coding.h 26.7 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, 2009, 2010  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, 2009, 2010
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
  };

59 60 61 62 63 64
enum define_coding_utf8_arg_index
  {
    coding_arg_utf8_bom = coding_arg_max,
    coding_arg_utf8_max
  };

Kenichi Handa's avatar
Kenichi Handa committed
65
enum define_coding_utf16_arg_index
Karl Heuer's avatar
Karl Heuer committed
66
  {
Kenichi Handa's avatar
Kenichi Handa committed
67 68 69
    coding_arg_utf16_bom = coding_arg_max,
    coding_arg_utf16_endian,
    coding_arg_utf16_max
Karl Heuer's avatar
Karl Heuer committed
70 71
  };

Kenichi Handa's avatar
Kenichi Handa committed
72 73
enum define_coding_ccl_arg_index
  {
74
    coding_arg_ccl_decoder = coding_arg_max,
Kenichi Handa's avatar
Kenichi Handa committed
75 76 77 78
    coding_arg_ccl_encoder,
    coding_arg_ccl_valids,
    coding_arg_ccl_max
  };
Karl Heuer's avatar
Karl Heuer committed
79

Kenichi Handa's avatar
Kenichi Handa committed
80 81 82 83 84 85 86
/* 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
87
extern Lisp_Object Vcoding_system_hash_table;
Karl Heuer's avatar
Karl Heuer committed
88

Kenichi Handa's avatar
Kenichi Handa committed
89

Kenichi Handa's avatar
Kenichi Handa committed
90
/* Enumeration of coding system type.  */
Karl Heuer's avatar
Karl Heuer committed
91

Kenichi Handa's avatar
Kenichi Handa committed
92 93 94 95 96 97 98 99 100 101 102 103 104
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
105 106


Kenichi Handa's avatar
Kenichi Handa committed
107
/* Enumeration of end-of-line format type.  */
Karl Heuer's avatar
Karl Heuer committed
108

Kenichi Handa's avatar
Kenichi Handa committed
109 110 111 112 113 114 115 116 117 118 119 120 121
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
122

Kenichi Handa's avatar
Kenichi Handa committed
123
/* Enumeration of index to an attribute vector of a coding system.  */
Karl Heuer's avatar
Karl Heuer committed
124

Kenichi Handa's avatar
Kenichi Handa committed
125 126 127 128 129 130 131 132 133 134
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,
135
    coding_attr_trans_tbl,
Kenichi Handa's avatar
Kenichi Handa committed
136 137 138
    coding_attr_post_read,
    coding_attr_pre_write,
    coding_attr_default_char,
Kenichi Handa's avatar
Kenichi Handa committed
139
    coding_attr_for_unibyte,
Kenichi Handa's avatar
Kenichi Handa committed
140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
    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,

157
    coding_attr_utf_bom,
Kenichi Handa's avatar
Kenichi Handa committed
158 159 160 161 162 163
    coding_attr_utf_16_endian,

    coding_attr_emacs_mule_full,

    coding_attr_last_index
  };
Karl Heuer's avatar
Karl Heuer committed
164 165


Kenichi Handa's avatar
Kenichi Handa committed
166 167
/* Macros to access an element of an attribute vector.  */

Juanma Barranquero's avatar
Juanma Barranquero committed
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184
#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)
185 186


Kenichi Handa's avatar
Kenichi Handa committed
187 188 189 190 191 192
/* 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
193 194
#define CODING_ID_ATTRS(id)	\
  (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 0))
195

Kenichi Handa's avatar
Kenichi Handa committed
196 197
/* Return the list of aliases of a coding system specified by ID.  */

Kenichi Handa's avatar
Kenichi Handa committed
198 199
#define CODING_ID_ALIASES(id)	\
  (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 1))
200

Kenichi Handa's avatar
Kenichi Handa committed
201 202
/* Return the eol-type of a coding system specified by ID.  */

Kenichi Handa's avatar
Kenichi Handa committed
203 204
#define CODING_ID_EOL_TYPE(id)	\
  (AREF (HASH_VALUE (XHASH_TABLE (Vcoding_system_hash_table), id), 2))
205

Kenichi Handa's avatar
Kenichi Handa committed
206 207

/* Return the spec vector of CODING_SYSTEM_SYMBOL.  */
208

Kenichi Handa's avatar
Kenichi Handa committed
209 210
#define CODING_SYSTEM_SPEC(coding_system_symbol)	\
  (Fgethash (coding_system_symbol, Vcoding_system_hash_table, Qnil))
Karl Heuer's avatar
Karl Heuer committed
211

Kenichi Handa's avatar
Kenichi Handa committed
212 213 214

/* Return the ID of CODING_SYSTEM_SYMBOL.  */

Kenichi Handa's avatar
Kenichi Handa committed
215 216 217
#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
218

Miles Bader's avatar
Miles Bader committed
219
/* Return 1 if CODING_SYSTEM_SYMBOL is a coding system.  */
Kenichi Handa's avatar
Kenichi Handa committed
220

221 222 223 224
#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
225

Kenichi Handa's avatar
Kenichi Handa committed
226 227
/* Check if X is a coding system or not.  */

Kenichi Handa's avatar
Kenichi Handa committed
228
#define CHECK_CODING_SYSTEM(x)				\
Kenichi Handa's avatar
Kenichi Handa committed
229
  do {							\
230 231
    if (CODING_SYSTEM_ID (x) < 0			\
	&& NILP (Fcheck_coding_system (x)))		\
Kenichi Handa's avatar
Kenichi Handa committed
232
      wrong_type_argument (Qcoding_system_p, (x));	\
Kenichi Handa's avatar
Kenichi Handa committed
233
  } while (0)
234

Karl Heuer's avatar
Karl Heuer committed
235

Kenichi Handa's avatar
Kenichi Handa committed
236 237 238
/* 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
239 240 241
#define CHECK_CODING_SYSTEM_GET_SPEC(x, spec)		\
  do {							\
    spec = CODING_SYSTEM_SPEC (x);			\
242 243 244 245 246
    if (NILP (spec))					\
      {							\
	Fcheck_coding_system (x);			\
	spec = CODING_SYSTEM_SPEC (x);			\
      }							\
Kenichi Handa's avatar
Kenichi Handa committed
247
    if (NILP (spec))					\
248
      wrong_type_argument (Qcoding_system_p, (x));	\
Kenichi Handa's avatar
Kenichi Handa committed
249
  } while (0)
250

251

Kenichi Handa's avatar
Kenichi Handa committed
252 253 254
/* 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
255 256 257 258
#define CHECK_CODING_SYSTEM_GET_ID(x, id)			\
  do								\
    {								\
      id = CODING_SYSTEM_ID (x);				\
259 260 261 262 263
      if (id < 0)						\
	{							\
	  Fcheck_coding_system (x);				\
	  id = CODING_SYSTEM_ID (x);				\
	}							\
Kenichi Handa's avatar
Kenichi Handa committed
264
      if (id < 0)						\
265
	wrong_type_argument (Qcoding_system_p, (x));	\
Kenichi Handa's avatar
Kenichi Handa committed
266
    } while (0)
Karl Heuer's avatar
Karl Heuer committed
267 268 269 270


/*** GENERAL section ***/

Kenichi Handa's avatar
Kenichi Handa committed
271 272
/* Enumeration of result code of code conversion.  */
enum coding_result_code
Karl Heuer's avatar
Karl Heuer committed
273
  {
Kenichi Handa's avatar
Kenichi Handa committed
274 275 276 277
    CODING_RESULT_SUCCESS,
    CODING_RESULT_INSUFFICIENT_SRC,
    CODING_RESULT_INSUFFICIENT_DST,
    CODING_RESULT_INCONSISTENT_EOL,
278
    CODING_RESULT_INVALID_SRC,
Kenichi Handa's avatar
Kenichi Handa committed
279 280
    CODING_RESULT_INTERRUPT,
    CODING_RESULT_INSUFFICIENT_MEM
Karl Heuer's avatar
Karl Heuer committed
281 282
  };

283

Kenichi Handa's avatar
Kenichi Handa committed
284
/* Macros used for the member `mode' of the struct coding_system.  */
285 286 287 288 289 290

/* 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
291
   the last block of the whole text to be converted, and do the
Dave Love's avatar
Dave Love committed
292
   appropriate finishing job.  */
293 294 295 296 297 298 299 300 301 302
#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
303 304
#define CODING_MODE_FIXED_DESTINATION		0x10

Kenichi Handa's avatar
Kenichi Handa committed
305 306
/* If set, it means that the encoding routines produces some safe
   ASCII characters (usually '?') for unsupported characters.  */
Kenichi Handa's avatar
Kenichi Handa committed
307 308
#define CODING_MODE_SAFE_ENCODING		0x20

Kenichi Handa's avatar
Kenichi Handa committed
309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
  /* For handling composition sequence.  */
#include "composite.h"

enum composition_state
  {
    COMPOSING_NO,
    COMPOSING_CHAR,
    COMPOSING_RULE,
    COMPOSING_COMPONENT_CHAR,
    COMPOSING_COMPONENT_RULE
  };

/* Structure for the current composition status.  */
struct composition_status
{
  enum composition_state state;
  enum composition_method method;
  int old_form;		  /* 0:pre-21 form, 1:post-21 form */
  int length;		  /* number of elements produced in charbuf */
  int nchars;		  /* number of characters composed */
  int ncomps;		  /* number of composition components */
  /* Maximum carryover is for the case of COMPOSITION_WITH_RULE_ALTCHARS.
     See the comment in coding.c.  */
  int carryover[4 		/* annotation header */
		+ MAX_COMPOSITION_COMPONENTS * 3 - 2 /* ALTs and RULEs */
		+ 2				     /* intermediate -1 -1 */
		+ MAX_COMPOSITION_COMPONENTS	     /* CHARs */
		];
};


Kenichi Handa's avatar
Kenichi Handa committed
340 341 342
/* Structure of the field `spec.iso_2022' in the structure
   `coding_system'.  */
struct iso_2022_spec
Karl Heuer's avatar
Karl Heuer committed
343
{
Kenichi Handa's avatar
Kenichi Handa committed
344
  /* Bit-wise-or of CODING_ISO_FLAG_XXX.  */
Kenichi Handa's avatar
Kenichi Handa committed
345
  unsigned flags;
Karl Heuer's avatar
Karl Heuer committed
346

Kenichi Handa's avatar
Kenichi Handa committed
347 348
  /* The current graphic register invoked to each graphic plane.  */
  int current_invocation[2];
349

Kenichi Handa's avatar
Kenichi Handa committed
350 351 352 353
  /* 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
354

Kenichi Handa's avatar
Kenichi Handa committed
355 356 357
  /* 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
358

Kenichi Handa's avatar
Kenichi Handa committed
359 360
  /* Set to 1 temporarily only when processing at beginning of line.  */
  int bol;
Kenichi Handa's avatar
Kenichi Handa committed
361 362 363 364 365 366 367 368 369 370 371 372 373 374 375

  /* If positive, we are now scanning CTEXT extended segment.  */
  int ctext_extended_segment_len;

  /* If nonzero, we are now scanning embedded UTF-8 sequence.  */
  int embedded_utf_8;

  /* The current composition.  */
  struct composition_status cmp_status;
};

struct emacs_mule_spec
{
  int full_support;
  struct composition_status cmp_status;
Kenichi Handa's avatar
Kenichi Handa committed
376
};
Karl Heuer's avatar
Karl Heuer committed
377

Kenichi Handa's avatar
Kenichi Handa committed
378
struct ccl_spec;
Karl Heuer's avatar
Karl Heuer committed
379

380
enum utf_bom_type
Kenichi Handa's avatar
Kenichi Handa committed
381
  {
382 383 384
    utf_detect_bom,
    utf_without_bom,
    utf_with_bom
Kenichi Handa's avatar
Kenichi Handa committed
385
  };
386

Kenichi Handa's avatar
Kenichi Handa committed
387 388 389 390 391
enum utf_16_endian_type
  {
    utf_16_big_endian,
    utf_16_little_endian
  };
392

Kenichi Handa's avatar
Kenichi Handa committed
393 394
struct utf_16_spec
{
395
  enum utf_bom_type bom;
Kenichi Handa's avatar
Kenichi Handa committed
396 397 398
  enum utf_16_endian_type endian;
  int surrogate;
};
399

400 401 402 403 404 405 406 407 408 409
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;
};
410 411


Kenichi Handa's avatar
Kenichi Handa committed
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427
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;
428

Karl Heuer's avatar
Karl Heuer committed
429
  /* Detailed information specific to each type of coding system.  */
Kenichi Handa's avatar
Kenichi Handa committed
430
  union
Karl Heuer's avatar
Karl Heuer committed
431
    {
Kenichi Handa's avatar
Kenichi Handa committed
432 433 434
      struct iso_2022_spec iso_2022;
      struct ccl_spec *ccl;	/* Defined in ccl.h.  */
      struct utf_16_spec utf_16;
435
      enum utf_bom_type utf_8_bom;
Kenichi Handa's avatar
Kenichi Handa committed
436
      struct emacs_mule_spec emacs_mule;
Karl Heuer's avatar
Karl Heuer committed
437 438
    } spec;

Kenichi Handa's avatar
Kenichi Handa committed
439
  int max_charset_id;
440
  unsigned char *safe_charsets;
441

Kenichi Handa's avatar
Kenichi Handa committed
442 443 444 445
  /* 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).  */
446 447 448
  unsigned src_multibyte : 1;
  unsigned dst_multibyte : 1;

449 450 451 452
  /* 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
453
  int head_ascii;
454 455

  /* The following members are set by encoding/decoding routine.  */
Kenichi Handa's avatar
Kenichi Handa committed
456
  EMACS_INT produced, produced_char, consumed, consumed_char;
457

458 459 460
  /* Number of error source data found in a decoding routine.  */
  int errors;

Kenichi Handa's avatar
Kenichi Handa committed
461 462
  /* Store the positions of error source data. */
  EMACS_INT *error_positions;
463

Kenichi Handa's avatar
Kenichi Handa committed
464 465
  /* Finish status of code conversion.  */
  enum coding_result_code result;
466

Kenichi Handa's avatar
Kenichi Handa committed
467 468
  EMACS_INT src_pos, src_pos_byte, src_chars, src_bytes;
  Lisp_Object src_object;
Kenichi Handa's avatar
Kenichi Handa committed
469
  const unsigned char *source;
Karl Heuer's avatar
Karl Heuer committed
470

Kenichi Handa's avatar
Kenichi Handa committed
471 472 473
  EMACS_INT dst_pos, dst_pos_byte, dst_bytes;
  Lisp_Object dst_object;
  unsigned char *destination;
Karl Heuer's avatar
Karl Heuer committed
474

Miles Bader's avatar
Miles Bader committed
475
  /* Set to 1 if the source of conversion is not in the member
Kenichi Handa's avatar
Kenichi Handa committed
476
     `charbuf', but at `src_object'.  */
Kenichi Handa's avatar
Kenichi Handa committed
477
  int chars_at_source;
Karl Heuer's avatar
Karl Heuer committed
478

Kenichi Handa's avatar
Kenichi Handa committed
479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495
  /* 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
496

Kenichi Handa's avatar
Kenichi Handa committed
497 498
  unsigned char carryover[64];
  int carryover_bytes;
499

Kenichi Handa's avatar
Kenichi Handa committed
500 501
  int default_char;

Jan D's avatar
Jan D committed
502 503 504 505
  int (*detector) (struct coding_system *,
                   struct coding_detection_info *);
  void (*decoder) (struct coding_system *);
  int (*encoder) (struct coding_system *);
Kenichi Handa's avatar
Kenichi Handa committed
506 507 508 509 510 511 512 513
};

/* 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
514
#define CODING_ANNOTATE_CHARSET_MASK		0x0003
Kenichi Handa's avatar
Kenichi Handa committed
515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531
#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
532 533 534 535
   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
536
/* Return 1 if the coding context CODING requires code conversion on
537 538
   decoding.  */
#define CODING_REQUIRE_DECODING(coding)	\
539 540
  ((coding)->dst_multibyte		\
   || (coding)->common_flags & CODING_REQUIRE_DECODING_MASK)
541

Kenichi Handa's avatar
Kenichi Handa committed
542 543

/* Return 1 if the coding context CODING requires code conversion on
544 545 546 547
   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
548 549 550 551 552
#define CODING_REQUIRE_ENCODING(coding)				\
  ((coding)->src_multibyte					\
   || (coding)->common_flags & CODING_REQUIRE_ENCODING_MASK	\
   || (coding)->mode & CODING_MODE_SELECTIVE_DISPLAY)

553

Kenichi Handa's avatar
Kenichi Handa committed
554
/* Return 1 if the coding context CODING requires some kind of code
555 556 557 558
   detection.  */
#define CODING_REQUIRE_DETECTION(coding) \
  ((coding)->common_flags & CODING_REQUIRE_DETECTION_MASK)

Kenichi Handa's avatar
Kenichi Handa committed
559
/* Return 1 if the coding context CODING requires code conversion on
560
   decoding or some kind of code detection.  */
561
#define CODING_MAY_REQUIRE_DECODING(coding)	\
562 563
  (CODING_REQUIRE_DECODING (coding)		\
   || CODING_REQUIRE_DETECTION (coding))
Karl Heuer's avatar
Karl Heuer committed
564 565 566 567 568 569

/* 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
570 571 572 573 574 575 576 577 578 579 580 581 582
#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
583 584
  } while (0)

585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609
#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
610 611

#define JIS_TO_SJIS(code)				\
Karl Heuer's avatar
Karl Heuer committed
612
  do {							\
Kenichi Handa's avatar
Kenichi Handa committed
613 614 615 616 617 618
    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
619
    else						\
Kenichi Handa's avatar
Kenichi Handa committed
620 621
      (s1 = j1 / 2 + ((j1 < 0x5F) ? 0x70 : 0xB0),	\
       s2 = j2 + 0x7E);					\
622
    (code) = (s1 << 8) | s2;				\
Karl Heuer's avatar
Karl Heuer committed
623 624
  } while (0)

625 626 627 628 629 630 631 632
#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	\
633
	      : j1 <= 0x2F ? 0xF3 + (j1 - 0x2D) / 2	\
634 635 636 637 638 639 640 641 642 643 644 645 646 647
	      : 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
648

649 650 651 652
/* Encode the file name NAME using the specified coding system
   for file names, if any.  */
#define ENCODE_FILE(name)						   \
  (! NILP (Vfile_name_coding_system)					   \
653
   && !EQ (Vfile_name_coding_system, make_number (0))			   \
Karl Heuer's avatar
Karl Heuer committed
654
   ? code_convert_string_norecord (name, Vfile_name_coding_system, 1)	   \
655
   : (! NILP (Vdefault_file_name_coding_system)				   \
656
      && !EQ (Vdefault_file_name_coding_system, make_number (0))	   \
Karl Heuer's avatar
Karl Heuer committed
657
      ? code_convert_string_norecord (name, Vdefault_file_name_coding_system, 1) \
658 659
      : name))

Kenichi Handa's avatar
Kenichi Handa committed
660

661 662 663 664
/* Decode the file name NAME using the specified coding system
   for file names, if any.  */
#define DECODE_FILE(name)						   \
  (! NILP (Vfile_name_coding_system)					   \
665
   && !EQ (Vfile_name_coding_system, make_number (0))			   \
Karl Heuer's avatar
Karl Heuer committed
666
   ? code_convert_string_norecord (name, Vfile_name_coding_system, 0)	   \
667
   : (! NILP (Vdefault_file_name_coding_system)				   \
668
      && !EQ (Vdefault_file_name_coding_system, make_number (0))	   \
Karl Heuer's avatar
Karl Heuer committed
669
      ? code_convert_string_norecord (name, Vdefault_file_name_coding_system, 0) \
670 671
      : name))

Kenichi Handa's avatar
Kenichi Handa committed
672

673
/* Encode the string STR using the specified coding system
674
   for system functions, if any.  */
675
#define ENCODE_SYSTEM(str)						   \
676
  (! NILP (Vlocale_coding_system)					   \
677
   && !EQ (Vlocale_coding_system, make_number (0))			   \
678
   ? code_convert_string_norecord (str, Vlocale_coding_system, 1)	   \
679 680 681
   : str)

/* Decode the string STR using the specified coding system
682
   for system functions, if any.  */
683
#define DECODE_SYSTEM(str)						   \
684
  (! NILP (Vlocale_coding_system)					   \
685
   && !EQ (Vlocale_coding_system, make_number (0))			   \
686
   ? code_convert_string_norecord (str, Vlocale_coding_system, 0)	   \
687
   : str)
688

Dave Love's avatar
Dave Love committed
689 690 691 692
/* 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
693
/* Extern declarations.  */
Jan D's avatar
Jan D committed
694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720
extern Lisp_Object code_conversion_save (int, int);
extern int decoding_buffer_size (struct coding_system *, int);
extern int encoding_buffer_size (struct coding_system *, int);
extern void setup_coding_system (Lisp_Object, struct coding_system *);
extern Lisp_Object coding_charset_list (struct coding_system *);
extern Lisp_Object coding_system_charset_list (Lisp_Object);
extern void detect_coding (struct coding_system *);
extern Lisp_Object code_convert_region (Lisp_Object, Lisp_Object,
                                        Lisp_Object, Lisp_Object,
                                        int, int);
extern Lisp_Object code_convert_string (Lisp_Object, Lisp_Object,
                                        Lisp_Object, int, int, int);
extern Lisp_Object code_convert_string_norecord (Lisp_Object, Lisp_Object,
                                                 int);
extern Lisp_Object raw_text_coding_system (Lisp_Object);
extern Lisp_Object coding_inherit_eol_type (Lisp_Object, Lisp_Object);

extern int decode_coding_gap (struct coding_system *,
                              EMACS_INT, EMACS_INT);
extern int encode_coding_gap (struct coding_system *,
                              EMACS_INT, EMACS_INT);
extern void decode_coding_object (struct coding_system *,
                                  Lisp_Object, EMACS_INT, EMACS_INT,
                                  EMACS_INT, EMACS_INT, Lisp_Object);
extern void encode_coding_object (struct coding_system *,
                                  Lisp_Object, EMACS_INT, EMACS_INT,
                                  EMACS_INT, EMACS_INT, Lisp_Object);
Kenichi Handa's avatar
Kenichi Handa committed
721

Kenichi Handa's avatar
Kenichi Handa committed
722 723
/* Macros for backward compatibility.  */

Kenichi Handa's avatar
Kenichi Handa committed
724 725 726 727 728 729 730 731 732 733 734 735 736
#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)			\
737
  decode_coding_object (coding, string, 0, 0, SCHARS (string),		\
738
			SBYTES (string), Qt)
Kenichi Handa's avatar
Kenichi Handa committed
739 740

#define encode_coding_string(coding, string, nocopy)			\
741 742 743 744
  (STRING_MULTIBYTE(string) ?						\
    (encode_coding_object (coding, string, 0, 0, SCHARS (string),	\
			   SBYTES (string), Qt),			\
     (coding)->dst_object) : (string))
Kenichi Handa's avatar
Kenichi Handa committed
745 746 747 748 749 750 751 752 753 754 755


#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)


Jan D's avatar
Jan D committed
756
extern Lisp_Object preferred_coding_system ();
Kenichi Handa's avatar
Kenichi Handa committed
757 758


Kenichi Handa's avatar
Kenichi Handa committed
759 760
extern Lisp_Object Qutf_8, Qutf_8_emacs;

Karl Heuer's avatar
Karl Heuer committed
761
extern Lisp_Object Qcoding_system, Qeol_type, Qcoding_category_index;
Kenichi Handa's avatar
Kenichi Handa committed
762 763 764
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
765
extern Lisp_Object Qbuffer_file_coding_system;
Kenichi Handa's avatar
Kenichi Handa committed
766 767

extern Lisp_Object Qunix, Qdos, Qmac;
Karl Heuer's avatar
Karl Heuer committed
768

769 770
extern Lisp_Object Qtranslation_table;
extern Lisp_Object Qtranslation_table_id;
771

772 773 774 775
/* 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
776 777 778

#ifdef emacs
extern Lisp_Object Qfile_coding_system;
Kim F. Storm's avatar
Kim F. Storm committed
779
extern Lisp_Object Qcall_process, Qcall_process_region;
Karl Heuer's avatar
Karl Heuer committed
780
extern Lisp_Object Qstart_process, Qopen_network_stream;
781
extern Lisp_Object Qwrite_region;
Karl Heuer's avatar
Karl Heuer committed
782

Jan D's avatar
Jan D committed
783
extern char *emacs_strerror (int);
784

Karl Heuer's avatar
Karl Heuer committed
785 786 787 788 789 790
/* 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;
791 792
/* Coding-system to use with system messages (e.g. strerror).  */
extern Lisp_Object Vlocale_coding_system;
Karl Heuer's avatar
Karl Heuer committed
793

794 795 796 797
/* If non-zero, process buffer inherits the coding system used to decode
   the subprocess output.  */
extern int inherit_process_coding_system;

798 799 800 801
/* Coding system to be used to encode text for terminal display when
   terminal coding system is nil.  */
extern struct coding_system safe_terminal_coding;

802 803
/* Default coding systems used for process I/O.  */
extern Lisp_Object Vdefault_process_coding_system;
Karl Heuer's avatar
Karl Heuer committed
804

Kenichi Handa's avatar
Kenichi Handa committed
805
/* Function to call to force a user to force select a propert coding
806 807 808
   system.  */
extern Lisp_Object Vselect_safe_coding_system_function;

809 810 811 812
/* 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;

813 814 815 816 817 818
/* 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;
819

Karl Heuer's avatar
Karl Heuer committed
820 821
#endif

822 823 824
/* Error signaled when there's a problem with detecting coding system */
extern Lisp_Object Qcoding_system_error;

Kenichi Handa's avatar
Kenichi Handa committed
825
extern char emacs_mule_bytes[256];
Jan D's avatar
Jan D committed
826
extern int emacs_mule_string_char (unsigned char *);
Kenichi Handa's avatar
Kenichi Handa committed
827

828
#endif /* EMACS_CODING_H */
Kenichi Handa's avatar
Kenichi Handa committed
829 830 831

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