src/coding.h

   1 /* Header for coding system handler.
   2    Copyright (C) 2001-2013 Free Software Foundation, Inc.
   3    Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
   4      2005, 2006, 2007, 2008, 2009, 2010, 2011
   5      National Institute of Advanced Industrial Science and Technology (AIST)
   6      Registration Number H14PRO021
   7    Copyright (C) 2003
   8      National Institute of Advanced Industrial Science and Technology (AIST)
   9      Registration Number H13PRO009
  10
  11 This file is part of GNU Emacs.
  12
  13 GNU Emacs is free software: you can redistribute it and/or modify
  14 it under the terms of the GNU General Public License as published by
  15 the Free Software Foundation, either version 3 of the License, or
  16 (at your option) any later version.
  17
  18 GNU Emacs is distributed in the hope that it will be useful,
  19 but WITHOUT ANY WARRANTY; without even the implied warranty of
  20 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  21 GNU General Public License for more details.
  22
  23 You should have received a copy of the GNU General Public License
  24 along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
  25
  26 #ifndef EMACS_CODING_H
  27 #define EMACS_CODING_H
  28
  29 /* Index to arguments of Fdefine_coding_system_internal.  */
  30
  31 enum define_coding_system_arg_index
  32   {
  33     coding_arg_name,
  34     coding_arg_mnemonic,
  35     coding_arg_coding_type,
  36     coding_arg_charset_list,
  37     coding_arg_ascii_compatible_p,
  38     coding_arg_decode_translation_table,
  39     coding_arg_encode_translation_table,
  40     coding_arg_post_read_conversion,
  41     coding_arg_pre_write_conversion,
  42     coding_arg_default_char,
  43     coding_arg_for_unibyte,
  44     coding_arg_plist,
  45     coding_arg_eol_type,
  46     coding_arg_max
  47   };
  48
  49 enum define_coding_iso2022_arg_index
  50   {
  51     coding_arg_iso2022_initial = coding_arg_max,
  52     coding_arg_iso2022_reg_usage,
  53     coding_arg_iso2022_request,
  54     coding_arg_iso2022_flags,
  55     coding_arg_iso2022_max
  56   };
  57
  58 enum define_coding_utf8_arg_index
  59   {
  60     coding_arg_utf8_bom = coding_arg_max,
  61     coding_arg_utf8_max
  62   };
  63
  64 enum define_coding_utf16_arg_index
  65   {
  66     coding_arg_utf16_bom = coding_arg_max,
  67     coding_arg_utf16_endian,
  68     coding_arg_utf16_max
  69   };
  70
  71 enum define_coding_ccl_arg_index
  72   {
  73     coding_arg_ccl_decoder = coding_arg_max,
  74     coding_arg_ccl_encoder,
  75     coding_arg_ccl_valids,
  76     coding_arg_ccl_max
  77   };
  78
  79 /* Hash table for all coding systems.  Keys are coding system symbols
  80    and values are spec vectors of the corresponding coding system.  A
  81    spec vector has the form [ ATTRS ALIASES EOL-TYPE ].  ATTRS is a
  82    vector of attribute of the coding system.  ALIASES is a list of
  83    aliases (symbols) of the coding system.  EOL-TYPE is `unix', `dos',
  84    `mac' or a vector of coding systems (symbols).  */
  85
  86 extern Lisp_Object Vcoding_system_hash_table;
  87
  88
  89 /* Enumeration of coding system type.  */
  90
  91 enum coding_system_type
  92   {
  93     coding_type_charset,
  94     coding_type_utf_8,
  95     coding_type_utf_16,
  96     coding_type_iso_2022,
  97     coding_type_emacs_mule,
  98     coding_type_sjis,
  99     coding_type_ccl,
 100     coding_type_raw_text,
 101     coding_type_undecided,
 102     coding_type_max
 103   };
 104
 105
 106 /* Enumeration of end-of-line format type.  */
 107
 108 enum end_of_line_type
 109   {
 110     eol_lf,             /* Line-feed only, same as Emacs' internal
 111                            format.  */
 112     eol_crlf,           /* Sequence of carriage-return and
 113                            line-feed.  */
 114     eol_cr,             /* Carriage-return only.  */
 115     eol_any,            /* Accept any of above.  Produce line-feed
 116                            only.  */
 117     eol_undecided,      /* This value is used to denote that the
 118                            eol-type is not yet undecided.  */
 119     eol_type_max
 120   };
 121
 122 /* Enumeration of index to an attribute vector of a coding system.  */
 123
 124 enum coding_attr_index
 125   {
 126     coding_attr_base_name,
 127     coding_attr_docstring,
 128     coding_attr_mnemonic,
 129     coding_attr_type,
 130     coding_attr_charset_list,
 131     coding_attr_ascii_compat,
 132     coding_attr_decode_tbl,
 133     coding_attr_encode_tbl,
 134     coding_attr_trans_tbl,
 135     coding_attr_post_read,
 136     coding_attr_pre_write,
 137     coding_attr_default_char,
 138     coding_attr_for_unibyte,
 139     coding_attr_plist,
 140
 141     coding_attr_category,
 142     coding_attr_safe_charsets,
 143
 144     /* The followings are extra attributes for each type.  */
 145     coding_attr_charset_valids,
 146
 147     coding_attr_ccl_decoder,
 148     coding_attr_ccl_encoder,
 149     coding_attr_ccl_valids,
 150
 151     coding_attr_iso_initial,
 152     coding_attr_iso_usage,
 153     coding_attr_iso_request,
 154     coding_attr_iso_flags,
 155
 156     coding_attr_utf_bom,
 157     coding_attr_utf_16_endian,
 158
 159     coding_attr_emacs_mule_full,
 160
 161     coding_attr_last_index
 162   };
 163
 164
 165 /* Macros to access an element of an attribute vector.  */
 166
 167 #define CODING_ATTR_BASE_NAME(attrs)    AREF (attrs, coding_attr_base_name)
 168 #define CODING_ATTR_TYPE(attrs)         AREF (attrs, coding_attr_type)
 169 #define CODING_ATTR_CHARSET_LIST(attrs) AREF (attrs, coding_attr_charset_list)
 170 #define CODING_ATTR_MNEMONIC(attrs)     AREF (attrs, coding_attr_mnemonic)
 171 #define CODING_ATTR_DOCSTRING(attrs)    AREF (attrs, coding_attr_docstring)
 172 #define CODING_ATTR_ASCII_COMPAT(attrs) AREF (attrs, coding_attr_ascii_compat)
 173 #define CODING_ATTR_DECODE_TBL(attrs)   AREF (attrs, coding_attr_decode_tbl)
 174 #define CODING_ATTR_ENCODE_TBL(attrs)   AREF (attrs, coding_attr_encode_tbl)
 175 #define CODING_ATTR_TRANS_TBL(attrs)    AREF (attrs, coding_attr_trans_tbl)
 176 #define CODING_ATTR_POST_READ(attrs)    AREF (attrs, coding_attr_post_read)
 177 #define CODING_ATTR_PRE_WRITE(attrs)    AREF (attrs, coding_attr_pre_write)
 178 #define CODING_ATTR_DEFAULT_CHAR(attrs) AREF (attrs, coding_attr_default_char)
 179 #define CODING_ATTR_FOR_UNIBYTE(attrs)  AREF (attrs, coding_attr_for_unibyte)
 180 #define CODING_ATTR_PLIST(attrs)        AREF (attrs, coding_attr_plist)
 181 #define CODING_ATTR_CATEGORY(attrs)     AREF (attrs, coding_attr_category)
 182 #define CODING_ATTR_SAFE_CHARSETS(attrs)AREF (attrs, coding_attr_safe_charsets)
 183
 184
 185 /* Return the name of a coding system specified by ID.  */
 186 #define CODING_ID_NAME(id) \
 187   (HASH_KEY (XHASH_TABLE (Vcoding_system_hash_table), id))
 188
 189 /* Return the attribute vector of a coding system specified by ID.  */
 190
 191 #define CODING_ID_ATTRS(id)     \
 192   (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 0))
 193
 194 /* Return the list of aliases of a coding system specified by ID.  */
 195
 196 #define CODING_ID_ALIASES(id)   \
 197   (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 1))
 198
 199 /* Return the eol-type of a coding system specified by ID.  */
 200
 201 #define CODING_ID_EOL_TYPE(id)  \
 202   (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 2))
 203
 204
 205 /* Return the spec vector of CODING_SYSTEM_SYMBOL.  */
 206
 207 #define CODING_SYSTEM_SPEC(coding_system_symbol)        \
 208   (Fgethash (coding_system_symbol, Vcoding_system_hash_table, Qnil))
 209
 210
 211 /* Return the ID of CODING_SYSTEM_SYMBOL.  */
 212
 213 #define CODING_SYSTEM_ID(coding_system_symbol)                  \
 214   hash_lookup (XHASH_TABLE (Vcoding_system_hash_table),         \
 215                coding_system_symbol, NULL)
 216
 217 /* Return 1 if CODING_SYSTEM_SYMBOL is a coding system.  */
 218
 219 #define CODING_SYSTEM_P(coding_system_symbol)           \
 220   (CODING_SYSTEM_ID (coding_system_symbol) >= 0         \
 221    || (! NILP (coding_system_symbol)                    \
 222        && ! NILP (Fcoding_system_p (coding_system_symbol))))
 223
 224 /* Check if X is a coding system or not.  */
 225
 226 #define CHECK_CODING_SYSTEM(x)                          \
 227   do {                                                  \
 228     if (CODING_SYSTEM_ID (x) < 0                        \
 229         && NILP (Fcheck_coding_system (x)))             \
 230       wrong_type_argument (Qcoding_system_p, (x));      \
 231   } while (0)
 232
 233
 234 /* Check if X is a coding system or not.  If it is, set SEPC to the
 235    spec vector of the coding system.  */
 236
 237 #define CHECK_CODING_SYSTEM_GET_SPEC(x, spec)           \
 238   do {                                                  \
 239     spec = CODING_SYSTEM_SPEC (x);                      \
 240     if (NILP (spec))                                    \
 241       {                                                 \
 242         Fcheck_coding_system (x);                       \
 243         spec = CODING_SYSTEM_SPEC (x);                  \
 244       }                                                 \
 245     if (NILP (spec))                                    \
 246       wrong_type_argument (Qcoding_system_p, (x));      \
 247   } while (0)
 248
 249
 250 /* Check if X is a coding system or not.  If it is, set ID to the
 251    ID of the coding system.  */
 252
 253 #define CHECK_CODING_SYSTEM_GET_ID(x, id)                       \
 254   do                                                            \
 255     {                                                           \
 256       id = CODING_SYSTEM_ID (x);                                \
 257       if (id < 0)                                               \
 258         {                                                       \
 259           Fcheck_coding_system (x);                             \
 260           id = CODING_SYSTEM_ID (x);                            \
 261         }                                                       \
 262       if (id < 0)                                               \
 263         wrong_type_argument (Qcoding_system_p, (x));    \
 264     } while (0)
 265
 266
 267 /*** GENERAL section ***/
 268
 269 /* Enumeration of result code of code conversion.  */
 270 enum coding_result_code
 271   {
 272     CODING_RESULT_SUCCESS,
 273     CODING_RESULT_INSUFFICIENT_SRC,
 274     CODING_RESULT_INSUFFICIENT_DST,
 275     CODING_RESULT_INVALID_SRC,
 276     CODING_RESULT_INTERRUPT
 277   };
 278
 279
 280 /* Macros used for the member `mode' of the struct coding_system.  */
 281
 282 /* If set, the decoding/encoding routines treat the current data as
 283    the last block of the whole text to be converted, and do the
 284    appropriate finishing job.  */
 285 #define CODING_MODE_LAST_BLOCK                  0x01
 286
 287 /* If set, it means that the current source text is in a buffer which
 288    enables selective display.  */
 289 #define CODING_MODE_SELECTIVE_DISPLAY           0x02
 290
 291 /* This flag is used by the decoding/encoding routines on the fly.  If
 292    set, it means that right-to-left text is being processed.  */
 293 #define CODING_MODE_DIRECTION                   0x04
 294
 295 #define CODING_MODE_FIXED_DESTINATION           0x08
 296
 297 /* If set, it means that the encoding routines produces some safe
 298    ASCII characters (usually '?') for unsupported characters.  */
 299 #define CODING_MODE_SAFE_ENCODING               0x10
 300
 301   /* For handling composition sequence.  */
 302 #include "composite.h"
 303
 304 enum composition_state
 305   {
 306     COMPOSING_NO,
 307     COMPOSING_CHAR,
 308     COMPOSING_RULE,
 309     COMPOSING_COMPONENT_CHAR,
 310     COMPOSING_COMPONENT_RULE
 311   };
 312
 313 /* Structure for the current composition status.  */
 314 struct composition_status
 315 {
 316   enum composition_state state;
 317   enum composition_method method;
 318   bool old_form;          /* true if pre-21 form */
 319   int length;             /* number of elements produced in charbuf */
 320   int nchars;             /* number of characters composed */
 321   int ncomps;             /* number of composition components */
 322   /* Maximum carryover is for the case of COMPOSITION_WITH_RULE_ALTCHARS.
 323      See the comment in coding.c.  */
 324   int carryover[4               /* annotation header */
 325                 + MAX_COMPOSITION_COMPONENTS * 3 - 2 /* ALTs and RULEs */
 326                 + 2                                  /* intermediate -1 -1 */
 327                 + MAX_COMPOSITION_COMPONENTS         /* CHARs */
 328                 ];
 329 };
 330
 331
 332 /* Structure of the field `spec.iso_2022' in the structure
 333    `coding_system'.  */
 334 struct iso_2022_spec
 335 {
 336   /* Bit-wise-or of CODING_ISO_FLAG_XXX.  */
 337   unsigned flags;
 338
 339   /* The current graphic register invoked to each graphic plane.  */
 340   int current_invocation[2];
 341
 342   /* The current charset designated to each graphic register.  The
 343      value -1 means that not charset is designated, -2 means that
 344      there was an invalid designation previously.  */
 345   int current_designation[4];
 346
 347   /* If positive, we are now scanning CTEXT extended segment.  */
 348   int ctext_extended_segment_len;
 349
 350   /* True temporarily only when graphic register 2 or 3 is invoked by
 351      single-shift while encoding.  */
 352   unsigned single_shifting : 1;
 353
 354   /* True temporarily only when processing at beginning of line.  */
 355   unsigned bol : 1;
 356
 357   /* If true, we are now scanning embedded UTF-8 sequence.  */
 358   unsigned embedded_utf_8 : 1;
 359
 360   /* The current composition.  */
 361   struct composition_status cmp_status;
 362 };
 363
 364 struct emacs_mule_spec
 365 {
 366   struct composition_status cmp_status;
 367 };
 368
 369 struct ccl_spec;
 370
 371 enum utf_bom_type
 372   {
 373     utf_detect_bom,
 374     utf_without_bom,
 375     utf_with_bom
 376   };
 377
 378 enum utf_16_endian_type
 379   {
 380     utf_16_big_endian,
 381     utf_16_little_endian
 382   };
 383
 384 struct utf_16_spec
 385 {
 386   enum utf_bom_type bom;
 387   enum utf_16_endian_type endian;
 388   int surrogate;
 389 };
 390
 391 struct coding_detection_info
 392 {
 393   /* Values of these members are bitwise-OR of CATEGORY_MASK_XXXs.  */
 394   /* Which categories are already checked.  */
 395   int checked;
 396   /* Which categories are strongly found.  */
 397   int found;
 398   /* Which categories are rejected.  */
 399   int rejected;
 400 };
 401
 402
 403 struct coding_system
 404 {
 405   /* ID number of the coding system.  This is an index to
 406      Vcoding_system_hash_table.  This value is set by
 407      setup_coding_system.  At the early stage of building time, this
 408      value is -1 in the array coding_categories to indicate that no
 409      coding-system of that category is yet defined.  */
 410   ptrdiff_t id;
 411
 412   /* Flag bits of the coding system.  The meaning of each bit is common
 413      to all types of coding systems.  */
 414   int common_flags;
 415
 416   /* Mode bits of the coding system.  See the comments of the macros
 417      CODING_MODE_XXX.  */
 418   unsigned int mode;
 419
 420   /* Detailed information specific to each type of coding system.  */
 421   union
 422     {
 423       struct iso_2022_spec iso_2022;
 424       struct ccl_spec *ccl;     /* Defined in ccl.h.  */
 425       struct utf_16_spec utf_16;
 426       enum utf_bom_type utf_8_bom;
 427       struct emacs_mule_spec emacs_mule;
 428     } spec;
 429
 430   int max_charset_id;
 431   unsigned char *safe_charsets;
 432
 433   /* The following two members specify how binary 8-bit code 128..255
 434      are represented in source and destination text respectively.  1
 435      means they are represented by 2-byte sequence, 0 means they are
 436      represented by 1-byte as is (see the comment in character.h).  */
 437   unsigned src_multibyte : 1;
 438   unsigned dst_multibyte : 1;
 439
 440   /* How may heading bytes we can skip for decoding.  This is set to
 441      -1 in setup_coding_system, and updated by detect_coding.  So,
 442      when this is equal to the byte length of the text being
 443      converted, we can skip the actual conversion process except for
 444      the eol format.  */
 445   ptrdiff_t head_ascii;
 446
 447   ptrdiff_t detected_utf8_chars;
 448
 449   /* Used internally in coding.c.  See the comment of detect_ascii.  */
 450   int eol_seen;
 451
 452   /* The following members are set by encoding/decoding routine.  */
 453   ptrdiff_t produced, produced_char, consumed, consumed_char;
 454
 455   /* Number of error source data found in a decoding routine.  */
 456   int errors;
 457
 458   /* Store the positions of error source data.  */
 459   ptrdiff_t *error_positions;
 460
 461   /* Finish status of code conversion.  */
 462   enum coding_result_code result;
 463
 464   ptrdiff_t src_pos, src_pos_byte, src_chars, src_bytes;
 465   Lisp_Object src_object;
 466   const unsigned char *source;
 467
 468   ptrdiff_t dst_pos, dst_pos_byte, dst_bytes;
 469   Lisp_Object dst_object;
 470   unsigned char *destination;
 471
 472   /* If an element is non-negative, it is a character code.
 473
 474      If it is in the range -128..-1, it is a 8-bit character code
 475      minus 256.
 476
 477      If it is less than -128, it specifies the start of an annotation
 478      chunk.  The length of the chunk is -128 minus the value of the
 479      element.  The following elements are OFFSET, ANNOTATION-TYPE, and
 480      a sequence of actual data for the annotation.  OFFSET is a
 481      character position offset from dst_pos or src_pos,
 482      ANNOTATION-TYPE specifies the meaning of the annotation and how to
 483      handle the following data..  */
 484   int *charbuf;
 485   int charbuf_size, charbuf_used;
 486
 487   /* True if the source of conversion is not in the member
 488      `charbuf', but at `src_object'.  */
 489   unsigned chars_at_source : 1;
 490
 491   /* Set to 1 if charbuf contains an annotation.  */
 492   unsigned annotated : 1;
 493
 494   unsigned char carryover[64];
 495   int carryover_bytes;
 496
 497   int default_char;
 498
 499   bool (*detector) (struct coding_system *, struct coding_detection_info *);
 500   void (*decoder) (struct coding_system *);
 501   bool (*encoder) (struct coding_system *);
 502 };
 503
 504 /* Meanings of bits in the member `common_flags' of the structure
 505    coding_system.  The lowest 8 bits are reserved for various kind of
 506    annotations (currently two of them are used).  */
 507 #define CODING_ANNOTATION_MASK                  0x00FF
 508 #define CODING_ANNOTATE_COMPOSITION_MASK        0x0001
 509 #define CODING_ANNOTATE_DIRECTION_MASK          0x0002
 510 #define CODING_ANNOTATE_CHARSET_MASK            0x0003
 511 #define CODING_FOR_UNIBYTE_MASK                 0x0100
 512 #define CODING_REQUIRE_FLUSHING_MASK            0x0200
 513 #define CODING_REQUIRE_DECODING_MASK            0x0400
 514 #define CODING_REQUIRE_ENCODING_MASK            0x0800
 515 #define CODING_REQUIRE_DETECTION_MASK           0x1000
 516 #define CODING_RESET_AT_BOL_MASK                0x2000
 517
 518 /* Return 1 if the coding context CODING requires annotation
 519    handling.  */
 520 #define CODING_REQUIRE_ANNOTATION(coding) \
 521   ((coding)->common_flags & CODING_ANNOTATION_MASK)
 522
 523 /* Return 1 if the coding context CODING prefers decoding into unibyte.  */
 524 #define CODING_FOR_UNIBYTE(coding) \
 525   ((coding)->common_flags & CODING_FOR_UNIBYTE_MASK)
 526
 527 /* Return 1 if the coding context CODING requires specific code to be
 528    attached at the tail of converted text.  */
 529 #define CODING_REQUIRE_FLUSHING(coding) \
 530   ((coding)->common_flags & CODING_REQUIRE_FLUSHING_MASK)
 531
 532 /* Return 1 if the coding context CODING requires code conversion on
 533    decoding.  */
 534 #define CODING_REQUIRE_DECODING(coding) \
 535   ((coding)->dst_multibyte              \
 536    || (coding)->common_flags & CODING_REQUIRE_DECODING_MASK)
 537
 538
 539 /* Return 1 if the coding context CODING requires code conversion on
 540    encoding.
 541    The non-multibyte part of the condition is to support encoding of
 542    unibyte strings/buffers generated by string-as-unibyte or
 543    (set-buffer-multibyte nil) from multibyte strings/buffers.  */
 544 #define CODING_REQUIRE_ENCODING(coding)                         \
 545   ((coding)->src_multibyte                                      \
 546    || (coding)->common_flags & CODING_REQUIRE_ENCODING_MASK     \
 547    || (coding)->mode & CODING_MODE_SELECTIVE_DISPLAY)
 548
 549
 550 /* Return 1 if the coding context CODING requires some kind of code
 551    detection.  */
 552 #define CODING_REQUIRE_DETECTION(coding) \
 553   ((coding)->common_flags & CODING_REQUIRE_DETECTION_MASK)
 554
 555 /* Return 1 if the coding context CODING requires code conversion on
 556    decoding or some kind of code detection.  */
 557 #define CODING_MAY_REQUIRE_DECODING(coding)     \
 558   (CODING_REQUIRE_DECODING (coding)             \
 559    || CODING_REQUIRE_DETECTION (coding))
 560
 561 /* Macros to decode or encode a character of JISX0208 in SJIS.  S1 and
 562    S2 are the 1st and 2nd position-codes of JISX0208 in SJIS coding
 563    system.  C1 and C2 are the 1st and 2nd position codes of Emacs'
 564    internal format.  */
 565
 566 #define SJIS_TO_JIS(code)                               \
 567   do {                                                  \
 568     int s1, s2, j1, j2;                                 \
 569                                                         \
 570     s1 = (code) >> 8, s2 = (code) & 0xFF;               \
 571                                                         \
 572     if (s2 >= 0x9F)                                     \
 573       (j1 = s1 * 2 - (s1 >= 0xE0 ? 0x160 : 0xE0),       \
 574        j2 = s2 - 0x7E);                                 \
 575     else                                                \
 576       (j1 = s1 * 2 - ((s1 >= 0xE0) ? 0x161 : 0xE1),     \
 577        j2 = s2 - ((s2 >= 0x7F) ? 0x20 : 0x1F));         \
 578     (code) = (j1 << 8) | j2;                            \
 579   } while (0)
 580
 581 #define SJIS_TO_JIS2(code)                              \
 582   do {                                                  \
 583     int s1, s2, j1, j2;                                 \
 584                                                         \
 585     s1 = (code) >> 8, s2 = (code) & 0xFF;               \
 586                                                         \
 587     if (s2 >= 0x9F)                                     \
 588       {                                                 \
 589         j1 = (s1 == 0xF0 ? 0x28                         \
 590               : s1 == 0xF1 ? 0x24                       \
 591               : s1 == 0xF2 ? 0x2C                       \
 592               : s1 == 0xF3 ? 0x2E                       \
 593               : 0x6E + (s1 - 0xF4) * 2);                \
 594         j2 = s2 - 0x7E;                                 \
 595       }                                                 \
 596     else                                                \
 597       {                                                 \
 598         j1 = (s1 <= 0xF2 ? 0x21 + (s1 - 0xF0) * 2       \
 599               : s1 <= 0xF4 ? 0x2D + (s1 - 0xF3) * 2     \
 600               : 0x6F + (s1 - 0xF5) * 2);                \
 601         j2 = s2 - ((s2 >= 0x7F ? 0x20 : 0x1F));         \
 602       }                                                 \
 603     (code) = (j1 << 8) | j2;                            \
 604   } while (0)
 605
 606
 607 #define JIS_TO_SJIS(code)                               \
 608   do {                                                  \
 609     int s1, s2, j1, j2;                                 \
 610                                                         \
 611     j1 = (code) >> 8, j2 = (code) & 0xFF;               \
 612     if (j1 & 1)                                         \
 613       (s1 = j1 / 2 + ((j1 < 0x5F) ? 0x71 : 0xB1),       \
 614        s2 = j2 + ((j2 >= 0x60) ? 0x20 : 0x1F));         \
 615     else                                                \
 616       (s1 = j1 / 2 + ((j1 < 0x5F) ? 0x70 : 0xB0),       \
 617        s2 = j2 + 0x7E);                                 \
 618     (code) = (s1 << 8) | s2;                            \
 619   } while (0)
 620
 621 #define JIS_TO_SJIS2(code)                              \
 622   do {                                                  \
 623     int s1, s2, j1, j2;                                 \
 624                                                         \
 625     j1 = (code) >> 8, j2 = (code) & 0xFF;               \
 626     if (j1 & 1)                                         \
 627       {                                                 \
 628         s1 = (j1 <= 0x25 ? 0xF0 + (j1 - 0x21) / 2       \
 629               : j1 <= 0x2F ? 0xF3 + (j1 - 0x2D) / 2     \
 630               : 0xF5 + (j1 - 0x6F) / 2);                \
 631         s2 = j2 + ((j2 >= 0x60) ? 0x20 : 0x1F);         \
 632       }                                                 \
 633     else                                                \
 634       {                                                 \
 635         s1 = (j1 == 0x28 ? 0xF0                         \
 636               : j1 == 0x24 ? 0xF1                       \
 637               : j1 == 0x2C ? 0xF2                       \
 638               : j1 == 0x2E ? 0xF3                       \
 639               : 0xF4 + (j1 - 0x6E) / 2);                \
 640         s2 = j2 + 0x7E;                                 \
 641       }                                                 \
 642     (code) = (s1 << 8) | s2;                            \
 643   } while (0)
 644
 645 /* Encode the file name NAME using the specified coding system
 646    for file names, if any.  */
 647 #define ENCODE_FILE(name)                                                  \
 648   (! NILP (Vfile_name_coding_system)                                       \
 649    ? code_convert_string_norecord (name, Vfile_name_coding_system, 1)      \
 650    : (! NILP (Vdefault_file_name_coding_system)                            \
 651       ? code_convert_string_norecord (name, Vdefault_file_name_coding_system, 1) \
 652       : name))
 653
 654
 655 /* Decode the file name NAME using the specified coding system
 656    for file names, if any.  */
 657 #define DECODE_FILE(name)                                                  \
 658   (! NILP (Vfile_name_coding_system)                                       \
 659    ? code_convert_string_norecord (name, Vfile_name_coding_system, 0)      \
 660    : (! NILP (Vdefault_file_name_coding_system)                            \
 661       ? code_convert_string_norecord (name, Vdefault_file_name_coding_system, 0) \
 662       : name))
 663
 664
 665 /* Encode the string STR using the specified coding system
 666    for system functions, if any.  */
 667 #define ENCODE_SYSTEM(str)                                                 \
 668   (! NILP (Vlocale_coding_system)                                          \
 669    ? code_convert_string_norecord (str, Vlocale_coding_system, 1)          \
 670    : str)
 671
 672 /* Decode the string STR using the specified coding system
 673    for system functions, if any.  */
 674 #define DECODE_SYSTEM(str)                                                 \
 675   (! NILP (Vlocale_coding_system)                                          \
 676    ? code_convert_string_norecord (str, Vlocale_coding_system, 0)          \
 677    : str)
 678
 679 /* Note that this encodes utf-8, not utf-8-emacs, so it's not a no-op.  */
 680 #define ENCODE_UTF_8(str) code_convert_string_norecord (str, Qutf_8, 1)
 681
 682 /* Extern declarations.  */
 683 extern Lisp_Object code_conversion_save (bool, bool);
 684 extern void setup_coding_system (Lisp_Object, struct coding_system *);
 685 extern Lisp_Object coding_charset_list (struct coding_system *);
 686 extern Lisp_Object coding_system_charset_list (Lisp_Object);
 687 extern Lisp_Object code_convert_string (Lisp_Object, Lisp_Object,
 688                                         Lisp_Object, bool, bool, bool);
 689 extern Lisp_Object code_convert_string_norecord (Lisp_Object, Lisp_Object,
 690                                                  bool);
 691 extern Lisp_Object raw_text_coding_system (Lisp_Object);
 692 extern Lisp_Object coding_inherit_eol_type (Lisp_Object, Lisp_Object);
 693 extern Lisp_Object complement_process_encoding_system (Lisp_Object);
 694
 695 extern void decode_coding_gap (struct coding_system *,
 696                                ptrdiff_t, ptrdiff_t);
 697 extern void decode_coding_object (struct coding_system *,
 698                                   Lisp_Object, ptrdiff_t, ptrdiff_t,
 699                                   ptrdiff_t, ptrdiff_t, Lisp_Object);
 700 extern void encode_coding_object (struct coding_system *,
 701                                   Lisp_Object, ptrdiff_t, ptrdiff_t,
 702                                   ptrdiff_t, ptrdiff_t, Lisp_Object);
 703
 704 #if defined (WINDOWSNT) || defined (CYGWIN)
 705
 706 /* These functions use Lisp string objects to store the UTF-16LE
 707    strings that modern versions of Windows expect.  These strings are
 708    not particularly useful to Lisp, and all Lisp strings should be
 709    native Emacs multibyte.  */
 710
 711 /* Access the wide-character string stored in a Lisp string object.  */
 712 #define WCSDATA(x) ((wchar_t *) SDATA (x))
 713
 714 /* Convert the multi-byte string in STR to UTF-16LE encoded unibyte
 715    string, and store it in *BUF.  BUF may safely point to STR on entry.  */
 716 extern wchar_t *to_unicode (Lisp_Object str, Lisp_Object *buf);
 717
 718 /* Convert STR, a UTF-16LE encoded string embedded in a unibyte string
 719    object, to a multi-byte Emacs string and return it.  This function
 720    calls code_convert_string_norecord internally and has all its
 721    failure modes.  STR itself is not modified.  */
 722 extern Lisp_Object from_unicode (Lisp_Object str);
 723
 724 /* Convert WSTR to an Emacs string.  */
 725 extern Lisp_Object from_unicode_buffer (const wchar_t* wstr);
 726
 727 #endif /* WINDOWSNT || CYGWIN */
 728
 729 /* Macros for backward compatibility.  */
 730
 731 #define encode_coding_string(coding, string, nocopy)                    \
 732   (STRING_MULTIBYTE(string) ?                                           \
 733     (encode_coding_object (coding, string, 0, 0, SCHARS (string),       \
 734                            SBYTES (string), Qt),                        \
 735      (coding)->dst_object) : (string))
 736
 737
 738 #define decode_coding_c_string(coding, src, bytes, dst_object)          \
 739   do {                                                                  \
 740     (coding)->source = (src);                                           \
 741     (coding)->src_chars = (coding)->src_bytes = (bytes);                \
 742     decode_coding_object ((coding), Qnil, 0, 0, (bytes), (bytes),       \
 743                           (dst_object));                                \
 744   } while (0)
 745
 746
 747 extern Lisp_Object preferred_coding_system (void);
 748
 749
 750 extern Lisp_Object Qutf_8, Qutf_8_emacs;
 751
 752 extern Lisp_Object Qcoding_category_index;
 753 extern Lisp_Object Qcoding_system_p;
 754 extern Lisp_Object Qraw_text, Qemacs_mule, Qno_conversion, Qundecided;
 755 extern Lisp_Object Qbuffer_file_coding_system;
 756
 757 extern Lisp_Object Qunix, Qdos;
 758
 759 extern Lisp_Object Qtranslation_table;
 760 extern Lisp_Object Qtranslation_table_id;
 761
 762 #ifdef emacs
 763 extern Lisp_Object Qfile_coding_system;
 764 extern Lisp_Object Qcall_process, Qcall_process_region;
 765 extern Lisp_Object Qstart_process, Qopen_network_stream;
 766 extern Lisp_Object Qwrite_region;
 767
 768 extern char *emacs_strerror (int);
 769
 770 /* Coding system to be used to encode text for terminal display when
 771    terminal coding system is nil.  */
 772 extern struct coding_system safe_terminal_coding;
 773
 774 #endif
 775
 776 /* Error signaled when there's a problem with detecting coding system */
 777 extern Lisp_Object Qcoding_system_error;
 778
 779 extern char emacs_mule_bytes[256];
 780
 781 #endif /* EMACS_CODING_H */