coding.h 25.1 KB
Newer Older
Karl Heuer's avatar
Karl Heuer committed
1
/* Header for coding system handler.
2
   Copyright (C) 2001, 2002, 2003, 2004, 2005,
Glenn Morris's avatar
Glenn Morris committed
3
                 2006, 2007, 2008  Free Software Foundation, Inc.
Kenichi Handa's avatar
Kenichi Handa committed
4
   Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
Glenn Morris's avatar
Glenn Morris committed
5
     2005, 2006, 2007, 2008
Kenichi Handa's avatar
Kenichi Handa committed
6 7
     National Institute of Advanced Industrial Science and Technology (AIST)
     Registration Number H14PRO021
Kenichi Handa's avatar
Kenichi Handa committed
8
   Copyright (C) 2003
Kenichi Handa's avatar
Kenichi Handa committed
9 10
     National Institute of Advanced Industrial Science and Technology (AIST)
     Registration Number H13PRO009
Karl Heuer's avatar
Karl Heuer committed
11

Karl Heuer's avatar
Karl Heuer committed
12 13
This file is part of GNU Emacs.

14
GNU Emacs is free software: you can redistribute it and/or modify
Karl Heuer's avatar
Karl Heuer committed
15
it under the terms of the GNU General Public License as published by
16 17
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Karl Heuer's avatar
Karl Heuer committed
18

Karl Heuer's avatar
Karl Heuer committed
19 20 21 22
GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.
Karl Heuer's avatar
Karl Heuer committed
23

Karl Heuer's avatar
Karl Heuer committed
24
You should have received a copy of the GNU General Public License
25
along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
Karl Heuer's avatar
Karl Heuer committed
26

27 28
#ifndef EMACS_CODING_H
#define EMACS_CODING_H
Karl Heuer's avatar
Karl Heuer committed
29

Kenichi Handa's avatar
Kenichi Handa committed
30
/* Index to arguments of Fdefine_coding_system_internal.  */
Karl Heuer's avatar
Karl Heuer committed
31

Kenichi Handa's avatar
Kenichi Handa committed
32 33 34 35 36 37 38 39 40 41 42 43
enum define_coding_system_arg_index
  {
    coding_arg_name,
    coding_arg_mnemonic,
    coding_arg_coding_type,
    coding_arg_charset_list,
    coding_arg_ascii_compatible_p,
    coding_arg_decode_translation_table,
    coding_arg_encode_translation_table,
    coding_arg_post_read_conversion,
    coding_arg_pre_write_conversion,
    coding_arg_default_char,
Kenichi Handa's avatar
Kenichi Handa committed
44
    coding_arg_for_unibyte,
Kenichi Handa's avatar
Kenichi Handa committed
45 46 47 48
    coding_arg_plist,
    coding_arg_eol_type,
    coding_arg_max
  };
Karl Heuer's avatar
Karl Heuer committed
49

Kenichi Handa's avatar
Kenichi Handa committed
50
enum define_coding_iso2022_arg_index
Karl Heuer's avatar
Karl Heuer committed
51
  {
Kenichi Handa's avatar
Kenichi Handa committed
52 53 54 55 56
    coding_arg_iso2022_initial = coding_arg_max,
    coding_arg_iso2022_reg_usage,
    coding_arg_iso2022_request,
    coding_arg_iso2022_flags,
    coding_arg_iso2022_max
Karl Heuer's avatar
Karl Heuer committed
57 58
  };

Kenichi Handa's avatar
Kenichi Handa committed
59
enum define_coding_utf16_arg_index
Karl Heuer's avatar
Karl Heuer committed
60
  {
Kenichi Handa's avatar
Kenichi Handa committed
61 62 63
    coding_arg_utf16_bom = coding_arg_max,
    coding_arg_utf16_endian,
    coding_arg_utf16_max
Karl Heuer's avatar
Karl Heuer committed
64 65
  };

Kenichi Handa's avatar
Kenichi Handa committed
66 67
enum define_coding_ccl_arg_index
  {
68
    coding_arg_ccl_decoder = coding_arg_max,
Kenichi Handa's avatar
Kenichi Handa committed
69 70 71 72
    coding_arg_ccl_encoder,
    coding_arg_ccl_valids,
    coding_arg_ccl_max
  };
Karl Heuer's avatar
Karl Heuer committed
73

Kenichi Handa's avatar
Kenichi Handa committed
74 75 76 77 78 79 80
/* Hash table for all coding systems.  Keys are coding system symbols
   and values are spec vectors of the corresponding coding system.  A
   spec vector has the form [ ATTRS ALIASES EOL-TYPE ].  ATTRS is a
   vector of attribute of the coding system.  ALIASES is a list of
   aliases (symbols) of the coding system.  EOL-TYPE is `unix', `dos',
   `mac' or a vector of coding systems (symbols).  */

Kenichi Handa's avatar
Kenichi Handa committed
81
extern Lisp_Object Vcoding_system_hash_table;
Karl Heuer's avatar
Karl Heuer committed
82

Kenichi Handa's avatar
Kenichi Handa committed
83

Kenichi Handa's avatar
Kenichi Handa committed
84
/* Enumeration of coding system type.  */
Karl Heuer's avatar
Karl Heuer committed
85

Kenichi Handa's avatar
Kenichi Handa committed
86 87 88 89 90 91 92 93 94 95 96 97 98
enum coding_system_type
  {
    coding_type_charset,
    coding_type_utf_8,
    coding_type_utf_16,
    coding_type_iso_2022,
    coding_type_emacs_mule,
    coding_type_sjis,
    coding_type_ccl,
    coding_type_raw_text,
    coding_type_undecided,
    coding_type_max
  };
Karl Heuer's avatar
Karl Heuer committed
99 100


Kenichi Handa's avatar
Kenichi Handa committed
101
/* Enumeration of end-of-line format type.  */
Karl Heuer's avatar
Karl Heuer committed
102

Kenichi Handa's avatar
Kenichi Handa committed
103 104 105 106 107 108 109 110 111 112 113 114 115
enum end_of_line_type
  {
    eol_lf,		/* Line-feed only, same as Emacs' internal
			   format.  */
    eol_crlf,		/* Sequence of carriage-return and
			   line-feed.  */
    eol_cr,		/* Carriage-return only.  */
    eol_any,		/* Accept any of above.  Produce line-feed
			   only.  */
    eol_undecided,	/* This value is used to denote that the
			   eol-type is not yet undecided.  */
    eol_type_max
  };
Karl Heuer's avatar
Karl Heuer committed
116

Kenichi Handa's avatar
Kenichi Handa committed
117
/* Enumeration of index to an attribute vector of a coding system.  */
Karl Heuer's avatar
Karl Heuer committed
118

Kenichi Handa's avatar
Kenichi Handa committed
119 120 121 122 123 124 125 126 127 128
enum coding_attr_index
  {
    coding_attr_base_name,
    coding_attr_docstring,
    coding_attr_mnemonic,
    coding_attr_type,
    coding_attr_charset_list,
    coding_attr_ascii_compat,
    coding_attr_decode_tbl,
    coding_attr_encode_tbl,
129
    coding_attr_trans_tbl,
Kenichi Handa's avatar
Kenichi Handa committed
130 131 132
    coding_attr_post_read,
    coding_attr_pre_write,
    coding_attr_default_char,
Kenichi Handa's avatar
Kenichi Handa committed
133
    coding_attr_for_unibyte,
Kenichi Handa's avatar
Kenichi Handa committed
134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
    coding_attr_plist,

    coding_attr_category,
    coding_attr_safe_charsets,

    /* The followings are extra attributes for each type.  */
    coding_attr_charset_valids,

    coding_attr_ccl_decoder,
    coding_attr_ccl_encoder,
    coding_attr_ccl_valids,

    coding_attr_iso_initial,
    coding_attr_iso_usage,
    coding_attr_iso_request,
    coding_attr_iso_flags,

    coding_attr_utf_16_bom,
    coding_attr_utf_16_endian,

    coding_attr_emacs_mule_full,

    coding_attr_last_index
  };
Karl Heuer's avatar
Karl Heuer committed
158 159


Kenichi Handa's avatar
Kenichi Handa committed
160 161
/* Macros to access an element of an attribute vector.  */

Juanma Barranquero's avatar
Juanma Barranquero committed
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
#define CODING_ATTR_BASE_NAME(attrs)	AREF (attrs, coding_attr_base_name)
#define CODING_ATTR_TYPE(attrs)		AREF (attrs, coding_attr_type)
#define CODING_ATTR_CHARSET_LIST(attrs)	AREF (attrs, coding_attr_charset_list)
#define CODING_ATTR_MNEMONIC(attrs)	AREF (attrs, coding_attr_mnemonic)
#define CODING_ATTR_DOCSTRING(attrs)	AREF (attrs, coding_attr_docstring)
#define CODING_ATTR_ASCII_COMPAT(attrs)	AREF (attrs, coding_attr_ascii_compat)
#define CODING_ATTR_DECODE_TBL(attrs)	AREF (attrs, coding_attr_decode_tbl)
#define CODING_ATTR_ENCODE_TBL(attrs)	AREF (attrs, coding_attr_encode_tbl)
#define CODING_ATTR_TRANS_TBL(attrs)	AREF (attrs, coding_attr_trans_tbl)
#define CODING_ATTR_POST_READ(attrs)	AREF (attrs, coding_attr_post_read)
#define CODING_ATTR_PRE_WRITE(attrs)	AREF (attrs, coding_attr_pre_write)
#define CODING_ATTR_DEFAULT_CHAR(attrs)	AREF (attrs, coding_attr_default_char)
#define CODING_ATTR_FOR_UNIBYTE(attrs)	AREF (attrs, coding_attr_for_unibyte)
#define CODING_ATTR_FLUSHING(attrs)	AREF (attrs, coding_attr_flushing)
#define CODING_ATTR_PLIST(attrs)	AREF (attrs, coding_attr_plist)
#define CODING_ATTR_CATEGORY(attrs)	AREF (attrs, coding_attr_category)
#define CODING_ATTR_SAFE_CHARSETS(attrs)AREF (attrs, coding_attr_safe_charsets)
179 180


Kenichi Handa's avatar
Kenichi Handa committed
181 182 183 184 185 186
/* Return the name of a coding system specified by ID.  */
#define CODING_ID_NAME(id) \
  (HASH_KEY (XHASH_TABLE (Vcoding_system_hash_table), id))

/* Return the attribute vector of a coding system specified by ID.  */

Kenichi Handa's avatar
Kenichi Handa committed
187 188
#define CODING_ID_ATTRS(id)	\
  (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 0))
189

Kenichi Handa's avatar
Kenichi Handa committed
190 191
/* Return the list of aliases of a coding system specified by ID.  */

Kenichi Handa's avatar
Kenichi Handa committed
192 193
#define CODING_ID_ALIASES(id)	\
  (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 1))
194

Kenichi Handa's avatar
Kenichi Handa committed
195 196
/* Return the eol-type of a coding system specified by ID.  */

Kenichi Handa's avatar
Kenichi Handa committed
197 198
#define CODING_ID_EOL_TYPE(id)	\
  (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 2))
199

Kenichi Handa's avatar
Kenichi Handa committed
200 201

/* Return the spec vector of CODING_SYSTEM_SYMBOL.  */
202

Kenichi Handa's avatar
Kenichi Handa committed
203 204
#define CODING_SYSTEM_SPEC(coding_system_symbol)	\
  (Fgethash (coding_system_symbol, Vcoding_system_hash_table, Qnil))
Karl Heuer's avatar
Karl Heuer committed
205

Kenichi Handa's avatar
Kenichi Handa committed
206 207 208

/* Return the ID of CODING_SYSTEM_SYMBOL.  */

Kenichi Handa's avatar
Kenichi Handa committed
209 210 211
#define CODING_SYSTEM_ID(coding_system_symbol)			\
  hash_lookup (XHASH_TABLE (Vcoding_system_hash_table),		\
	       coding_system_symbol, NULL)
Karl Heuer's avatar
Karl Heuer committed
212

Miles Bader's avatar
Miles Bader committed
213
/* Return 1 if CODING_SYSTEM_SYMBOL is a coding system.  */
Kenichi Handa's avatar
Kenichi Handa committed
214

215 216 217 218
#define CODING_SYSTEM_P(coding_system_symbol)		\
  (CODING_SYSTEM_ID (coding_system_symbol) >= 0		\
   || (! NILP (coding_system_symbol)			\
       && ! NILP (Fcoding_system_p (coding_system_symbol))))
Karl Heuer's avatar
Karl Heuer committed
219

Kenichi Handa's avatar
Kenichi Handa committed
220 221
/* Check if X is a coding system or not.  */

Kenichi Handa's avatar
Kenichi Handa committed
222
#define CHECK_CODING_SYSTEM(x)				\
Kenichi Handa's avatar
Kenichi Handa committed
223
  do {							\
224 225
    if (CODING_SYSTEM_ID (x) < 0			\
	&& NILP (Fcheck_coding_system (x)))		\
Kenichi Handa's avatar
Kenichi Handa committed
226
      wrong_type_argument (Qcoding_system_p, (x));	\
Kenichi Handa's avatar
Kenichi Handa committed
227
  } while (0)
228

Karl Heuer's avatar
Karl Heuer committed
229

Kenichi Handa's avatar
Kenichi Handa committed
230 231 232
/* Check if X is a coding system or not.  If it is, set SEPC to the
   spec vector of the coding system.  */

Kenichi Handa's avatar
Kenichi Handa committed
233 234 235
#define CHECK_CODING_SYSTEM_GET_SPEC(x, spec)		\
  do {							\
    spec = CODING_SYSTEM_SPEC (x);			\
236 237 238 239 240
    if (NILP (spec))					\
      {							\
	Fcheck_coding_system (x);			\
	spec = CODING_SYSTEM_SPEC (x);			\
      }							\
Kenichi Handa's avatar
Kenichi Handa committed
241
    if (NILP (spec))					\
242
      wrong_type_argument (Qcoding_system_p, (x));	\
Kenichi Handa's avatar
Kenichi Handa committed
243
  } while (0)
244

245

Kenichi Handa's avatar
Kenichi Handa committed
246 247 248
/* Check if X is a coding system or not.  If it is, set ID to the
   ID of the coding system.  */

Kenichi Handa's avatar
Kenichi Handa committed
249 250 251 252
#define CHECK_CODING_SYSTEM_GET_ID(x, id)			\
  do								\
    {								\
      id = CODING_SYSTEM_ID (x);				\
253 254 255 256 257
      if (id < 0)						\
	{							\
	  Fcheck_coding_system (x);				\
	  id = CODING_SYSTEM_ID (x);				\
	}							\
Kenichi Handa's avatar
Kenichi Handa committed
258
      if (id < 0)						\
259
	wrong_type_argument (Qcoding_system_p, (x));	\
Kenichi Handa's avatar
Kenichi Handa committed
260
    } while (0)
Karl Heuer's avatar
Karl Heuer committed
261 262 263 264


/*** GENERAL section ***/

Kenichi Handa's avatar
Kenichi Handa committed
265 266
/* Enumeration of result code of code conversion.  */
enum coding_result_code
Karl Heuer's avatar
Karl Heuer committed
267
  {
Kenichi Handa's avatar
Kenichi Handa committed
268 269 270 271
    CODING_RESULT_SUCCESS,
    CODING_RESULT_INSUFFICIENT_SRC,
    CODING_RESULT_INSUFFICIENT_DST,
    CODING_RESULT_INCONSISTENT_EOL,
272
    CODING_RESULT_INVALID_SRC,
Kenichi Handa's avatar
Kenichi Handa committed
273 274
    CODING_RESULT_INTERRUPT,
    CODING_RESULT_INSUFFICIENT_MEM
Karl Heuer's avatar
Karl Heuer committed
275 276
  };

277

Kenichi Handa's avatar
Kenichi Handa committed
278
/* Macros used for the member `mode' of the struct coding_system.  */
279 280 281 282 283 284

/* If set, recover the original CR or LF of the already decoded text
   when the decoding routine encounters an inconsistent eol format.  */
#define CODING_MODE_INHIBIT_INCONSISTENT_EOL	0x01

/* If set, the decoding/encoding routines treat the current data as
Dave Love's avatar
Dave Love committed
285
   the last block of the whole text to be converted, and do the
Dave Love's avatar
Dave Love committed
286
   appropriate finishing job.  */
287 288 289 290 291 292 293 294 295 296
#define CODING_MODE_LAST_BLOCK			0x02

/* If set, it means that the current source text is in a buffer which
   enables selective display.  */
#define CODING_MODE_SELECTIVE_DISPLAY		0x04

/* This flag is used by the decoding/encoding routines on the fly.  If
   set, it means that right-to-left text is being processed.  */
#define CODING_MODE_DIRECTION			0x08

Kenichi Handa's avatar
Kenichi Handa committed
297 298
#define CODING_MODE_FIXED_DESTINATION		0x10

Kenichi Handa's avatar
Kenichi Handa committed
299 300
/* If set, it means that the encoding routines produces some safe
   ASCII characters (usually '?') for unsupported characters.  */
Kenichi Handa's avatar
Kenichi Handa committed
301 302 303 304 305
#define CODING_MODE_SAFE_ENCODING		0x20

/* Structure of the field `spec.iso_2022' in the structure
   `coding_system'.  */
struct iso_2022_spec
Karl Heuer's avatar
Karl Heuer committed
306
{
Kenichi Handa's avatar
Kenichi Handa committed
307
  /* Bit-wise-or of CODING_ISO_FLAG_XXX.  */
Kenichi Handa's avatar
Kenichi Handa committed
308
  unsigned flags;
Karl Heuer's avatar
Karl Heuer committed
309

Kenichi Handa's avatar
Kenichi Handa committed
310 311
  /* The current graphic register invoked to each graphic plane.  */
  int current_invocation[2];
312

Kenichi Handa's avatar
Kenichi Handa committed
313 314 315 316
  /* The current charset designated to each graphic register.  The
     value -1 means that not charset is designated, -2 means that
     there was an invalid designation previously.  */
  int current_designation[4];
Karl Heuer's avatar
Karl Heuer committed
317

Kenichi Handa's avatar
Kenichi Handa committed
318 319 320
  /* Set to 1 temporarily only when graphic register 2 or 3 is invoked
     by single-shift while encoding.  */
  int single_shifting;
Karl Heuer's avatar
Karl Heuer committed
321

Kenichi Handa's avatar
Kenichi Handa committed
322 323 324
  /* Set to 1 temporarily only when processing at beginning of line.  */
  int bol;
};
Karl Heuer's avatar
Karl Heuer committed
325

Kenichi Handa's avatar
Kenichi Handa committed
326
struct ccl_spec;
Karl Heuer's avatar
Karl Heuer committed
327

Kenichi Handa's avatar
Kenichi Handa committed
328 329 330 331 332 333
enum utf_16_bom_type
  {
    utf_16_detect_bom,
    utf_16_without_bom,
    utf_16_with_bom
  };
334

Kenichi Handa's avatar
Kenichi Handa committed
335 336 337 338 339
enum utf_16_endian_type
  {
    utf_16_big_endian,
    utf_16_little_endian
  };
340

Kenichi Handa's avatar
Kenichi Handa committed
341 342 343 344 345 346
struct utf_16_spec
{
  enum utf_16_bom_type bom;
  enum utf_16_endian_type endian;
  int surrogate;
};
347

348 349 350 351 352 353 354 355 356 357
struct coding_detection_info
{
  /* Values of these members are bitwise-OR of CATEGORY_MASK_XXXs.  */
  /* Which categories are already checked.  */
  int checked;
  /* Which categories are strongly found.  */
  int found;
  /* Which categories are rejected.  */
  int rejected;
};
358 359


Kenichi Handa's avatar
Kenichi Handa committed
360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375
struct coding_system
{
  /* ID number of the coding system.  This is an index to
     Vcoding_system_hash_table.  This value is set by
     setup_coding_system.  At the early stage of building time, this
     value is -1 in the array coding_categories to indicate that no
     coding-system of that category is yet defined.  */
  int id;

  /* Flag bits of the coding system.  The meaning of each bit is common
     to all types of coding systems.  */
  int common_flags;

  /* Mode bits of the coding system.  See the comments of the macros
     CODING_MODE_XXX.  */
  unsigned int mode;
376

Karl Heuer's avatar
Karl Heuer committed
377
  /* Detailed information specific to each type of coding system.  */
Kenichi Handa's avatar
Kenichi Handa committed
378
  union
Karl Heuer's avatar
Karl Heuer committed
379
    {
Kenichi Handa's avatar
Kenichi Handa committed
380 381 382 383
      struct iso_2022_spec iso_2022;
      struct ccl_spec *ccl;	/* Defined in ccl.h.  */
      struct utf_16_spec utf_16;
      int emacs_mule_full_support;
Karl Heuer's avatar
Karl Heuer committed
384 385
    } spec;

Kenichi Handa's avatar
Kenichi Handa committed
386 387
  int max_charset_id;
  char *safe_charsets;
388

Kenichi Handa's avatar
Kenichi Handa committed
389 390 391 392
  /* The following two members specify how binary 8-bit code 128..255
     are represented in source and destination text respectively.  1
     means they are represented by 2-byte sequence, 0 means they are
     represented by 1-byte as is (see the comment in character.h).  */
393 394 395
  unsigned src_multibyte : 1;
  unsigned dst_multibyte : 1;

396 397 398 399
  /* How may heading bytes we can skip for decoding.  This is set to
     -1 in setup_coding_system, and updated by detect_coding.  So,
     when this is equal to the byte length of the text being
     converted, we can skip the actual conversion process.  */
Kenichi Handa's avatar
Kenichi Handa committed
400
  int head_ascii;
401 402

  /* The following members are set by encoding/decoding routine.  */
Kenichi Handa's avatar
Kenichi Handa committed
403
  EMACS_INT produced, produced_char, consumed, consumed_char;
404

405 406 407
  /* Number of error source data found in a decoding routine.  */
  int errors;

Kenichi Handa's avatar
Kenichi Handa committed
408 409
  /* Store the positions of error source data. */
  EMACS_INT *error_positions;
410

Kenichi Handa's avatar
Kenichi Handa committed
411 412
  /* Finish status of code conversion.  */
  enum coding_result_code result;
413

Kenichi Handa's avatar
Kenichi Handa committed
414 415
  EMACS_INT src_pos, src_pos_byte, src_chars, src_bytes;
  Lisp_Object src_object;
Kenichi Handa's avatar
Kenichi Handa committed
416
  const unsigned char *source;
Karl Heuer's avatar
Karl Heuer committed
417

Kenichi Handa's avatar
Kenichi Handa committed
418 419 420
  EMACS_INT dst_pos, dst_pos_byte, dst_bytes;
  Lisp_Object dst_object;
  unsigned char *destination;
Karl Heuer's avatar
Karl Heuer committed
421

Miles Bader's avatar
Miles Bader committed
422
  /* Set to 1 if the source of conversion is not in the member
Kenichi Handa's avatar
Kenichi Handa committed
423
     `charbuf', but at `src_object'.  */
Kenichi Handa's avatar
Kenichi Handa committed
424
  int chars_at_source;
Karl Heuer's avatar
Karl Heuer committed
425

Kenichi Handa's avatar
Kenichi Handa committed
426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
  /* If an element is non-negative, it is a character code.

     If it is in the range -128..-1, it is a 8-bit character code
     minus 256.

     If it is less than -128, it specifies the start of an annotation
     chunk.  The length of the chunk is -128 minus the value of the
     element.  The following elements are OFFSET, ANNOTATION-TYPE, and
     a sequence of actual data for the annotation.  OFFSET is a
     character position offset from dst_pos or src_pos,
     ANNOTATION-TYPE specfies the meaning of the annotation and how to
     handle the following data..  */
  int *charbuf;
  int charbuf_size, charbuf_used;

  /* Set to 1 if charbuf contains an annotation.  */
  int annotated;
Karl Heuer's avatar
Karl Heuer committed
443

Kenichi Handa's avatar
Kenichi Handa committed
444 445
  unsigned char carryover[64];
  int carryover_bytes;
446

Kenichi Handa's avatar
Kenichi Handa committed
447 448
  int default_char;

449 450
  int (*detector) P_ ((struct coding_system *,
		       struct coding_detection_info *));
Kenichi Handa's avatar
Kenichi Handa committed
451 452 453 454 455 456 457 458 459 460
  void (*decoder) P_ ((struct coding_system *));
  int (*encoder) P_ ((struct coding_system *));
};

/* Meanings of bits in the member `common_flags' of the structure
   coding_system.  The lowest 8 bits are reserved for various kind of
   annotations (currently two of them are used).  */
#define CODING_ANNOTATION_MASK			0x00FF
#define CODING_ANNOTATE_COMPOSITION_MASK	0x0001
#define CODING_ANNOTATE_DIRECTION_MASK		0x0002
461
#define CODING_ANNOTATE_CHARSET_MASK		0x0003
Kenichi Handa's avatar
Kenichi Handa committed
462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478
#define CODING_FOR_UNIBYTE_MASK			0x0100
#define CODING_REQUIRE_FLUSHING_MASK		0x0200
#define CODING_REQUIRE_DECODING_MASK		0x0400
#define CODING_REQUIRE_ENCODING_MASK		0x0800
#define CODING_REQUIRE_DETECTION_MASK		0x1000
#define CODING_RESET_AT_BOL_MASK		0x2000

/* Return 1 if the coding context CODING requires annotaion
   handling.  */
#define CODING_REQUIRE_ANNOTATION(coding) \
  ((coding)->common_flags & CODING_ANNOTATION_MASK)

/* Return 1 if the coding context CODING prefers decoding into unibyte.  */
#define CODING_FOR_UNIBYTE(coding) \
  ((coding)->common_flags & CODING_FOR_UNIBYTE_MASK)

/* Return 1 if the coding context CODING requires specific code to be
479 480 481 482
   attached at the tail of converted text.  */
#define CODING_REQUIRE_FLUSHING(coding) \
  ((coding)->common_flags & CODING_REQUIRE_FLUSHING_MASK)

Kenichi Handa's avatar
Kenichi Handa committed
483
/* Return 1 if the coding context CODING requires code conversion on
484 485
   decoding.  */
#define CODING_REQUIRE_DECODING(coding)	\
486 487
  ((coding)->dst_multibyte		\
   || (coding)->common_flags & CODING_REQUIRE_DECODING_MASK)
488

Kenichi Handa's avatar
Kenichi Handa committed
489 490

/* Return 1 if the coding context CODING requires code conversion on
491 492 493 494
   encoding.
   The non-multibyte part of the condition is to support encoding of
   unibyte strings/buffers generated by string-as-unibyte or
   (set-buffer-multibyte nil) from multibyte strings/buffers.  */
Kenichi Handa's avatar
Kenichi Handa committed
495 496 497 498 499
#define CODING_REQUIRE_ENCODING(coding)				\
  ((coding)->src_multibyte					\
   || (coding)->common_flags & CODING_REQUIRE_ENCODING_MASK	\
   || (coding)->mode & CODING_MODE_SELECTIVE_DISPLAY)

500

Kenichi Handa's avatar
Kenichi Handa committed
501
/* Return 1 if the coding context CODING requires some kind of code
502 503 504 505
   detection.  */
#define CODING_REQUIRE_DETECTION(coding) \
  ((coding)->common_flags & CODING_REQUIRE_DETECTION_MASK)

Kenichi Handa's avatar
Kenichi Handa committed
506
/* Return 1 if the coding context CODING requires code conversion on
507
   decoding or some kind of code detection.  */
508
#define CODING_MAY_REQUIRE_DECODING(coding)	\
509 510
  (CODING_REQUIRE_DECODING (coding)		\
   || CODING_REQUIRE_DETECTION (coding))
Karl Heuer's avatar
Karl Heuer committed
511 512 513 514 515 516

/* Macros to decode or encode a character of JISX0208 in SJIS.  S1 and
   S2 are the 1st and 2nd position-codes of JISX0208 in SJIS coding
   system.  C1 and C2 are the 1st and 2nd position codes of Emacs'
   internal format.  */

Kenichi Handa's avatar
Kenichi Handa committed
517 518 519 520 521 522 523 524 525 526 527 528 529
#define SJIS_TO_JIS(code)				\
  do {							\
    int s1, s2, j1, j2;					\
							\
    s1 = (code) >> 8, s2 = (code) & 0xFF;		\
							\
    if (s2 >= 0x9F)					\
      (j1 = s1 * 2 - (s1 >= 0xE0 ? 0x160 : 0xE0),	\
       j2 = s2 - 0x7E);					\
    else						\
      (j1 = s1 * 2 - ((s1 >= 0xE0) ? 0x161 : 0xE1),	\
       j2 = s2 - ((s2 >= 0x7F) ? 0x20 : 0x1F));		\
    (code) = (j1 << 8) | j2;				\
Karl Heuer's avatar
Karl Heuer committed
530 531
  } while (0)

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
#define SJIS_TO_JIS2(code)				\
  do {							\
    int s1, s2, j1, j2;					\
							\
    s1 = (code) >> 8, s2 = (code) & 0xFF;		\
							\
    if (s2 >= 0x9F)					\
      {							\
	j1 = (s1 == 0xF0 ? 0x28				\
	      : s1 == 0xF1 ? 0x24			\
	      : s1 == 0xF2 ? 0x2C			\
	      : s1 == 0xF3 ? 0x2E			\
	      : 0x6E + (s1 - 0xF4) * 2);		\
	j2 = s2 - 0x7E;					\
      }							\
    else						\
      {							\
	j1 = (s1 <= 0xF2 ? 0x21 + (s1 - 0xF0) * 2	\
	      : s1 <= 0xF4 ? 0x2D + (s1 - 0xF3) * 2	\
	      : 0x6F + (s1 - 0xF5) * 2);		\
	j2 = s2 - ((s2 >= 0x7F ? 0x20 : 0x1F));		\
      }							\
    (code) = (j1 << 8) | j2;				\
  } while (0)

Kenichi Handa's avatar
Kenichi Handa committed
557 558

#define JIS_TO_SJIS(code)				\
Karl Heuer's avatar
Karl Heuer committed
559
  do {							\
Kenichi Handa's avatar
Kenichi Handa committed
560 561 562 563 564 565
    int s1, s2, j1, j2;					\
							\
    j1 = (code) >> 8, j2 = (code) & 0xFF;		\
    if (j1 & 1)						\
      (s1 = j1 / 2 + ((j1 < 0x5F) ? 0x71 : 0xB1),	\
       s2 = j2 + ((j2 >= 0x60) ? 0x20 : 0x1F));		\
Karl Heuer's avatar
Karl Heuer committed
566
    else						\
Kenichi Handa's avatar
Kenichi Handa committed
567 568
      (s1 = j1 / 2 + ((j1 < 0x5F) ? 0x70 : 0xB0),	\
       s2 = j2 + 0x7E);					\
569
    (code) = (s1 << 8) | s2;				\
Karl Heuer's avatar
Karl Heuer committed
570 571
  } while (0)

572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594
#define JIS_TO_SJIS2(code)				\
  do {							\
    int s1, s2, j1, j2;					\
							\
    j1 = (code) >> 8, j2 = (code) & 0xFF;		\
    if (j1 & 1)						\
      {							\
	s1 = (j1 <= 0x25 ? 0xF0 + (j1 - 0x21) / 2	\
	      : j1 <= 0x27 ? 0xF3 + (j1 - 0x2D) / 2	\
	      : 0xF5 + (j1 - 0x6F) / 2);		\
	s2 = j2 + ((j2 >= 0x60) ? 0x20 : 0x1F);		\
      }							\
    else						\
      {							\
	s1 = (j1 == 0x28 ? 0xF0				\
	      : j1 == 0x24 ? 0xF1			\
	      : j1 == 0x2C ? 0xF2			\
	      : j1 == 0x2E ? 0xF3			\
	      : 0xF4 + (j1 - 0x6E) / 2);		\
	s2 = j2 + 0x7E;					\
      }							\
    (code) = (s1 << 8) | s2;				\
  } while (0)
Kenichi Handa's avatar
Kenichi Handa committed
595

596 597 598 599
/* Encode the file name NAME using the specified coding system
   for file names, if any.  */
#define ENCODE_FILE(name)						   \
  (! NILP (Vfile_name_coding_system)					   \
600
   && !EQ (Vfile_name_coding_system, make_number (0))			   \
Karl Heuer's avatar
Karl Heuer committed
601
   ? code_convert_string_norecord (name, Vfile_name_coding_system, 1)	   \
602
   : (! NILP (Vdefault_file_name_coding_system)				   \
603
      && !EQ (Vdefault_file_name_coding_system, make_number (0))	   \
Karl Heuer's avatar
Karl Heuer committed
604
      ? code_convert_string_norecord (name, Vdefault_file_name_coding_system, 1) \
605 606
      : name))

Kenichi Handa's avatar
Kenichi Handa committed
607

608 609 610 611
/* Decode the file name NAME using the specified coding system
   for file names, if any.  */
#define DECODE_FILE(name)						   \
  (! NILP (Vfile_name_coding_system)					   \
612
   && !EQ (Vfile_name_coding_system, make_number (0))			   \
Karl Heuer's avatar
Karl Heuer committed
613
   ? code_convert_string_norecord (name, Vfile_name_coding_system, 0)	   \
614
   : (! NILP (Vdefault_file_name_coding_system)				   \
615
      && !EQ (Vdefault_file_name_coding_system, make_number (0))	   \
Karl Heuer's avatar
Karl Heuer committed
616
      ? code_convert_string_norecord (name, Vdefault_file_name_coding_system, 0) \
617 618
      : name))

Kenichi Handa's avatar
Kenichi Handa committed
619

620
/* Encode the string STR using the specified coding system
621
   for system functions, if any.  */
622
#define ENCODE_SYSTEM(str)						   \
623
  (! NILP (Vlocale_coding_system)					   \
624
   && !EQ (Vlocale_coding_system, make_number (0))			   \
625
   ? code_convert_string_norecord (str, Vlocale_coding_system, 1)	   \
626 627 628
   : str)

/* Decode the string STR using the specified coding system
629
   for system functions, if any.  */
630
#define DECODE_SYSTEM(str)						   \
631
  (! NILP (Vlocale_coding_system)					   \
632
   && !EQ (Vlocale_coding_system, make_number (0))			   \
633
   ? code_convert_string_norecord (str, Vlocale_coding_system, 0)	   \
634
   : str)
635

Dave Love's avatar
Dave Love committed
636 637 638 639
/* Used by the gtk menu code.  Note that this encodes utf-8, not
   utf-8-emacs, so it's not a no-op.  */
#define ENCODE_UTF_8(str) code_convert_string_norecord (str, Qutf_8, 1)

Karl Heuer's avatar
Karl Heuer committed
640
/* Extern declarations.  */
641
extern Lisp_Object code_conversion_save P_ ((int, int));
642 643
extern int decoding_buffer_size P_ ((struct coding_system *, int));
extern int encoding_buffer_size P_ ((struct coding_system *, int));
Kenichi Handa's avatar
Kenichi Handa committed
644
extern void setup_coding_system P_ ((Lisp_Object, struct coding_system *));
645
extern Lisp_Object coding_charset_list P_ ((struct coding_system *));
Kenichi Handa's avatar
Kenichi Handa committed
646
extern void detect_coding P_ ((struct coding_system *));
647
extern Lisp_Object code_convert_region P_ ((Lisp_Object, Lisp_Object,
Kenichi Handa's avatar
Kenichi Handa committed
648 649 650 651
					    Lisp_Object, Lisp_Object,
					    int, int));
extern Lisp_Object code_convert_string P_ ((Lisp_Object, Lisp_Object,
					    Lisp_Object, int, int, int));
652 653
extern Lisp_Object code_convert_string_norecord P_ ((Lisp_Object, Lisp_Object,
						     int));
Kenichi Handa's avatar
Kenichi Handa committed
654 655 656 657 658 659 660 661 662 663 664 665 666 667
extern Lisp_Object raw_text_coding_system P_ ((Lisp_Object));
extern Lisp_Object coding_inherit_eol_type P_ ((Lisp_Object, Lisp_Object));

extern int decode_coding_gap P_ ((struct coding_system *,
				  EMACS_INT, EMACS_INT));
extern int encode_coding_gap P_ ((struct coding_system *,
				  EMACS_INT, EMACS_INT));
extern void decode_coding_object P_ ((struct coding_system *,
				      Lisp_Object, EMACS_INT, EMACS_INT,
				      EMACS_INT, EMACS_INT, Lisp_Object));
extern void encode_coding_object P_ ((struct coding_system *,
				      Lisp_Object, EMACS_INT, EMACS_INT,
				      EMACS_INT, EMACS_INT, Lisp_Object));

Kenichi Handa's avatar
Kenichi Handa committed
668 669
/* Macros for backward compatibility.  */

Kenichi Handa's avatar
Kenichi Handa committed
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
#define decode_coding_region(coding, from, to)		\
  decode_coding_object (coding, Fcurrent_buffer (),	\
			from, CHAR_TO_BYTE (from),	\
			to, CHAR_TO_BYTE (to), Fcurrent_buffer ())


#define encode_coding_region(coding, from, to)		\
  encode_coding_object (coding, Fcurrent_buffer (),	\
			from, CHAR_TO_BYTE (from),	\
			to, CHAR_TO_BYTE (to), Fcurrent_buffer ())


#define decode_coding_string(coding, string, nocopy)			\
  decode_coding_object (coding, string, 0, 0, XSTRING (string)->size,	\
			STRING_BYTES (XSTRING (string)), Qt)

#define encode_coding_string(coding, string, nocopy)			\
  (encode_coding_object (coding, string, 0, 0, XSTRING (string)->size,	\
			 STRING_BYTES (XSTRING (string)), Qt),		\
   (coding)->dst_object)


#define decode_coding_c_string(coding, src, bytes, dst_object)		\
  do {									\
    (coding)->source = (src);						\
    (coding)->src_chars = (coding)->src_bytes = (bytes);		\
    decode_coding_object ((coding), Qnil, 0, 0, (bytes), (bytes),	\
			  (dst_object));				\
  } while (0)


extern Lisp_Object preferred_coding_system P_ (());


Kenichi Handa's avatar
Kenichi Handa committed
704 705
extern Lisp_Object Qutf_8, Qutf_8_emacs;

Karl Heuer's avatar
Karl Heuer committed
706
extern Lisp_Object Qcoding_system, Qeol_type, Qcoding_category_index;
Kenichi Handa's avatar
Kenichi Handa committed
707 708 709
extern Lisp_Object Qcoding_system_p;
extern Lisp_Object Qraw_text, Qemacs_mule, Qno_conversion, Qundecided;
extern Lisp_Object Qiso_2022;
Karl Heuer's avatar
Karl Heuer committed
710
extern Lisp_Object Qbuffer_file_coding_system;
Kenichi Handa's avatar
Kenichi Handa committed
711 712

extern Lisp_Object Qunix, Qdos, Qmac;
Karl Heuer's avatar
Karl Heuer committed
713

714 715
extern Lisp_Object Qtranslation_table;
extern Lisp_Object Qtranslation_table_id;
716

717 718 719 720
/* Mnemonic strings to indicate each type of end-of-line.  */
extern Lisp_Object eol_mnemonic_unix, eol_mnemonic_dos, eol_mnemonic_mac;
/* Mnemonic string to indicate type of end-of-line is not yet decided.  */
extern Lisp_Object eol_mnemonic_undecided;
Karl Heuer's avatar
Karl Heuer committed
721 722 723

#ifdef emacs
extern Lisp_Object Qfile_coding_system;
Kim F. Storm's avatar
Kim F. Storm committed
724
extern Lisp_Object Qcall_process, Qcall_process_region;
Karl Heuer's avatar
Karl Heuer committed
725
extern Lisp_Object Qstart_process, Qopen_network_stream;
726
extern Lisp_Object Qwrite_region;
Karl Heuer's avatar
Karl Heuer committed
727

728 729
extern char *emacs_strerror P_ ((int));

Karl Heuer's avatar
Karl Heuer committed
730 731 732 733 734 735
/* Coding-system for reading files and receiving data from process.  */
extern Lisp_Object Vcoding_system_for_read;
/* Coding-system for writing files and sending data to process.  */
extern Lisp_Object Vcoding_system_for_write;
/* Coding-system actually used in the latest I/O.  */
extern Lisp_Object Vlast_coding_system_used;
736 737
/* Coding-system to use with system messages (e.g. strerror).  */
extern Lisp_Object Vlocale_coding_system;
Karl Heuer's avatar
Karl Heuer committed
738

739 740 741 742
/* If non-zero, process buffer inherits the coding system used to decode
   the subprocess output.  */
extern int inherit_process_coding_system;

743 744 745 746
/* Coding system to be used to encode text for terminal display when
   terminal coding system is nil.  */
extern struct coding_system safe_terminal_coding;

747 748
/* Default coding systems used for process I/O.  */
extern Lisp_Object Vdefault_process_coding_system;
Karl Heuer's avatar
Karl Heuer committed
749

Kenichi Handa's avatar
Kenichi Handa committed
750
/* Function to call to force a user to force select a propert coding
751 752 753
   system.  */
extern Lisp_Object Vselect_safe_coding_system_function;

754 755 756 757
/* If nonzero, on writing a file, Vselect_safe_coding_system_function
   is called even if Vcoding_system_for_write is non-nil.  */
extern int coding_system_require_warning;

758 759 760 761 762 763
/* Coding system for file names, or nil if none.  */
extern Lisp_Object Vfile_name_coding_system;

/* Coding system for file names used only when
   Vfile_name_coding_system is nil.  */
extern Lisp_Object Vdefault_file_name_coding_system;
764

Karl Heuer's avatar
Karl Heuer committed
765 766
#endif

767 768 769
/* Error signaled when there's a problem with detecting coding system */
extern Lisp_Object Qcoding_system_error;

Kenichi Handa's avatar
Kenichi Handa committed
770 771 772
extern char emacs_mule_bytes[256];
extern int emacs_mule_string_char P_ ((unsigned char *));

773
#endif /* EMACS_CODING_H */
Kenichi Handa's avatar
Kenichi Handa committed
774 775 776

/* arch-tag: 2bc3b4fa-6870-4f64-8135-b962b2d290e4
   (do not change this comment) */