 425           OUTPUT= STRING_CHAR_AND_LENGTH (ptr, len);            \
 426           BYTEIDX += len;                                       \
 427         }                                                       \
 428       else                                                      \
 429         {                                                       \
 430           OUTPUT = *(BYTE_POS_ADDR (BYTEIDX));                  \
 431           BYTEIDX++;                                            \
 432         }                                                       \
 433     }                                                           \
 434   while (0)
 435
 436
 437 /* Like FETCH_CHAR_ADVANCE, but assumes the current buffer is multibyte.  */
 438
 439 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX)   \
 440   do                                                            \
 441     {                                                           \
 442       unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX);             \
 443       int len;                                                  \
 444                                                                 \
 445       OUTPUT = STRING_CHAR_AND_LENGTH (ptr, len);               \
 446       BYTEIDX += len;                                           \
 447       CHARIDX++;                                                \
 448     }                                                           \
 449   while (0)
 450
 451
 452 /* Increment the buffer byte position POS_BYTE of the current buffer to
 453    the next character boundary.  No range checking of POS.  */
 454
 455 #define INC_POS(pos_byte)                               \
 456   do {                                                  \
 457     unsigned char *p = BYTE_POS_ADDR (pos_byte);        \
 458     pos_byte += BYTES_BY_CHAR_HEAD (*p);                \
 459   } while (0)
 460
 461
 462 /* Decrement the buffer byte position POS_BYTE of the current buffer to
 463    the previous character boundary.  No range checking of POS.  */
 464
 465 #define DEC_POS(pos_byte)                       \
 466   do {                                          \
 467     unsigned char *p;                           \
 468                                                 \
 469     pos_byte--;                                 \
 470     if (pos_byte < GPT_BYTE)                    \
 471       p = BEG_ADDR + pos_byte - BEG_BYTE;       \
 472     else                                        \
 473       p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\
 474     while (!CHAR_HEAD_P (*p))                   \
 475       {                                         \
 476         p--;                                    \
 477         pos_byte--;                             \
 478       }                                         \
 479   } while (0)
 480
 481 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way.  */
 482
 483 #define INC_BOTH(charpos, bytepos)                              \
 484   do                                                            \
 485     {                                                           \
 486       (charpos)++;                                              \
 487       if (NILP (current_buffer->enable_multibyte_characters))   \
 488         (bytepos)++;                                            \
 489       else                                                      \
 490         INC_POS ((bytepos));                                    \
 491     }                                                           \
 492   while (0)
 493
 494
 495 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way.  */
 496
 497 #define DEC_BOTH(charpos, bytepos)                              \
 498   do                                                            \
 499     {                                                           \
 500       (charpos)--;                                              \
 501       if (NILP (current_buffer->enable_multibyte_characters))   \
 502         (bytepos)--;                                            \
 503       else                                                      \
 504         DEC_POS ((bytepos));                                    \
 505     }                                                           \
 506   while (0)
 507
 508
 509 /* Increment the buffer byte position POS_BYTE of the current buffer to
 510    the next character boundary.  This macro relies on the fact that
 511    *GPT_ADDR and *Z_ADDR are always accessible and the values are
 512    '\0'.  No range checking of POS_BYTE.  */
 513
 514 #define BUF_INC_POS(buf, pos_byte)                              \
 515   do {                                                          \
 516     unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte);        \
 517     pos_byte += BYTES_BY_CHAR_HEAD (*p);                        \
 518   } while (0)
 519
 520
 521 /* Decrement the buffer byte position POS_BYTE of the current buffer to
 522    the previous character boundary.  No range checking of POS_BYTE.  */
 523
 524 #define BUF_DEC_POS(buf, pos_byte)                                      \
 525   do {                                                                  \
 526     unsigned char *p;                                                   \
 527     pos_byte--;                                                         \
 528     if (pos_byte < BUF_GPT_BYTE (buf))                                  \
 529       p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE;                     \
 530     else                                                                \
 531       p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\
 532     while (!CHAR_HEAD_P (*p))                                           \
 533       {                                                                 \
 534         p--;                                                            \
 535         pos_byte--;                                                     \
 536       }                                                                 \
 537   } while (0)
 538
 539
 540 /* If C is a character to be unified with a Unicode character, return
 541    the unified Unicode character.  */
 542
 543 #define MAYBE_UNIFY_CHAR(c)                             \
 544   do {                                                  \
 545     if (c > MAX_UNICODE_CHAR && c <= MAX_5_BYTE_CHAR)   \
 546       {                                                 \
 547         Lisp_Object val;                                \
 548         val = CHAR_TABLE_REF (Vchar_unify_table, c);    \
 549         if (INTEGERP (val))                             \
 550           c = XINT (val);                               \
 551         else if (! NILP (val))                          \
 552           c = maybe_unify_char (c, val);                \
 553       }                                                 \
 554   } while (0)
 555
 556
 557 /* Return the width of ASCII character C.  The width is measured by
 558    how many columns C will occupy on the screen when displayed in the
 559    current buffer.  */
 560
 561 #define ASCII_CHAR_WIDTH(c)                                             \
 562   (c < 0x20                                                             \
 563    ? (c == '\t'                                                         \
 564       ? XFASTINT (current_buffer->tab_width)                            \
 565       : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2)))   \
 566    : (c < 0x7f                                                          \
 567       ? 1                                                               \
 568       : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
 569
 570 /* Return the width of character C.  The width is measured by how many
 571    columns C will occupy on the screen when displayed in the current
 572    buffer.  */
 573
 574 #define CHAR_WIDTH(c)           \
 575   (ASCII_CHAR_P (c)             \
 576    ? ASCII_CHAR_WIDTH (c)       \
 577    : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
 578
 579 /* If C is a variation selector, return the index numnber of the
 580    variation selector (1..256).  Otherwise, return 0.  */
 581
 582 #define CHAR_VARIATION_SELECTOR_P(c)            \
 583   ((c) < 0xFE00 ? 0                             \
 584    : (c) <= 0xFE0F ? (c) - 0xFE00 + 1           \
 585    : (c) < 0xE0100 ? 0                          \
 586    : (c) <= 0xE01EF ? (c) - 0xE0100 + 17        \
 587    : 0)
 588
 589 /* If C is a high surrogate, return 1.  If C is a low surrogate,
 590    return 0.  Otherwise, return 0.  */
 591
 592 #define CHAR_SURROGATE_PAIR_P(c)        \
 593   ((c) < 0xD800 ? 0                     \
 594    : (c) <= 0xDBFF ? 1                  \
 595    : (c) <= 0xDFFF ? 2                  \
 596    : 0)
 597
 598
 599 extern int char_resolve_modifier_mask (int);
 600 extern int char_string (unsigned, unsigned char *);
 601 extern int string_char (const unsigned char *,
 602                         const unsigned char **, int *);
 603
 604 extern int translate_char (Lisp_Object, int c);
 605 extern int char_printable_p (int c);
 606 extern void parse_str_as_multibyte (const unsigned char *,
 607                                     EMACS_INT, EMACS_INT *, EMACS_INT *);
 608 extern EMACS_INT parse_str_to_multibyte (const unsigned char *, EMACS_INT);
 609 extern EMACS_INT str_as_multibyte (unsigned char *, EMACS_INT, EMACS_INT,
 610                              EMACS_INT *);
 611 extern EMACS_INT str_to_multibyte (unsigned char *, EMACS_INT, EMACS_INT);
 612 extern EMACS_INT str_as_unibyte (unsigned char *, EMACS_INT);
 613 extern EMACS_INT str_to_unibyte (const unsigned char *, unsigned char *,
 614                                  EMACS_INT, int);
 615 extern EMACS_INT strwidth (const unsigned char *, EMACS_INT);
 616 extern EMACS_INT c_string_width (const unsigned char *, EMACS_INT, int,
 617                                  EMACS_INT *, EMACS_INT *);
 618 extern EMACS_INT lisp_string_width (Lisp_Object, int,
 619                                     EMACS_INT *, EMACS_INT *);
 620
 621 extern Lisp_Object Qcharacterp, Qauto_fill_chars;
 622 extern Lisp_Object Vchar_unify_table;
 623 extern Lisp_Object string_escape_byte8 (Lisp_Object);
 624
 625 /* Return a translation table of id number ID.  */
 626 #define GET_TRANSLATION_TABLE(id) \
 627   (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
 628
 629 #define DEFSYM(sym, name)       \
 630   do { (sym) = intern_c_string ((name)); staticpro (&(sym)); } while (0)
 631
 632 #endif /* EMACS_CHARACTER_H */
 633