1 /* Coding system handler (conversion, detection, etc).
2 Copyright (C) 2001-2015 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
8 National Institute of Advanced Industrial Science and Technology (AIST)
9 Registration Number H13PRO009
11 This file is part of GNU Emacs.
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.
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.
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/>. */
26 /*** TABLE OF CONTENTS ***
30 2. Emacs' internal format (emacs-utf-8) handlers
33 5. Charset-base coding systems handlers
34 6. emacs-mule (old Emacs' internal format) handlers
36 8. Shift-JIS and BIG5 handlers
38 10. C library functions
39 11. Emacs Lisp library functions
44 /*** 0. General comments ***
49 A coding system is an object for an encoding mechanism that contains
50 information about how to convert byte sequences to character
51 sequences and vice versa. When we say "decode", it means converting
52 a byte sequence of a specific coding system into a character
53 sequence that is represented by Emacs' internal coding system
54 `emacs-utf-8', and when we say "encode", it means converting a
55 character sequence of emacs-utf-8 to a byte sequence of a specific
58 In Emacs Lisp, a coding system is represented by a Lisp symbol. On
59 the C level, a coding system is represented by a vector of attributes
60 stored in the hash table Vcharset_hash_table. The conversion from
61 coding system symbol to attributes vector is done by looking up
62 Vcharset_hash_table by the symbol.
64 Coding systems are classified into the following types depending on
65 the encoding mechanism. Here's a brief description of the types.
71 o Charset-base coding system
73 A coding system defined by one or more (coded) character sets.
74 Decoding and encoding are done by a code converter defined for each
77 o Old Emacs internal format (emacs-mule)
79 The coding system adopted by old versions of Emacs (20 and 21).
81 o ISO2022-base coding system
83 The most famous coding system for multiple character sets. X's
84 Compound Text, various EUCs (Extended Unix Code), and coding systems
85 used in the Internet communication such as ISO-2022-JP are all
88 o SJIS (or Shift-JIS or MS-Kanji-Code)
90 A coding system to encode character sets: ASCII, JISX0201, and
91 JISX0208. Widely used for PC's in Japan. Details are described in
96 A coding system to encode character sets: ASCII and Big5. Widely
97 used for Chinese (mainly in Taiwan and Hong Kong). Details are
98 described in section 8. In this file, when we write "big5" (all
99 lowercase), we mean the coding system, and when we write "Big5"
100 (capitalized), we mean the character set.
104 If a user wants to decode/encode text encoded in a coding system
105 not listed above, he can supply a decoder and an encoder for it in
106 CCL (Code Conversion Language) programs. Emacs executes the CCL
107 program while decoding/encoding.
111 A coding system for text containing raw eight-bit data. Emacs
112 treats each byte of source text as a character (except for
113 end-of-line conversion).
117 Like raw text, but don't do end-of-line conversion.
122 How text end-of-line is encoded depends on operating system. For
123 instance, Unix's format is just one byte of LF (line-feed) code,
124 whereas DOS's format is two-byte sequence of `carriage-return' and
125 `line-feed' codes. MacOS's format is usually one byte of
128 Since text character encoding and end-of-line encoding are
129 independent, any coding system described above can take any format
130 of end-of-line (except for no-conversion).
134 Before using a coding system for code conversion (i.e. decoding and
135 encoding), we setup a structure of type `struct coding_system'.
136 This structure keeps various information about a specific code
137 conversion (e.g. the location of source and destination data).
144 /*** GENERAL NOTES on `detect_coding_XXX ()' functions ***
146 These functions check if a byte sequence specified as a source in
147 CODING conforms to the format of XXX, and update the members of
150 Return true if the byte sequence conforms to XXX.
152 Below is the template of these functions. */
156 detect_coding_XXX (struct coding_system
*coding
,
157 struct coding_detection_info
*detect_info
)
159 const unsigned char *src
= coding
->source
;
160 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
161 bool multibytep
= coding
->src_multibyte
;
162 ptrdiff_t consumed_chars
= 0;
168 /* Get one byte from the source. If the source is exhausted, jump
169 to no_more_source:. */
172 if (! __C_conforms_to_XXX___ (c
))
174 if (! __C_strongly_suggests_XXX__ (c
))
175 found
= CATEGORY_MASK_XXX
;
177 /* The byte sequence is invalid for XXX. */
178 detect_info
->rejected
|= CATEGORY_MASK_XXX
;
182 /* The source exhausted successfully. */
183 detect_info
->found
|= found
;
188 /*** GENERAL NOTES on `decode_coding_XXX ()' functions ***
190 These functions decode a byte sequence specified as a source by
191 CODING. The resulting multibyte text goes to a place pointed to by
192 CODING->charbuf, the length of which should not exceed
193 CODING->charbuf_size;
195 These functions set the information of original and decoded texts in
196 CODING->consumed, CODING->consumed_char, and CODING->charbuf_used.
197 They also set CODING->result to one of CODING_RESULT_XXX indicating
198 how the decoding is finished.
200 Below is the template of these functions. */
204 decode_coding_XXXX (struct coding_system
*coding
)
206 const unsigned char *src
= coding
->source
+ coding
->consumed
;
207 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
208 /* SRC_BASE remembers the start position in source in each loop.
209 The loop will be exited when there's not enough source code, or
210 when there's no room in CHARBUF for a decoded character. */
211 const unsigned char *src_base
;
212 /* A buffer to produce decoded characters. */
213 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
214 int *charbuf_end
= coding
->charbuf
+ coding
->charbuf_size
;
215 bool multibytep
= coding
->src_multibyte
;
220 if (charbuf
< charbuf_end
)
221 /* No more room to produce a decoded character. */
228 if (src_base
< src_end
229 && coding
->mode
& CODING_MODE_LAST_BLOCK
)
230 /* If the source ends by partial bytes to construct a character,
231 treat them as eight-bit raw data. */
232 while (src_base
< src_end
&& charbuf
< charbuf_end
)
233 *charbuf
++ = *src_base
++;
234 /* Remember how many bytes and characters we consumed. If the
235 source is multibyte, the bytes and chars are not identical. */
236 coding
->consumed
= coding
->consumed_char
= src_base
- coding
->source
;
237 /* Remember how many characters we produced. */
238 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
242 /*** GENERAL NOTES on `encode_coding_XXX ()' functions ***
244 These functions encode SRC_BYTES length text at SOURCE of Emacs'
245 internal multibyte format by CODING. The resulting byte sequence
246 goes to a place pointed to by DESTINATION, the length of which
247 should not exceed DST_BYTES.
249 These functions set the information of original and encoded texts in
250 the members produced, produced_char, consumed, and consumed_char of
251 the structure *CODING. They also set the member result to one of
252 CODING_RESULT_XXX indicating how the encoding finished.
254 DST_BYTES zero means that source area and destination area are
255 overlapped, which means that we can produce a encoded text until it
256 reaches at the head of not-yet-encoded source text.
258 Below is a template of these functions. */
261 encode_coding_XXX (struct coding_system
*coding
)
263 bool multibytep
= coding
->dst_multibyte
;
264 int *charbuf
= coding
->charbuf
;
265 int *charbuf_end
= charbuf
->charbuf
+ coding
->charbuf_used
;
266 unsigned char *dst
= coding
->destination
+ coding
->produced
;
267 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
268 unsigned char *adjusted_dst_end
= dst_end
- _MAX_BYTES_PRODUCED_IN_LOOP_
;
269 ptrdiff_t produced_chars
= 0;
271 for (; charbuf
< charbuf_end
&& dst
< adjusted_dst_end
; charbuf
++)
274 /* Encode C into DST, and increment DST. */
276 label_no_more_destination
:
277 /* How many chars and bytes we produced. */
278 coding
->produced_char
+= produced_chars
;
279 coding
->produced
= dst
- coding
->destination
;
284 /*** 1. Preamble ***/
291 #endif /* HAVE_WCHAR_H */
294 #include "character.h"
298 #include "composite.h"
302 #include "termhooks.h"
304 Lisp_Object Vcoding_system_hash_table
;
306 /* Format of end-of-line decided by system. This is Qunix on
307 Unix and Mac, Qdos on DOS/Windows.
308 This has an effect only for external encoding (i.e. for output to
309 file and process), not for in-buffer or Lisp string encoding. */
310 static Lisp_Object system_eol_type
;
314 /* Coding-systems are handed between Emacs Lisp programs and C internal
315 routines by the following three variables. */
316 /* Coding system to be used to encode text for terminal display when
317 terminal coding system is nil. */
318 struct coding_system safe_terminal_coding
;
322 /* Two special coding systems. */
323 static Lisp_Object Vsjis_coding_system
;
324 static Lisp_Object Vbig5_coding_system
;
326 /* ISO2022 section */
328 #define CODING_ISO_INITIAL(coding, reg) \
329 (XINT (AREF (AREF (CODING_ID_ATTRS ((coding)->id), \
330 coding_attr_iso_initial), \
334 #define CODING_ISO_REQUEST(coding, charset_id) \
335 (((charset_id) <= (coding)->max_charset_id \
336 ? ((coding)->safe_charsets[charset_id] != 255 \
337 ? (coding)->safe_charsets[charset_id] \
342 #define CODING_ISO_FLAGS(coding) \
343 ((coding)->spec.iso_2022.flags)
344 #define CODING_ISO_DESIGNATION(coding, reg) \
345 ((coding)->spec.iso_2022.current_designation[reg])
346 #define CODING_ISO_INVOCATION(coding, plane) \
347 ((coding)->spec.iso_2022.current_invocation[plane])
348 #define CODING_ISO_SINGLE_SHIFTING(coding) \
349 ((coding)->spec.iso_2022.single_shifting)
350 #define CODING_ISO_BOL(coding) \
351 ((coding)->spec.iso_2022.bol)
352 #define CODING_ISO_INVOKED_CHARSET(coding, plane) \
353 (CODING_ISO_INVOCATION (coding, plane) < 0 ? -1 \
354 : CODING_ISO_DESIGNATION (coding, CODING_ISO_INVOCATION (coding, plane)))
355 #define CODING_ISO_CMP_STATUS(coding) \
356 (&(coding)->spec.iso_2022.cmp_status)
357 #define CODING_ISO_EXTSEGMENT_LEN(coding) \
358 ((coding)->spec.iso_2022.ctext_extended_segment_len)
359 #define CODING_ISO_EMBEDDED_UTF_8(coding) \
360 ((coding)->spec.iso_2022.embedded_utf_8)
362 /* Control characters of ISO2022. */
363 /* code */ /* function */
364 #define ISO_CODE_SO 0x0E /* shift-out */
365 #define ISO_CODE_SI 0x0F /* shift-in */
366 #define ISO_CODE_SS2_7 0x19 /* single-shift-2 for 7-bit code */
367 #define ISO_CODE_ESC 0x1B /* escape */
368 #define ISO_CODE_SS2 0x8E /* single-shift-2 */
369 #define ISO_CODE_SS3 0x8F /* single-shift-3 */
370 #define ISO_CODE_CSI 0x9B /* control-sequence-introducer */
372 /* All code (1-byte) of ISO2022 is classified into one of the
374 enum iso_code_class_type
376 ISO_control_0
, /* Control codes in the range
377 0x00..0x1F and 0x7F, except for the
378 following 5 codes. */
379 ISO_shift_out
, /* ISO_CODE_SO (0x0E) */
380 ISO_shift_in
, /* ISO_CODE_SI (0x0F) */
381 ISO_single_shift_2_7
, /* ISO_CODE_SS2_7 (0x19) */
382 ISO_escape
, /* ISO_CODE_ESC (0x1B) */
383 ISO_control_1
, /* Control codes in the range
384 0x80..0x9F, except for the
385 following 3 codes. */
386 ISO_single_shift_2
, /* ISO_CODE_SS2 (0x8E) */
387 ISO_single_shift_3
, /* ISO_CODE_SS3 (0x8F) */
388 ISO_control_sequence_introducer
, /* ISO_CODE_CSI (0x9B) */
389 ISO_0x20_or_0x7F
, /* Codes of the values 0x20 or 0x7F. */
390 ISO_graphic_plane_0
, /* Graphic codes in the range 0x21..0x7E. */
391 ISO_0xA0_or_0xFF
, /* Codes of the values 0xA0 or 0xFF. */
392 ISO_graphic_plane_1
/* Graphic codes in the range 0xA1..0xFE. */
395 /** The macros CODING_ISO_FLAG_XXX defines a flag bit of the
396 `iso-flags' attribute of an iso2022 coding system. */
398 /* If set, produce long-form designation sequence (e.g. ESC $ ( A)
399 instead of the correct short-form sequence (e.g. ESC $ A). */
400 #define CODING_ISO_FLAG_LONG_FORM 0x0001
402 /* If set, reset graphic planes and registers at end-of-line to the
404 #define CODING_ISO_FLAG_RESET_AT_EOL 0x0002
406 /* If set, reset graphic planes and registers before any control
407 characters to the initial state. */
408 #define CODING_ISO_FLAG_RESET_AT_CNTL 0x0004
410 /* If set, encode by 7-bit environment. */
411 #define CODING_ISO_FLAG_SEVEN_BITS 0x0008
413 /* If set, use locking-shift function. */
414 #define CODING_ISO_FLAG_LOCKING_SHIFT 0x0010
416 /* If set, use single-shift function. Overwrite
417 CODING_ISO_FLAG_LOCKING_SHIFT. */
418 #define CODING_ISO_FLAG_SINGLE_SHIFT 0x0020
420 /* If set, use designation escape sequence. */
421 #define CODING_ISO_FLAG_DESIGNATION 0x0040
423 /* If set, produce revision number sequence. */
424 #define CODING_ISO_FLAG_REVISION 0x0080
426 /* If set, produce ISO6429's direction specifying sequence. */
427 #define CODING_ISO_FLAG_DIRECTION 0x0100
429 /* If set, assume designation states are reset at beginning of line on
431 #define CODING_ISO_FLAG_INIT_AT_BOL 0x0200
433 /* If set, designation sequence should be placed at beginning of line
435 #define CODING_ISO_FLAG_DESIGNATE_AT_BOL 0x0400
437 /* If set, do not encode unsafe characters on output. */
438 #define CODING_ISO_FLAG_SAFE 0x0800
440 /* If set, extra latin codes (128..159) are accepted as a valid code
442 #define CODING_ISO_FLAG_LATIN_EXTRA 0x1000
444 #define CODING_ISO_FLAG_COMPOSITION 0x2000
446 /* #define CODING_ISO_FLAG_EUC_TW_SHIFT 0x4000 */
448 #define CODING_ISO_FLAG_USE_ROMAN 0x8000
450 #define CODING_ISO_FLAG_USE_OLDJIS 0x10000
452 #define CODING_ISO_FLAG_LEVEL_4 0x20000
454 #define CODING_ISO_FLAG_FULL_SUPPORT 0x100000
456 /* A character to be produced on output if encoding of the original
457 character is prohibited by CODING_ISO_FLAG_SAFE. */
458 #define CODING_INHIBIT_CHARACTER_SUBSTITUTION '?'
461 #define CODING_UTF_8_BOM(coding) \
462 ((coding)->spec.utf_8_bom)
465 #define CODING_UTF_16_BOM(coding) \
466 ((coding)->spec.utf_16.bom)
468 #define CODING_UTF_16_ENDIAN(coding) \
469 ((coding)->spec.utf_16.endian)
471 #define CODING_UTF_16_SURROGATE(coding) \
472 ((coding)->spec.utf_16.surrogate)
476 #define CODING_CCL_DECODER(coding) \
477 AREF (CODING_ID_ATTRS ((coding)->id), coding_attr_ccl_decoder)
478 #define CODING_CCL_ENCODER(coding) \
479 AREF (CODING_ID_ATTRS ((coding)->id), coding_attr_ccl_encoder)
480 #define CODING_CCL_VALIDS(coding) \
481 (SDATA (AREF (CODING_ID_ATTRS ((coding)->id), coding_attr_ccl_valids)))
483 /* Index for each coding category in `coding_categories' */
487 coding_category_iso_7
,
488 coding_category_iso_7_tight
,
489 coding_category_iso_8_1
,
490 coding_category_iso_8_2
,
491 coding_category_iso_7_else
,
492 coding_category_iso_8_else
,
493 coding_category_utf_8_auto
,
494 coding_category_utf_8_nosig
,
495 coding_category_utf_8_sig
,
496 coding_category_utf_16_auto
,
497 coding_category_utf_16_be
,
498 coding_category_utf_16_le
,
499 coding_category_utf_16_be_nosig
,
500 coding_category_utf_16_le_nosig
,
501 coding_category_charset
,
502 coding_category_sjis
,
503 coding_category_big5
,
505 coding_category_emacs_mule
,
506 /* All above are targets of code detection. */
507 coding_category_raw_text
,
508 coding_category_undecided
,
512 /* Definitions of flag bits used in detect_coding_XXXX. */
513 #define CATEGORY_MASK_ISO_7 (1 << coding_category_iso_7)
514 #define CATEGORY_MASK_ISO_7_TIGHT (1 << coding_category_iso_7_tight)
515 #define CATEGORY_MASK_ISO_8_1 (1 << coding_category_iso_8_1)
516 #define CATEGORY_MASK_ISO_8_2 (1 << coding_category_iso_8_2)
517 #define CATEGORY_MASK_ISO_7_ELSE (1 << coding_category_iso_7_else)
518 #define CATEGORY_MASK_ISO_8_ELSE (1 << coding_category_iso_8_else)
519 #define CATEGORY_MASK_UTF_8_AUTO (1 << coding_category_utf_8_auto)
520 #define CATEGORY_MASK_UTF_8_NOSIG (1 << coding_category_utf_8_nosig)
521 #define CATEGORY_MASK_UTF_8_SIG (1 << coding_category_utf_8_sig)
522 #define CATEGORY_MASK_UTF_16_AUTO (1 << coding_category_utf_16_auto)
523 #define CATEGORY_MASK_UTF_16_BE (1 << coding_category_utf_16_be)
524 #define CATEGORY_MASK_UTF_16_LE (1 << coding_category_utf_16_le)
525 #define CATEGORY_MASK_UTF_16_BE_NOSIG (1 << coding_category_utf_16_be_nosig)
526 #define CATEGORY_MASK_UTF_16_LE_NOSIG (1 << coding_category_utf_16_le_nosig)
527 #define CATEGORY_MASK_CHARSET (1 << coding_category_charset)
528 #define CATEGORY_MASK_SJIS (1 << coding_category_sjis)
529 #define CATEGORY_MASK_BIG5 (1 << coding_category_big5)
530 #define CATEGORY_MASK_CCL (1 << coding_category_ccl)
531 #define CATEGORY_MASK_EMACS_MULE (1 << coding_category_emacs_mule)
532 #define CATEGORY_MASK_RAW_TEXT (1 << coding_category_raw_text)
534 /* This value is returned if detect_coding_mask () find nothing other
535 than ASCII characters. */
536 #define CATEGORY_MASK_ANY \
537 (CATEGORY_MASK_ISO_7 \
538 | CATEGORY_MASK_ISO_7_TIGHT \
539 | CATEGORY_MASK_ISO_8_1 \
540 | CATEGORY_MASK_ISO_8_2 \
541 | CATEGORY_MASK_ISO_7_ELSE \
542 | CATEGORY_MASK_ISO_8_ELSE \
543 | CATEGORY_MASK_UTF_8_AUTO \
544 | CATEGORY_MASK_UTF_8_NOSIG \
545 | CATEGORY_MASK_UTF_8_SIG \
546 | CATEGORY_MASK_UTF_16_AUTO \
547 | CATEGORY_MASK_UTF_16_BE \
548 | CATEGORY_MASK_UTF_16_LE \
549 | CATEGORY_MASK_UTF_16_BE_NOSIG \
550 | CATEGORY_MASK_UTF_16_LE_NOSIG \
551 | CATEGORY_MASK_CHARSET \
552 | CATEGORY_MASK_SJIS \
553 | CATEGORY_MASK_BIG5 \
554 | CATEGORY_MASK_CCL \
555 | CATEGORY_MASK_EMACS_MULE)
558 #define CATEGORY_MASK_ISO_7BIT \
559 (CATEGORY_MASK_ISO_7 | CATEGORY_MASK_ISO_7_TIGHT)
561 #define CATEGORY_MASK_ISO_8BIT \
562 (CATEGORY_MASK_ISO_8_1 | CATEGORY_MASK_ISO_8_2)
564 #define CATEGORY_MASK_ISO_ELSE \
565 (CATEGORY_MASK_ISO_7_ELSE | CATEGORY_MASK_ISO_8_ELSE)
567 #define CATEGORY_MASK_ISO_ESCAPE \
568 (CATEGORY_MASK_ISO_7 \
569 | CATEGORY_MASK_ISO_7_TIGHT \
570 | CATEGORY_MASK_ISO_7_ELSE \
571 | CATEGORY_MASK_ISO_8_ELSE)
573 #define CATEGORY_MASK_ISO \
574 ( CATEGORY_MASK_ISO_7BIT \
575 | CATEGORY_MASK_ISO_8BIT \
576 | CATEGORY_MASK_ISO_ELSE)
578 #define CATEGORY_MASK_UTF_16 \
579 (CATEGORY_MASK_UTF_16_AUTO \
580 | CATEGORY_MASK_UTF_16_BE \
581 | CATEGORY_MASK_UTF_16_LE \
582 | CATEGORY_MASK_UTF_16_BE_NOSIG \
583 | CATEGORY_MASK_UTF_16_LE_NOSIG)
585 #define CATEGORY_MASK_UTF_8 \
586 (CATEGORY_MASK_UTF_8_AUTO \
587 | CATEGORY_MASK_UTF_8_NOSIG \
588 | CATEGORY_MASK_UTF_8_SIG)
590 /* Table of coding categories (Lisp symbols). This variable is for
591 internal use only. */
592 static Lisp_Object Vcoding_category_table
;
594 /* Table of coding-categories ordered by priority. */
595 static enum coding_category coding_priorities
[coding_category_max
];
597 /* Nth element is a coding context for the coding system bound to the
598 Nth coding category. */
599 static struct coding_system coding_categories
[coding_category_max
];
601 /* Encode a flag that can be nil, something else, or t as -1, 0, 1. */
604 encode_inhibit_flag (Lisp_Object flag
)
606 return NILP (flag
) ? -1 : EQ (flag
, Qt
);
609 /* True if the value of ENCODED_FLAG says a flag should be treated as set.
610 1 means yes, -1 means no, 0 means ask the user variable VAR. */
613 inhibit_flag (int encoded_flag
, bool var
)
615 return 0 < encoded_flag
+ var
;
618 #define CODING_GET_INFO(coding, attrs, charset_list) \
620 (attrs) = CODING_ID_ATTRS ((coding)->id); \
621 (charset_list) = CODING_ATTR_CHARSET_LIST (attrs); \
625 CHECK_NATNUM_CAR (Lisp_Object x
)
627 Lisp_Object tmp
= XCAR (x
);
633 CHECK_NATNUM_CDR (Lisp_Object x
)
635 Lisp_Object tmp
= XCDR (x
);
640 /* True if CODING's destination can be grown. */
643 growable_destination (struct coding_system
*coding
)
645 return STRINGP (coding
->dst_object
) || BUFFERP (coding
->dst_object
);
649 /* Safely get one byte from the source text pointed by SRC which ends
650 at SRC_END, and set C to that byte. If there are not enough bytes
651 in the source, it jumps to 'no_more_source'. If MULTIBYTEP,
652 and a multibyte character is found at SRC, set C to the
653 negative value of the character code. The caller should declare
654 and set these variables appropriately in advance:
655 src, src_end, multibytep */
657 #define ONE_MORE_BYTE(c) \
659 if (src == src_end) \
661 if (src_base < src) \
662 record_conversion_result \
663 (coding, CODING_RESULT_INSUFFICIENT_SRC); \
664 goto no_more_source; \
667 if (multibytep && (c & 0x80)) \
669 if ((c & 0xFE) == 0xC0) \
670 c = ((c & 1) << 6) | *src++; \
674 c = - string_char (src, &src, NULL); \
675 record_conversion_result \
676 (coding, CODING_RESULT_INVALID_SRC); \
682 /* Safely get two bytes from the source text pointed by SRC which ends
683 at SRC_END, and set C1 and C2 to those bytes while skipping the
684 heading multibyte characters. If there are not enough bytes in the
685 source, it jumps to 'no_more_source'. If MULTIBYTEP and
686 a multibyte character is found for C2, set C2 to the negative value
687 of the character code. The caller should declare and set these
688 variables appropriately in advance:
689 src, src_end, multibytep
690 It is intended that this macro is used in detect_coding_utf_16. */
692 #define TWO_MORE_BYTES(c1, c2) \
695 if (src == src_end) \
696 goto no_more_source; \
698 if (multibytep && (c1 & 0x80)) \
700 if ((c1 & 0xFE) == 0xC0) \
701 c1 = ((c1 & 1) << 6) | *src++; \
704 src += BYTES_BY_CHAR_HEAD (c1) - 1; \
709 if (src == src_end) \
710 goto no_more_source; \
712 if (multibytep && (c2 & 0x80)) \
714 if ((c2 & 0xFE) == 0xC0) \
715 c2 = ((c2 & 1) << 6) | *src++; \
722 /* Store a byte C in the place pointed by DST and increment DST to the
723 next free point, and increment PRODUCED_CHARS. The caller should
724 assure that C is 0..127, and declare and set the variable `dst'
725 appropriately in advance.
729 #define EMIT_ONE_ASCII_BYTE(c) \
736 /* Like EMIT_ONE_ASCII_BYTE but store two bytes; C1 and C2. */
738 #define EMIT_TWO_ASCII_BYTES(c1, c2) \
740 produced_chars += 2; \
741 *dst++ = (c1), *dst++ = (c2); \
745 /* Store a byte C in the place pointed by DST and increment DST to the
746 next free point, and increment PRODUCED_CHARS. If MULTIBYTEP,
747 store in an appropriate multibyte form. The caller should
748 declare and set the variables `dst' and `multibytep' appropriately
751 #define EMIT_ONE_BYTE(c) \
758 ch = BYTE8_TO_CHAR (ch); \
759 CHAR_STRING_ADVANCE (ch, dst); \
766 /* Like EMIT_ONE_BYTE, but emit two bytes; C1 and C2. */
768 #define EMIT_TWO_BYTES(c1, c2) \
770 produced_chars += 2; \
777 ch = BYTE8_TO_CHAR (ch); \
778 CHAR_STRING_ADVANCE (ch, dst); \
781 ch = BYTE8_TO_CHAR (ch); \
782 CHAR_STRING_ADVANCE (ch, dst); \
792 #define EMIT_THREE_BYTES(c1, c2, c3) \
794 EMIT_ONE_BYTE (c1); \
795 EMIT_TWO_BYTES (c2, c3); \
799 #define EMIT_FOUR_BYTES(c1, c2, c3, c4) \
801 EMIT_TWO_BYTES (c1, c2); \
802 EMIT_TWO_BYTES (c3, c4); \
807 record_conversion_result (struct coding_system
*coding
,
808 enum coding_result_code result
)
810 coding
->result
= result
;
813 case CODING_RESULT_INSUFFICIENT_SRC
:
814 Vlast_code_conversion_error
= Qinsufficient_source
;
816 case CODING_RESULT_INVALID_SRC
:
817 Vlast_code_conversion_error
= Qinvalid_source
;
819 case CODING_RESULT_INTERRUPT
:
820 Vlast_code_conversion_error
= Qinterrupted
;
822 case CODING_RESULT_INSUFFICIENT_DST
:
823 /* Don't record this error in Vlast_code_conversion_error
824 because it happens just temporarily and is resolved when the
825 whole conversion is finished. */
827 case CODING_RESULT_SUCCESS
:
830 Vlast_code_conversion_error
= intern ("Unknown error");
834 /* These wrapper macros are used to preserve validity of pointers into
835 buffer text across calls to decode_char, encode_char, etc, which
836 could cause relocation of buffers if it loads a charset map,
837 because loading a charset map allocates large structures. */
839 #define CODING_DECODE_CHAR(coding, src, src_base, src_end, charset, code, c) \
843 charset_map_loaded = 0; \
844 c = DECODE_CHAR (charset, code); \
845 if (charset_map_loaded \
846 && (offset = coding_change_source (coding))) \
849 src_base += offset; \
854 #define CODING_ENCODE_CHAR(coding, dst, dst_end, charset, c, code) \
858 charset_map_loaded = 0; \
859 code = ENCODE_CHAR (charset, c); \
860 if (charset_map_loaded \
861 && (offset = coding_change_destination (coding))) \
868 #define CODING_CHAR_CHARSET(coding, dst, dst_end, c, charset_list, code_return, charset) \
872 charset_map_loaded = 0; \
873 charset = char_charset (c, charset_list, code_return); \
874 if (charset_map_loaded \
875 && (offset = coding_change_destination (coding))) \
882 #define CODING_CHAR_CHARSET_P(coding, dst, dst_end, c, charset, result) \
886 charset_map_loaded = 0; \
887 result = CHAR_CHARSET_P (c, charset); \
888 if (charset_map_loaded \
889 && (offset = coding_change_destination (coding))) \
897 /* If there are at least BYTES length of room at dst, allocate memory
898 for coding->destination and update dst and dst_end. We don't have
899 to take care of coding->source which will be relocated. It is
900 handled by calling coding_set_source in encode_coding. */
902 #define ASSURE_DESTINATION(bytes) \
904 if (dst + (bytes) >= dst_end) \
906 ptrdiff_t more_bytes = charbuf_end - charbuf + (bytes); \
908 dst = alloc_destination (coding, more_bytes, dst); \
909 dst_end = coding->destination + coding->dst_bytes; \
914 /* Store multibyte form of the character C in P, and advance P to the
915 end of the multibyte form. This used to be like CHAR_STRING_ADVANCE
916 without ever calling MAYBE_UNIFY_CHAR, but nowadays we don't call
917 MAYBE_UNIFY_CHAR in CHAR_STRING_ADVANCE. */
919 #define CHAR_STRING_ADVANCE_NO_UNIFY(c, p) CHAR_STRING_ADVANCE(c, p)
921 /* Return the character code of character whose multibyte form is at
922 P, and advance P to the end of the multibyte form. This used to be
923 like STRING_CHAR_ADVANCE without ever calling MAYBE_UNIFY_CHAR, but
924 nowadays STRING_CHAR_ADVANCE doesn't call MAYBE_UNIFY_CHAR. */
926 #define STRING_CHAR_ADVANCE_NO_UNIFY(p) STRING_CHAR_ADVANCE(p)
928 /* Set coding->source from coding->src_object. */
931 coding_set_source (struct coding_system
*coding
)
933 if (BUFFERP (coding
->src_object
))
935 struct buffer
*buf
= XBUFFER (coding
->src_object
);
937 if (coding
->src_pos
< 0)
938 coding
->source
= BUF_GAP_END_ADDR (buf
) + coding
->src_pos_byte
;
940 coding
->source
= BUF_BYTE_ADDRESS (buf
, coding
->src_pos_byte
);
942 else if (STRINGP (coding
->src_object
))
944 coding
->source
= SDATA (coding
->src_object
) + coding
->src_pos_byte
;
948 /* Otherwise, the source is C string and is never relocated
949 automatically. Thus we don't have to update anything. */
954 /* Set coding->source from coding->src_object, and return how many
955 bytes coding->source was changed. */
958 coding_change_source (struct coding_system
*coding
)
960 const unsigned char *orig
= coding
->source
;
961 coding_set_source (coding
);
962 return coding
->source
- orig
;
966 /* Set coding->destination from coding->dst_object. */
969 coding_set_destination (struct coding_system
*coding
)
971 if (BUFFERP (coding
->dst_object
))
973 if (BUFFERP (coding
->src_object
) && coding
->src_pos
< 0)
975 coding
->destination
= BEG_ADDR
+ coding
->dst_pos_byte
- BEG_BYTE
;
976 coding
->dst_bytes
= (GAP_END_ADDR
977 - (coding
->src_bytes
- coding
->consumed
)
978 - coding
->destination
);
982 /* We are sure that coding->dst_pos_byte is before the gap
984 coding
->destination
= (BUF_BEG_ADDR (XBUFFER (coding
->dst_object
))
985 + coding
->dst_pos_byte
- BEG_BYTE
);
986 coding
->dst_bytes
= (BUF_GAP_END_ADDR (XBUFFER (coding
->dst_object
))
987 - coding
->destination
);
992 /* Otherwise, the destination is C string and is never relocated
993 automatically. Thus we don't have to update anything. */
998 /* Set coding->destination from coding->dst_object, and return how
999 many bytes coding->destination was changed. */
1002 coding_change_destination (struct coding_system
*coding
)
1004 const unsigned char *orig
= coding
->destination
;
1005 coding_set_destination (coding
);
1006 return coding
->destination
- orig
;
1011 coding_alloc_by_realloc (struct coding_system
*coding
, ptrdiff_t bytes
)
1013 if (STRING_BYTES_BOUND
- coding
->dst_bytes
< bytes
)
1015 coding
->destination
= xrealloc (coding
->destination
,
1016 coding
->dst_bytes
+ bytes
);
1017 coding
->dst_bytes
+= bytes
;
1021 coding_alloc_by_making_gap (struct coding_system
*coding
,
1022 ptrdiff_t gap_head_used
, ptrdiff_t bytes
)
1024 if (EQ (coding
->src_object
, coding
->dst_object
))
1026 /* The gap may contain the produced data at the head and not-yet
1027 consumed data at the tail. To preserve those data, we at
1028 first make the gap size to zero, then increase the gap
1030 ptrdiff_t add
= GAP_SIZE
;
1032 GPT
+= gap_head_used
, GPT_BYTE
+= gap_head_used
;
1033 GAP_SIZE
= 0; ZV
+= add
; Z
+= add
; ZV_BYTE
+= add
; Z_BYTE
+= add
;
1035 GAP_SIZE
+= add
; ZV
-= add
; Z
-= add
; ZV_BYTE
-= add
; Z_BYTE
-= add
;
1036 GPT
-= gap_head_used
, GPT_BYTE
-= gap_head_used
;
1039 make_gap_1 (XBUFFER (coding
->dst_object
), bytes
);
1043 static unsigned char *
1044 alloc_destination (struct coding_system
*coding
, ptrdiff_t nbytes
,
1047 ptrdiff_t offset
= dst
- coding
->destination
;
1049 if (BUFFERP (coding
->dst_object
))
1051 struct buffer
*buf
= XBUFFER (coding
->dst_object
);
1053 coding_alloc_by_making_gap (coding
, dst
- BUF_GPT_ADDR (buf
), nbytes
);
1056 coding_alloc_by_realloc (coding
, nbytes
);
1057 coding_set_destination (coding
);
1058 dst
= coding
->destination
+ offset
;
1062 /** Macros for annotations. */
1064 /* An annotation data is stored in the array coding->charbuf in this
1066 [ -LENGTH ANNOTATION_MASK NCHARS ... ]
1067 LENGTH is the number of elements in the annotation.
1068 ANNOTATION_MASK is one of CODING_ANNOTATE_XXX_MASK.
1069 NCHARS is the number of characters in the text annotated.
1071 The format of the following elements depend on ANNOTATION_MASK.
1073 In the case of CODING_ANNOTATE_COMPOSITION_MASK, these elements
1075 ... NBYTES METHOD [ COMPOSITION-COMPONENTS ... ]
1077 NBYTES is the number of bytes specified in the header part of
1078 old-style emacs-mule encoding, or 0 for the other kind of
1081 METHOD is one of enum composition_method.
1083 Optional COMPOSITION-COMPONENTS are characters and composition
1086 In the case of CODING_ANNOTATE_CHARSET_MASK, one element CHARSET-ID
1089 If ANNOTATION_MASK is 0, this annotation is just a space holder to
1090 recover from an invalid annotation, and should be skipped by
1091 produce_annotation. */
1093 /* Maximum length of the header of annotation data. */
1094 #define MAX_ANNOTATION_LENGTH 5
1096 #define ADD_ANNOTATION_DATA(buf, len, mask, nchars) \
1098 *(buf)++ = -(len); \
1099 *(buf)++ = (mask); \
1100 *(buf)++ = (nchars); \
1101 coding->annotated = 1; \
1104 #define ADD_COMPOSITION_DATA(buf, nchars, nbytes, method) \
1106 ADD_ANNOTATION_DATA (buf, 5, CODING_ANNOTATE_COMPOSITION_MASK, nchars); \
1112 #define ADD_CHARSET_DATA(buf, nchars, id) \
1114 ADD_ANNOTATION_DATA (buf, 4, CODING_ANNOTATE_CHARSET_MASK, nchars); \
1119 /* Bitmasks for coding->eol_seen. */
1121 #define EOL_SEEN_NONE 0
1122 #define EOL_SEEN_LF 1
1123 #define EOL_SEEN_CR 2
1124 #define EOL_SEEN_CRLF 4
1127 /*** 2. Emacs' internal format (emacs-utf-8) ***/
1134 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
1135 Return true if a text is encoded in UTF-8. */
1137 #define UTF_8_1_OCTET_P(c) ((c) < 0x80)
1138 #define UTF_8_EXTRA_OCTET_P(c) (((c) & 0xC0) == 0x80)
1139 #define UTF_8_2_OCTET_LEADING_P(c) (((c) & 0xE0) == 0xC0)
1140 #define UTF_8_3_OCTET_LEADING_P(c) (((c) & 0xF0) == 0xE0)
1141 #define UTF_8_4_OCTET_LEADING_P(c) (((c) & 0xF8) == 0xF0)
1142 #define UTF_8_5_OCTET_LEADING_P(c) (((c) & 0xFC) == 0xF8)
1144 #define UTF_8_BOM_1 0xEF
1145 #define UTF_8_BOM_2 0xBB
1146 #define UTF_8_BOM_3 0xBF
1148 /* Unlike the other detect_coding_XXX, this function counts the number
1149 of characters and checks the EOL format. */
1152 detect_coding_utf_8 (struct coding_system
*coding
,
1153 struct coding_detection_info
*detect_info
)
1155 const unsigned char *src
= coding
->source
, *src_base
;
1156 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
1157 bool multibytep
= coding
->src_multibyte
;
1158 ptrdiff_t consumed_chars
= 0;
1160 ptrdiff_t nchars
= coding
->head_ascii
;
1161 int eol_seen
= coding
->eol_seen
;
1163 detect_info
->checked
|= CATEGORY_MASK_UTF_8
;
1164 /* A coding system of this category is always ASCII compatible. */
1167 if (src
== coding
->source
/* BOM should be at the head. */
1168 && src
+ 3 < src_end
/* BOM is 3-byte long. */
1169 && src
[0] == UTF_8_BOM_1
1170 && src
[1] == UTF_8_BOM_2
1171 && src
[2] == UTF_8_BOM_3
)
1180 int c
, c1
, c2
, c3
, c4
;
1184 if (c
< 0 || UTF_8_1_OCTET_P (c
))
1189 if (src
< src_end
&& *src
== '\n')
1191 eol_seen
|= EOL_SEEN_CRLF
;
1196 eol_seen
|= EOL_SEEN_CR
;
1199 eol_seen
|= EOL_SEEN_LF
;
1203 if (c1
< 0 || ! UTF_8_EXTRA_OCTET_P (c1
))
1205 if (UTF_8_2_OCTET_LEADING_P (c
))
1211 if (c2
< 0 || ! UTF_8_EXTRA_OCTET_P (c2
))
1213 if (UTF_8_3_OCTET_LEADING_P (c
))
1219 if (c3
< 0 || ! UTF_8_EXTRA_OCTET_P (c3
))
1221 if (UTF_8_4_OCTET_LEADING_P (c
))
1227 if (c4
< 0 || ! UTF_8_EXTRA_OCTET_P (c4
))
1229 if (UTF_8_5_OCTET_LEADING_P (c
))
1236 detect_info
->rejected
|= CATEGORY_MASK_UTF_8
;
1240 if (src_base
< src
&& coding
->mode
& CODING_MODE_LAST_BLOCK
)
1242 detect_info
->rejected
|= CATEGORY_MASK_UTF_8
;
1247 /* The first character 0xFFFE doesn't necessarily mean a BOM. */
1248 detect_info
->found
|= CATEGORY_MASK_UTF_8_AUTO
| CATEGORY_MASK_UTF_8_SIG
| CATEGORY_MASK_UTF_8_NOSIG
;
1252 detect_info
->rejected
|= CATEGORY_MASK_UTF_8_SIG
;
1253 if (nchars
< src_end
- coding
->source
)
1254 /* The found characters are less than source bytes, which
1255 means that we found a valid non-ASCII characters. */
1256 detect_info
->found
|= CATEGORY_MASK_UTF_8_AUTO
| CATEGORY_MASK_UTF_8_NOSIG
;
1258 coding
->detected_utf8_bytes
= src_base
- coding
->source
;
1259 coding
->detected_utf8_chars
= nchars
;
1265 decode_coding_utf_8 (struct coding_system
*coding
)
1267 const unsigned char *src
= coding
->source
+ coding
->consumed
;
1268 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
1269 const unsigned char *src_base
;
1270 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
1271 int *charbuf_end
= coding
->charbuf
+ coding
->charbuf_size
;
1272 ptrdiff_t consumed_chars
= 0, consumed_chars_base
= 0;
1273 bool multibytep
= coding
->src_multibyte
;
1274 enum utf_bom_type bom
= CODING_UTF_8_BOM (coding
);
1276 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
1277 int byte_after_cr
= -1;
1279 if (bom
!= utf_without_bom
)
1285 if (! UTF_8_3_OCTET_LEADING_P (c1
))
1290 if (! UTF_8_EXTRA_OCTET_P (c2
))
1295 if (! UTF_8_EXTRA_OCTET_P (c3
))
1299 if ((c1
!= UTF_8_BOM_1
)
1300 || (c2
!= UTF_8_BOM_2
) || (c3
!= UTF_8_BOM_3
))
1303 CODING_UTF_8_BOM (coding
) = utf_without_bom
;
1308 CODING_UTF_8_BOM (coding
) = utf_without_bom
;
1312 int c
, c1
, c2
, c3
, c4
, c5
;
1315 consumed_chars_base
= consumed_chars
;
1317 if (charbuf
>= charbuf_end
)
1319 if (byte_after_cr
>= 0)
1324 /* In the simple case, rapidly handle ordinary characters */
1325 if (multibytep
&& ! eol_dos
1326 && charbuf
< charbuf_end
- 6 && src
< src_end
- 6)
1328 while (charbuf
< charbuf_end
- 6 && src
< src_end
- 6)
1358 /* If we handled at least one character, restart the main loop. */
1359 if (src
!= src_base
)
1363 if (byte_after_cr
>= 0)
1364 c1
= byte_after_cr
, byte_after_cr
= -1;
1371 else if (UTF_8_1_OCTET_P (c1
))
1373 if (eol_dos
&& c1
== '\r')
1374 ONE_MORE_BYTE (byte_after_cr
);
1380 if (c2
< 0 || ! UTF_8_EXTRA_OCTET_P (c2
))
1382 if (UTF_8_2_OCTET_LEADING_P (c1
))
1384 c
= ((c1
& 0x1F) << 6) | (c2
& 0x3F);
1385 /* Reject overlong sequences here and below. Encoders
1386 producing them are incorrect, they can be misleading,
1387 and they mess up read/write invariance. */
1394 if (c3
< 0 || ! UTF_8_EXTRA_OCTET_P (c3
))
1396 if (UTF_8_3_OCTET_LEADING_P (c1
))
1398 c
= (((c1
& 0xF) << 12)
1399 | ((c2
& 0x3F) << 6) | (c3
& 0x3F));
1401 || (c
>= 0xd800 && c
< 0xe000)) /* surrogates (invalid) */
1407 if (c4
< 0 || ! UTF_8_EXTRA_OCTET_P (c4
))
1409 if (UTF_8_4_OCTET_LEADING_P (c1
))
1411 c
= (((c1
& 0x7) << 18) | ((c2
& 0x3F) << 12)
1412 | ((c3
& 0x3F) << 6) | (c4
& 0x3F));
1419 if (c5
< 0 || ! UTF_8_EXTRA_OCTET_P (c5
))
1421 if (UTF_8_5_OCTET_LEADING_P (c1
))
1423 c
= (((c1
& 0x3) << 24) | ((c2
& 0x3F) << 18)
1424 | ((c3
& 0x3F) << 12) | ((c4
& 0x3F) << 6)
1426 if ((c
> MAX_CHAR
) || (c
< 0x200000))
1441 consumed_chars
= consumed_chars_base
;
1443 *charbuf
++ = ASCII_CHAR_P (c
) ? c
: BYTE8_TO_CHAR (c
);
1447 coding
->consumed_char
+= consumed_chars_base
;
1448 coding
->consumed
= src_base
- coding
->source
;
1449 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
1454 encode_coding_utf_8 (struct coding_system
*coding
)
1456 bool multibytep
= coding
->dst_multibyte
;
1457 int *charbuf
= coding
->charbuf
;
1458 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
1459 unsigned char *dst
= coding
->destination
+ coding
->produced
;
1460 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
1461 ptrdiff_t produced_chars
= 0;
1464 if (CODING_UTF_8_BOM (coding
) == utf_with_bom
)
1466 ASSURE_DESTINATION (3);
1467 EMIT_THREE_BYTES (UTF_8_BOM_1
, UTF_8_BOM_2
, UTF_8_BOM_3
);
1468 CODING_UTF_8_BOM (coding
) = utf_without_bom
;
1473 int safe_room
= MAX_MULTIBYTE_LENGTH
* 2;
1475 while (charbuf
< charbuf_end
)
1477 unsigned char str
[MAX_MULTIBYTE_LENGTH
], *p
, *pend
= str
;
1479 ASSURE_DESTINATION (safe_room
);
1481 if (CHAR_BYTE8_P (c
))
1483 c
= CHAR_TO_BYTE8 (c
);
1488 CHAR_STRING_ADVANCE_NO_UNIFY (c
, pend
);
1489 for (p
= str
; p
< pend
; p
++)
1496 int safe_room
= MAX_MULTIBYTE_LENGTH
;
1498 while (charbuf
< charbuf_end
)
1500 ASSURE_DESTINATION (safe_room
);
1502 if (CHAR_BYTE8_P (c
))
1503 *dst
++ = CHAR_TO_BYTE8 (c
);
1505 CHAR_STRING_ADVANCE_NO_UNIFY (c
, dst
);
1507 produced_chars
= dst
- (coding
->destination
+ coding
->produced
);
1509 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
1510 coding
->produced_char
+= produced_chars
;
1511 coding
->produced
= dst
- coding
->destination
;
1516 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
1517 Return true if a text is encoded in one of UTF-16 based coding systems. */
1519 #define UTF_16_HIGH_SURROGATE_P(val) \
1520 (((val) & 0xFC00) == 0xD800)
1522 #define UTF_16_LOW_SURROGATE_P(val) \
1523 (((val) & 0xFC00) == 0xDC00)
1527 detect_coding_utf_16 (struct coding_system
*coding
,
1528 struct coding_detection_info
*detect_info
)
1530 const unsigned char *src
= coding
->source
;
1531 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
1532 bool multibytep
= coding
->src_multibyte
;
1535 detect_info
->checked
|= CATEGORY_MASK_UTF_16
;
1536 if (coding
->mode
& CODING_MODE_LAST_BLOCK
1537 && (coding
->src_chars
& 1))
1539 detect_info
->rejected
|= CATEGORY_MASK_UTF_16
;
1543 TWO_MORE_BYTES (c1
, c2
);
1544 if ((c1
== 0xFF) && (c2
== 0xFE))
1546 detect_info
->found
|= (CATEGORY_MASK_UTF_16_LE
1547 | CATEGORY_MASK_UTF_16_AUTO
);
1548 detect_info
->rejected
|= (CATEGORY_MASK_UTF_16_BE
1549 | CATEGORY_MASK_UTF_16_BE_NOSIG
1550 | CATEGORY_MASK_UTF_16_LE_NOSIG
);
1552 else if ((c1
== 0xFE) && (c2
== 0xFF))
1554 detect_info
->found
|= (CATEGORY_MASK_UTF_16_BE
1555 | CATEGORY_MASK_UTF_16_AUTO
);
1556 detect_info
->rejected
|= (CATEGORY_MASK_UTF_16_LE
1557 | CATEGORY_MASK_UTF_16_BE_NOSIG
1558 | CATEGORY_MASK_UTF_16_LE_NOSIG
);
1562 detect_info
->rejected
|= CATEGORY_MASK_UTF_16
;
1567 /* We check the dispersion of Eth and Oth bytes where E is even and
1568 O is odd. If both are high, we assume binary data.*/
1569 unsigned char e
[256], o
[256];
1570 unsigned e_num
= 1, o_num
= 1;
1577 detect_info
->rejected
|= (CATEGORY_MASK_UTF_16_AUTO
1578 |CATEGORY_MASK_UTF_16_BE
1579 | CATEGORY_MASK_UTF_16_LE
);
1581 while ((detect_info
->rejected
& CATEGORY_MASK_UTF_16
)
1582 != CATEGORY_MASK_UTF_16
)
1584 TWO_MORE_BYTES (c1
, c2
);
1592 detect_info
->rejected
|= CATEGORY_MASK_UTF_16_BE_NOSIG
;
1599 detect_info
->rejected
|= CATEGORY_MASK_UTF_16_LE_NOSIG
;
1610 decode_coding_utf_16 (struct coding_system
*coding
)
1612 const unsigned char *src
= coding
->source
+ coding
->consumed
;
1613 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
1614 const unsigned char *src_base
;
1615 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
1616 /* We may produces at most 3 chars in one loop. */
1617 int *charbuf_end
= coding
->charbuf
+ coding
->charbuf_size
- 2;
1618 ptrdiff_t consumed_chars
= 0, consumed_chars_base
= 0;
1619 bool multibytep
= coding
->src_multibyte
;
1620 enum utf_bom_type bom
= CODING_UTF_16_BOM (coding
);
1621 enum utf_16_endian_type endian
= CODING_UTF_16_ENDIAN (coding
);
1622 int surrogate
= CODING_UTF_16_SURROGATE (coding
);
1624 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
1625 int byte_after_cr1
= -1, byte_after_cr2
= -1;
1627 if (bom
== utf_with_bom
)
1636 if (endian
== utf_16_big_endian
1637 ? c
!= 0xFEFF : c
!= 0xFFFE)
1639 /* The first two bytes are not BOM. Treat them as bytes
1640 for a normal character. */
1643 CODING_UTF_16_BOM (coding
) = utf_without_bom
;
1645 else if (bom
== utf_detect_bom
)
1647 /* We have already tried to detect BOM and failed in
1649 CODING_UTF_16_BOM (coding
) = utf_without_bom
;
1657 consumed_chars_base
= consumed_chars
;
1659 if (charbuf
>= charbuf_end
)
1661 if (byte_after_cr1
>= 0)
1666 if (byte_after_cr1
>= 0)
1667 c1
= byte_after_cr1
, byte_after_cr1
= -1;
1675 if (byte_after_cr2
>= 0)
1676 c2
= byte_after_cr2
, byte_after_cr2
= -1;
1681 *charbuf
++ = ASCII_CHAR_P (c1
) ? c1
: BYTE8_TO_CHAR (c1
);
1685 c
= (endian
== utf_16_big_endian
1686 ? ((c1
<< 8) | c2
) : ((c2
<< 8) | c1
));
1690 if (! UTF_16_LOW_SURROGATE_P (c
))
1692 if (endian
== utf_16_big_endian
)
1693 c1
= surrogate
>> 8, c2
= surrogate
& 0xFF;
1695 c1
= surrogate
& 0xFF, c2
= surrogate
>> 8;
1698 if (UTF_16_HIGH_SURROGATE_P (c
))
1699 CODING_UTF_16_SURROGATE (coding
) = surrogate
= c
;
1705 c
= ((surrogate
- 0xD800) << 10) | (c
- 0xDC00);
1706 CODING_UTF_16_SURROGATE (coding
) = surrogate
= 0;
1707 *charbuf
++ = 0x10000 + c
;
1712 if (UTF_16_HIGH_SURROGATE_P (c
))
1713 CODING_UTF_16_SURROGATE (coding
) = surrogate
= c
;
1716 if (eol_dos
&& c
== '\r')
1718 ONE_MORE_BYTE (byte_after_cr1
);
1719 ONE_MORE_BYTE (byte_after_cr2
);
1727 coding
->consumed_char
+= consumed_chars_base
;
1728 coding
->consumed
= src_base
- coding
->source
;
1729 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
1733 encode_coding_utf_16 (struct coding_system
*coding
)
1735 bool multibytep
= coding
->dst_multibyte
;
1736 int *charbuf
= coding
->charbuf
;
1737 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
1738 unsigned char *dst
= coding
->destination
+ coding
->produced
;
1739 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
1741 enum utf_bom_type bom
= CODING_UTF_16_BOM (coding
);
1742 bool big_endian
= CODING_UTF_16_ENDIAN (coding
) == utf_16_big_endian
;
1743 ptrdiff_t produced_chars
= 0;
1746 if (bom
!= utf_without_bom
)
1748 ASSURE_DESTINATION (safe_room
);
1750 EMIT_TWO_BYTES (0xFE, 0xFF);
1752 EMIT_TWO_BYTES (0xFF, 0xFE);
1753 CODING_UTF_16_BOM (coding
) = utf_without_bom
;
1756 while (charbuf
< charbuf_end
)
1758 ASSURE_DESTINATION (safe_room
);
1760 if (c
> MAX_UNICODE_CHAR
)
1761 c
= coding
->default_char
;
1766 EMIT_TWO_BYTES (c
>> 8, c
& 0xFF);
1768 EMIT_TWO_BYTES (c
& 0xFF, c
>> 8);
1775 c1
= (c
>> 10) + 0xD800;
1776 c2
= (c
& 0x3FF) + 0xDC00;
1778 EMIT_FOUR_BYTES (c1
>> 8, c1
& 0xFF, c2
>> 8, c2
& 0xFF);
1780 EMIT_FOUR_BYTES (c1
& 0xFF, c1
>> 8, c2
& 0xFF, c2
>> 8);
1783 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
1784 coding
->produced
= dst
- coding
->destination
;
1785 coding
->produced_char
+= produced_chars
;
1790 /*** 6. Old Emacs' internal format (emacs-mule) ***/
1792 /* Emacs' internal format for representation of multiple character
1793 sets is a kind of multi-byte encoding, i.e. characters are
1794 represented by variable-length sequences of one-byte codes.
1796 ASCII characters and control characters (e.g. `tab', `newline') are
1797 represented by one-byte sequences which are their ASCII codes, in
1798 the range 0x00 through 0x7F.
1800 8-bit characters of the range 0x80..0x9F are represented by
1801 two-byte sequences of LEADING_CODE_8_BIT_CONTROL and (their 8-bit
1804 8-bit characters of the range 0xA0..0xFF are represented by
1805 one-byte sequences which are their 8-bit code.
1807 The other characters are represented by a sequence of `base
1808 leading-code', optional `extended leading-code', and one or two
1809 `position-code's. The length of the sequence is determined by the
1810 base leading-code. Leading-code takes the range 0x81 through 0x9D,
1811 whereas extended leading-code and position-code take the range 0xA0
1812 through 0xFF. See `charset.h' for more details about leading-code
1815 --- CODE RANGE of Emacs' internal format ---
1819 eight-bit-control LEADING_CODE_8_BIT_CONTROL + 0xA0..0xBF
1820 eight-bit-graphic 0xA0..0xBF
1821 ELSE 0x81..0x9D + [0xA0..0xFF]+
1822 ---------------------------------------------
1824 As this is the internal character representation, the format is
1825 usually not used externally (i.e. in a file or in a data sent to a
1826 process). But, it is possible to have a text externally in this
1827 format (i.e. by encoding by the coding system `emacs-mule').
1829 In that case, a sequence of one-byte codes has a slightly different
1832 At first, all characters in eight-bit-control are represented by
1833 one-byte sequences which are their 8-bit code.
1835 Next, character composition data are represented by the byte
1836 sequence of the form: 0x80 METHOD BYTES CHARS COMPONENT ...,
1838 METHOD is 0xF2 plus one of composition method (enum
1839 composition_method),
1841 BYTES is 0xA0 plus a byte length of this composition data,
1843 CHARS is 0xA0 plus a number of characters composed by this
1846 COMPONENTs are characters of multibyte form or composition
1847 rules encoded by two-byte of ASCII codes.
1849 In addition, for backward compatibility, the following formats are
1850 also recognized as composition data on decoding.
1853 0x80 0xFF MSEQ RULE MSEQ RULE ... MSEQ
1856 MSEQ is a multibyte form but in these special format:
1857 ASCII: 0xA0 ASCII_CODE+0x80,
1858 other: LEADING_CODE+0x20 FOLLOWING-BYTE ...,
1859 RULE is a one byte code of the range 0xA0..0xF0 that
1860 represents a composition rule.
1863 char emacs_mule_bytes
[256];
1866 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
1867 Return true if a text is encoded in 'emacs-mule'. */
1870 detect_coding_emacs_mule (struct coding_system
*coding
,
1871 struct coding_detection_info
*detect_info
)
1873 const unsigned char *src
= coding
->source
, *src_base
;
1874 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
1875 bool multibytep
= coding
->src_multibyte
;
1876 ptrdiff_t consumed_chars
= 0;
1880 detect_info
->checked
|= CATEGORY_MASK_EMACS_MULE
;
1881 /* A coding system of this category is always ASCII compatible. */
1882 src
+= coding
->head_ascii
;
1892 /* Perhaps the start of composite character. We simply skip
1893 it because analyzing it is too heavy for detecting. But,
1894 at least, we check that the composite character
1895 constitutes of more than 4 bytes. */
1896 const unsigned char *src_start
;
1906 if (src
- src_start
<= 4)
1908 found
= CATEGORY_MASK_EMACS_MULE
;
1916 && (c
== ISO_CODE_ESC
|| c
== ISO_CODE_SI
|| c
== ISO_CODE_SO
))
1921 int more_bytes
= emacs_mule_bytes
[c
] - 1;
1923 while (more_bytes
> 0)
1928 src
--; /* Unread the last byte. */
1933 if (more_bytes
!= 0)
1935 found
= CATEGORY_MASK_EMACS_MULE
;
1938 detect_info
->rejected
|= CATEGORY_MASK_EMACS_MULE
;
1942 if (src_base
< src
&& coding
->mode
& CODING_MODE_LAST_BLOCK
)
1944 detect_info
->rejected
|= CATEGORY_MASK_EMACS_MULE
;
1947 detect_info
->found
|= found
;
1952 /* Parse emacs-mule multibyte sequence at SRC and return the decoded
1953 character. If CMP_STATUS indicates that we must expect MSEQ or
1954 RULE described above, decode it and return the negative value of
1955 the decoded character or rule. If an invalid byte is found, return
1956 -1. If SRC is too short, return -2. */
1959 emacs_mule_char (struct coding_system
*coding
, const unsigned char *src
,
1960 int *nbytes
, int *nchars
, int *id
,
1961 struct composition_status
*cmp_status
)
1963 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
1964 const unsigned char *src_base
= src
;
1965 bool multibytep
= coding
->src_multibyte
;
1969 ptrdiff_t consumed_chars
= 0;
1970 bool mseq_found
= 0;
1976 charset_ID
= emacs_mule_charset
[0];
1982 if (cmp_status
->state
!= COMPOSING_NO
1983 && cmp_status
->old_form
)
1985 if (cmp_status
->state
== COMPOSING_CHAR
)
2000 *nbytes
= src
- src_base
;
2001 *nchars
= consumed_chars
;
2009 switch (emacs_mule_bytes
[c
])
2012 if ((charset_ID
= emacs_mule_charset
[c
]) < 0)
2021 if (c
== EMACS_MULE_LEADING_CODE_PRIVATE_11
2022 || c
== EMACS_MULE_LEADING_CODE_PRIVATE_12
)
2025 if (c
< 0xA0 || (charset_ID
= emacs_mule_charset
[c
]) < 0)
2034 if ((charset_ID
= emacs_mule_charset
[c
]) < 0)
2039 code
= (c
& 0x7F) << 8;
2049 if (c
< 0 || (charset_ID
= emacs_mule_charset
[c
]) < 0)
2054 code
= (c
& 0x7F) << 8;
2063 charset_ID
= ASCII_CHAR_P (code
) ? charset_ascii
: charset_eight_bit
;
2069 CODING_DECODE_CHAR (coding
, src
, src_base
, src_end
,
2070 CHARSET_FROM_ID (charset_ID
), code
, c
);
2074 *nbytes
= src
- src_base
;
2075 *nchars
= consumed_chars
;
2078 return (mseq_found
? -c
: c
);
2088 /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
2090 /* Handle these composition sequence ('|': the end of header elements,
2091 BYTES and CHARS >= 0xA0):
2093 (1) relative composition: 0x80 0xF2 BYTES CHARS | CHAR ...
2094 (2) altchar composition: 0x80 0xF4 BYTES CHARS | ALT ... ALT CHAR ...
2095 (3) alt&rule composition: 0x80 0xF5 BYTES CHARS | ALT RULE ... ALT CHAR ...
2099 (4) relative composition: 0x80 | MSEQ ... MSEQ
2100 (5) rulebase composition: 0x80 0xFF | MSEQ MRULE ... MSEQ
2102 When the starter 0x80 and the following header elements are found,
2103 this annotation header is produced.
2105 [ -LENGTH(==-5) CODING_ANNOTATE_COMPOSITION_MASK NCHARS NBYTES METHOD ]
2107 NCHARS is CHARS - 0xA0 for (1), (2), (3), and 0 for (4), (5).
2108 NBYTES is BYTES - 0xA0 for (1), (2), (3), and 0 for (4), (5).
2110 Then, upon reading the following elements, these codes are produced
2111 until the composition end is found:
2114 (2) ALT ... ALT CHAR ... CHAR
2115 (3) ALT -2 DECODED-RULE ALT -2 DECODED-RULE ... ALT CHAR ... CHAR
2117 (5) CHAR -2 DECODED-RULE CHAR -2 DECODED-RULE ... CHAR
2119 When the composition end is found, LENGTH and NCHARS in the
2120 annotation header is updated as below:
2122 (1) LENGTH: unchanged, NCHARS: unchanged
2123 (2) LENGTH: length of the whole sequence minus NCHARS, NCHARS: unchanged
2124 (3) LENGTH: length of the whole sequence minus NCHARS, NCHARS: unchanged
2125 (4) LENGTH: unchanged, NCHARS: number of CHARs
2126 (5) LENGTH: unchanged, NCHARS: number of CHARs
2128 If an error is found while composing, the annotation header is
2129 changed to the original composition header (plus filler -1s) as
2132 (1),(2),(3) [ 0x80 0xF2+METHOD BYTES CHARS -1 ]
2133 (5) [ 0x80 0xFF -1 -1- -1 ]
2135 and the sequence [ -2 DECODED-RULE ] is changed to the original
2136 byte sequence as below:
2137 o the original byte sequence is B: [ B -1 ]
2138 o the original byte sequence is B1 B2: [ B1 B2 ]
2140 Most of the routines are implemented by macros because many
2141 variables and labels in the caller decode_coding_emacs_mule must be
2142 accessible, and they are usually called just once (thus doesn't
2143 increase the size of compiled object). */
2145 /* Decode a composition rule represented by C as a component of
2146 composition sequence of Emacs 20 style. Set RULE to the decoded
2149 #define DECODE_EMACS_MULE_COMPOSITION_RULE_20(c, rule) \
2154 if (c < 0 || c >= 81) \
2155 goto invalid_code; \
2156 gref = c / 9, nref = c % 9; \
2157 if (gref == 4) gref = 10; \
2158 if (nref == 4) nref = 10; \
2159 rule = COMPOSITION_ENCODE_RULE (gref, nref); \
2163 /* Decode a composition rule represented by C and the following byte
2164 at SRC as a component of composition sequence of Emacs 21 style.
2165 Set RULE to the decoded rule. */
2167 #define DECODE_EMACS_MULE_COMPOSITION_RULE_21(c, rule) \
2172 if (gref < 0 || gref >= 81) \
2173 goto invalid_code; \
2174 ONE_MORE_BYTE (c); \
2176 if (nref < 0 || nref >= 81) \
2177 goto invalid_code; \
2178 rule = COMPOSITION_ENCODE_RULE (gref, nref); \
2182 /* Start of Emacs 21 style format. The first three bytes at SRC are
2183 (METHOD - 0xF2), (BYTES - 0xA0), (CHARS - 0xA0), where BYTES is the
2184 byte length of this composition information, CHARS is the number of
2185 characters composed by this composition. */
2187 #define DECODE_EMACS_MULE_21_COMPOSITION() \
2189 enum composition_method method = c - 0xF2; \
2190 int nbytes, nchars; \
2192 ONE_MORE_BYTE (c); \
2194 goto invalid_code; \
2195 nbytes = c - 0xA0; \
2196 if (nbytes < 3 || (method == COMPOSITION_RELATIVE && nbytes != 4)) \
2197 goto invalid_code; \
2198 ONE_MORE_BYTE (c); \
2199 nchars = c - 0xA0; \
2200 if (nchars <= 0 || nchars >= MAX_COMPOSITION_COMPONENTS) \
2201 goto invalid_code; \
2202 cmp_status->old_form = 0; \
2203 cmp_status->method = method; \
2204 if (method == COMPOSITION_RELATIVE) \
2205 cmp_status->state = COMPOSING_CHAR; \
2207 cmp_status->state = COMPOSING_COMPONENT_CHAR; \
2208 cmp_status->length = MAX_ANNOTATION_LENGTH; \
2209 cmp_status->nchars = nchars; \
2210 cmp_status->ncomps = nbytes - 4; \
2211 ADD_COMPOSITION_DATA (charbuf, nchars, nbytes, method); \
2215 /* Start of Emacs 20 style format for relative composition. */
2217 #define DECODE_EMACS_MULE_20_RELATIVE_COMPOSITION() \
2219 cmp_status->old_form = 1; \
2220 cmp_status->method = COMPOSITION_RELATIVE; \
2221 cmp_status->state = COMPOSING_CHAR; \
2222 cmp_status->length = MAX_ANNOTATION_LENGTH; \
2223 cmp_status->nchars = cmp_status->ncomps = 0; \
2224 ADD_COMPOSITION_DATA (charbuf, 0, 0, cmp_status->method); \
2228 /* Start of Emacs 20 style format for rule-base composition. */
2230 #define DECODE_EMACS_MULE_20_RULEBASE_COMPOSITION() \
2232 cmp_status->old_form = 1; \
2233 cmp_status->method = COMPOSITION_WITH_RULE; \
2234 cmp_status->state = COMPOSING_CHAR; \
2235 cmp_status->length = MAX_ANNOTATION_LENGTH; \
2236 cmp_status->nchars = cmp_status->ncomps = 0; \
2237 ADD_COMPOSITION_DATA (charbuf, 0, 0, cmp_status->method); \
2241 #define DECODE_EMACS_MULE_COMPOSITION_START() \
2243 const unsigned char *current_src = src; \
2245 ONE_MORE_BYTE (c); \
2247 goto invalid_code; \
2248 if (c - 0xF2 >= COMPOSITION_RELATIVE \
2249 && c - 0xF2 <= COMPOSITION_WITH_RULE_ALTCHARS) \
2250 DECODE_EMACS_MULE_21_COMPOSITION (); \
2251 else if (c < 0xA0) \
2252 goto invalid_code; \
2253 else if (c < 0xC0) \
2255 DECODE_EMACS_MULE_20_RELATIVE_COMPOSITION (); \
2256 /* Re-read C as a composition component. */ \
2257 src = current_src; \
2259 else if (c == 0xFF) \
2260 DECODE_EMACS_MULE_20_RULEBASE_COMPOSITION (); \
2262 goto invalid_code; \
2265 #define EMACS_MULE_COMPOSITION_END() \
2267 int idx = - cmp_status->length; \
2269 if (cmp_status->old_form) \
2270 charbuf[idx + 2] = cmp_status->nchars; \
2271 else if (cmp_status->method > COMPOSITION_RELATIVE) \
2272 charbuf[idx] = charbuf[idx + 2] - cmp_status->length; \
2273 cmp_status->state = COMPOSING_NO; \
2278 emacs_mule_finish_composition (int *charbuf
,
2279 struct composition_status
*cmp_status
)
2281 int idx
= - cmp_status
->length
;
2284 if (cmp_status
->old_form
&& cmp_status
->nchars
> 0)
2286 charbuf
[idx
+ 2] = cmp_status
->nchars
;
2288 if (cmp_status
->method
== COMPOSITION_WITH_RULE
2289 && cmp_status
->state
== COMPOSING_CHAR
)
2291 /* The last rule was invalid. */
2292 int rule
= charbuf
[-1] + 0xA0;
2294 charbuf
[-2] = BYTE8_TO_CHAR (rule
);
2301 charbuf
[idx
++] = BYTE8_TO_CHAR (0x80);
2303 if (cmp_status
->method
== COMPOSITION_WITH_RULE
)
2305 charbuf
[idx
++] = BYTE8_TO_CHAR (0xFF);
2306 charbuf
[idx
++] = -3;
2312 int nchars
= charbuf
[idx
+ 1] + 0xA0;
2313 int nbytes
= charbuf
[idx
+ 2] + 0xA0;
2315 charbuf
[idx
++] = BYTE8_TO_CHAR (0xF2 + cmp_status
->method
);
2316 charbuf
[idx
++] = BYTE8_TO_CHAR (nbytes
);
2317 charbuf
[idx
++] = BYTE8_TO_CHAR (nchars
);
2318 charbuf
[idx
++] = -1;
2322 cmp_status
->state
= COMPOSING_NO
;
2326 #define EMACS_MULE_MAYBE_FINISH_COMPOSITION() \
2328 if (cmp_status->state != COMPOSING_NO) \
2329 char_offset += emacs_mule_finish_composition (charbuf, cmp_status); \
2334 decode_coding_emacs_mule (struct coding_system
*coding
)
2336 const unsigned char *src
= coding
->source
+ coding
->consumed
;
2337 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
2338 const unsigned char *src_base
;
2339 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
2340 /* We may produce two annotations (charset and composition) in one
2341 loop and one more charset annotation at the end. */
2343 = coding
->charbuf
+ coding
->charbuf_size
- (MAX_ANNOTATION_LENGTH
* 3)
2344 /* We can produce up to 2 characters in a loop. */
2346 ptrdiff_t consumed_chars
= 0, consumed_chars_base
;
2347 bool multibytep
= coding
->src_multibyte
;
2348 ptrdiff_t char_offset
= coding
->produced_char
;
2349 ptrdiff_t last_offset
= char_offset
;
2350 int last_id
= charset_ascii
;
2352 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
2353 int byte_after_cr
= -1;
2354 struct composition_status
*cmp_status
= &coding
->spec
.emacs_mule
.cmp_status
;
2356 if (cmp_status
->state
!= COMPOSING_NO
)
2360 if (charbuf_end
- charbuf
< cmp_status
->length
)
2362 for (i
= 0; i
< cmp_status
->length
; i
++)
2363 *charbuf
++ = cmp_status
->carryover
[i
];
2364 coding
->annotated
= 1;
2369 int c
, id
IF_LINT (= 0);
2372 consumed_chars_base
= consumed_chars
;
2374 if (charbuf
>= charbuf_end
)
2376 if (byte_after_cr
>= 0)
2381 if (byte_after_cr
>= 0)
2382 c
= byte_after_cr
, byte_after_cr
= -1;
2386 if (c
< 0 || c
== 0x80)
2388 EMACS_MULE_MAYBE_FINISH_COMPOSITION ();
2395 DECODE_EMACS_MULE_COMPOSITION_START ();
2401 if (eol_dos
&& c
== '\r')
2402 ONE_MORE_BYTE (byte_after_cr
);
2404 if (cmp_status
->state
!= COMPOSING_NO
)
2406 if (cmp_status
->old_form
)
2407 EMACS_MULE_MAYBE_FINISH_COMPOSITION ();
2408 else if (cmp_status
->state
>= COMPOSING_COMPONENT_CHAR
)
2409 cmp_status
->ncomps
--;
2414 int nchars
IF_LINT (= 0), nbytes
IF_LINT (= 0);
2415 /* emacs_mule_char can load a charset map from a file, which
2416 allocates a large structure and might cause buffer text
2417 to be relocated as result. Thus, we need to remember the
2418 original pointer to buffer text, and fix up all related
2419 pointers after the call. */
2420 const unsigned char *orig
= coding
->source
;
2423 c
= emacs_mule_char (coding
, src_base
, &nbytes
, &nchars
, &id
,
2425 offset
= coding
->source
- orig
;
2439 src
= src_base
+ nbytes
;
2440 consumed_chars
= consumed_chars_base
+ nchars
;
2441 if (cmp_status
->state
>= COMPOSING_COMPONENT_CHAR
)
2442 cmp_status
->ncomps
-= nchars
;
2445 /* Now if C >= 0, we found a normally encoded character, if C <
2446 0, we found an old-style composition component character or
2449 if (cmp_status
->state
== COMPOSING_NO
)
2453 if (last_id
!= charset_ascii
)
2454 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
,
2457 last_offset
= char_offset
;
2462 else if (cmp_status
->state
== COMPOSING_CHAR
)
2464 if (cmp_status
->old_form
)
2468 EMACS_MULE_MAYBE_FINISH_COMPOSITION ();
2475 cmp_status
->nchars
++;
2476 cmp_status
->length
++;
2477 if (cmp_status
->nchars
== MAX_COMPOSITION_COMPONENTS
)
2478 EMACS_MULE_COMPOSITION_END ();
2479 else if (cmp_status
->method
== COMPOSITION_WITH_RULE
)
2480 cmp_status
->state
= COMPOSING_RULE
;
2486 cmp_status
->length
++;
2487 cmp_status
->nchars
--;
2488 if (cmp_status
->nchars
== 0)
2489 EMACS_MULE_COMPOSITION_END ();
2492 else if (cmp_status
->state
== COMPOSING_RULE
)
2498 EMACS_MULE_COMPOSITION_END ();
2505 DECODE_EMACS_MULE_COMPOSITION_RULE_20 (c
, rule
);
2510 cmp_status
->length
+= 2;
2511 cmp_status
->state
= COMPOSING_CHAR
;
2514 else if (cmp_status
->state
== COMPOSING_COMPONENT_CHAR
)
2517 cmp_status
->length
++;
2518 if (cmp_status
->ncomps
== 0)
2519 cmp_status
->state
= COMPOSING_CHAR
;
2520 else if (cmp_status
->ncomps
> 0)
2522 if (cmp_status
->method
== COMPOSITION_WITH_RULE_ALTCHARS
)
2523 cmp_status
->state
= COMPOSING_COMPONENT_RULE
;
2526 EMACS_MULE_MAYBE_FINISH_COMPOSITION ();
2528 else /* COMPOSING_COMPONENT_RULE */
2532 DECODE_EMACS_MULE_COMPOSITION_RULE_21 (c
, rule
);
2537 cmp_status
->length
+= 2;
2538 cmp_status
->ncomps
--;
2539 if (cmp_status
->ncomps
> 0)
2540 cmp_status
->state
= COMPOSING_COMPONENT_CHAR
;
2542 EMACS_MULE_MAYBE_FINISH_COMPOSITION ();
2547 EMACS_MULE_MAYBE_FINISH_COMPOSITION ();
2549 consumed_chars
= consumed_chars_base
;
2551 *charbuf
++ = ASCII_CHAR_P (c
) ? c
: BYTE8_TO_CHAR (c
);
2556 if (cmp_status
->state
!= COMPOSING_NO
)
2558 if (coding
->mode
& CODING_MODE_LAST_BLOCK
)
2559 EMACS_MULE_MAYBE_FINISH_COMPOSITION ();
2564 charbuf
-= cmp_status
->length
;
2565 for (i
= 0; i
< cmp_status
->length
; i
++)
2566 cmp_status
->carryover
[i
] = charbuf
[i
];
2569 if (last_id
!= charset_ascii
)
2570 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
2571 coding
->consumed_char
+= consumed_chars_base
;
2572 coding
->consumed
= src_base
- coding
->source
;
2573 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
2577 #define EMACS_MULE_LEADING_CODES(id, codes) \
2580 codes[0] = id, codes[1] = 0; \
2581 else if (id < 0xE0) \
2582 codes[0] = 0x9A, codes[1] = id; \
2583 else if (id < 0xF0) \
2584 codes[0] = 0x9B, codes[1] = id; \
2585 else if (id < 0xF5) \
2586 codes[0] = 0x9C, codes[1] = id; \
2588 codes[0] = 0x9D, codes[1] = id; \
2593 encode_coding_emacs_mule (struct coding_system
*coding
)
2595 bool multibytep
= coding
->dst_multibyte
;
2596 int *charbuf
= coding
->charbuf
;
2597 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
2598 unsigned char *dst
= coding
->destination
+ coding
->produced
;
2599 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
2601 ptrdiff_t produced_chars
= 0;
2602 Lisp_Object attrs
, charset_list
;
2604 int preferred_charset_id
= -1;
2606 CODING_GET_INFO (coding
, attrs
, charset_list
);
2607 if (! EQ (charset_list
, Vemacs_mule_charset_list
))
2609 charset_list
= Vemacs_mule_charset_list
;
2610 ASET (attrs
, coding_attr_charset_list
, charset_list
);
2613 while (charbuf
< charbuf_end
)
2615 ASSURE_DESTINATION (safe_room
);
2620 /* Handle an annotation. */
2623 case CODING_ANNOTATE_COMPOSITION_MASK
:
2624 /* Not yet implemented. */
2626 case CODING_ANNOTATE_CHARSET_MASK
:
2627 preferred_charset_id
= charbuf
[3];
2628 if (preferred_charset_id
>= 0
2629 && NILP (Fmemq (make_number (preferred_charset_id
),
2631 preferred_charset_id
= -1;
2640 if (ASCII_CHAR_P (c
))
2641 EMIT_ONE_ASCII_BYTE (c
);
2642 else if (CHAR_BYTE8_P (c
))
2644 c
= CHAR_TO_BYTE8 (c
);
2649 struct charset
*charset
;
2653 unsigned char leading_codes
[2];
2655 if (preferred_charset_id
>= 0)
2659 charset
= CHARSET_FROM_ID (preferred_charset_id
);
2660 CODING_CHAR_CHARSET_P (coding
, dst
, dst_end
, c
, charset
, result
);
2662 code
= ENCODE_CHAR (charset
, c
);
2664 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
2668 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
2672 c
= coding
->default_char
;
2673 if (ASCII_CHAR_P (c
))
2675 EMIT_ONE_ASCII_BYTE (c
);
2678 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
2681 dimension
= CHARSET_DIMENSION (charset
);
2682 emacs_mule_id
= CHARSET_EMACS_MULE_ID (charset
);
2683 EMACS_MULE_LEADING_CODES (emacs_mule_id
, leading_codes
);
2684 EMIT_ONE_BYTE (leading_codes
[0]);
2685 if (leading_codes
[1])
2686 EMIT_ONE_BYTE (leading_codes
[1]);
2688 EMIT_ONE_BYTE (code
| 0x80);
2692 EMIT_ONE_BYTE (code
>> 8);
2693 EMIT_ONE_BYTE (code
& 0xFF);
2697 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
2698 coding
->produced_char
+= produced_chars
;
2699 coding
->produced
= dst
- coding
->destination
;
2704 /*** 7. ISO2022 handlers ***/
2706 /* The following note describes the coding system ISO2022 briefly.
2707 Since the intention of this note is to help understand the
2708 functions in this file, some parts are NOT ACCURATE or are OVERLY
2709 SIMPLIFIED. For thorough understanding, please refer to the
2710 original document of ISO2022. This is equivalent to the standard
2711 ECMA-35, obtainable from <URL:http://www.ecma.ch/> (*).
2713 ISO2022 provides many mechanisms to encode several character sets
2714 in 7-bit and 8-bit environments. For 7-bit environments, all text
2715 is encoded using bytes less than 128. This may make the encoded
2716 text a little bit longer, but the text passes more easily through
2717 several types of gateway, some of which strip off the MSB (Most
2720 There are two kinds of character sets: control character sets and
2721 graphic character sets. The former contain control characters such
2722 as `newline' and `escape' to provide control functions (control
2723 functions are also provided by escape sequences). The latter
2724 contain graphic characters such as 'A' and '-'. Emacs recognizes
2725 two control character sets and many graphic character sets.
2727 Graphic character sets are classified into one of the following
2728 four classes, according to the number of bytes (DIMENSION) and
2729 number of characters in one dimension (CHARS) of the set:
2730 - DIMENSION1_CHARS94
2731 - DIMENSION1_CHARS96
2732 - DIMENSION2_CHARS94
2733 - DIMENSION2_CHARS96
2735 In addition, each character set is assigned an identification tag,
2736 unique for each set, called the "final character" (denoted as <F>
2737 hereafter). The <F> of each character set is decided by ECMA(*)
2738 when it is registered in ISO. The code range of <F> is 0x30..0x7F
2739 (0x30..0x3F are for private use only).
2741 Note (*): ECMA = European Computer Manufacturers Association
2743 Here are examples of graphic character sets [NAME(<F>)]:
2744 o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ...
2745 o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ...
2746 o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ...
2747 o DIMENSION2_CHARS96 -- none for the moment
2749 A code area (1 byte=8 bits) is divided into 4 areas, C0, GL, C1, and GR.
2750 C0 [0x00..0x1F] -- control character plane 0
2751 GL [0x20..0x7F] -- graphic character plane 0
2752 C1 [0x80..0x9F] -- control character plane 1
2753 GR [0xA0..0xFF] -- graphic character plane 1
2755 A control character set is directly designated and invoked to C0 or
2756 C1 by an escape sequence. The most common case is that:
2757 - ISO646's control character set is designated/invoked to C0, and
2758 - ISO6429's control character set is designated/invoked to C1,
2759 and usually these designations/invocations are omitted in encoded
2760 text. In a 7-bit environment, only C0 can be used, and a control
2761 character for C1 is encoded by an appropriate escape sequence to
2762 fit into the environment. All control characters for C1 are
2763 defined to have corresponding escape sequences.
2765 A graphic character set is at first designated to one of four
2766 graphic registers (G0 through G3), then these graphic registers are
2767 invoked to GL or GR. These designations and invocations can be
2768 done independently. The most common case is that G0 is invoked to
2769 GL, G1 is invoked to GR, and ASCII is designated to G0. Usually
2770 these invocations and designations are omitted in encoded text.
2771 In a 7-bit environment, only GL can be used.
2773 When a graphic character set of CHARS94 is invoked to GL, codes
2774 0x20 and 0x7F of the GL area work as control characters SPACE and
2775 DEL respectively, and codes 0xA0 and 0xFF of the GR area should not
2778 There are two ways of invocation: locking-shift and single-shift.
2779 With locking-shift, the invocation lasts until the next different
2780 invocation, whereas with single-shift, the invocation affects the
2781 following character only and doesn't affect the locking-shift
2782 state. Invocations are done by the following control characters or
2785 ----------------------------------------------------------------------
2786 abbrev function cntrl escape seq description
2787 ----------------------------------------------------------------------
2788 SI/LS0 (shift-in) 0x0F none invoke G0 into GL
2789 SO/LS1 (shift-out) 0x0E none invoke G1 into GL
2790 LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL
2791 LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL
2792 LS1R (locking-shift-1 right) none ESC '~' invoke G1 into GR (*)
2793 LS2R (locking-shift-2 right) none ESC '}' invoke G2 into GR (*)
2794 LS3R (locking-shift 3 right) none ESC '|' invoke G3 into GR (*)
2795 SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 for one char
2796 SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 for one char
2797 ----------------------------------------------------------------------
2798 (*) These are not used by any known coding system.
2800 Control characters for these functions are defined by macros
2801 ISO_CODE_XXX in `coding.h'.
2803 Designations are done by the following escape sequences:
2804 ----------------------------------------------------------------------
2805 escape sequence description
2806 ----------------------------------------------------------------------
2807 ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0
2808 ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1
2809 ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2
2810 ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3
2811 ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*)
2812 ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1
2813 ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2
2814 ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3
2815 ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**)
2816 ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1
2817 ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2
2818 ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3
2819 ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*)
2820 ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1
2821 ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2
2822 ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3
2823 ----------------------------------------------------------------------
2825 In this list, "DIMENSION1_CHARS94<F>" means a graphic character set
2826 of dimension 1, chars 94, and final character <F>, etc...
2828 Note (*): Although these designations are not allowed in ISO2022,
2829 Emacs accepts them on decoding, and produces them on encoding
2830 CHARS96 character sets in a coding system which is characterized as
2831 7-bit environment, non-locking-shift, and non-single-shift.
2833 Note (**): If <F> is '@', 'A', or 'B', the intermediate character
2834 '(' must be omitted. We refer to this as "short-form" hereafter.
2836 Now you may notice that there are a lot of ways of encoding the
2837 same multilingual text in ISO2022. Actually, there exist many
2838 coding systems such as Compound Text (used in X11's inter client
2839 communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR
2840 (used in Korean Internet), EUC (Extended UNIX Code, used in Asian
2841 localized platforms), and all of these are variants of ISO2022.
2843 In addition to the above, Emacs handles two more kinds of escape
2844 sequences: ISO6429's direction specification and Emacs' private
2845 sequence for specifying character composition.
2847 ISO6429's direction specification takes the following form:
2848 o CSI ']' -- end of the current direction
2849 o CSI '0' ']' -- end of the current direction
2850 o CSI '1' ']' -- start of left-to-right text
2851 o CSI '2' ']' -- start of right-to-left text
2852 The control character CSI (0x9B: control sequence introducer) is
2853 abbreviated to the escape sequence ESC '[' in a 7-bit environment.
2855 Character composition specification takes the following form:
2856 o ESC '0' -- start relative composition
2857 o ESC '1' -- end composition
2858 o ESC '2' -- start rule-base composition (*)
2859 o ESC '3' -- start relative composition with alternate chars (**)
2860 o ESC '4' -- start rule-base composition with alternate chars (**)
2861 Since these are not standard escape sequences of any ISO standard,
2862 the use of them with these meanings is restricted to Emacs only.
2864 (*) This form is used only in Emacs 20.7 and older versions,
2865 but newer versions can safely decode it.
2866 (**) This form is used only in Emacs 21.1 and newer versions,
2867 and older versions can't decode it.
2869 Here's a list of example usages of these composition escape
2870 sequences (categorized by `enum composition_method').
2872 COMPOSITION_RELATIVE:
2873 ESC 0 CHAR [ CHAR ] ESC 1
2874 COMPOSITION_WITH_RULE:
2875 ESC 2 CHAR [ RULE CHAR ] ESC 1
2876 COMPOSITION_WITH_ALTCHARS:
2877 ESC 3 ALTCHAR [ ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1
2878 COMPOSITION_WITH_RULE_ALTCHARS:
2879 ESC 4 ALTCHAR [ RULE ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1 */
2881 static enum iso_code_class_type iso_code_class
[256];
2883 #define SAFE_CHARSET_P(coding, id) \
2884 ((id) <= (coding)->max_charset_id \
2885 && (coding)->safe_charsets[id] != 255)
2888 setup_iso_safe_charsets (Lisp_Object attrs
)
2890 Lisp_Object charset_list
, safe_charsets
;
2891 Lisp_Object request
;
2892 Lisp_Object reg_usage
;
2894 EMACS_INT reg94
, reg96
;
2895 int flags
= XINT (AREF (attrs
, coding_attr_iso_flags
));
2898 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
2899 if ((flags
& CODING_ISO_FLAG_FULL_SUPPORT
)
2900 && ! EQ (charset_list
, Viso_2022_charset_list
))
2902 charset_list
= Viso_2022_charset_list
;
2903 ASET (attrs
, coding_attr_charset_list
, charset_list
);
2904 ASET (attrs
, coding_attr_safe_charsets
, Qnil
);
2907 if (STRINGP (AREF (attrs
, coding_attr_safe_charsets
)))
2911 for (tail
= charset_list
; CONSP (tail
); tail
= XCDR (tail
))
2913 int id
= XINT (XCAR (tail
));
2914 if (max_charset_id
< id
)
2915 max_charset_id
= id
;
2918 safe_charsets
= make_uninit_string (max_charset_id
+ 1);
2919 memset (SDATA (safe_charsets
), 255, max_charset_id
+ 1);
2920 request
= AREF (attrs
, coding_attr_iso_request
);
2921 reg_usage
= AREF (attrs
, coding_attr_iso_usage
);
2922 reg94
= XINT (XCAR (reg_usage
));
2923 reg96
= XINT (XCDR (reg_usage
));
2925 for (tail
= charset_list
; CONSP (tail
); tail
= XCDR (tail
))
2929 struct charset
*charset
;
2932 charset
= CHARSET_FROM_ID (XINT (id
));
2933 reg
= Fcdr (Fassq (id
, request
));
2935 SSET (safe_charsets
, XINT (id
), XINT (reg
));
2936 else if (charset
->iso_chars_96
)
2939 SSET (safe_charsets
, XINT (id
), reg96
);
2944 SSET (safe_charsets
, XINT (id
), reg94
);
2947 ASET (attrs
, coding_attr_safe_charsets
, safe_charsets
);
2951 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
2952 Return true if a text is encoded in one of ISO-2022 based coding
2956 detect_coding_iso_2022 (struct coding_system
*coding
,
2957 struct coding_detection_info
*detect_info
)
2959 const unsigned char *src
= coding
->source
, *src_base
= src
;
2960 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
2961 bool multibytep
= coding
->src_multibyte
;
2962 bool single_shifting
= 0;
2965 ptrdiff_t consumed_chars
= 0;
2969 int composition_count
= -1;
2971 detect_info
->checked
|= CATEGORY_MASK_ISO
;
2973 for (i
= coding_category_iso_7
; i
<= coding_category_iso_8_else
; i
++)
2975 struct coding_system
*this = &(coding_categories
[i
]);
2976 Lisp_Object attrs
, val
;
2980 attrs
= CODING_ID_ATTRS (this->id
);
2981 if (CODING_ISO_FLAGS (this) & CODING_ISO_FLAG_FULL_SUPPORT
2982 && ! EQ (CODING_ATTR_CHARSET_LIST (attrs
), Viso_2022_charset_list
))
2983 setup_iso_safe_charsets (attrs
);
2984 val
= CODING_ATTR_SAFE_CHARSETS (attrs
);
2985 this->max_charset_id
= SCHARS (val
) - 1;
2986 this->safe_charsets
= SDATA (val
);
2989 /* A coding system of this category is always ASCII compatible. */
2990 src
+= coding
->head_ascii
;
2992 while (rejected
!= CATEGORY_MASK_ISO
)
2999 if (inhibit_iso_escape_detection
)
3001 single_shifting
= 0;
3003 if (c
== 'N' || c
== 'O')
3005 /* ESC <Fe> for SS2 or SS3. */
3006 single_shifting
= 1;
3007 rejected
|= CATEGORY_MASK_ISO_7BIT
| CATEGORY_MASK_ISO_8BIT
;
3011 /* End of composition. */
3012 if (composition_count
< 0
3013 || composition_count
> MAX_COMPOSITION_COMPONENTS
)
3016 composition_count
= -1;
3017 found
|= CATEGORY_MASK_ISO
;
3019 else if (c
>= '0' && c
<= '4')
3021 /* ESC <Fp> for start/end composition. */
3022 composition_count
= 0;
3026 if (c
>= '(' && c
<= '/')
3028 /* Designation sequence for a charset of dimension 1. */
3030 if (c1
< ' ' || c1
>= 0x80
3031 || (id
= iso_charset_table
[0][c
>= ','][c1
]) < 0)
3033 /* Invalid designation sequence. Just ignore. */
3035 rejected
|= (CATEGORY_MASK_ISO_7BIT
3036 | CATEGORY_MASK_ISO_7_ELSE
);
3042 /* Designation sequence for a charset of dimension 2. */
3044 if (c
>= '@' && c
<= 'B')
3045 /* Designation for JISX0208.1978, GB2312, or JISX0208. */
3046 id
= iso_charset_table
[1][0][c
];
3047 else if (c
>= '(' && c
<= '/')
3050 if (c1
< ' ' || c1
>= 0x80
3051 || (id
= iso_charset_table
[1][c
>= ','][c1
]) < 0)
3053 /* Invalid designation sequence. Just ignore. */
3055 rejected
|= (CATEGORY_MASK_ISO_7BIT
3056 | CATEGORY_MASK_ISO_7_ELSE
);
3062 /* Invalid designation sequence. Just ignore it. */
3064 rejected
|= (CATEGORY_MASK_ISO_7BIT
3065 | CATEGORY_MASK_ISO_7_ELSE
);
3071 /* Invalid escape sequence. Just ignore it. */
3073 rejected
|= (CATEGORY_MASK_ISO_7BIT
3074 | CATEGORY_MASK_ISO_7_ELSE
);
3078 /* We found a valid designation sequence for CHARSET. */
3079 rejected
|= CATEGORY_MASK_ISO_8BIT
;
3080 if (SAFE_CHARSET_P (&coding_categories
[coding_category_iso_7
],
3082 found
|= CATEGORY_MASK_ISO_7
;
3084 rejected
|= CATEGORY_MASK_ISO_7
;
3085 if (SAFE_CHARSET_P (&coding_categories
[coding_category_iso_7_tight
],
3087 found
|= CATEGORY_MASK_ISO_7_TIGHT
;
3089 rejected
|= CATEGORY_MASK_ISO_7_TIGHT
;
3090 if (SAFE_CHARSET_P (&coding_categories
[coding_category_iso_7_else
],
3092 found
|= CATEGORY_MASK_ISO_7_ELSE
;
3094 rejected
|= CATEGORY_MASK_ISO_7_ELSE
;
3095 if (SAFE_CHARSET_P (&coding_categories
[coding_category_iso_8_else
],
3097 found
|= CATEGORY_MASK_ISO_8_ELSE
;
3099 rejected
|= CATEGORY_MASK_ISO_8_ELSE
;
3105 /* Locking shift out/in. */
3106 if (inhibit_iso_escape_detection
)
3108 single_shifting
= 0;
3109 rejected
|= CATEGORY_MASK_ISO_7BIT
| CATEGORY_MASK_ISO_8BIT
;
3113 /* Control sequence introducer. */
3114 single_shifting
= 0;
3115 rejected
|= CATEGORY_MASK_ISO_7BIT
| CATEGORY_MASK_ISO_7_ELSE
;
3116 found
|= CATEGORY_MASK_ISO_8_ELSE
;
3117 goto check_extra_latin
;
3122 if (inhibit_iso_escape_detection
)
3124 single_shifting
= 0;
3125 rejected
|= CATEGORY_MASK_ISO_7BIT
| CATEGORY_MASK_ISO_7_ELSE
;
3126 if (CODING_ISO_FLAGS (&coding_categories
[coding_category_iso_8_1
])
3127 & CODING_ISO_FLAG_SINGLE_SHIFT
)
3129 found
|= CATEGORY_MASK_ISO_8_1
;
3130 single_shifting
= 1;
3132 if (CODING_ISO_FLAGS (&coding_categories
[coding_category_iso_8_2
])
3133 & CODING_ISO_FLAG_SINGLE_SHIFT
)
3135 found
|= CATEGORY_MASK_ISO_8_2
;
3136 single_shifting
= 1;
3138 if (single_shifting
)
3140 goto check_extra_latin
;
3147 if (composition_count
>= 0)
3148 composition_count
++;
3149 single_shifting
= 0;
3152 rejected
|= CATEGORY_MASK_ISO_7BIT
| CATEGORY_MASK_ISO_7_ELSE
;
3155 found
|= CATEGORY_MASK_ISO_8_1
;
3156 /* Check the length of succeeding codes of the range
3157 0xA0..0FF. If the byte length is even, we include
3158 CATEGORY_MASK_ISO_8_2 in `found'. We can check this
3159 only when we are not single shifting. */
3160 if (! single_shifting
3161 && ! (rejected
& CATEGORY_MASK_ISO_8_2
))
3164 while (src
< src_end
)
3176 if (len
& 1 && src
< src_end
)
3178 rejected
|= CATEGORY_MASK_ISO_8_2
;
3179 if (composition_count
>= 0)
3180 composition_count
+= len
;
3184 found
|= CATEGORY_MASK_ISO_8_2
;
3185 if (composition_count
>= 0)
3186 composition_count
+= len
/ 2;
3192 if (! VECTORP (Vlatin_extra_code_table
)
3193 || NILP (AREF (Vlatin_extra_code_table
, c
)))
3195 rejected
= CATEGORY_MASK_ISO
;
3198 if (CODING_ISO_FLAGS (&coding_categories
[coding_category_iso_8_1
])
3199 & CODING_ISO_FLAG_LATIN_EXTRA
)
3200 found
|= CATEGORY_MASK_ISO_8_1
;
3202 rejected
|= CATEGORY_MASK_ISO_8_1
;
3203 rejected
|= CATEGORY_MASK_ISO_8_2
;
3207 detect_info
->rejected
|= CATEGORY_MASK_ISO
;
3211 detect_info
->rejected
|= rejected
;
3212 detect_info
->found
|= (found
& ~rejected
);
3217 /* Set designation state into CODING. Set CHARS_96 to -1 if the
3218 escape sequence should be kept. */
3219 #define DECODE_DESIGNATION(reg, dim, chars_96, final) \
3223 if (final < '0' || final >= 128 \
3224 || ((id = ISO_CHARSET_TABLE (dim, chars_96, final)) < 0) \
3225 || !SAFE_CHARSET_P (coding, id)) \
3227 CODING_ISO_DESIGNATION (coding, reg) = -2; \
3231 prev = CODING_ISO_DESIGNATION (coding, reg); \
3232 if (id == charset_jisx0201_roman) \
3234 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_USE_ROMAN) \
3235 id = charset_ascii; \
3237 else if (id == charset_jisx0208_1978) \
3239 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_USE_OLDJIS) \
3240 id = charset_jisx0208; \
3242 CODING_ISO_DESIGNATION (coding, reg) = id; \
3243 /* If there was an invalid designation to REG previously, and this \
3244 designation is ASCII to REG, we should keep this designation \
3246 if (prev == -2 && id == charset_ascii) \
3251 /* Handle these composition sequence (ALT: alternate char):
3253 (1) relative composition: ESC 0 CHAR ... ESC 1
3254 (2) rulebase composition: ESC 2 CHAR RULE CHAR RULE ... CHAR ESC 1
3255 (3) altchar composition: ESC 3 ALT ... ALT ESC 0 CHAR ... ESC 1
3256 (4) alt&rule composition: ESC 4 ALT RULE ... ALT ESC 0 CHAR ... ESC 1
3258 When the start sequence (ESC 0/2/3/4) is found, this annotation
3261 [ -LENGTH(==-5) CODING_ANNOTATE_COMPOSITION_MASK NCHARS(==0) 0 METHOD ]
3263 Then, upon reading CHAR or RULE (one or two bytes), these codes are
3264 produced until the end sequence (ESC 1) is found:
3267 (2) CHAR -2 DECODED-RULE CHAR -2 DECODED-RULE ... CHAR
3268 (3) ALT ... ALT -1 -1 CHAR ... CHAR
3269 (4) ALT -2 DECODED-RULE ALT -2 DECODED-RULE ... ALT -1 -1 CHAR ... CHAR
3271 When the end sequence (ESC 1) is found, LENGTH and NCHARS in the
3272 annotation header is updated as below:
3274 (1) LENGTH: unchanged, NCHARS: number of CHARs
3275 (2) LENGTH: unchanged, NCHARS: number of CHARs
3276 (3) LENGTH: += number of ALTs + 2, NCHARS: number of CHARs
3277 (4) LENGTH: += number of ALTs * 3, NCHARS: number of CHARs
3279 If an error is found while composing, the annotation header is
3282 [ ESC '0'/'2'/'3'/'4' -2 0 ]
3284 and the sequence [ -2 DECODED-RULE ] is changed to the original
3285 byte sequence as below:
3286 o the original byte sequence is B: [ B -1 ]
3287 o the original byte sequence is B1 B2: [ B1 B2 ]
3288 and the sequence [ -1 -1 ] is changed to the original byte
3293 /* Decode a composition rule C1 and maybe one more byte from the
3294 source, and set RULE to the encoded composition rule. If the rule
3295 is invalid, goto invalid_code. */
3297 #define DECODE_COMPOSITION_RULE(rule) \
3301 goto invalid_code; \
3302 if (rule < 81) /* old format (before ver.21) */ \
3304 int gref = (rule) / 9; \
3305 int nref = (rule) % 9; \
3306 if (gref == 4) gref = 10; \
3307 if (nref == 4) nref = 10; \
3308 rule = COMPOSITION_ENCODE_RULE (gref, nref); \
3310 else /* new format (after ver.21) */ \
3314 ONE_MORE_BYTE (b); \
3315 if (! COMPOSITION_ENCODE_RULE_VALID (rule - 81, b - 32)) \
3316 goto invalid_code; \
3317 rule = COMPOSITION_ENCODE_RULE (rule - 81, b - 32); \
3318 rule += 0x100; /* Distinguish it from the old format. */ \
3322 #define ENCODE_COMPOSITION_RULE(rule) \
3324 int gref = (rule % 0x100) / 12, nref = (rule % 0x100) % 12; \
3326 if (rule < 0x100) /* old format */ \
3328 if (gref == 10) gref = 4; \
3329 if (nref == 10) nref = 4; \
3330 charbuf[idx] = 32 + gref * 9 + nref; \
3331 charbuf[idx + 1] = -1; \
3334 else /* new format */ \
3336 charbuf[idx] = 32 + 81 + gref; \
3337 charbuf[idx + 1] = 32 + nref; \
3342 /* Finish the current composition as invalid. */
3345 finish_composition (int *charbuf
, struct composition_status
*cmp_status
)
3347 int idx
= - cmp_status
->length
;
3350 /* Recover the original ESC sequence */
3351 charbuf
[idx
++] = ISO_CODE_ESC
;
3352 charbuf
[idx
++] = (cmp_status
->method
== COMPOSITION_RELATIVE
? '0'
3353 : cmp_status
->method
== COMPOSITION_WITH_RULE
? '2'
3354 : cmp_status
->method
== COMPOSITION_WITH_ALTCHARS
? '3'
3355 /* cmp_status->method == COMPOSITION_WITH_RULE_ALTCHARS */
3357 charbuf
[idx
++] = -2;
3359 charbuf
[idx
++] = -1;
3360 new_chars
= cmp_status
->nchars
;
3361 if (cmp_status
->method
>= COMPOSITION_WITH_RULE
)
3362 for (; idx
< 0; idx
++)
3364 int elt
= charbuf
[idx
];
3368 ENCODE_COMPOSITION_RULE (charbuf
[idx
+ 1]);
3373 charbuf
[idx
++] = ISO_CODE_ESC
;
3378 cmp_status
->state
= COMPOSING_NO
;
3382 /* If characters are under composition, finish the composition. */
3383 #define MAYBE_FINISH_COMPOSITION() \
3385 if (cmp_status->state != COMPOSING_NO) \
3386 char_offset += finish_composition (charbuf, cmp_status); \
3389 /* Handle composition start sequence ESC 0, ESC 2, ESC 3, or ESC 4.
3391 ESC 0 : relative composition : ESC 0 CHAR ... ESC 1
3392 ESC 2 : rulebase composition : ESC 2 CHAR RULE CHAR RULE ... CHAR ESC 1
3393 ESC 3 : altchar composition : ESC 3 CHAR ... ESC 0 CHAR ... ESC 1
3394 ESC 4 : alt&rule composition : ESC 4 CHAR RULE ... CHAR ESC 0 CHAR ... ESC 1
3396 Produce this annotation sequence now:
3398 [ -LENGTH(==-4) CODING_ANNOTATE_COMPOSITION_MASK NCHARS(==0) METHOD ]
3401 #define DECODE_COMPOSITION_START(c1) \
3404 && ((cmp_status->state == COMPOSING_COMPONENT_CHAR \
3405 && cmp_status->method == COMPOSITION_WITH_ALTCHARS) \
3406 || (cmp_status->state == COMPOSING_COMPONENT_RULE \
3407 && cmp_status->method == COMPOSITION_WITH_RULE_ALTCHARS))) \
3411 cmp_status->state = COMPOSING_CHAR; \
3412 cmp_status->length += 2; \
3416 MAYBE_FINISH_COMPOSITION (); \
3417 cmp_status->method = (c1 == '0' ? COMPOSITION_RELATIVE \
3418 : c1 == '2' ? COMPOSITION_WITH_RULE \
3419 : c1 == '3' ? COMPOSITION_WITH_ALTCHARS \
3420 : COMPOSITION_WITH_RULE_ALTCHARS); \
3422 = (c1 <= '2' ? COMPOSING_CHAR : COMPOSING_COMPONENT_CHAR); \
3423 ADD_COMPOSITION_DATA (charbuf, 0, 0, cmp_status->method); \
3424 cmp_status->length = MAX_ANNOTATION_LENGTH; \
3425 cmp_status->nchars = cmp_status->ncomps = 0; \
3426 coding->annotated = 1; \
3431 /* Handle composition end sequence ESC 1. */
3433 #define DECODE_COMPOSITION_END() \
3435 if (cmp_status->nchars == 0 \
3436 || ((cmp_status->state == COMPOSING_CHAR) \
3437 == (cmp_status->method == COMPOSITION_WITH_RULE))) \
3439 MAYBE_FINISH_COMPOSITION (); \
3440 goto invalid_code; \
3442 if (cmp_status->method == COMPOSITION_WITH_ALTCHARS) \
3443 charbuf[- cmp_status->length] -= cmp_status->ncomps + 2; \
3444 else if (cmp_status->method == COMPOSITION_WITH_RULE_ALTCHARS) \
3445 charbuf[- cmp_status->length] -= cmp_status->ncomps * 3; \
3446 charbuf[- cmp_status->length + 2] = cmp_status->nchars; \
3447 char_offset += cmp_status->nchars; \
3448 cmp_status->state = COMPOSING_NO; \
3451 /* Store a composition rule RULE in charbuf, and update cmp_status. */
3453 #define STORE_COMPOSITION_RULE(rule) \
3456 *charbuf++ = rule; \
3457 cmp_status->length += 2; \
3458 cmp_status->state--; \
3461 /* Store a composed char or a component char C in charbuf, and update
3464 #define STORE_COMPOSITION_CHAR(c) \
3467 cmp_status->length++; \
3468 if (cmp_status->state == COMPOSING_CHAR) \
3469 cmp_status->nchars++; \
3471 cmp_status->ncomps++; \
3472 if (cmp_status->method == COMPOSITION_WITH_RULE \
3473 || (cmp_status->method == COMPOSITION_WITH_RULE_ALTCHARS \
3474 && cmp_status->state == COMPOSING_COMPONENT_CHAR)) \
3475 cmp_status->state++; \
3479 /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
3482 decode_coding_iso_2022 (struct coding_system
*coding
)
3484 const unsigned char *src
= coding
->source
+ coding
->consumed
;
3485 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
3486 const unsigned char *src_base
;
3487 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
3488 /* We may produce two annotations (charset and composition) in one
3489 loop and one more charset annotation at the end. */
3491 = coding
->charbuf
+ coding
->charbuf_size
- (MAX_ANNOTATION_LENGTH
* 3);
3492 ptrdiff_t consumed_chars
= 0, consumed_chars_base
;
3493 bool multibytep
= coding
->src_multibyte
;
3494 /* Charsets invoked to graphic plane 0 and 1 respectively. */
3495 int charset_id_0
= CODING_ISO_INVOKED_CHARSET (coding
, 0);
3496 int charset_id_1
= CODING_ISO_INVOKED_CHARSET (coding
, 1);
3497 int charset_id_2
, charset_id_3
;
3498 struct charset
*charset
;
3500 struct composition_status
*cmp_status
= CODING_ISO_CMP_STATUS (coding
);
3501 Lisp_Object attrs
= CODING_ID_ATTRS (coding
->id
);
3502 ptrdiff_t char_offset
= coding
->produced_char
;
3503 ptrdiff_t last_offset
= char_offset
;
3504 int last_id
= charset_ascii
;
3506 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
3507 int byte_after_cr
= -1;
3510 setup_iso_safe_charsets (attrs
);
3511 coding
->safe_charsets
= SDATA (CODING_ATTR_SAFE_CHARSETS (attrs
));
3513 if (cmp_status
->state
!= COMPOSING_NO
)
3515 if (charbuf_end
- charbuf
< cmp_status
->length
)
3517 for (i
= 0; i
< cmp_status
->length
; i
++)
3518 *charbuf
++ = cmp_status
->carryover
[i
];
3519 coding
->annotated
= 1;
3527 consumed_chars_base
= consumed_chars
;
3529 if (charbuf
>= charbuf_end
)
3531 if (byte_after_cr
>= 0)
3536 if (byte_after_cr
>= 0)
3537 c1
= byte_after_cr
, byte_after_cr
= -1;
3543 if (CODING_ISO_EXTSEGMENT_LEN (coding
) > 0)
3545 *charbuf
++ = ASCII_CHAR_P (c1
) ? c1
: BYTE8_TO_CHAR (c1
);
3547 CODING_ISO_EXTSEGMENT_LEN (coding
)--;
3551 if (CODING_ISO_EMBEDDED_UTF_8 (coding
))
3553 if (c1
== ISO_CODE_ESC
)
3555 if (src
+ 1 >= src_end
)
3556 goto no_more_source
;
3557 *charbuf
++ = ISO_CODE_ESC
;
3559 if (src
[0] == '%' && src
[1] == '@')
3562 consumed_chars
+= 2;
3564 /* We are sure charbuf can contain two more chars. */
3567 CODING_ISO_EMBEDDED_UTF_8 (coding
) = 0;
3572 *charbuf
++ = ASCII_CHAR_P (c1
) ? c1
: BYTE8_TO_CHAR (c1
);
3578 if ((cmp_status
->state
== COMPOSING_RULE
3579 || cmp_status
->state
== COMPOSING_COMPONENT_RULE
)
3580 && c1
!= ISO_CODE_ESC
)
3584 DECODE_COMPOSITION_RULE (rule
);
3585 STORE_COMPOSITION_RULE (rule
);
3589 /* We produce at most one character. */
3590 switch (iso_code_class
[c1
])
3592 case ISO_0x20_or_0x7F
:
3593 if (charset_id_0
< 0
3594 || ! CHARSET_ISO_CHARS_96 (CHARSET_FROM_ID (charset_id_0
)))
3595 /* This is SPACE or DEL. */
3596 charset
= CHARSET_FROM_ID (charset_ascii
);
3598 charset
= CHARSET_FROM_ID (charset_id_0
);
3601 case ISO_graphic_plane_0
:
3602 if (charset_id_0
< 0)
3603 charset
= CHARSET_FROM_ID (charset_ascii
);
3605 charset
= CHARSET_FROM_ID (charset_id_0
);
3608 case ISO_0xA0_or_0xFF
:
3609 if (charset_id_1
< 0
3610 || ! CHARSET_ISO_CHARS_96 (CHARSET_FROM_ID (charset_id_1
))
3611 || CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SEVEN_BITS
)
3613 /* This is a graphic character, we fall down ... */
3615 case ISO_graphic_plane_1
:
3616 if (charset_id_1
< 0)
3618 charset
= CHARSET_FROM_ID (charset_id_1
);
3622 if (eol_dos
&& c1
== '\r')
3623 ONE_MORE_BYTE (byte_after_cr
);
3624 MAYBE_FINISH_COMPOSITION ();
3625 charset
= CHARSET_FROM_ID (charset_ascii
);
3632 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_LOCKING_SHIFT
)
3633 || CODING_ISO_DESIGNATION (coding
, 1) < 0)
3635 CODING_ISO_INVOCATION (coding
, 0) = 1;
3636 charset_id_0
= CODING_ISO_INVOKED_CHARSET (coding
, 0);
3640 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_LOCKING_SHIFT
))
3642 CODING_ISO_INVOCATION (coding
, 0) = 0;
3643 charset_id_0
= CODING_ISO_INVOKED_CHARSET (coding
, 0);
3646 case ISO_single_shift_2_7
:
3647 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SEVEN_BITS
))
3649 case ISO_single_shift_2
:
3650 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SINGLE_SHIFT
))
3652 /* SS2 is handled as an escape sequence of ESC 'N' */
3654 goto label_escape_sequence
;
3656 case ISO_single_shift_3
:
3657 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SINGLE_SHIFT
))
3659 /* SS2 is handled as an escape sequence of ESC 'O' */
3661 goto label_escape_sequence
;
3663 case ISO_control_sequence_introducer
:
3664 /* CSI is handled as an escape sequence of ESC '[' ... */
3666 goto label_escape_sequence
;
3670 label_escape_sequence
:
3671 /* Escape sequences handled here are invocation,
3672 designation, direction specification, and character
3673 composition specification. */
3676 case '&': /* revision of following character set */
3678 if (!(c1
>= '@' && c1
<= '~'))
3681 if (c1
!= ISO_CODE_ESC
)
3684 goto label_escape_sequence
;
3686 case '$': /* designation of 2-byte character set */
3687 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_DESIGNATION
))
3693 if (c1
>= '@' && c1
<= 'B')
3694 { /* designation of JISX0208.1978, GB2312.1980,
3696 reg
= 0, chars96
= 0;
3698 else if (c1
>= 0x28 && c1
<= 0x2B)
3699 { /* designation of DIMENSION2_CHARS94 character set */
3700 reg
= c1
- 0x28, chars96
= 0;
3703 else if (c1
>= 0x2C && c1
<= 0x2F)
3704 { /* designation of DIMENSION2_CHARS96 character set */
3705 reg
= c1
- 0x2C, chars96
= 1;
3710 DECODE_DESIGNATION (reg
, 2, chars96
, c1
);
3711 /* We must update these variables now. */
3713 charset_id_0
= CODING_ISO_INVOKED_CHARSET (coding
, 0);
3715 charset_id_1
= CODING_ISO_INVOKED_CHARSET (coding
, 1);
3721 case 'n': /* invocation of locking-shift-2 */
3722 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_LOCKING_SHIFT
)
3723 || CODING_ISO_DESIGNATION (coding
, 2) < 0)
3725 CODING_ISO_INVOCATION (coding
, 0) = 2;
3726 charset_id_0
= CODING_ISO_INVOKED_CHARSET (coding
, 0);
3729 case 'o': /* invocation of locking-shift-3 */
3730 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_LOCKING_SHIFT
)
3731 || CODING_ISO_DESIGNATION (coding
, 3) < 0)
3733 CODING_ISO_INVOCATION (coding
, 0) = 3;
3734 charset_id_0
= CODING_ISO_INVOKED_CHARSET (coding
, 0);
3737 case 'N': /* invocation of single-shift-2 */
3738 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SINGLE_SHIFT
)
3739 || CODING_ISO_DESIGNATION (coding
, 2) < 0)
3741 charset_id_2
= CODING_ISO_DESIGNATION (coding
, 2);
3742 if (charset_id_2
< 0)
3743 charset
= CHARSET_FROM_ID (charset_ascii
);
3745 charset
= CHARSET_FROM_ID (charset_id_2
);
3747 if (c1
< 0x20 || (c1
>= 0x80 && c1
< 0xA0)
3748 || (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SEVEN_BITS
)
3749 && ((CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_LEVEL_4
)
3750 ? c1
>= 0x80 : c1
< 0x80)))
3754 case 'O': /* invocation of single-shift-3 */
3755 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SINGLE_SHIFT
)
3756 || CODING_ISO_DESIGNATION (coding
, 3) < 0)
3758 charset_id_3
= CODING_ISO_DESIGNATION (coding
, 3);
3759 if (charset_id_3
< 0)
3760 charset
= CHARSET_FROM_ID (charset_ascii
);
3762 charset
= CHARSET_FROM_ID (charset_id_3
);
3764 if (c1
< 0x20 || (c1
>= 0x80 && c1
< 0xA0)
3765 || (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SEVEN_BITS
)
3766 && ((CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_LEVEL_4
)
3767 ? c1
>= 0x80 : c1
< 0x80)))
3771 case '0': case '2': case '3': case '4': /* start composition */
3772 if (! (coding
->common_flags
& CODING_ANNOTATE_COMPOSITION_MASK
))
3774 if (last_id
!= charset_ascii
)
3776 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
3777 last_id
= charset_ascii
;
3778 last_offset
= char_offset
;
3780 DECODE_COMPOSITION_START (c1
);
3783 case '1': /* end composition */
3784 if (cmp_status
->state
== COMPOSING_NO
)
3786 DECODE_COMPOSITION_END ();
3789 case '[': /* specification of direction */
3790 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_DIRECTION
))
3792 /* For the moment, nested direction is not supported.
3793 So, `coding->mode & CODING_MODE_DIRECTION' zero means
3794 left-to-right, and nonzero means right-to-left. */
3798 case ']': /* end of the current direction */
3799 coding
->mode
&= ~CODING_MODE_DIRECTION
;
3801 case '0': /* end of the current direction */
3802 case '1': /* start of left-to-right direction */
3805 coding
->mode
&= ~CODING_MODE_DIRECTION
;
3810 case '2': /* start of right-to-left direction */
3813 coding
->mode
|= CODING_MODE_DIRECTION
;
3827 /* CTEXT extended segment:
3828 ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
3829 We keep these bytes as is for the moment.
3830 They may be decoded by post-read-conversion. */
3834 ONE_MORE_BYTE (dim
);
3835 if (dim
< '0' || dim
> '4')
3843 size
= ((M
- 128) * 128) + (L
- 128);
3844 if (charbuf
+ 6 > charbuf_end
)
3846 *charbuf
++ = ISO_CODE_ESC
;
3850 *charbuf
++ = BYTE8_TO_CHAR (M
);
3851 *charbuf
++ = BYTE8_TO_CHAR (L
);
3852 CODING_ISO_EXTSEGMENT_LEN (coding
) = size
;
3856 /* XFree86 extension for embedding UTF-8 in CTEXT:
3857 ESC % G --UTF-8-BYTES-- ESC % @
3858 We keep these bytes as is for the moment.
3859 They may be decoded by post-read-conversion. */
3860 if (charbuf
+ 3 > charbuf_end
)
3862 *charbuf
++ = ISO_CODE_ESC
;
3865 CODING_ISO_EMBEDDED_UTF_8 (coding
) = 1;
3873 if (! (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_DESIGNATION
))
3878 if (c1
>= 0x28 && c1
<= 0x2B)
3879 { /* designation of DIMENSION1_CHARS94 character set */
3880 reg
= c1
- 0x28, chars96
= 0;
3883 else if (c1
>= 0x2C && c1
<= 0x2F)
3884 { /* designation of DIMENSION1_CHARS96 character set */
3885 reg
= c1
- 0x2C, chars96
= 1;
3890 DECODE_DESIGNATION (reg
, 1, chars96
, c1
);
3891 /* We must update these variables now. */
3893 charset_id_0
= CODING_ISO_INVOKED_CHARSET (coding
, 0);
3895 charset_id_1
= CODING_ISO_INVOKED_CHARSET (coding
, 1);
3907 if (cmp_status
->state
== COMPOSING_NO
3908 && charset
->id
!= charset_ascii
3909 && last_id
!= charset
->id
)
3911 if (last_id
!= charset_ascii
)
3912 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
3913 last_id
= charset
->id
;
3914 last_offset
= char_offset
;
3917 /* Now we know CHARSET and 1st position code C1 of a character.
3918 Produce a decoded character while getting 2nd and 3rd
3919 position codes C2, C3 if necessary. */
3920 if (CHARSET_DIMENSION (charset
) > 1)
3923 if (c2
< 0x20 || (c2
>= 0x80 && c2
< 0xA0)
3924 || ((c1
& 0x80) != (c2
& 0x80)))
3925 /* C2 is not in a valid range. */
3927 if (CHARSET_DIMENSION (charset
) == 2)
3928 c1
= (c1
<< 8) | c2
;
3932 if (c3
< 0x20 || (c3
>= 0x80 && c3
< 0xA0)
3933 || ((c1
& 0x80) != (c3
& 0x80)))
3934 /* C3 is not in a valid range. */
3936 c1
= (c1
<< 16) | (c2
<< 8) | c2
;
3940 CODING_DECODE_CHAR (coding
, src
, src_base
, src_end
, charset
, c1
, c
);
3943 MAYBE_FINISH_COMPOSITION ();
3944 for (; src_base
< src
; src_base
++, char_offset
++)
3946 if (ASCII_CHAR_P (*src_base
))
3947 *charbuf
++ = *src_base
;
3949 *charbuf
++ = BYTE8_TO_CHAR (*src_base
);
3952 else if (cmp_status
->state
== COMPOSING_NO
)
3957 else if ((cmp_status
->state
== COMPOSING_CHAR
3958 ? cmp_status
->nchars
3959 : cmp_status
->ncomps
)
3960 >= MAX_COMPOSITION_COMPONENTS
)
3962 /* Too long composition. */
3963 MAYBE_FINISH_COMPOSITION ();
3968 STORE_COMPOSITION_CHAR (c
);
3972 MAYBE_FINISH_COMPOSITION ();
3974 consumed_chars
= consumed_chars_base
;
3976 *charbuf
++ = c
< 0 ? -c
: ASCII_CHAR_P (c
) ? c
: BYTE8_TO_CHAR (c
);
3978 /* Reset the invocation and designation status to the safest
3979 one; i.e. designate ASCII to the graphic register 0, and
3980 invoke that register to the graphic plane 0. This typically
3981 helps the case that an designation sequence for ASCII "ESC (
3982 B" is somehow broken (e.g. broken by a newline). */
3983 CODING_ISO_INVOCATION (coding
, 0) = 0;
3984 CODING_ISO_DESIGNATION (coding
, 0) = charset_ascii
;
3985 charset_id_0
= charset_ascii
;
3993 if (cmp_status
->state
!= COMPOSING_NO
)
3995 if (coding
->mode
& CODING_MODE_LAST_BLOCK
)
3996 MAYBE_FINISH_COMPOSITION ();
3999 charbuf
-= cmp_status
->length
;
4000 for (i
= 0; i
< cmp_status
->length
; i
++)
4001 cmp_status
->carryover
[i
] = charbuf
[i
];
4004 else if (last_id
!= charset_ascii
)
4005 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
4006 coding
->consumed_char
+= consumed_chars_base
;
4007 coding
->consumed
= src_base
- coding
->source
;
4008 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
4012 /* ISO2022 encoding stuff. */
4015 It is not enough to say just "ISO2022" on encoding, we have to
4016 specify more details. In Emacs, each coding system of ISO2022
4017 variant has the following specifications:
4018 1. Initial designation to G0 thru G3.
4019 2. Allows short-form designation?
4020 3. ASCII should be designated to G0 before control characters?
4021 4. ASCII should be designated to G0 at end of line?
4022 5. 7-bit environment or 8-bit environment?
4023 6. Use locking-shift?
4024 7. Use Single-shift?
4025 And the following two are only for Japanese:
4026 8. Use ASCII in place of JIS0201-1976-Roman?
4027 9. Use JISX0208-1983 in place of JISX0208-1978?
4028 These specifications are encoded in CODING_ISO_FLAGS (coding) as flag bits
4029 defined by macros CODING_ISO_FLAG_XXX. See `coding.h' for more
4033 /* Produce codes (escape sequence) for designating CHARSET to graphic
4034 register REG at DST, and increment DST. If <final-char> of CHARSET is
4035 '@', 'A', or 'B' and the coding system CODING allows, produce
4036 designation sequence of short-form. */
4038 #define ENCODE_DESIGNATION(charset, reg, coding) \
4040 unsigned char final_char = CHARSET_ISO_FINAL (charset); \
4041 const char *intermediate_char_94 = "()*+"; \
4042 const char *intermediate_char_96 = ",-./"; \
4043 int revision = -1; \
4045 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_REVISION) \
4046 revision = CHARSET_ISO_REVISION (charset); \
4048 if (revision >= 0) \
4050 EMIT_TWO_ASCII_BYTES (ISO_CODE_ESC, '&'); \
4051 EMIT_ONE_BYTE ('@' + revision); \
4053 EMIT_ONE_ASCII_BYTE (ISO_CODE_ESC); \
4054 if (CHARSET_DIMENSION (charset) == 1) \
4057 if (! CHARSET_ISO_CHARS_96 (charset)) \
4058 b = intermediate_char_94[reg]; \
4060 b = intermediate_char_96[reg]; \
4061 EMIT_ONE_ASCII_BYTE (b); \
4065 EMIT_ONE_ASCII_BYTE ('$'); \
4066 if (! CHARSET_ISO_CHARS_96 (charset)) \
4068 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_LONG_FORM \
4070 || final_char < '@' || final_char > 'B') \
4071 EMIT_ONE_ASCII_BYTE (intermediate_char_94[reg]); \
4074 EMIT_ONE_ASCII_BYTE (intermediate_char_96[reg]); \
4076 EMIT_ONE_ASCII_BYTE (final_char); \
4078 CODING_ISO_DESIGNATION (coding, reg) = CHARSET_ID (charset); \
4082 /* The following two macros produce codes (control character or escape
4083 sequence) for ISO2022 single-shift functions (single-shift-2 and
4086 #define ENCODE_SINGLE_SHIFT_2 \
4088 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_SEVEN_BITS) \
4089 EMIT_TWO_ASCII_BYTES (ISO_CODE_ESC, 'N'); \
4091 EMIT_ONE_BYTE (ISO_CODE_SS2); \
4092 CODING_ISO_SINGLE_SHIFTING (coding) = 1; \
4096 #define ENCODE_SINGLE_SHIFT_3 \
4098 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_SEVEN_BITS) \
4099 EMIT_TWO_ASCII_BYTES (ISO_CODE_ESC, 'O'); \
4101 EMIT_ONE_BYTE (ISO_CODE_SS3); \
4102 CODING_ISO_SINGLE_SHIFTING (coding) = 1; \
4106 /* The following four macros produce codes (control character or
4107 escape sequence) for ISO2022 locking-shift functions (shift-in,
4108 shift-out, locking-shift-2, and locking-shift-3). */
4110 #define ENCODE_SHIFT_IN \
4112 EMIT_ONE_ASCII_BYTE (ISO_CODE_SI); \
4113 CODING_ISO_INVOCATION (coding, 0) = 0; \
4117 #define ENCODE_SHIFT_OUT \
4119 EMIT_ONE_ASCII_BYTE (ISO_CODE_SO); \
4120 CODING_ISO_INVOCATION (coding, 0) = 1; \
4124 #define ENCODE_LOCKING_SHIFT_2 \
4126 EMIT_TWO_ASCII_BYTES (ISO_CODE_ESC, 'n'); \
4127 CODING_ISO_INVOCATION (coding, 0) = 2; \
4131 #define ENCODE_LOCKING_SHIFT_3 \
4133 EMIT_TWO_ASCII_BYTES (ISO_CODE_ESC, 'n'); \
4134 CODING_ISO_INVOCATION (coding, 0) = 3; \
4138 /* Produce codes for a DIMENSION1 character whose character set is
4139 CHARSET and whose position-code is C1. Designation and invocation
4140 sequences are also produced in advance if necessary. */
4142 #define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \
4144 int id = CHARSET_ID (charset); \
4146 if ((CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_USE_ROMAN) \
4147 && id == charset_ascii) \
4149 id = charset_jisx0201_roman; \
4150 charset = CHARSET_FROM_ID (id); \
4153 if (CODING_ISO_SINGLE_SHIFTING (coding)) \
4155 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_SEVEN_BITS) \
4156 EMIT_ONE_ASCII_BYTE (c1 & 0x7F); \
4158 EMIT_ONE_BYTE (c1 | 0x80); \
4159 CODING_ISO_SINGLE_SHIFTING (coding) = 0; \
4162 else if (id == CODING_ISO_INVOKED_CHARSET (coding, 0)) \
4164 EMIT_ONE_ASCII_BYTE (c1 & 0x7F); \
4167 else if (id == CODING_ISO_INVOKED_CHARSET (coding, 1)) \
4169 EMIT_ONE_BYTE (c1 | 0x80); \
4173 /* Since CHARSET is not yet invoked to any graphic planes, we \
4174 must invoke it, or, at first, designate it to some graphic \
4175 register. Then repeat the loop to actually produce the \
4177 dst = encode_invocation_designation (charset, coding, dst, \
4182 /* Produce codes for a DIMENSION2 character whose character set is
4183 CHARSET and whose position-codes are C1 and C2. Designation and
4184 invocation codes are also produced in advance if necessary. */
4186 #define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2) \
4188 int id = CHARSET_ID (charset); \
4190 if ((CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_USE_OLDJIS) \
4191 && id == charset_jisx0208) \
4193 id = charset_jisx0208_1978; \
4194 charset = CHARSET_FROM_ID (id); \
4197 if (CODING_ISO_SINGLE_SHIFTING (coding)) \
4199 if (CODING_ISO_FLAGS (coding) & CODING_ISO_FLAG_SEVEN_BITS) \
4200 EMIT_TWO_ASCII_BYTES ((c1) & 0x7F, (c2) & 0x7F); \
4202 EMIT_TWO_BYTES ((c1) | 0x80, (c2) | 0x80); \
4203 CODING_ISO_SINGLE_SHIFTING (coding) = 0; \
4206 else if (id == CODING_ISO_INVOKED_CHARSET (coding, 0)) \
4208 EMIT_TWO_ASCII_BYTES ((c1) & 0x7F, (c2) & 0x7F); \
4211 else if (id == CODING_ISO_INVOKED_CHARSET (coding, 1)) \
4213 EMIT_TWO_BYTES ((c1) | 0x80, (c2) | 0x80); \
4217 /* Since CHARSET is not yet invoked to any graphic planes, we \
4218 must invoke it, or, at first, designate it to some graphic \
4219 register. Then repeat the loop to actually produce the \
4221 dst = encode_invocation_designation (charset, coding, dst, \
4226 #define ENCODE_ISO_CHARACTER(charset, c) \
4229 CODING_ENCODE_CHAR (coding, dst, dst_end, (charset), (c), code); \
4231 if (CHARSET_DIMENSION (charset) == 1) \
4232 ENCODE_ISO_CHARACTER_DIMENSION1 ((charset), code); \
4234 ENCODE_ISO_CHARACTER_DIMENSION2 ((charset), code >> 8, code & 0xFF); \
4238 /* Produce designation and invocation codes at a place pointed by DST
4239 to use CHARSET. The element `spec.iso_2022' of *CODING is updated.
4242 static unsigned char *
4243 encode_invocation_designation (struct charset
*charset
,
4244 struct coding_system
*coding
,
4245 unsigned char *dst
, ptrdiff_t *p_nchars
)
4247 bool multibytep
= coding
->dst_multibyte
;
4248 ptrdiff_t produced_chars
= *p_nchars
;
4249 int reg
; /* graphic register number */
4250 int id
= CHARSET_ID (charset
);
4252 /* At first, check designations. */
4253 for (reg
= 0; reg
< 4; reg
++)
4254 if (id
== CODING_ISO_DESIGNATION (coding
, reg
))
4259 /* CHARSET is not yet designated to any graphic registers. */
4260 /* At first check the requested designation. */
4261 reg
= CODING_ISO_REQUEST (coding
, id
);
4263 /* Since CHARSET requests no special designation, designate it
4264 to graphic register 0. */
4267 ENCODE_DESIGNATION (charset
, reg
, coding
);
4270 if (CODING_ISO_INVOCATION (coding
, 0) != reg
4271 && CODING_ISO_INVOCATION (coding
, 1) != reg
)
4273 /* Since the graphic register REG is not invoked to any graphic
4274 planes, invoke it to graphic plane 0. */
4277 case 0: /* graphic register 0 */
4281 case 1: /* graphic register 1 */
4285 case 2: /* graphic register 2 */
4286 if (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SINGLE_SHIFT
)
4287 ENCODE_SINGLE_SHIFT_2
;
4289 ENCODE_LOCKING_SHIFT_2
;
4292 case 3: /* graphic register 3 */
4293 if (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_SINGLE_SHIFT
)
4294 ENCODE_SINGLE_SHIFT_3
;
4296 ENCODE_LOCKING_SHIFT_3
;
4304 *p_nchars
= produced_chars
;
4309 /* Produce codes for designation and invocation to reset the graphic
4310 planes and registers to initial state. */
4311 #define ENCODE_RESET_PLANE_AND_REGISTER() \
4314 struct charset *charset; \
4316 if (CODING_ISO_INVOCATION (coding, 0) != 0) \
4318 for (reg = 0; reg < 4; reg++) \
4319 if (CODING_ISO_INITIAL (coding, reg) >= 0 \
4320 && (CODING_ISO_DESIGNATION (coding, reg) \
4321 != CODING_ISO_INITIAL (coding, reg))) \
4323 charset = CHARSET_FROM_ID (CODING_ISO_INITIAL (coding, reg)); \
4324 ENCODE_DESIGNATION (charset, reg, coding); \
4329 /* Produce designation sequences of charsets in the line started from
4330 CHARBUF to a place pointed by DST, and return the number of
4331 produced bytes. DST should not directly point a buffer text area
4332 which may be relocated by char_charset call.
4334 If the current block ends before any end-of-line, we may fail to
4335 find all the necessary designations. */
4338 encode_designation_at_bol (struct coding_system
*coding
,
4339 int *charbuf
, int *charbuf_end
,
4342 unsigned char *orig
= dst
;
4343 struct charset
*charset
;
4344 /* Table of charsets to be designated to each graphic register. */
4346 int c
, found
= 0, reg
;
4347 ptrdiff_t produced_chars
= 0;
4348 bool multibytep
= coding
->dst_multibyte
;
4350 Lisp_Object charset_list
;
4352 attrs
= CODING_ID_ATTRS (coding
->id
);
4353 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
4354 if (EQ (charset_list
, Qiso_2022
))
4355 charset_list
= Viso_2022_charset_list
;
4357 for (reg
= 0; reg
< 4; reg
++)
4360 while (charbuf
< charbuf_end
&& found
< 4)
4367 charset
= char_charset (c
, charset_list
, NULL
);
4368 id
= CHARSET_ID (charset
);
4369 reg
= CODING_ISO_REQUEST (coding
, id
);
4370 if (reg
>= 0 && r
[reg
] < 0)
4379 for (reg
= 0; reg
< 4; reg
++)
4381 && CODING_ISO_DESIGNATION (coding
, reg
) != r
[reg
])
4382 ENCODE_DESIGNATION (CHARSET_FROM_ID (r
[reg
]), reg
, coding
);
4388 /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */
4391 encode_coding_iso_2022 (struct coding_system
*coding
)
4393 bool multibytep
= coding
->dst_multibyte
;
4394 int *charbuf
= coding
->charbuf
;
4395 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
4396 unsigned char *dst
= coding
->destination
+ coding
->produced
;
4397 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
4399 bool bol_designation
4400 = (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_DESIGNATE_AT_BOL
4401 && CODING_ISO_BOL (coding
));
4402 ptrdiff_t produced_chars
= 0;
4403 Lisp_Object attrs
, eol_type
, charset_list
;
4404 bool ascii_compatible
;
4406 int preferred_charset_id
= -1;
4408 CODING_GET_INFO (coding
, attrs
, charset_list
);
4409 eol_type
= inhibit_eol_conversion
? Qunix
: CODING_ID_EOL_TYPE (coding
->id
);
4410 if (VECTORP (eol_type
))
4413 setup_iso_safe_charsets (attrs
);
4414 /* Charset list may have been changed. */
4415 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
4416 coding
->safe_charsets
= SDATA (CODING_ATTR_SAFE_CHARSETS (attrs
));
4419 = (! NILP (CODING_ATTR_ASCII_COMPAT (attrs
))
4420 && ! (CODING_ISO_FLAGS (coding
) & (CODING_ISO_FLAG_DESIGNATION
4421 | CODING_ISO_FLAG_LOCKING_SHIFT
)));
4423 while (charbuf
< charbuf_end
)
4425 ASSURE_DESTINATION (safe_room
);
4427 if (bol_designation
)
4429 /* We have to produce designation sequences if any now. */
4430 unsigned char desig_buf
[16];
4434 charset_map_loaded
= 0;
4435 nbytes
= encode_designation_at_bol (coding
, charbuf
, charbuf_end
,
4437 if (charset_map_loaded
4438 && (offset
= coding_change_destination (coding
)))
4443 memcpy (dst
, desig_buf
, nbytes
);
4445 /* We are sure that designation sequences are all ASCII bytes. */
4446 produced_chars
+= nbytes
;
4447 bol_designation
= 0;
4448 ASSURE_DESTINATION (safe_room
);
4455 /* Handle an annotation. */
4458 case CODING_ANNOTATE_COMPOSITION_MASK
:
4459 /* Not yet implemented. */
4461 case CODING_ANNOTATE_CHARSET_MASK
:
4462 preferred_charset_id
= charbuf
[2];
4463 if (preferred_charset_id
>= 0
4464 && NILP (Fmemq (make_number (preferred_charset_id
),
4466 preferred_charset_id
= -1;
4475 /* Now encode the character C. */
4476 if (c
< 0x20 || c
== 0x7F)
4479 || (c
== '\r' && EQ (eol_type
, Qmac
)))
4481 if (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_RESET_AT_EOL
)
4482 ENCODE_RESET_PLANE_AND_REGISTER ();
4483 if (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_INIT_AT_BOL
)
4487 for (i
= 0; i
< 4; i
++)
4488 CODING_ISO_DESIGNATION (coding
, i
)
4489 = CODING_ISO_INITIAL (coding
, i
);
4491 bol_designation
= ((CODING_ISO_FLAGS (coding
)
4492 & CODING_ISO_FLAG_DESIGNATE_AT_BOL
)
4495 else if (CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_RESET_AT_CNTL
)
4496 ENCODE_RESET_PLANE_AND_REGISTER ();
4497 EMIT_ONE_ASCII_BYTE (c
);
4499 else if (ASCII_CHAR_P (c
))
4501 if (ascii_compatible
)
4502 EMIT_ONE_ASCII_BYTE (c
);
4505 struct charset
*charset
= CHARSET_FROM_ID (charset_ascii
);
4506 ENCODE_ISO_CHARACTER (charset
, c
);
4509 else if (CHAR_BYTE8_P (c
))
4511 c
= CHAR_TO_BYTE8 (c
);
4516 struct charset
*charset
;
4518 if (preferred_charset_id
>= 0)
4522 charset
= CHARSET_FROM_ID (preferred_charset_id
);
4523 CODING_CHAR_CHARSET_P (coding
, dst
, dst_end
, c
, charset
, result
);
4525 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
4529 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
4533 if (coding
->mode
& CODING_MODE_SAFE_ENCODING
)
4535 c
= CODING_INHIBIT_CHARACTER_SUBSTITUTION
;
4536 charset
= CHARSET_FROM_ID (charset_ascii
);
4540 c
= coding
->default_char
;
4541 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
,
4542 charset_list
, NULL
, charset
);
4545 ENCODE_ISO_CHARACTER (charset
, c
);
4549 if (coding
->mode
& CODING_MODE_LAST_BLOCK
4550 && CODING_ISO_FLAGS (coding
) & CODING_ISO_FLAG_RESET_AT_EOL
)
4552 ASSURE_DESTINATION (safe_room
);
4553 ENCODE_RESET_PLANE_AND_REGISTER ();
4555 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
4556 CODING_ISO_BOL (coding
) = bol_designation
;
4557 coding
->produced_char
+= produced_chars
;
4558 coding
->produced
= dst
- coding
->destination
;
4563 /*** 8,9. SJIS and BIG5 handlers ***/
4565 /* Although SJIS and BIG5 are not ISO's coding system, they are used
4566 quite widely. So, for the moment, Emacs supports them in the bare
4567 C code. But, in the future, they may be supported only by CCL. */
4569 /* SJIS is a coding system encoding three character sets: ASCII, right
4570 half of JISX0201-Kana, and JISX0208. An ASCII character is encoded
4571 as is. A character of charset katakana-jisx0201 is encoded by
4572 "position-code + 0x80". A character of charset japanese-jisx0208
4573 is encoded in 2-byte but two position-codes are divided and shifted
4574 so that it fit in the range below.
4576 --- CODE RANGE of SJIS ---
4577 (character set) (range)
4579 KATAKANA-JISX0201 0xA0 .. 0xDF
4580 JISX0208 (1st byte) 0x81 .. 0x9F and 0xE0 .. 0xEF
4581 (2nd byte) 0x40 .. 0x7E and 0x80 .. 0xFC
4582 -------------------------------
4586 /* BIG5 is a coding system encoding two character sets: ASCII and
4587 Big5. An ASCII character is encoded as is. Big5 is a two-byte
4588 character set and is encoded in two-byte.
4590 --- CODE RANGE of BIG5 ---
4591 (character set) (range)
4593 Big5 (1st byte) 0xA1 .. 0xFE
4594 (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE
4595 --------------------------
4599 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
4600 Return true if a text is encoded in SJIS. */
4603 detect_coding_sjis (struct coding_system
*coding
,
4604 struct coding_detection_info
*detect_info
)
4606 const unsigned char *src
= coding
->source
, *src_base
;
4607 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
4608 bool multibytep
= coding
->src_multibyte
;
4609 ptrdiff_t consumed_chars
= 0;
4612 Lisp_Object attrs
, charset_list
;
4613 int max_first_byte_of_2_byte_code
;
4615 CODING_GET_INFO (coding
, attrs
, charset_list
);
4616 max_first_byte_of_2_byte_code
4617 = (XINT (Flength (charset_list
)) > 3 ? 0xFC : 0xEF);
4619 detect_info
->checked
|= CATEGORY_MASK_SJIS
;
4620 /* A coding system of this category is always ASCII compatible. */
4621 src
+= coding
->head_ascii
;
4629 if ((c
>= 0x81 && c
<= 0x9F)
4630 || (c
>= 0xE0 && c
<= max_first_byte_of_2_byte_code
))
4633 if (c
< 0x40 || c
== 0x7F || c
> 0xFC)
4635 found
= CATEGORY_MASK_SJIS
;
4637 else if (c
>= 0xA0 && c
< 0xE0)
4638 found
= CATEGORY_MASK_SJIS
;
4642 detect_info
->rejected
|= CATEGORY_MASK_SJIS
;
4646 if (src_base
< src
&& coding
->mode
& CODING_MODE_LAST_BLOCK
)
4648 detect_info
->rejected
|= CATEGORY_MASK_SJIS
;
4651 detect_info
->found
|= found
;
4655 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
4656 Return true if a text is encoded in BIG5. */
4659 detect_coding_big5 (struct coding_system
*coding
,
4660 struct coding_detection_info
*detect_info
)
4662 const unsigned char *src
= coding
->source
, *src_base
;
4663 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
4664 bool multibytep
= coding
->src_multibyte
;
4665 ptrdiff_t consumed_chars
= 0;
4669 detect_info
->checked
|= CATEGORY_MASK_BIG5
;
4670 /* A coding system of this category is always ASCII compatible. */
4671 src
+= coding
->head_ascii
;
4682 if (c
< 0x40 || (c
>= 0x7F && c
<= 0xA0))
4684 found
= CATEGORY_MASK_BIG5
;
4689 detect_info
->rejected
|= CATEGORY_MASK_BIG5
;
4693 if (src_base
< src
&& coding
->mode
& CODING_MODE_LAST_BLOCK
)
4695 detect_info
->rejected
|= CATEGORY_MASK_BIG5
;
4698 detect_info
->found
|= found
;
4702 /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
4705 decode_coding_sjis (struct coding_system
*coding
)
4707 const unsigned char *src
= coding
->source
+ coding
->consumed
;
4708 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
4709 const unsigned char *src_base
;
4710 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
4711 /* We may produce one charset annotation in one loop and one more at
4714 = coding
->charbuf
+ coding
->charbuf_size
- (MAX_ANNOTATION_LENGTH
* 2);
4715 ptrdiff_t consumed_chars
= 0, consumed_chars_base
;
4716 bool multibytep
= coding
->src_multibyte
;
4717 struct charset
*charset_roman
, *charset_kanji
, *charset_kana
;
4718 struct charset
*charset_kanji2
;
4719 Lisp_Object attrs
, charset_list
, val
;
4720 ptrdiff_t char_offset
= coding
->produced_char
;
4721 ptrdiff_t last_offset
= char_offset
;
4722 int last_id
= charset_ascii
;
4724 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
4725 int byte_after_cr
= -1;
4727 CODING_GET_INFO (coding
, attrs
, charset_list
);
4730 charset_roman
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
4731 charset_kana
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
4732 charset_kanji
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
4733 charset_kanji2
= NILP (val
) ? NULL
: CHARSET_FROM_ID (XINT (XCAR (val
)));
4738 struct charset
*charset
;
4741 consumed_chars_base
= consumed_chars
;
4743 if (charbuf
>= charbuf_end
)
4745 if (byte_after_cr
>= 0)
4750 if (byte_after_cr
>= 0)
4751 c
= byte_after_cr
, byte_after_cr
= -1;
4758 if (eol_dos
&& c
== '\r')
4759 ONE_MORE_BYTE (byte_after_cr
);
4760 charset
= charset_roman
;
4762 else if (c
== 0x80 || c
== 0xA0)
4764 else if (c
>= 0xA1 && c
<= 0xDF)
4766 /* SJIS -> JISX0201-Kana */
4768 charset
= charset_kana
;
4772 /* SJIS -> JISX0208 */
4774 if (c1
< 0x40 || c1
== 0x7F || c1
> 0xFC)
4778 charset
= charset_kanji
;
4780 else if (c
<= 0xFC && charset_kanji2
)
4782 /* SJIS -> JISX0213-2 */
4784 if (c1
< 0x40 || c1
== 0x7F || c1
> 0xFC)
4788 charset
= charset_kanji2
;
4792 if (charset
->id
!= charset_ascii
4793 && last_id
!= charset
->id
)
4795 if (last_id
!= charset_ascii
)
4796 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
4797 last_id
= charset
->id
;
4798 last_offset
= char_offset
;
4800 CODING_DECODE_CHAR (coding
, src
, src_base
, src_end
, charset
, c
, c
);
4807 consumed_chars
= consumed_chars_base
;
4809 *charbuf
++ = c
< 0 ? -c
: BYTE8_TO_CHAR (c
);
4814 if (last_id
!= charset_ascii
)
4815 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
4816 coding
->consumed_char
+= consumed_chars_base
;
4817 coding
->consumed
= src_base
- coding
->source
;
4818 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
4822 decode_coding_big5 (struct coding_system
*coding
)
4824 const unsigned char *src
= coding
->source
+ coding
->consumed
;
4825 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
4826 const unsigned char *src_base
;
4827 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
4828 /* We may produce one charset annotation in one loop and one more at
4831 = coding
->charbuf
+ coding
->charbuf_size
- (MAX_ANNOTATION_LENGTH
* 2);
4832 ptrdiff_t consumed_chars
= 0, consumed_chars_base
;
4833 bool multibytep
= coding
->src_multibyte
;
4834 struct charset
*charset_roman
, *charset_big5
;
4835 Lisp_Object attrs
, charset_list
, val
;
4836 ptrdiff_t char_offset
= coding
->produced_char
;
4837 ptrdiff_t last_offset
= char_offset
;
4838 int last_id
= charset_ascii
;
4840 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
4841 int byte_after_cr
= -1;
4843 CODING_GET_INFO (coding
, attrs
, charset_list
);
4845 charset_roman
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
4846 charset_big5
= CHARSET_FROM_ID (XINT (XCAR (val
)));
4851 struct charset
*charset
;
4854 consumed_chars_base
= consumed_chars
;
4856 if (charbuf
>= charbuf_end
)
4858 if (byte_after_cr
>= 0)
4863 if (byte_after_cr
>= 0)
4864 c
= byte_after_cr
, byte_after_cr
= -1;
4872 if (eol_dos
&& c
== '\r')
4873 ONE_MORE_BYTE (byte_after_cr
);
4874 charset
= charset_roman
;
4879 if (c
< 0xA1 || c
> 0xFE)
4882 if (c1
< 0x40 || (c1
> 0x7E && c1
< 0xA1) || c1
> 0xFE)
4885 charset
= charset_big5
;
4887 if (charset
->id
!= charset_ascii
4888 && last_id
!= charset
->id
)
4890 if (last_id
!= charset_ascii
)
4891 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
4892 last_id
= charset
->id
;
4893 last_offset
= char_offset
;
4895 CODING_DECODE_CHAR (coding
, src
, src_base
, src_end
, charset
, c
, c
);
4902 consumed_chars
= consumed_chars_base
;
4904 *charbuf
++ = c
< 0 ? -c
: BYTE8_TO_CHAR (c
);
4909 if (last_id
!= charset_ascii
)
4910 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
4911 coding
->consumed_char
+= consumed_chars_base
;
4912 coding
->consumed
= src_base
- coding
->source
;
4913 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
4916 /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions".
4917 This function can encode charsets `ascii', `katakana-jisx0201',
4918 `japanese-jisx0208', `chinese-big5-1', and `chinese-big5-2'. We
4919 are sure that all these charsets are registered as official charset
4920 (i.e. do not have extended leading-codes). Characters of other
4921 charsets are produced without any encoding. */
4924 encode_coding_sjis (struct coding_system
*coding
)
4926 bool multibytep
= coding
->dst_multibyte
;
4927 int *charbuf
= coding
->charbuf
;
4928 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
4929 unsigned char *dst
= coding
->destination
+ coding
->produced
;
4930 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
4932 ptrdiff_t produced_chars
= 0;
4933 Lisp_Object attrs
, charset_list
, val
;
4934 bool ascii_compatible
;
4935 struct charset
*charset_kanji
, *charset_kana
;
4936 struct charset
*charset_kanji2
;
4939 CODING_GET_INFO (coding
, attrs
, charset_list
);
4940 val
= XCDR (charset_list
);
4941 charset_kana
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
4942 charset_kanji
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
4943 charset_kanji2
= NILP (val
) ? NULL
: CHARSET_FROM_ID (XINT (XCAR (val
)));
4945 ascii_compatible
= ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
));
4947 while (charbuf
< charbuf_end
)
4949 ASSURE_DESTINATION (safe_room
);
4951 /* Now encode the character C. */
4952 if (ASCII_CHAR_P (c
) && ascii_compatible
)
4953 EMIT_ONE_ASCII_BYTE (c
);
4954 else if (CHAR_BYTE8_P (c
))
4956 c
= CHAR_TO_BYTE8 (c
);
4962 struct charset
*charset
;
4963 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
4968 if (coding
->mode
& CODING_MODE_SAFE_ENCODING
)
4970 code
= CODING_INHIBIT_CHARACTER_SUBSTITUTION
;
4971 charset
= CHARSET_FROM_ID (charset_ascii
);
4975 c
= coding
->default_char
;
4976 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
,
4977 charset_list
, &code
, charset
);
4980 if (code
== CHARSET_INVALID_CODE (charset
))
4982 if (charset
== charset_kanji
)
4986 c1
= code
>> 8, c2
= code
& 0xFF;
4987 EMIT_TWO_BYTES (c1
, c2
);
4989 else if (charset
== charset_kana
)
4990 EMIT_ONE_BYTE (code
| 0x80);
4991 else if (charset_kanji2
&& charset
== charset_kanji2
)
4996 if (c1
== 0x21 || (c1
>= 0x23 && c1
<= 0x25)
4998 || (c1
>= 0x2C && c1
<= 0x2F) || c1
>= 0x6E)
5000 JIS_TO_SJIS2 (code
);
5001 c1
= code
>> 8, c2
= code
& 0xFF;
5002 EMIT_TWO_BYTES (c1
, c2
);
5005 EMIT_ONE_ASCII_BYTE (code
& 0x7F);
5008 EMIT_ONE_ASCII_BYTE (code
& 0x7F);
5011 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
5012 coding
->produced_char
+= produced_chars
;
5013 coding
->produced
= dst
- coding
->destination
;
5018 encode_coding_big5 (struct coding_system
*coding
)
5020 bool multibytep
= coding
->dst_multibyte
;
5021 int *charbuf
= coding
->charbuf
;
5022 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
5023 unsigned char *dst
= coding
->destination
+ coding
->produced
;
5024 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
5026 ptrdiff_t produced_chars
= 0;
5027 Lisp_Object attrs
, charset_list
, val
;
5028 bool ascii_compatible
;
5029 struct charset
*charset_big5
;
5032 CODING_GET_INFO (coding
, attrs
, charset_list
);
5033 val
= XCDR (charset_list
);
5034 charset_big5
= CHARSET_FROM_ID (XINT (XCAR (val
)));
5035 ascii_compatible
= ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
));
5037 while (charbuf
< charbuf_end
)
5039 ASSURE_DESTINATION (safe_room
);
5041 /* Now encode the character C. */
5042 if (ASCII_CHAR_P (c
) && ascii_compatible
)
5043 EMIT_ONE_ASCII_BYTE (c
);
5044 else if (CHAR_BYTE8_P (c
))
5046 c
= CHAR_TO_BYTE8 (c
);
5052 struct charset
*charset
;
5053 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
5058 if (coding
->mode
& CODING_MODE_SAFE_ENCODING
)
5060 code
= CODING_INHIBIT_CHARACTER_SUBSTITUTION
;
5061 charset
= CHARSET_FROM_ID (charset_ascii
);
5065 c
= coding
->default_char
;
5066 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
,
5067 charset_list
, &code
, charset
);
5070 if (code
== CHARSET_INVALID_CODE (charset
))
5072 if (charset
== charset_big5
)
5076 c1
= code
>> 8, c2
= code
& 0xFF;
5077 EMIT_TWO_BYTES (c1
, c2
);
5080 EMIT_ONE_ASCII_BYTE (code
& 0x7F);
5083 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
5084 coding
->produced_char
+= produced_chars
;
5085 coding
->produced
= dst
- coding
->destination
;
5090 /*** 10. CCL handlers ***/
5092 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
5093 Return true if a text is encoded in a coding system of which
5094 encoder/decoder are written in CCL program. */
5097 detect_coding_ccl (struct coding_system
*coding
,
5098 struct coding_detection_info
*detect_info
)
5100 const unsigned char *src
= coding
->source
, *src_base
;
5101 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
5102 bool multibytep
= coding
->src_multibyte
;
5103 ptrdiff_t consumed_chars
= 0;
5105 unsigned char *valids
;
5106 ptrdiff_t head_ascii
= coding
->head_ascii
;
5109 detect_info
->checked
|= CATEGORY_MASK_CCL
;
5111 coding
= &coding_categories
[coding_category_ccl
];
5112 valids
= CODING_CCL_VALIDS (coding
);
5113 attrs
= CODING_ID_ATTRS (coding
->id
);
5114 if (! NILP (CODING_ATTR_ASCII_COMPAT (attrs
)))
5123 if (c
< 0 || ! valids
[c
])
5125 if ((valids
[c
] > 1))
5126 found
= CATEGORY_MASK_CCL
;
5128 detect_info
->rejected
|= CATEGORY_MASK_CCL
;
5132 detect_info
->found
|= found
;
5137 decode_coding_ccl (struct coding_system
*coding
)
5139 const unsigned char *src
= coding
->source
+ coding
->consumed
;
5140 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
5141 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
5142 int *charbuf_end
= coding
->charbuf
+ coding
->charbuf_size
;
5143 ptrdiff_t consumed_chars
= 0;
5144 bool multibytep
= coding
->src_multibyte
;
5145 struct ccl_program
*ccl
= &coding
->spec
.ccl
->ccl
;
5146 int source_charbuf
[1024];
5147 int source_byteidx
[1025];
5148 Lisp_Object attrs
, charset_list
;
5150 CODING_GET_INFO (coding
, attrs
, charset_list
);
5154 const unsigned char *p
= src
;
5160 while (i
< 1024 && p
< src_end
)
5162 source_byteidx
[i
] = p
- src
;
5163 source_charbuf
[i
++] = STRING_CHAR_ADVANCE (p
);
5165 source_byteidx
[i
] = p
- src
;
5168 while (i
< 1024 && p
< src_end
)
5169 source_charbuf
[i
++] = *p
++;
5171 if (p
== src_end
&& coding
->mode
& CODING_MODE_LAST_BLOCK
)
5172 ccl
->last_block
= true;
5173 /* As ccl_driver calls DECODE_CHAR, buffer may be relocated. */
5174 charset_map_loaded
= 0;
5175 ccl_driver (ccl
, source_charbuf
, charbuf
, i
, charbuf_end
- charbuf
,
5177 if (charset_map_loaded
5178 && (offset
= coding_change_source (coding
)))
5184 charbuf
+= ccl
->produced
;
5186 src
+= source_byteidx
[ccl
->consumed
];
5188 src
+= ccl
->consumed
;
5189 consumed_chars
+= ccl
->consumed
;
5190 if (p
== src_end
|| ccl
->status
!= CCL_STAT_SUSPEND_BY_SRC
)
5194 switch (ccl
->status
)
5196 case CCL_STAT_SUSPEND_BY_SRC
:
5197 record_conversion_result (coding
, CODING_RESULT_INSUFFICIENT_SRC
);
5199 case CCL_STAT_SUSPEND_BY_DST
:
5200 record_conversion_result (coding
, CODING_RESULT_INSUFFICIENT_DST
);
5203 case CCL_STAT_INVALID_CMD
:
5204 record_conversion_result (coding
, CODING_RESULT_INTERRUPT
);
5207 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
5210 coding
->consumed_char
+= consumed_chars
;
5211 coding
->consumed
= src
- coding
->source
;
5212 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
5216 encode_coding_ccl (struct coding_system
*coding
)
5218 struct ccl_program
*ccl
= &coding
->spec
.ccl
->ccl
;
5219 bool multibytep
= coding
->dst_multibyte
;
5220 int *charbuf
= coding
->charbuf
;
5221 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
5222 unsigned char *dst
= coding
->destination
+ coding
->produced
;
5223 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
5224 int destination_charbuf
[1024];
5225 ptrdiff_t produced_chars
= 0;
5227 Lisp_Object attrs
, charset_list
;
5229 CODING_GET_INFO (coding
, attrs
, charset_list
);
5230 if (coding
->consumed_char
== coding
->src_chars
5231 && coding
->mode
& CODING_MODE_LAST_BLOCK
)
5232 ccl
->last_block
= true;
5238 /* As ccl_driver calls DECODE_CHAR, buffer may be relocated. */
5239 charset_map_loaded
= 0;
5240 ccl_driver (ccl
, charbuf
, destination_charbuf
,
5241 charbuf_end
- charbuf
, 1024, charset_list
);
5242 if (charset_map_loaded
5243 && (offset
= coding_change_destination (coding
)))
5247 ASSURE_DESTINATION (ccl
->produced
* 2);
5248 for (i
= 0; i
< ccl
->produced
; i
++)
5249 EMIT_ONE_BYTE (destination_charbuf
[i
] & 0xFF);
5253 ASSURE_DESTINATION (ccl
->produced
);
5254 for (i
= 0; i
< ccl
->produced
; i
++)
5255 *dst
++ = destination_charbuf
[i
] & 0xFF;
5256 produced_chars
+= ccl
->produced
;
5258 charbuf
+= ccl
->consumed
;
5259 if (ccl
->status
== CCL_STAT_QUIT
5260 || ccl
->status
== CCL_STAT_INVALID_CMD
)
5263 while (charbuf
< charbuf_end
);
5265 switch (ccl
->status
)
5267 case CCL_STAT_SUSPEND_BY_SRC
:
5268 record_conversion_result (coding
, CODING_RESULT_INSUFFICIENT_SRC
);
5270 case CCL_STAT_SUSPEND_BY_DST
:
5271 record_conversion_result (coding
, CODING_RESULT_INSUFFICIENT_DST
);
5274 case CCL_STAT_INVALID_CMD
:
5275 record_conversion_result (coding
, CODING_RESULT_INTERRUPT
);
5278 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
5282 coding
->produced_char
+= produced_chars
;
5283 coding
->produced
= dst
- coding
->destination
;
5288 /*** 10, 11. no-conversion handlers ***/
5290 /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
5293 decode_coding_raw_text (struct coding_system
*coding
)
5296 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
5298 coding
->chars_at_source
= 1;
5299 coding
->consumed_char
= coding
->src_chars
;
5300 coding
->consumed
= coding
->src_bytes
;
5301 if (eol_dos
&& coding
->source
[coding
->src_bytes
- 1] == '\r')
5303 coding
->consumed_char
--;
5305 record_conversion_result (coding
, CODING_RESULT_INSUFFICIENT_SRC
);
5308 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
5312 encode_coding_raw_text (struct coding_system
*coding
)
5314 bool multibytep
= coding
->dst_multibyte
;
5315 int *charbuf
= coding
->charbuf
;
5316 int *charbuf_end
= coding
->charbuf
+ coding
->charbuf_used
;
5317 unsigned char *dst
= coding
->destination
+ coding
->produced
;
5318 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
5319 ptrdiff_t produced_chars
= 0;
5324 int safe_room
= MAX_MULTIBYTE_LENGTH
* 2;
5326 if (coding
->src_multibyte
)
5327 while (charbuf
< charbuf_end
)
5329 ASSURE_DESTINATION (safe_room
);
5331 if (ASCII_CHAR_P (c
))
5332 EMIT_ONE_ASCII_BYTE (c
);
5333 else if (CHAR_BYTE8_P (c
))
5335 c
= CHAR_TO_BYTE8 (c
);
5340 unsigned char str
[MAX_MULTIBYTE_LENGTH
], *p0
= str
, *p1
= str
;
5342 CHAR_STRING_ADVANCE (c
, p1
);
5345 EMIT_ONE_BYTE (*p0
);
5352 while (charbuf
< charbuf_end
)
5354 ASSURE_DESTINATION (safe_room
);
5361 if (coding
->src_multibyte
)
5363 int safe_room
= MAX_MULTIBYTE_LENGTH
;
5365 while (charbuf
< charbuf_end
)
5367 ASSURE_DESTINATION (safe_room
);
5369 if (ASCII_CHAR_P (c
))
5371 else if (CHAR_BYTE8_P (c
))
5372 *dst
++ = CHAR_TO_BYTE8 (c
);
5374 CHAR_STRING_ADVANCE (c
, dst
);
5379 ASSURE_DESTINATION (charbuf_end
- charbuf
);
5380 while (charbuf
< charbuf_end
&& dst
< dst_end
)
5381 *dst
++ = *charbuf
++;
5383 produced_chars
= dst
- (coding
->destination
+ coding
->produced
);
5385 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
5386 coding
->produced_char
+= produced_chars
;
5387 coding
->produced
= dst
- coding
->destination
;
5391 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
5392 Return true if a text is encoded in a charset-based coding system. */
5395 detect_coding_charset (struct coding_system
*coding
,
5396 struct coding_detection_info
*detect_info
)
5398 const unsigned char *src
= coding
->source
, *src_base
;
5399 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
5400 bool multibytep
= coding
->src_multibyte
;
5401 ptrdiff_t consumed_chars
= 0;
5402 Lisp_Object attrs
, valids
, name
;
5404 ptrdiff_t head_ascii
= coding
->head_ascii
;
5405 bool check_latin_extra
= 0;
5407 detect_info
->checked
|= CATEGORY_MASK_CHARSET
;
5409 coding
= &coding_categories
[coding_category_charset
];
5410 attrs
= CODING_ID_ATTRS (coding
->id
);
5411 valids
= AREF (attrs
, coding_attr_charset_valids
);
5412 name
= CODING_ID_NAME (coding
->id
);
5413 if (strncmp (SSDATA (SYMBOL_NAME (name
)),
5414 "iso-8859-", sizeof ("iso-8859-") - 1) == 0
5415 || strncmp (SSDATA (SYMBOL_NAME (name
)),
5416 "iso-latin-", sizeof ("iso-latin-") - 1) == 0)
5417 check_latin_extra
= 1;
5419 if (! NILP (CODING_ATTR_ASCII_COMPAT (attrs
)))
5426 struct charset
*charset
;
5433 val
= AREF (valids
, c
);
5439 && check_latin_extra
5440 && (!VECTORP (Vlatin_extra_code_table
)
5441 || NILP (AREF (Vlatin_extra_code_table
, c
))))
5443 found
= CATEGORY_MASK_CHARSET
;
5447 charset
= CHARSET_FROM_ID (XFASTINT (val
));
5448 dim
= CHARSET_DIMENSION (charset
);
5449 for (idx
= 1; idx
< dim
; idx
++)
5454 if (c
< charset
->code_space
[(dim
- 1 - idx
) * 4]
5455 || c
> charset
->code_space
[(dim
- 1 - idx
) * 4 + 1])
5464 for (; CONSP (val
); val
= XCDR (val
))
5466 charset
= CHARSET_FROM_ID (XFASTINT (XCAR (val
)));
5467 dim
= CHARSET_DIMENSION (charset
);
5473 if (c
< charset
->code_space
[(dim
- 1 - idx
) * 4]
5474 || c
> charset
->code_space
[(dim
- 1 - idx
) * 4 + 1])
5489 detect_info
->rejected
|= CATEGORY_MASK_CHARSET
;
5493 detect_info
->found
|= found
;
5498 decode_coding_charset (struct coding_system
*coding
)
5500 const unsigned char *src
= coding
->source
+ coding
->consumed
;
5501 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
5502 const unsigned char *src_base
;
5503 int *charbuf
= coding
->charbuf
+ coding
->charbuf_used
;
5504 /* We may produce one charset annotation in one loop and one more at
5507 = coding
->charbuf
+ coding
->charbuf_size
- (MAX_ANNOTATION_LENGTH
* 2);
5508 ptrdiff_t consumed_chars
= 0, consumed_chars_base
;
5509 bool multibytep
= coding
->src_multibyte
;
5510 Lisp_Object attrs
= CODING_ID_ATTRS (coding
->id
);
5512 ptrdiff_t char_offset
= coding
->produced_char
;
5513 ptrdiff_t last_offset
= char_offset
;
5514 int last_id
= charset_ascii
;
5516 = !inhibit_eol_conversion
&& EQ (CODING_ID_EOL_TYPE (coding
->id
), Qdos
);
5517 int byte_after_cr
= -1;
5519 valids
= AREF (attrs
, coding_attr_charset_valids
);
5525 struct charset
*charset
;
5531 consumed_chars_base
= consumed_chars
;
5533 if (charbuf
>= charbuf_end
)
5535 if (byte_after_cr
>= 0)
5540 if (byte_after_cr
>= 0)
5548 if (eol_dos
&& c
== '\r')
5549 ONE_MORE_BYTE (byte_after_cr
);
5555 val
= AREF (valids
, c
);
5556 if (! INTEGERP (val
) && ! CONSP (val
))
5560 charset
= CHARSET_FROM_ID (XFASTINT (val
));
5561 dim
= CHARSET_DIMENSION (charset
);
5565 code
= (code
<< 8) | c
;
5568 CODING_DECODE_CHAR (coding
, src
, src_base
, src_end
,
5573 /* VAL is a list of charset IDs. It is assured that the
5574 list is sorted by charset dimensions (smaller one
5578 charset
= CHARSET_FROM_ID (XFASTINT (XCAR (val
)));
5579 dim
= CHARSET_DIMENSION (charset
);
5583 code
= (code
<< 8) | c
;
5586 CODING_DECODE_CHAR (coding
, src
, src_base
,
5587 src_end
, charset
, code
, c
);
5595 if (charset
->id
!= charset_ascii
5596 && last_id
!= charset
->id
)
5598 if (last_id
!= charset_ascii
)
5599 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
5600 last_id
= charset
->id
;
5601 last_offset
= char_offset
;
5610 consumed_chars
= consumed_chars_base
;
5612 *charbuf
++ = c
< 0 ? -c
: ASCII_CHAR_P (c
) ? c
: BYTE8_TO_CHAR (c
);
5617 if (last_id
!= charset_ascii
)
5618 ADD_CHARSET_DATA (charbuf
, char_offset
- last_offset
, last_id
);
5619 coding
->consumed_char
+= consumed_chars_base
;
5620 coding
->consumed
= src_base
- coding
->source
;
5621 coding
->charbuf_used
= charbuf
- coding
->charbuf
;
5625 encode_coding_charset (struct coding_system
*coding
)
5627 bool multibytep
= coding
->dst_multibyte
;
5628 int *charbuf
= coding
->charbuf
;
5629 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
5630 unsigned char *dst
= coding
->destination
+ coding
->produced
;
5631 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
5632 int safe_room
= MAX_MULTIBYTE_LENGTH
;
5633 ptrdiff_t produced_chars
= 0;
5634 Lisp_Object attrs
, charset_list
;
5635 bool ascii_compatible
;
5638 CODING_GET_INFO (coding
, attrs
, charset_list
);
5639 ascii_compatible
= ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
));
5641 while (charbuf
< charbuf_end
)
5643 struct charset
*charset
;
5646 ASSURE_DESTINATION (safe_room
);
5648 if (ascii_compatible
&& ASCII_CHAR_P (c
))
5649 EMIT_ONE_ASCII_BYTE (c
);
5650 else if (CHAR_BYTE8_P (c
))
5652 c
= CHAR_TO_BYTE8 (c
);
5657 CODING_CHAR_CHARSET (coding
, dst
, dst_end
, c
, charset_list
,
5662 if (CHARSET_DIMENSION (charset
) == 1)
5663 EMIT_ONE_BYTE (code
);
5664 else if (CHARSET_DIMENSION (charset
) == 2)
5665 EMIT_TWO_BYTES (code
>> 8, code
& 0xFF);
5666 else if (CHARSET_DIMENSION (charset
) == 3)
5667 EMIT_THREE_BYTES (code
>> 16, (code
>> 8) & 0xFF, code
& 0xFF);
5669 EMIT_FOUR_BYTES (code
>> 24, (code
>> 16) & 0xFF,
5670 (code
>> 8) & 0xFF, code
& 0xFF);
5674 if (coding
->mode
& CODING_MODE_SAFE_ENCODING
)
5675 c
= CODING_INHIBIT_CHARACTER_SUBSTITUTION
;
5677 c
= coding
->default_char
;
5683 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
5684 coding
->produced_char
+= produced_chars
;
5685 coding
->produced
= dst
- coding
->destination
;
5690 /*** 7. C library functions ***/
5692 /* Setup coding context CODING from information about CODING_SYSTEM.
5693 If CODING_SYSTEM is nil, `no-conversion' is assumed. If
5694 CODING_SYSTEM is invalid, signal an error. */
5697 setup_coding_system (Lisp_Object coding_system
, struct coding_system
*coding
)
5700 Lisp_Object eol_type
;
5701 Lisp_Object coding_type
;
5704 if (NILP (coding_system
))
5705 coding_system
= Qundecided
;
5707 CHECK_CODING_SYSTEM_GET_ID (coding_system
, coding
->id
);
5709 attrs
= CODING_ID_ATTRS (coding
->id
);
5710 eol_type
= inhibit_eol_conversion
? Qunix
: CODING_ID_EOL_TYPE (coding
->id
);
5713 if (VECTORP (eol_type
))
5714 coding
->common_flags
= (CODING_REQUIRE_DECODING_MASK
5715 | CODING_REQUIRE_DETECTION_MASK
);
5716 else if (! EQ (eol_type
, Qunix
))
5717 coding
->common_flags
= (CODING_REQUIRE_DECODING_MASK
5718 | CODING_REQUIRE_ENCODING_MASK
);
5720 coding
->common_flags
= 0;
5721 if (! NILP (CODING_ATTR_POST_READ (attrs
)))
5722 coding
->common_flags
|= CODING_REQUIRE_DECODING_MASK
;
5723 if (! NILP (CODING_ATTR_PRE_WRITE (attrs
)))
5724 coding
->common_flags
|= CODING_REQUIRE_ENCODING_MASK
;
5725 if (! NILP (CODING_ATTR_FOR_UNIBYTE (attrs
)))
5726 coding
->common_flags
|= CODING_FOR_UNIBYTE_MASK
;
5728 val
= CODING_ATTR_SAFE_CHARSETS (attrs
);
5729 coding
->max_charset_id
= SCHARS (val
) - 1;
5730 coding
->safe_charsets
= SDATA (val
);
5731 coding
->default_char
= XINT (CODING_ATTR_DEFAULT_CHAR (attrs
));
5732 coding
->carryover_bytes
= 0;
5733 coding
->raw_destination
= 0;
5735 coding_type
= CODING_ATTR_TYPE (attrs
);
5736 if (EQ (coding_type
, Qundecided
))
5738 coding
->detector
= NULL
;
5739 coding
->decoder
= decode_coding_raw_text
;
5740 coding
->encoder
= encode_coding_raw_text
;
5741 coding
->common_flags
|= CODING_REQUIRE_DETECTION_MASK
;
5742 coding
->spec
.undecided
.inhibit_nbd
5743 = (encode_inhibit_flag
5744 (AREF (attrs
, coding_attr_undecided_inhibit_null_byte_detection
)));
5745 coding
->spec
.undecided
.inhibit_ied
5746 = (encode_inhibit_flag
5747 (AREF (attrs
, coding_attr_undecided_inhibit_iso_escape_detection
)));
5748 coding
->spec
.undecided
.prefer_utf_8
5749 = ! NILP (AREF (attrs
, coding_attr_undecided_prefer_utf_8
));
5751 else if (EQ (coding_type
, Qiso_2022
))
5754 int flags
= XINT (AREF (attrs
, coding_attr_iso_flags
));
5756 /* Invoke graphic register 0 to plane 0. */
5757 CODING_ISO_INVOCATION (coding
, 0) = 0;
5758 /* Invoke graphic register 1 to plane 1 if we can use 8-bit. */
5759 CODING_ISO_INVOCATION (coding
, 1)
5760 = (flags
& CODING_ISO_FLAG_SEVEN_BITS
? -1 : 1);
5761 /* Setup the initial status of designation. */
5762 for (i
= 0; i
< 4; i
++)
5763 CODING_ISO_DESIGNATION (coding
, i
) = CODING_ISO_INITIAL (coding
, i
);
5764 /* Not single shifting initially. */
5765 CODING_ISO_SINGLE_SHIFTING (coding
) = 0;
5766 /* Beginning of buffer should also be regarded as bol. */
5767 CODING_ISO_BOL (coding
) = 1;
5768 coding
->detector
= detect_coding_iso_2022
;
5769 coding
->decoder
= decode_coding_iso_2022
;
5770 coding
->encoder
= encode_coding_iso_2022
;
5771 if (flags
& CODING_ISO_FLAG_SAFE
)
5772 coding
->mode
|= CODING_MODE_SAFE_ENCODING
;
5773 coding
->common_flags
5774 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
5775 | CODING_REQUIRE_FLUSHING_MASK
);
5776 if (flags
& CODING_ISO_FLAG_COMPOSITION
)
5777 coding
->common_flags
|= CODING_ANNOTATE_COMPOSITION_MASK
;
5778 if (flags
& CODING_ISO_FLAG_DESIGNATION
)
5779 coding
->common_flags
|= CODING_ANNOTATE_CHARSET_MASK
;
5780 if (flags
& CODING_ISO_FLAG_FULL_SUPPORT
)
5782 setup_iso_safe_charsets (attrs
);
5783 val
= CODING_ATTR_SAFE_CHARSETS (attrs
);
5784 coding
->max_charset_id
= SCHARS (val
) - 1;
5785 coding
->safe_charsets
= SDATA (val
);
5787 CODING_ISO_FLAGS (coding
) = flags
;
5788 CODING_ISO_CMP_STATUS (coding
)->state
= COMPOSING_NO
;
5789 CODING_ISO_CMP_STATUS (coding
)->method
= COMPOSITION_NO
;
5790 CODING_ISO_EXTSEGMENT_LEN (coding
) = 0;
5791 CODING_ISO_EMBEDDED_UTF_8 (coding
) = 0;
5793 else if (EQ (coding_type
, Qcharset
))
5795 coding
->detector
= detect_coding_charset
;
5796 coding
->decoder
= decode_coding_charset
;
5797 coding
->encoder
= encode_coding_charset
;
5798 coding
->common_flags
5799 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
);
5801 else if (EQ (coding_type
, Qutf_8
))
5803 val
= AREF (attrs
, coding_attr_utf_bom
);
5804 CODING_UTF_8_BOM (coding
) = (CONSP (val
) ? utf_detect_bom
5805 : EQ (val
, Qt
) ? utf_with_bom
5807 coding
->detector
= detect_coding_utf_8
;
5808 coding
->decoder
= decode_coding_utf_8
;
5809 coding
->encoder
= encode_coding_utf_8
;
5810 coding
->common_flags
5811 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
);
5812 if (CODING_UTF_8_BOM (coding
) == utf_detect_bom
)
5813 coding
->common_flags
|= CODING_REQUIRE_DETECTION_MASK
;
5815 else if (EQ (coding_type
, Qutf_16
))
5817 val
= AREF (attrs
, coding_attr_utf_bom
);
5818 CODING_UTF_16_BOM (coding
) = (CONSP (val
) ? utf_detect_bom
5819 : EQ (val
, Qt
) ? utf_with_bom
5821 val
= AREF (attrs
, coding_attr_utf_16_endian
);
5822 CODING_UTF_16_ENDIAN (coding
) = (EQ (val
, Qbig
) ? utf_16_big_endian
5823 : utf_16_little_endian
);
5824 CODING_UTF_16_SURROGATE (coding
) = 0;
5825 coding
->detector
= detect_coding_utf_16
;
5826 coding
->decoder
= decode_coding_utf_16
;
5827 coding
->encoder
= encode_coding_utf_16
;
5828 coding
->common_flags
5829 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
);
5830 if (CODING_UTF_16_BOM (coding
) == utf_detect_bom
)
5831 coding
->common_flags
|= CODING_REQUIRE_DETECTION_MASK
;
5833 else if (EQ (coding_type
, Qccl
))
5835 coding
->detector
= detect_coding_ccl
;
5836 coding
->decoder
= decode_coding_ccl
;
5837 coding
->encoder
= encode_coding_ccl
;
5838 coding
->common_flags
5839 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
5840 | CODING_REQUIRE_FLUSHING_MASK
);
5842 else if (EQ (coding_type
, Qemacs_mule
))
5844 coding
->detector
= detect_coding_emacs_mule
;
5845 coding
->decoder
= decode_coding_emacs_mule
;
5846 coding
->encoder
= encode_coding_emacs_mule
;
5847 coding
->common_flags
5848 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
);
5849 if (! NILP (AREF (attrs
, coding_attr_emacs_mule_full
))
5850 && ! EQ (CODING_ATTR_CHARSET_LIST (attrs
), Vemacs_mule_charset_list
))
5852 Lisp_Object tail
, safe_charsets
;
5853 int max_charset_id
= 0;
5855 for (tail
= Vemacs_mule_charset_list
; CONSP (tail
);
5857 if (max_charset_id
< XFASTINT (XCAR (tail
)))
5858 max_charset_id
= XFASTINT (XCAR (tail
));
5859 safe_charsets
= make_uninit_string (max_charset_id
+ 1);
5860 memset (SDATA (safe_charsets
), 255, max_charset_id
+ 1);
5861 for (tail
= Vemacs_mule_charset_list
; CONSP (tail
);
5863 SSET (safe_charsets
, XFASTINT (XCAR (tail
)), 0);
5864 coding
->max_charset_id
= max_charset_id
;
5865 coding
->safe_charsets
= SDATA (safe_charsets
);
5867 coding
->spec
.emacs_mule
.cmp_status
.state
= COMPOSING_NO
;
5868 coding
->spec
.emacs_mule
.cmp_status
.method
= COMPOSITION_NO
;
5870 else if (EQ (coding_type
, Qshift_jis
))
5872 coding
->detector
= detect_coding_sjis
;
5873 coding
->decoder
= decode_coding_sjis
;
5874 coding
->encoder
= encode_coding_sjis
;
5875 coding
->common_flags
5876 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
);
5878 else if (EQ (coding_type
, Qbig5
))
5880 coding
->detector
= detect_coding_big5
;
5881 coding
->decoder
= decode_coding_big5
;
5882 coding
->encoder
= encode_coding_big5
;
5883 coding
->common_flags
5884 |= (CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
);
5886 else /* EQ (coding_type, Qraw_text) */
5888 coding
->detector
= NULL
;
5889 coding
->decoder
= decode_coding_raw_text
;
5890 coding
->encoder
= encode_coding_raw_text
;
5891 if (! EQ (eol_type
, Qunix
))
5893 coding
->common_flags
|= CODING_REQUIRE_DECODING_MASK
;
5894 if (! VECTORP (eol_type
))
5895 coding
->common_flags
|= CODING_REQUIRE_ENCODING_MASK
;
5903 /* Return a list of charsets supported by CODING. */
5906 coding_charset_list (struct coding_system
*coding
)
5908 Lisp_Object attrs
, charset_list
;
5910 CODING_GET_INFO (coding
, attrs
, charset_list
);
5911 if (EQ (CODING_ATTR_TYPE (attrs
), Qiso_2022
))
5913 int flags
= XINT (AREF (attrs
, coding_attr_iso_flags
));
5915 if (flags
& CODING_ISO_FLAG_FULL_SUPPORT
)
5916 charset_list
= Viso_2022_charset_list
;
5918 else if (EQ (CODING_ATTR_TYPE (attrs
), Qemacs_mule
))
5920 charset_list
= Vemacs_mule_charset_list
;
5922 return charset_list
;
5926 /* Return a list of charsets supported by CODING-SYSTEM. */
5929 coding_system_charset_list (Lisp_Object coding_system
)
5932 Lisp_Object attrs
, charset_list
;
5934 CHECK_CODING_SYSTEM_GET_ID (coding_system
, id
);
5935 attrs
= CODING_ID_ATTRS (id
);
5937 if (EQ (CODING_ATTR_TYPE (attrs
), Qiso_2022
))
5939 int flags
= XINT (AREF (attrs
, coding_attr_iso_flags
));
5941 if (flags
& CODING_ISO_FLAG_FULL_SUPPORT
)
5942 charset_list
= Viso_2022_charset_list
;
5944 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
5946 else if (EQ (CODING_ATTR_TYPE (attrs
), Qemacs_mule
))
5948 charset_list
= Vemacs_mule_charset_list
;
5952 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
5954 return charset_list
;
5958 /* Return raw-text or one of its subsidiaries that has the same
5959 eol_type as CODING-SYSTEM. */
5962 raw_text_coding_system (Lisp_Object coding_system
)
5964 Lisp_Object spec
, attrs
;
5965 Lisp_Object eol_type
, raw_text_eol_type
;
5967 if (NILP (coding_system
))
5969 spec
= CODING_SYSTEM_SPEC (coding_system
);
5970 attrs
= AREF (spec
, 0);
5972 if (EQ (CODING_ATTR_TYPE (attrs
), Qraw_text
))
5973 return coding_system
;
5975 eol_type
= AREF (spec
, 2);
5976 if (VECTORP (eol_type
))
5978 spec
= CODING_SYSTEM_SPEC (Qraw_text
);
5979 raw_text_eol_type
= AREF (spec
, 2);
5980 return (EQ (eol_type
, Qunix
) ? AREF (raw_text_eol_type
, 0)
5981 : EQ (eol_type
, Qdos
) ? AREF (raw_text_eol_type
, 1)
5982 : AREF (raw_text_eol_type
, 2));
5985 /* Return true if CODING corresponds to raw-text coding-system. */
5988 raw_text_coding_system_p (struct coding_system
*coding
)
5990 return (coding
->decoder
== decode_coding_raw_text
5991 && coding
->encoder
== encode_coding_raw_text
) ? true : false;
5995 /* If CODING_SYSTEM doesn't specify end-of-line format, return one of
5996 the subsidiary that has the same eol-spec as PARENT (if it is not
5997 nil and specifies end-of-line format) or the system's setting
5998 (system_eol_type). */
6001 coding_inherit_eol_type (Lisp_Object coding_system
, Lisp_Object parent
)
6003 Lisp_Object spec
, eol_type
;
6005 if (NILP (coding_system
))
6006 coding_system
= Qraw_text
;
6007 CHECK_CODING_SYSTEM (coding_system
);
6008 spec
= CODING_SYSTEM_SPEC (coding_system
);
6009 eol_type
= AREF (spec
, 2);
6010 if (VECTORP (eol_type
))
6012 Lisp_Object parent_eol_type
;
6014 if (! NILP (parent
))
6016 Lisp_Object parent_spec
;
6018 CHECK_CODING_SYSTEM (parent
);
6019 parent_spec
= CODING_SYSTEM_SPEC (parent
);
6020 parent_eol_type
= AREF (parent_spec
, 2);
6021 if (VECTORP (parent_eol_type
))
6022 parent_eol_type
= system_eol_type
;
6025 parent_eol_type
= system_eol_type
;
6026 if (EQ (parent_eol_type
, Qunix
))
6027 coding_system
= AREF (eol_type
, 0);
6028 else if (EQ (parent_eol_type
, Qdos
))
6029 coding_system
= AREF (eol_type
, 1);
6030 else if (EQ (parent_eol_type
, Qmac
))
6031 coding_system
= AREF (eol_type
, 2);
6033 return coding_system
;
6037 /* Check if text-conversion and eol-conversion of CODING_SYSTEM are
6038 decided for writing to a process. If not, complement them, and
6039 return a new coding system. */
6042 complement_process_encoding_system (Lisp_Object coding_system
)
6044 Lisp_Object coding_base
= Qnil
, eol_base
= Qnil
;
6045 Lisp_Object spec
, attrs
;
6048 for (i
= 0; i
< 3; i
++)
6051 coding_system
= CDR_SAFE (Vdefault_process_coding_system
);
6053 coding_system
= preferred_coding_system ();
6054 spec
= CODING_SYSTEM_SPEC (coding_system
);
6057 attrs
= AREF (spec
, 0);
6058 if (NILP (coding_base
) && ! EQ (CODING_ATTR_TYPE (attrs
), Qundecided
))
6059 coding_base
= CODING_ATTR_BASE_NAME (attrs
);
6060 if (NILP (eol_base
) && ! VECTORP (AREF (spec
, 2)))
6061 eol_base
= coding_system
;
6062 if (! NILP (coding_base
) && ! NILP (eol_base
))
6067 /* The original CODING_SYSTEM didn't specify text-conversion or
6068 eol-conversion. Be sure that we return a fully complemented
6070 coding_system
= coding_inherit_eol_type (coding_base
, eol_base
);
6071 return coding_system
;
6075 /* Emacs has a mechanism to automatically detect a coding system if it
6076 is one of Emacs' internal format, ISO2022, SJIS, and BIG5. But,
6077 it's impossible to distinguish some coding systems accurately
6078 because they use the same range of codes. So, at first, coding
6079 systems are categorized into 7, those are:
6081 o coding-category-emacs-mule
6083 The category for a coding system which has the same code range
6084 as Emacs' internal format. Assigned the coding-system (Lisp
6085 symbol) `emacs-mule' by default.
6087 o coding-category-sjis
6089 The category for a coding system which has the same code range
6090 as SJIS. Assigned the coding-system (Lisp
6091 symbol) `japanese-shift-jis' by default.
6093 o coding-category-iso-7
6095 The category for a coding system which has the same code range
6096 as ISO2022 of 7-bit environment. This doesn't use any locking
6097 shift and single shift functions. This can encode/decode all
6098 charsets. Assigned the coding-system (Lisp symbol)
6099 `iso-2022-7bit' by default.
6101 o coding-category-iso-7-tight
6103 Same as coding-category-iso-7 except that this can
6104 encode/decode only the specified charsets.
6106 o coding-category-iso-8-1
6108 The category for a coding system which has the same code range
6109 as ISO2022 of 8-bit environment and graphic plane 1 used only
6110 for DIMENSION1 charset. This doesn't use any locking shift
6111 and single shift functions. Assigned the coding-system (Lisp
6112 symbol) `iso-latin-1' by default.
6114 o coding-category-iso-8-2
6116 The category for a coding system which has the same code range
6117 as ISO2022 of 8-bit environment and graphic plane 1 used only
6118 for DIMENSION2 charset. This doesn't use any locking shift
6119 and single shift functions. Assigned the coding-system (Lisp
6120 symbol) `japanese-iso-8bit' by default.
6122 o coding-category-iso-7-else
6124 The category for a coding system which has the same code range
6125 as ISO2022 of 7-bit environment but uses locking shift or
6126 single shift functions. Assigned the coding-system (Lisp
6127 symbol) `iso-2022-7bit-lock' by default.
6129 o coding-category-iso-8-else
6131 The category for a coding system which has the same code range
6132 as ISO2022 of 8-bit environment but uses locking shift or
6133 single shift functions. Assigned the coding-system (Lisp
6134 symbol) `iso-2022-8bit-ss2' by default.
6136 o coding-category-big5
6138 The category for a coding system which has the same code range
6139 as BIG5. Assigned the coding-system (Lisp symbol)
6140 `cn-big5' by default.
6142 o coding-category-utf-8
6144 The category for a coding system which has the same code range
6145 as UTF-8 (cf. RFC3629). Assigned the coding-system (Lisp
6146 symbol) `utf-8' by default.
6148 o coding-category-utf-16-be
6150 The category for a coding system in which a text has an
6151 Unicode signature (cf. Unicode Standard) in the order of BIG
6152 endian at the head. Assigned the coding-system (Lisp symbol)
6153 `utf-16-be' by default.
6155 o coding-category-utf-16-le
6157 The category for a coding system in which a text has an
6158 Unicode signature (cf. Unicode Standard) in the order of
6159 LITTLE endian at the head. Assigned the coding-system (Lisp
6160 symbol) `utf-16-le' by default.
6162 o coding-category-ccl
6164 The category for a coding system of which encoder/decoder is
6165 written in CCL programs. The default value is nil, i.e., no
6166 coding system is assigned.
6168 o coding-category-binary
6170 The category for a coding system not categorized in any of the
6171 above. Assigned the coding-system (Lisp symbol)
6172 `no-conversion' by default.
6174 Each of them is a Lisp symbol and the value is an actual
6175 `coding-system's (this is also a Lisp symbol) assigned by a user.
6176 What Emacs does actually is to detect a category of coding system.
6177 Then, it uses a `coding-system' assigned to it. If Emacs can't
6178 decide only one possible category, it selects a category of the
6179 highest priority. Priorities of categories are also specified by a
6180 user in a Lisp variable `coding-category-list'.
6184 static Lisp_Object
adjust_coding_eol_type (struct coding_system
*coding
,
6188 /* Return the number of ASCII characters at the head of the source.
6189 By side effects, set coding->head_ascii and update
6190 coding->eol_seen. The value of coding->eol_seen is "logical or" of
6191 EOL_SEEN_LF, EOL_SEEN_CR, and EOL_SEEN_CRLF, but the value is
6192 reliable only when all the source bytes are ASCII. */
6195 check_ascii (struct coding_system
*coding
)
6197 const unsigned char *src
, *end
;
6198 Lisp_Object eol_type
= CODING_ID_EOL_TYPE (coding
->id
);
6199 int eol_seen
= coding
->eol_seen
;
6201 coding_set_source (coding
);
6202 src
= coding
->source
;
6203 end
= src
+ coding
->src_bytes
;
6205 if (inhibit_eol_conversion
6206 || SYMBOLP (eol_type
))
6208 /* We don't have to check EOL format. */
6209 while (src
< end
&& !( *src
& 0x80))
6212 eol_seen
|= EOL_SEEN_LF
;
6217 end
--; /* We look ahead one byte for "CR LF". */
6229 eol_seen
|= EOL_SEEN_CRLF
;
6233 eol_seen
|= EOL_SEEN_CR
;
6236 eol_seen
|= EOL_SEEN_LF
;
6242 /* All bytes but the last one C are ASCII. */
6246 eol_seen
|= EOL_SEEN_CR
;
6248 eol_seen
|= EOL_SEEN_LF
;
6253 coding
->head_ascii
= src
- coding
->source
;
6254 coding
->eol_seen
= eol_seen
;
6255 return (coding
->head_ascii
);
6259 /* Return the number of characters at the source if all the bytes are
6260 valid UTF-8 (of Unicode range). Otherwise, return -1. By side
6261 effects, update coding->eol_seen. The value of coding->eol_seen is
6262 "logical or" of EOL_SEEN_LF, EOL_SEEN_CR, and EOL_SEEN_CRLF, but
6263 the value is reliable only when all the source bytes are valid
6267 check_utf_8 (struct coding_system
*coding
)
6269 const unsigned char *src
, *end
;
6271 ptrdiff_t nchars
= coding
->head_ascii
;
6273 if (coding
->head_ascii
< 0)
6274 check_ascii (coding
);
6276 coding_set_source (coding
);
6277 src
= coding
->source
+ coding
->head_ascii
;
6278 /* We look ahead one byte for CR LF. */
6279 end
= coding
->source
+ coding
->src_bytes
- 1;
6280 eol_seen
= coding
->eol_seen
;
6285 if (UTF_8_1_OCTET_P (*src
))
6294 eol_seen
|= EOL_SEEN_CRLF
;
6299 eol_seen
|= EOL_SEEN_CR
;
6302 eol_seen
|= EOL_SEEN_LF
;
6305 else if (UTF_8_2_OCTET_LEADING_P (c
))
6307 if (c
< 0xC2 /* overlong sequence */
6309 || ! UTF_8_EXTRA_OCTET_P (src
[1]))
6313 else if (UTF_8_3_OCTET_LEADING_P (c
))
6316 || ! (UTF_8_EXTRA_OCTET_P (src
[1])
6317 && UTF_8_EXTRA_OCTET_P (src
[2])))
6319 c
= (((c
& 0xF) << 12)
6320 | ((src
[1] & 0x3F) << 6) | (src
[2] & 0x3F));
6321 if (c
< 0x800 /* overlong sequence */
6322 || (c
>= 0xd800 && c
< 0xe000)) /* surrogates (invalid) */
6326 else if (UTF_8_4_OCTET_LEADING_P (c
))
6329 || ! (UTF_8_EXTRA_OCTET_P (src
[1])
6330 && UTF_8_EXTRA_OCTET_P (src
[2])
6331 && UTF_8_EXTRA_OCTET_P (src
[3])))
6333 c
= (((c
& 0x7) << 18) | ((src
[1] & 0x3F) << 12)
6334 | ((src
[2] & 0x3F) << 6) | (src
[3] & 0x3F));
6335 if (c
< 0x10000 /* overlong sequence */
6336 || c
>= 0x110000) /* non-Unicode character */
6347 if (! UTF_8_1_OCTET_P (*src
))
6351 eol_seen
|= EOL_SEEN_CR
;
6352 else if (*src
== '\n')
6353 eol_seen
|= EOL_SEEN_LF
;
6355 coding
->eol_seen
= eol_seen
;
6360 /* Detect how end-of-line of a text of length SRC_BYTES pointed by
6361 SOURCE is encoded. If CATEGORY is one of
6362 coding_category_utf_16_XXXX, assume that CR and LF are encoded by
6363 two-byte, else they are encoded by one-byte.
6365 Return one of EOL_SEEN_XXX. */
6367 #define MAX_EOL_CHECK_COUNT 3
6370 detect_eol (const unsigned char *source
, ptrdiff_t src_bytes
,
6371 enum coding_category category
)
6373 const unsigned char *src
= source
, *src_end
= src
+ src_bytes
;
6376 int eol_seen
= EOL_SEEN_NONE
;
6378 if ((1 << category
) & CATEGORY_MASK_UTF_16
)
6380 bool msb
= category
== (coding_category_utf_16_le
6381 | coding_category_utf_16_le_nosig
);
6384 while (src
+ 1 < src_end
)
6387 if (src
[msb
] == 0 && (c
== '\n' || c
== '\r'))
6392 this_eol
= EOL_SEEN_LF
;
6393 else if (src
+ 3 >= src_end
6394 || src
[msb
+ 2] != 0
6395 || src
[lsb
+ 2] != '\n')
6396 this_eol
= EOL_SEEN_CR
;
6399 this_eol
= EOL_SEEN_CRLF
;
6403 if (eol_seen
== EOL_SEEN_NONE
)
6404 /* This is the first end-of-line. */
6405 eol_seen
= this_eol
;
6406 else if (eol_seen
!= this_eol
)
6408 /* The found type is different from what found before.
6409 Allow for stray ^M characters in DOS EOL files. */
6410 if ((eol_seen
== EOL_SEEN_CR
&& this_eol
== EOL_SEEN_CRLF
)
6411 || (eol_seen
== EOL_SEEN_CRLF
6412 && this_eol
== EOL_SEEN_CR
))
6413 eol_seen
= EOL_SEEN_CRLF
;
6416 eol_seen
= EOL_SEEN_LF
;
6420 if (++total
== MAX_EOL_CHECK_COUNT
)
6427 while (src
< src_end
)
6430 if (c
== '\n' || c
== '\r')
6435 this_eol
= EOL_SEEN_LF
;
6436 else if (src
>= src_end
|| *src
!= '\n')
6437 this_eol
= EOL_SEEN_CR
;
6439 this_eol
= EOL_SEEN_CRLF
, src
++;
6441 if (eol_seen
== EOL_SEEN_NONE
)
6442 /* This is the first end-of-line. */
6443 eol_seen
= this_eol
;
6444 else if (eol_seen
!= this_eol
)
6446 /* The found type is different from what found before.
6447 Allow for stray ^M characters in DOS EOL files. */
6448 if ((eol_seen
== EOL_SEEN_CR
&& this_eol
== EOL_SEEN_CRLF
)
6449 || (eol_seen
== EOL_SEEN_CRLF
&& this_eol
== EOL_SEEN_CR
))
6450 eol_seen
= EOL_SEEN_CRLF
;
6453 eol_seen
= EOL_SEEN_LF
;
6457 if (++total
== MAX_EOL_CHECK_COUNT
)
6466 adjust_coding_eol_type (struct coding_system
*coding
, int eol_seen
)
6468 Lisp_Object eol_type
;
6470 eol_type
= CODING_ID_EOL_TYPE (coding
->id
);
6471 if (! VECTORP (eol_type
))
6472 /* Already adjusted. */
6474 if (eol_seen
& EOL_SEEN_LF
)
6476 coding
->id
= CODING_SYSTEM_ID (AREF (eol_type
, 0));
6479 else if (eol_seen
& EOL_SEEN_CRLF
)
6481 coding
->id
= CODING_SYSTEM_ID (AREF (eol_type
, 1));
6484 else if (eol_seen
& EOL_SEEN_CR
)
6486 coding
->id
= CODING_SYSTEM_ID (AREF (eol_type
, 2));
6492 /* Detect how a text specified in CODING is encoded. If a coding
6493 system is detected, update fields of CODING by the detected coding
6497 detect_coding (struct coding_system
*coding
)
6499 const unsigned char *src
, *src_end
;
6500 unsigned int saved_mode
= coding
->mode
;
6501 Lisp_Object found
= Qnil
;
6502 Lisp_Object eol_type
= CODING_ID_EOL_TYPE (coding
->id
);
6504 coding
->consumed
= coding
->consumed_char
= 0;
6505 coding
->produced
= coding
->produced_char
= 0;
6506 coding_set_source (coding
);
6508 src_end
= coding
->source
+ coding
->src_bytes
;
6510 coding
->eol_seen
= EOL_SEEN_NONE
;
6511 /* If we have not yet decided the text encoding type, detect it
6513 if (EQ (CODING_ATTR_TYPE (CODING_ID_ATTRS (coding
->id
)), Qundecided
))
6516 struct coding_detection_info detect_info
;
6517 bool null_byte_found
= 0, eight_bit_found
= 0;
6518 bool inhibit_nbd
= inhibit_flag (coding
->spec
.undecided
.inhibit_nbd
,
6519 inhibit_null_byte_detection
);
6520 bool inhibit_ied
= inhibit_flag (coding
->spec
.undecided
.inhibit_ied
,
6521 inhibit_iso_escape_detection
);
6522 bool prefer_utf_8
= coding
->spec
.undecided
.prefer_utf_8
;
6524 coding
->head_ascii
= 0;
6525 detect_info
.checked
= detect_info
.found
= detect_info
.rejected
= 0;
6526 for (src
= coding
->source
; src
< src_end
; src
++)
6531 eight_bit_found
= 1;
6532 if (null_byte_found
)
6537 if ((c
== ISO_CODE_ESC
|| c
== ISO_CODE_SI
|| c
== ISO_CODE_SO
)
6539 && ! detect_info
.checked
)
6541 if (detect_coding_iso_2022 (coding
, &detect_info
))
6543 /* We have scanned the whole data. */
6544 if (! (detect_info
.rejected
& CATEGORY_MASK_ISO_7_ELSE
))
6546 /* We didn't find an 8-bit code. We may
6547 have found a null-byte, but it's very
6548 rare that a binary file conforms to
6551 coding
->head_ascii
= src
- coding
->source
;
6553 detect_info
.rejected
|= ~CATEGORY_MASK_ISO_ESCAPE
;
6557 else if (! c
&& !inhibit_nbd
)
6559 null_byte_found
= 1;
6560 if (eight_bit_found
)
6563 else if (! disable_ascii_optimization
6564 && ! inhibit_eol_conversion
)
6568 if (src
< src_end
&& src
[1] == '\n')
6570 coding
->eol_seen
|= EOL_SEEN_CRLF
;
6572 if (! eight_bit_found
)
6573 coding
->head_ascii
++;
6576 coding
->eol_seen
|= EOL_SEEN_CR
;
6580 coding
->eol_seen
|= EOL_SEEN_LF
;
6584 if (! eight_bit_found
)
6585 coding
->head_ascii
++;
6587 else if (! eight_bit_found
)
6588 coding
->head_ascii
++;
6591 if (null_byte_found
|| eight_bit_found
6592 || coding
->head_ascii
< coding
->src_bytes
6593 || detect_info
.found
)
6595 enum coding_category category
;
6596 struct coding_system
*this;
6598 if (coding
->head_ascii
== coding
->src_bytes
)
6599 /* As all bytes are 7-bit, we can ignore non-ISO-2022 codings. */
6600 for (i
= 0; i
< coding_category_raw_text
; i
++)
6602 category
= coding_priorities
[i
];
6603 this = coding_categories
+ category
;
6604 if (detect_info
.found
& (1 << category
))
6609 if (null_byte_found
)
6611 detect_info
.checked
|= ~CATEGORY_MASK_UTF_16
;
6612 detect_info
.rejected
|= ~CATEGORY_MASK_UTF_16
;
6614 else if (prefer_utf_8
6615 && detect_coding_utf_8 (coding
, &detect_info
))
6617 detect_info
.checked
|= ~CATEGORY_MASK_UTF_8
;
6618 detect_info
.rejected
|= ~CATEGORY_MASK_UTF_8
;
6620 for (i
= 0; i
< coding_category_raw_text
; i
++)
6622 category
= coding_priorities
[i
];
6623 this = coding_categories
+ category
;
6624 /* Some of this->detector (e.g. detect_coding_sjis)
6625 require this information. */
6626 coding
->id
= this->id
;
6629 /* No coding system of this category is defined. */
6630 detect_info
.rejected
|= (1 << category
);
6632 else if (category
>= coding_category_raw_text
)
6634 else if (detect_info
.checked
& (1 << category
))
6636 if (detect_info
.found
& (1 << category
))
6639 else if ((*(this->detector
)) (coding
, &detect_info
)
6640 && detect_info
.found
& (1 << category
))
6645 if (i
< coding_category_raw_text
)
6647 if (category
== coding_category_utf_8_auto
)
6649 Lisp_Object coding_systems
;
6651 coding_systems
= AREF (CODING_ID_ATTRS (this->id
),
6652 coding_attr_utf_bom
);
6653 if (CONSP (coding_systems
))
6655 if (detect_info
.found
& CATEGORY_MASK_UTF_8_SIG
)
6656 found
= XCAR (coding_systems
);
6658 found
= XCDR (coding_systems
);
6661 found
= CODING_ID_NAME (this->id
);
6663 else if (category
== coding_category_utf_16_auto
)
6665 Lisp_Object coding_systems
;
6667 coding_systems
= AREF (CODING_ID_ATTRS (this->id
),
6668 coding_attr_utf_bom
);
6669 if (CONSP (coding_systems
))
6671 if (detect_info
.found
& CATEGORY_MASK_UTF_16_LE
)
6672 found
= XCAR (coding_systems
);
6673 else if (detect_info
.found
& CATEGORY_MASK_UTF_16_BE
)
6674 found
= XCDR (coding_systems
);
6677 found
= CODING_ID_NAME (this->id
);
6680 found
= CODING_ID_NAME (this->id
);
6682 else if (null_byte_found
)
6683 found
= Qno_conversion
;
6684 else if ((detect_info
.rejected
& CATEGORY_MASK_ANY
)
6685 == CATEGORY_MASK_ANY
)
6687 else if (detect_info
.rejected
)
6688 for (i
= 0; i
< coding_category_raw_text
; i
++)
6689 if (! (detect_info
.rejected
& (1 << coding_priorities
[i
])))
6691 this = coding_categories
+ coding_priorities
[i
];
6692 found
= CODING_ID_NAME (this->id
);
6697 else if (XINT (CODING_ATTR_CATEGORY (CODING_ID_ATTRS (coding
->id
)))
6698 == coding_category_utf_8_auto
)
6700 Lisp_Object coding_systems
;
6701 struct coding_detection_info detect_info
;
6704 = AREF (CODING_ID_ATTRS (coding
->id
), coding_attr_utf_bom
);
6705 detect_info
.found
= detect_info
.rejected
= 0;
6706 if (check_ascii (coding
) == coding
->src_bytes
)
6708 if (CONSP (coding_systems
))
6709 found
= XCDR (coding_systems
);
6713 if (CONSP (coding_systems
)
6714 && detect_coding_utf_8 (coding
, &detect_info
))
6716 if (detect_info
.found
& CATEGORY_MASK_UTF_8_SIG
)
6717 found
= XCAR (coding_systems
);
6719 found
= XCDR (coding_systems
);
6723 else if (XINT (CODING_ATTR_CATEGORY (CODING_ID_ATTRS (coding
->id
)))
6724 == coding_category_utf_16_auto
)
6726 Lisp_Object coding_systems
;
6727 struct coding_detection_info detect_info
;
6730 = AREF (CODING_ID_ATTRS (coding
->id
), coding_attr_utf_bom
);
6731 detect_info
.found
= detect_info
.rejected
= 0;
6732 coding
->head_ascii
= 0;
6733 if (CONSP (coding_systems
)
6734 && detect_coding_utf_16 (coding
, &detect_info
))
6736 if (detect_info
.found
& CATEGORY_MASK_UTF_16_LE
)
6737 found
= XCAR (coding_systems
);
6738 else if (detect_info
.found
& CATEGORY_MASK_UTF_16_BE
)
6739 found
= XCDR (coding_systems
);
6745 int specified_eol
= (VECTORP (eol_type
) ? EOL_SEEN_NONE
6746 : EQ (eol_type
, Qdos
) ? EOL_SEEN_CRLF
6747 : EQ (eol_type
, Qmac
) ? EOL_SEEN_CR
6750 setup_coding_system (found
, coding
);
6751 if (specified_eol
!= EOL_SEEN_NONE
)
6752 adjust_coding_eol_type (coding
, specified_eol
);
6755 coding
->mode
= saved_mode
;
6760 decode_eol (struct coding_system
*coding
)
6762 Lisp_Object eol_type
;
6763 unsigned char *p
, *pbeg
, *pend
;
6765 eol_type
= CODING_ID_EOL_TYPE (coding
->id
);
6766 if (EQ (eol_type
, Qunix
) || inhibit_eol_conversion
)
6769 if (NILP (coding
->dst_object
))
6770 pbeg
= coding
->destination
;
6772 pbeg
= BYTE_POS_ADDR (coding
->dst_pos_byte
);
6773 pend
= pbeg
+ coding
->produced
;
6775 if (VECTORP (eol_type
))
6777 int eol_seen
= EOL_SEEN_NONE
;
6779 for (p
= pbeg
; p
< pend
; p
++)
6782 eol_seen
|= EOL_SEEN_LF
;
6783 else if (*p
== '\r')
6785 if (p
+ 1 < pend
&& *(p
+ 1) == '\n')
6787 eol_seen
|= EOL_SEEN_CRLF
;
6791 eol_seen
|= EOL_SEEN_CR
;
6794 /* Handle DOS-style EOLs in a file with stray ^M characters. */
6795 if ((eol_seen
& EOL_SEEN_CRLF
) != 0
6796 && (eol_seen
& EOL_SEEN_CR
) != 0
6797 && (eol_seen
& EOL_SEEN_LF
) == 0)
6798 eol_seen
= EOL_SEEN_CRLF
;
6799 else if (eol_seen
!= EOL_SEEN_NONE
6800 && eol_seen
!= EOL_SEEN_LF
6801 && eol_seen
!= EOL_SEEN_CRLF
6802 && eol_seen
!= EOL_SEEN_CR
)
6803 eol_seen
= EOL_SEEN_LF
;
6804 if (eol_seen
!= EOL_SEEN_NONE
)
6805 eol_type
= adjust_coding_eol_type (coding
, eol_seen
);
6808 if (EQ (eol_type
, Qmac
))
6810 for (p
= pbeg
; p
< pend
; p
++)
6814 else if (EQ (eol_type
, Qdos
))
6818 if (NILP (coding
->dst_object
))
6820 /* Start deleting '\r' from the tail to minimize the memory
6822 for (p
= pend
- 2; p
>= pbeg
; p
--)
6825 memmove (p
, p
+ 1, pend
-- - p
- 1);
6831 ptrdiff_t pos_byte
= coding
->dst_pos_byte
;
6832 ptrdiff_t pos
= coding
->dst_pos
;
6833 ptrdiff_t pos_end
= pos
+ coding
->produced_char
- 1;
6835 while (pos
< pos_end
)
6837 p
= BYTE_POS_ADDR (pos_byte
);
6838 if (*p
== '\r' && p
[1] == '\n')
6840 del_range_2 (pos
, pos_byte
, pos
+ 1, pos_byte
+ 1, 0);
6845 if (coding
->dst_multibyte
)
6846 pos_byte
+= BYTES_BY_CHAR_HEAD (*p
);
6851 coding
->produced
-= n
;
6852 coding
->produced_char
-= n
;
6857 /* MAX_LOOKUP's maximum value. MAX_LOOKUP is an int and so cannot
6858 exceed INT_MAX. Also, MAX_LOOKUP is multiplied by sizeof (int) for
6859 alloca, so it cannot exceed MAX_ALLOCA / sizeof (int). */
6860 enum { MAX_LOOKUP_MAX
= min (INT_MAX
, MAX_ALLOCA
/ sizeof (int)) };
6862 /* Return a translation table (or list of them) from coding system
6863 attribute vector ATTRS for encoding (if ENCODEP) or decoding (if
6867 get_translation_table (Lisp_Object attrs
, bool encodep
, int *max_lookup
)
6869 Lisp_Object standard
, translation_table
;
6872 if (NILP (Venable_character_translation
))
6879 translation_table
= CODING_ATTR_ENCODE_TBL (attrs
),
6880 standard
= Vstandard_translation_table_for_encode
;
6882 translation_table
= CODING_ATTR_DECODE_TBL (attrs
),
6883 standard
= Vstandard_translation_table_for_decode
;
6884 if (NILP (translation_table
))
6885 translation_table
= standard
;
6888 if (SYMBOLP (translation_table
))
6889 translation_table
= Fget (translation_table
, Qtranslation_table
);
6890 else if (CONSP (translation_table
))
6892 translation_table
= Fcopy_sequence (translation_table
);
6893 for (val
= translation_table
; CONSP (val
); val
= XCDR (val
))
6894 if (SYMBOLP (XCAR (val
)))
6895 XSETCAR (val
, Fget (XCAR (val
), Qtranslation_table
));
6897 if (CHAR_TABLE_P (standard
))
6899 if (CONSP (translation_table
))
6900 translation_table
= nconc2 (translation_table
, list1 (standard
));
6902 translation_table
= list2 (translation_table
, standard
);
6909 if (CHAR_TABLE_P (translation_table
)
6910 && CHAR_TABLE_EXTRA_SLOTS (XCHAR_TABLE (translation_table
)) > 1)
6912 val
= XCHAR_TABLE (translation_table
)->extras
[1];
6913 if (NATNUMP (val
) && *max_lookup
< XFASTINT (val
))
6914 *max_lookup
= min (XFASTINT (val
), MAX_LOOKUP_MAX
);
6916 else if (CONSP (translation_table
))
6920 for (tail
= translation_table
; CONSP (tail
); tail
= XCDR (tail
))
6921 if (CHAR_TABLE_P (XCAR (tail
))
6922 && CHAR_TABLE_EXTRA_SLOTS (XCHAR_TABLE (XCAR (tail
))) > 1)
6924 Lisp_Object tailval
= XCHAR_TABLE (XCAR (tail
))->extras
[1];
6925 if (NATNUMP (tailval
) && *max_lookup
< XFASTINT (tailval
))
6926 *max_lookup
= min (XFASTINT (tailval
), MAX_LOOKUP_MAX
);
6930 return translation_table
;
6933 #define LOOKUP_TRANSLATION_TABLE(table, c, trans) \
6936 if (CHAR_TABLE_P (table)) \
6938 trans = CHAR_TABLE_REF (table, c); \
6939 if (CHARACTERP (trans)) \
6940 c = XFASTINT (trans), trans = Qnil; \
6942 else if (CONSP (table)) \
6946 for (tail = table; CONSP (tail); tail = XCDR (tail)) \
6947 if (CHAR_TABLE_P (XCAR (tail))) \
6949 trans = CHAR_TABLE_REF (XCAR (tail), c); \
6950 if (CHARACTERP (trans)) \
6951 c = XFASTINT (trans), trans = Qnil; \
6952 else if (! NILP (trans)) \
6959 /* Return a translation of character(s) at BUF according to TRANS.
6960 TRANS is TO-CHAR or ((FROM . TO) ...) where
6961 FROM = [FROM-CHAR ...], TO is TO-CHAR or [TO-CHAR ...].
6962 The return value is TO-CHAR or ([FROM-CHAR ...] . TO) if a
6963 translation is found, and Qnil if not found..
6964 If BUF is too short to lookup characters in FROM, return Qt. */
6967 get_translation (Lisp_Object trans
, int *buf
, int *buf_end
)
6970 if (INTEGERP (trans
))
6972 for (; CONSP (trans
); trans
= XCDR (trans
))
6974 Lisp_Object val
= XCAR (trans
);
6975 Lisp_Object from
= XCAR (val
);
6976 ptrdiff_t len
= ASIZE (from
);
6979 for (i
= 0; i
< len
; i
++)
6981 if (buf
+ i
== buf_end
)
6983 if (XINT (AREF (from
, i
)) != buf
[i
])
6994 produce_chars (struct coding_system
*coding
, Lisp_Object translation_table
,
6997 unsigned char *dst
= coding
->destination
+ coding
->produced
;
6998 unsigned char *dst_end
= coding
->destination
+ coding
->dst_bytes
;
7000 ptrdiff_t produced_chars
= 0;
7003 if (! coding
->chars_at_source
)
7005 /* Source characters are in coding->charbuf. */
7006 int *buf
= coding
->charbuf
;
7007 int *buf_end
= buf
+ coding
->charbuf_used
;
7009 if (EQ (coding
->src_object
, coding
->dst_object
)
7010 && ! NILP (coding
->dst_object
))
7012 eassert (growable_destination (coding
));
7013 coding_set_source (coding
);
7014 dst_end
= ((unsigned char *) coding
->source
) + coding
->consumed
;
7017 while (buf
< buf_end
)
7024 ptrdiff_t from_nchars
= 1, to_nchars
= 1;
7025 Lisp_Object trans
= Qnil
;
7027 LOOKUP_TRANSLATION_TABLE (translation_table
, c
, trans
);
7030 trans
= get_translation (trans
, buf
, buf_end
);
7031 if (INTEGERP (trans
))
7033 else if (CONSP (trans
))
7035 from_nchars
= ASIZE (XCAR (trans
));
7036 trans
= XCDR (trans
);
7037 if (INTEGERP (trans
))
7041 to_nchars
= ASIZE (trans
);
7042 c
= XINT (AREF (trans
, 0));
7045 else if (EQ (trans
, Qt
) && ! last_block
)
7049 if ((dst_end
- dst
) / MAX_MULTIBYTE_LENGTH
< to_nchars
)
7051 eassert (growable_destination (coding
));
7052 if (((min (PTRDIFF_MAX
, SIZE_MAX
) - (buf_end
- buf
))
7053 / MAX_MULTIBYTE_LENGTH
)
7055 memory_full (SIZE_MAX
);
7056 dst
= alloc_destination (coding
,
7058 + MAX_MULTIBYTE_LENGTH
* to_nchars
,
7060 if (EQ (coding
->src_object
, coding
->dst_object
))
7062 coding_set_source (coding
);
7063 dst_end
= (((unsigned char *) coding
->source
)
7064 + coding
->consumed
);
7067 dst_end
= coding
->destination
+ coding
->dst_bytes
;
7070 for (i
= 0; i
< to_nchars
; i
++)
7073 c
= XINT (AREF (trans
, i
));
7074 if (coding
->dst_multibyte
7075 || ! CHAR_BYTE8_P (c
))
7076 CHAR_STRING_ADVANCE_NO_UNIFY (c
, dst
);
7078 *dst
++ = CHAR_TO_BYTE8 (c
);
7080 produced_chars
+= to_nchars
;
7084 /* This is an annotation datum. (-C) is the length. */
7087 carryover
= buf_end
- buf
;
7091 /* Source characters are at coding->source. */
7092 const unsigned char *src
= coding
->source
;
7093 const unsigned char *src_end
= src
+ coding
->consumed
;
7095 if (EQ (coding
->dst_object
, coding
->src_object
))
7097 eassert (growable_destination (coding
));
7098 dst_end
= (unsigned char *) src
;
7100 if (coding
->src_multibyte
!= coding
->dst_multibyte
)
7102 if (coding
->src_multibyte
)
7104 bool multibytep
= 1;
7105 ptrdiff_t consumed_chars
= 0;
7109 const unsigned char *src_base
= src
;
7115 eassert (growable_destination (coding
));
7116 if (EQ (coding
->src_object
, coding
->dst_object
))
7117 dst_end
= (unsigned char *) src
;
7120 ptrdiff_t offset
= src
- coding
->source
;
7122 dst
= alloc_destination (coding
, src_end
- src
+ 1,
7124 dst_end
= coding
->destination
+ coding
->dst_bytes
;
7125 coding_set_source (coding
);
7126 src
= coding
->source
+ offset
;
7127 src_end
= coding
->source
+ coding
->consumed
;
7128 if (EQ (coding
->src_object
, coding
->dst_object
))
7129 dst_end
= (unsigned char *) src
;
7139 while (src
< src_end
)
7141 bool multibytep
= 1;
7144 if (dst
>= dst_end
- 1)
7146 eassert (growable_destination (coding
));
7147 if (EQ (coding
->src_object
, coding
->dst_object
))
7148 dst_end
= (unsigned char *) src
;
7149 if (dst
>= dst_end
- 1)
7151 ptrdiff_t offset
= src
- coding
->source
;
7152 ptrdiff_t more_bytes
;
7154 if (EQ (coding
->src_object
, coding
->dst_object
))
7155 more_bytes
= ((src_end
- src
) / 2) + 2;
7157 more_bytes
= src_end
- src
+ 2;
7158 dst
= alloc_destination (coding
, more_bytes
, dst
);
7159 dst_end
= coding
->destination
+ coding
->dst_bytes
;
7160 coding_set_source (coding
);
7161 src
= coding
->source
+ offset
;
7162 src_end
= coding
->source
+ coding
->consumed
;
7163 if (EQ (coding
->src_object
, coding
->dst_object
))
7164 dst_end
= (unsigned char *) src
;
7172 if (!EQ (coding
->src_object
, coding
->dst_object
))
7174 ptrdiff_t require
= coding
->src_bytes
- coding
->dst_bytes
;
7178 ptrdiff_t offset
= src
- coding
->source
;
7180 dst
= alloc_destination (coding
, require
, dst
);
7181 coding_set_source (coding
);
7182 src
= coding
->source
+ offset
;
7183 src_end
= coding
->source
+ coding
->consumed
;
7186 produced_chars
= coding
->consumed_char
;
7187 while (src
< src_end
)
7192 produced
= dst
- (coding
->destination
+ coding
->produced
);
7193 if (BUFFERP (coding
->dst_object
) && produced_chars
> 0)
7194 insert_from_gap (produced_chars
, produced
, 0);
7195 coding
->produced
+= produced
;
7196 coding
->produced_char
+= produced_chars
;
7200 /* Compose text in CODING->object according to the annotation data at
7201 CHARBUF. CHARBUF is an array:
7202 [ -LENGTH ANNOTATION_MASK NCHARS NBYTES METHOD [ COMPONENTS... ] ]
7206 produce_composition (struct coding_system
*coding
, int *charbuf
, ptrdiff_t pos
)
7210 enum composition_method method
;
7211 Lisp_Object components
;
7213 len
= -charbuf
[0] - MAX_ANNOTATION_LENGTH
;
7214 to
= pos
+ charbuf
[2];
7215 method
= (enum composition_method
) (charbuf
[4]);
7217 if (method
== COMPOSITION_RELATIVE
)
7221 Lisp_Object args
[MAX_COMPOSITION_COMPONENTS
* 2 - 1];
7224 if (method
== COMPOSITION_WITH_RULE
)
7225 len
= charbuf
[2] * 3 - 2;
7226 charbuf
+= MAX_ANNOTATION_LENGTH
;
7227 /* charbuf = [ CHRA ... CHAR] or [ CHAR -2 RULE ... CHAR ] */
7228 for (i
= j
= 0; i
< len
&& charbuf
[i
] != -1; i
++, j
++)
7230 if (charbuf
[i
] >= 0)
7231 args
[j
] = make_number (charbuf
[i
]);
7235 args
[j
] = make_number (charbuf
[i
] % 0x100);
7238 components
= (i
== j
? Fstring (j
, args
) : Fvector (j
, args
));
7240 compose_text (pos
, to
, components
, Qnil
, coding
->dst_object
);
7244 /* Put `charset' property on text in CODING->object according to
7245 the annotation data at CHARBUF. CHARBUF is an array:
7246 [ -LENGTH ANNOTATION_MASK NCHARS CHARSET-ID ]
7250 produce_charset (struct coding_system
*coding
, int *charbuf
, ptrdiff_t pos
)
7252 ptrdiff_t from
= pos
- charbuf
[2];
7253 struct charset
*charset
= CHARSET_FROM_ID (charbuf
[3]);
7255 Fput_text_property (make_number (from
), make_number (pos
),
7256 Qcharset
, CHARSET_NAME (charset
),
7257 coding
->dst_object
);
7260 #define MAX_CHARBUF_SIZE 0x4000
7261 /* How many units decoding functions expect in coding->charbuf at
7262 most. Currently, decode_coding_emacs_mule expects the following
7263 size, and that is the largest value. */
7264 #define MAX_CHARBUF_EXTRA_SIZE ((MAX_ANNOTATION_LENGTH * 3) + 1)
7266 #define ALLOC_CONVERSION_WORK_AREA(coding, size) \
7268 ptrdiff_t units = min ((size) + MAX_CHARBUF_EXTRA_SIZE, \
7269 MAX_CHARBUF_SIZE); \
7270 coding->charbuf = SAFE_ALLOCA (units * sizeof (int)); \
7271 coding->charbuf_size = units; \
7275 produce_annotation (struct coding_system
*coding
, ptrdiff_t pos
)
7277 int *charbuf
= coding
->charbuf
;
7278 int *charbuf_end
= charbuf
+ coding
->charbuf_used
;
7280 if (NILP (coding
->dst_object
))
7283 while (charbuf
< charbuf_end
)
7289 int len
= -*charbuf
;
7294 case CODING_ANNOTATE_COMPOSITION_MASK
:
7295 produce_composition (coding
, charbuf
, pos
);
7297 case CODING_ANNOTATE_CHARSET_MASK
:
7298 produce_charset (coding
, charbuf
, pos
);
7308 /* Decode the data at CODING->src_object into CODING->dst_object.
7309 CODING->src_object is a buffer, a string, or nil.
7310 CODING->dst_object is a buffer.
7312 If CODING->src_object is a buffer, it must be the current buffer.
7313 In this case, if CODING->src_pos is positive, it is a position of
7314 the source text in the buffer, otherwise, the source text is in the
7315 gap area of the buffer, and CODING->src_pos specifies the offset of
7316 the text from GPT (which must be the same as PT). If this is the
7317 same buffer as CODING->dst_object, CODING->src_pos must be
7320 If CODING->src_object is a string, CODING->src_pos is an index to
7323 If CODING->src_object is nil, CODING->source must already point to
7324 the non-relocatable memory area. In this case, CODING->src_pos is
7325 an offset from CODING->source.
7327 The decoded data is inserted at the current point of the buffer
7332 decode_coding (struct coding_system
*coding
)
7335 Lisp_Object undo_list
;
7336 Lisp_Object translation_table
;
7337 struct ccl_spec cclspec
;
7343 if (BUFFERP (coding
->src_object
)
7344 && coding
->src_pos
> 0
7345 && coding
->src_pos
< GPT
7346 && coding
->src_pos
+ coding
->src_chars
> GPT
)
7347 move_gap_both (coding
->src_pos
, coding
->src_pos_byte
);
7350 if (BUFFERP (coding
->dst_object
))
7352 set_buffer_internal (XBUFFER (coding
->dst_object
));
7354 move_gap_both (PT
, PT_BYTE
);
7356 /* We must disable undo_list in order to record the whole insert
7357 transaction via record_insert at the end. But doing so also
7358 disables the recording of the first change to the undo_list.
7359 Therefore we check for first change here and record it via
7360 record_first_change if needed. */
7361 if (MODIFF
<= SAVE_MODIFF
)
7362 record_first_change ();
7364 undo_list
= BVAR (current_buffer
, undo_list
);
7365 bset_undo_list (current_buffer
, Qt
);
7368 coding
->consumed
= coding
->consumed_char
= 0;
7369 coding
->produced
= coding
->produced_char
= 0;
7370 coding
->chars_at_source
= 0;
7371 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
7373 ALLOC_CONVERSION_WORK_AREA (coding
, coding
->src_bytes
);
7375 attrs
= CODING_ID_ATTRS (coding
->id
);
7376 translation_table
= get_translation_table (attrs
, 0, NULL
);
7379 if (coding
->decoder
== decode_coding_ccl
)
7381 coding
->spec
.ccl
= &cclspec
;
7382 setup_ccl_program (&cclspec
.ccl
, CODING_CCL_DECODER (coding
));
7386 ptrdiff_t pos
= coding
->dst_pos
+ coding
->produced_char
;
7388 coding_set_source (coding
);
7389 coding
->annotated
= 0;
7390 coding
->charbuf_used
= carryover
;
7391 (*(coding
->decoder
)) (coding
);
7392 coding_set_destination (coding
);
7393 carryover
= produce_chars (coding
, translation_table
, 0);
7394 if (coding
->annotated
)
7395 produce_annotation (coding
, pos
);
7396 for (i
= 0; i
< carryover
; i
++)
7398 = coding
->charbuf
[coding
->charbuf_used
- carryover
+ i
];
7400 while (coding
->result
== CODING_RESULT_INSUFFICIENT_DST
7401 || (coding
->consumed
< coding
->src_bytes
7402 && (coding
->result
== CODING_RESULT_SUCCESS
7403 || coding
->result
== CODING_RESULT_INVALID_SRC
)));
7407 coding_set_destination (coding
);
7408 coding
->charbuf_used
= carryover
;
7409 produce_chars (coding
, translation_table
, 1);
7412 coding
->carryover_bytes
= 0;
7413 if (coding
->consumed
< coding
->src_bytes
)
7415 ptrdiff_t nbytes
= coding
->src_bytes
- coding
->consumed
;
7416 const unsigned char *src
;
7418 coding_set_source (coding
);
7419 coding_set_destination (coding
);
7420 src
= coding
->source
+ coding
->consumed
;
7422 if (coding
->mode
& CODING_MODE_LAST_BLOCK
)
7424 /* Flush out unprocessed data as binary chars. We are sure
7425 that the number of data is less than the size of
7427 coding
->charbuf_used
= 0;
7428 coding
->chars_at_source
= 0;
7430 while (nbytes
-- > 0)
7435 c
= BYTE8_TO_CHAR (c
);
7436 coding
->charbuf
[coding
->charbuf_used
++] = c
;
7438 produce_chars (coding
, Qnil
, 1);
7442 /* Record unprocessed bytes in coding->carryover. We are
7443 sure that the number of data is less than the size of
7444 coding->carryover. */
7445 unsigned char *p
= coding
->carryover
;
7447 if (nbytes
> sizeof coding
->carryover
)
7448 nbytes
= sizeof coding
->carryover
;
7449 coding
->carryover_bytes
= nbytes
;
7450 while (nbytes
-- > 0)
7453 coding
->consumed
= coding
->src_bytes
;
7456 if (! EQ (CODING_ID_EOL_TYPE (coding
->id
), Qunix
)
7457 && !inhibit_eol_conversion
)
7458 decode_eol (coding
);
7459 if (BUFFERP (coding
->dst_object
))
7461 bset_undo_list (current_buffer
, undo_list
);
7462 record_insert (coding
->dst_pos
, coding
->produced_char
);
7469 /* Extract an annotation datum from a composition starting at POS and
7470 ending before LIMIT of CODING->src_object (buffer or string), store
7471 the data in BUF, set *STOP to a starting position of the next
7472 composition (if any) or to LIMIT, and return the address of the
7473 next element of BUF.
7475 If such an annotation is not found, set *STOP to a starting
7476 position of a composition after POS (if any) or to LIMIT, and
7480 handle_composition_annotation (ptrdiff_t pos
, ptrdiff_t limit
,
7481 struct coding_system
*coding
, int *buf
,
7484 ptrdiff_t start
, end
;
7487 if (! find_composition (pos
, limit
, &start
, &end
, &prop
, coding
->src_object
)
7490 else if (start
> pos
)
7496 /* We found a composition. Store the corresponding
7497 annotation data in BUF. */
7499 enum composition_method method
= composition_method (prop
);
7500 int nchars
= COMPOSITION_LENGTH (prop
);
7502 ADD_COMPOSITION_DATA (buf
, nchars
, 0, method
);
7503 if (method
!= COMPOSITION_RELATIVE
)
7505 Lisp_Object components
;
7506 ptrdiff_t i
, len
, i_byte
;
7508 components
= COMPOSITION_COMPONENTS (prop
);
7509 if (VECTORP (components
))
7511 len
= ASIZE (components
);
7512 for (i
= 0; i
< len
; i
++)
7513 *buf
++ = XINT (AREF (components
, i
));
7515 else if (STRINGP (components
))
7517 len
= SCHARS (components
);
7521 FETCH_STRING_CHAR_ADVANCE (*buf
, components
, i
, i_byte
);
7525 else if (INTEGERP (components
))
7528 *buf
++ = XINT (components
);
7530 else if (CONSP (components
))
7532 for (len
= 0; CONSP (components
);
7533 len
++, components
= XCDR (components
))
7534 *buf
++ = XINT (XCAR (components
));
7542 if (find_composition (end
, limit
, &start
, &end
, &prop
,
7553 /* Extract an annotation datum from a text property `charset' at POS of
7554 CODING->src_object (buffer of string), store the data in BUF, set
7555 *STOP to the position where the value of `charset' property changes
7556 (limiting by LIMIT), and return the address of the next element of
7559 If the property value is nil, set *STOP to the position where the
7560 property value is non-nil (limiting by LIMIT), and return BUF. */
7563 handle_charset_annotation (ptrdiff_t pos
, ptrdiff_t limit
,
7564 struct coding_system
*coding
, int *buf
,
7567 Lisp_Object val
, next
;
7570 val
= Fget_text_property (make_number (pos
), Qcharset
, coding
->src_object
);
7571 if (! NILP (val
) && CHARSETP (val
))
7572 id
= XINT (CHARSET_SYMBOL_ID (val
));
7575 ADD_CHARSET_DATA (buf
, 0, id
);
7576 next
= Fnext_single_property_change (make_number (pos
), Qcharset
,
7578 make_number (limit
));
7579 *stop
= XINT (next
);
7585 consume_chars (struct coding_system
*coding
, Lisp_Object translation_table
,
7588 int *buf
= coding
->charbuf
;
7589 int *buf_end
= coding
->charbuf
+ coding
->charbuf_size
;
7590 const unsigned char *src
= coding
->source
+ coding
->consumed
;
7591 const unsigned char *src_end
= coding
->source
+ coding
->src_bytes
;
7592 ptrdiff_t pos
= coding
->src_pos
+ coding
->consumed_char
;
7593 ptrdiff_t end_pos
= coding
->src_pos
+ coding
->src_chars
;
7594 bool multibytep
= coding
->src_multibyte
;
7595 Lisp_Object eol_type
;
7597 ptrdiff_t stop
, stop_composition
, stop_charset
;
7598 int *lookup_buf
= NULL
;
7600 if (! NILP (translation_table
))
7601 lookup_buf
= alloca (sizeof (int) * max_lookup
);
7603 eol_type
= inhibit_eol_conversion
? Qunix
: CODING_ID_EOL_TYPE (coding
->id
);
7604 if (VECTORP (eol_type
))
7607 /* Note: composition handling is not yet implemented. */
7608 coding
->common_flags
&= ~CODING_ANNOTATE_COMPOSITION_MASK
;
7610 if (NILP (coding
->src_object
))
7611 stop
= stop_composition
= stop_charset
= end_pos
;
7614 if (coding
->common_flags
& CODING_ANNOTATE_COMPOSITION_MASK
)
7615 stop
= stop_composition
= pos
;
7617 stop
= stop_composition
= end_pos
;
7618 if (coding
->common_flags
& CODING_ANNOTATE_CHARSET_MASK
)
7619 stop
= stop_charset
= pos
;
7621 stop_charset
= end_pos
;
7624 /* Compensate for CRLF and conversion. */
7625 buf_end
-= 1 + MAX_ANNOTATION_LENGTH
;
7626 while (buf
< buf_end
)
7634 if (pos
== stop_composition
)
7635 buf
= handle_composition_annotation (pos
, end_pos
, coding
,
7636 buf
, &stop_composition
);
7637 if (pos
== stop_charset
)
7638 buf
= handle_charset_annotation (pos
, end_pos
, coding
,
7639 buf
, &stop_charset
);
7640 stop
= (stop_composition
< stop_charset
7641 ? stop_composition
: stop_charset
);
7648 if (coding
->encoder
== encode_coding_raw_text
7649 || coding
->encoder
== encode_coding_ccl
)
7651 else if ((bytes
= MULTIBYTE_LENGTH (src
, src_end
)) > 0)
7652 c
= STRING_CHAR_ADVANCE_NO_UNIFY (src
), pos
+= bytes
;
7654 c
= BYTE8_TO_CHAR (*src
), src
++, pos
++;
7657 c
= STRING_CHAR_ADVANCE_NO_UNIFY (src
), pos
++;
7658 if ((c
== '\r') && (coding
->mode
& CODING_MODE_SELECTIVE_DISPLAY
))
7660 if (! EQ (eol_type
, Qunix
))
7664 if (EQ (eol_type
, Qdos
))
7672 LOOKUP_TRANSLATION_TABLE (translation_table
, c
, trans
);
7677 ptrdiff_t from_nchars
= 1, to_nchars
= 1;
7678 int *lookup_buf_end
;
7679 const unsigned char *p
= src
;
7683 for (i
= 1; i
< max_lookup
&& p
< src_end
; i
++)
7684 lookup_buf
[i
] = STRING_CHAR_ADVANCE (p
);
7685 lookup_buf_end
= lookup_buf
+ i
;
7686 trans
= get_translation (trans
, lookup_buf
, lookup_buf_end
);
7687 if (INTEGERP (trans
))
7689 else if (CONSP (trans
))
7691 from_nchars
= ASIZE (XCAR (trans
));
7692 trans
= XCDR (trans
);
7693 if (INTEGERP (trans
))
7697 to_nchars
= ASIZE (trans
);
7698 if (buf_end
- buf
< to_nchars
)
7700 c
= XINT (AREF (trans
, 0));
7706 for (i
= 1; i
< to_nchars
; i
++)
7707 *buf
++ = XINT (AREF (trans
, i
));
7708 for (i
= 1; i
< from_nchars
; i
++, pos
++)
7709 src
+= MULTIBYTE_LENGTH_NO_CHECK (src
);
7713 coding
->consumed
= src
- coding
->source
;
7714 coding
->consumed_char
= pos
- coding
->src_pos
;
7715 coding
->charbuf_used
= buf
- coding
->charbuf
;
7716 coding
->chars_at_source
= 0;
7720 /* Encode the text at CODING->src_object into CODING->dst_object.
7721 CODING->src_object is a buffer or a string.
7722 CODING->dst_object is a buffer or nil.
7724 If CODING->src_object is a buffer, it must be the current buffer.
7725 In this case, if CODING->src_pos is positive, it is a position of
7726 the source text in the buffer, otherwise. the source text is in the
7727 gap area of the buffer, and coding->src_pos specifies the offset of
7728 the text from GPT (which must be the same as PT). If this is the
7729 same buffer as CODING->dst_object, CODING->src_pos must be
7730 negative and CODING should not have `pre-write-conversion'.
7732 If CODING->src_object is a string, CODING should not have
7733 `pre-write-conversion'.
7735 If CODING->dst_object is a buffer, the encoded data is inserted at
7736 the current point of that buffer.
7738 If CODING->dst_object is nil, the encoded data is placed at the
7739 memory area specified by CODING->destination. */
7742 encode_coding (struct coding_system
*coding
)
7745 Lisp_Object translation_table
;
7747 struct ccl_spec cclspec
;
7751 attrs
= CODING_ID_ATTRS (coding
->id
);
7752 if (coding
->encoder
== encode_coding_raw_text
)
7753 translation_table
= Qnil
, max_lookup
= 0;
7755 translation_table
= get_translation_table (attrs
, 1, &max_lookup
);
7757 if (BUFFERP (coding
->dst_object
))
7759 set_buffer_internal (XBUFFER (coding
->dst_object
));
7760 coding
->dst_multibyte
7761 = ! NILP (BVAR (current_buffer
, enable_multibyte_characters
));
7764 coding
->consumed
= coding
->consumed_char
= 0;
7765 coding
->produced
= coding
->produced_char
= 0;
7766 record_conversion_result (coding
, CODING_RESULT_SUCCESS
);
7768 ALLOC_CONVERSION_WORK_AREA (coding
, coding
->src_chars
);
7770 if (coding
->encoder
== encode_coding_ccl
)
7772 coding
->spec
.ccl
= &cclspec
;
7773 setup_ccl_program (&cclspec
.ccl
, CODING_CCL_ENCODER (coding
));
7776 coding_set_source (coding
);
7777 consume_chars (coding
, translation_table
, max_lookup
);
7778 coding_set_destination (coding
);
7779 (*(coding
->encoder
)) (coding
);
7780 } while (coding
->consumed_char
< coding
->src_chars
);
7782 if (BUFFERP (coding
->dst_object
) && coding
->produced_char
> 0)
7783 insert_from_gap (coding
->produced_char
, coding
->produced
, 0);
7789 /* Name (or base name) of work buffer for code conversion. */
7790 static Lisp_Object Vcode_conversion_workbuf_name
;
7792 /* A working buffer used by the top level conversion. Once it is
7793 created, it is never destroyed. It has the name
7794 Vcode_conversion_workbuf_name. The other working buffers are
7795 destroyed after the use is finished, and their names are modified
7796 versions of Vcode_conversion_workbuf_name. */
7797 static Lisp_Object Vcode_conversion_reused_workbuf
;
7799 /* True iff Vcode_conversion_reused_workbuf is already in use. */
7800 static bool reused_workbuf_in_use
;
7803 /* Return a working buffer of code conversion. MULTIBYTE specifies the
7804 multibyteness of returning buffer. */
7807 make_conversion_work_buffer (bool multibyte
)
7809 Lisp_Object name
, workbuf
;
7810 struct buffer
*current
;
7812 if (reused_workbuf_in_use
)
7814 name
= Fgenerate_new_buffer_name (Vcode_conversion_workbuf_name
, Qnil
);
7815 workbuf
= Fget_buffer_create (name
);
7819 reused_workbuf_in_use
= 1;
7820 if (NILP (Fbuffer_live_p (Vcode_conversion_reused_workbuf
)))
7821 Vcode_conversion_reused_workbuf
7822 = Fget_buffer_create (Vcode_conversion_workbuf_name
);
7823 workbuf
= Vcode_conversion_reused_workbuf
;
7825 current
= current_buffer
;
7826 set_buffer_internal (XBUFFER (workbuf
));
7827 /* We can't allow modification hooks to run in the work buffer. For
7828 instance, directory_files_internal assumes that file decoding
7829 doesn't compile new regexps. */
7830 Fset (Fmake_local_variable (Qinhibit_modification_hooks
), Qt
);
7832 bset_undo_list (current_buffer
, Qt
);
7833 bset_enable_multibyte_characters (current_buffer
, multibyte
? Qt
: Qnil
);
7834 set_buffer_internal (current
);
7840 code_conversion_restore (Lisp_Object arg
)
7842 Lisp_Object current
, workbuf
;
7844 current
= XCAR (arg
);
7845 workbuf
= XCDR (arg
);
7846 if (! NILP (workbuf
))
7848 if (EQ (workbuf
, Vcode_conversion_reused_workbuf
))
7849 reused_workbuf_in_use
= 0;
7851 Fkill_buffer (workbuf
);
7853 set_buffer_internal (XBUFFER (current
));
7857 code_conversion_save (bool with_work_buf
, bool multibyte
)
7859 Lisp_Object workbuf
= Qnil
;
7862 workbuf
= make_conversion_work_buffer (multibyte
);
7863 record_unwind_protect (code_conversion_restore
,
7864 Fcons (Fcurrent_buffer (), workbuf
));
7869 decode_coding_gap (struct coding_system
*coding
,
7870 ptrdiff_t chars
, ptrdiff_t bytes
)
7872 ptrdiff_t count
= SPECPDL_INDEX ();
7875 coding
->src_object
= Fcurrent_buffer ();
7876 coding
->src_chars
= chars
;
7877 coding
->src_bytes
= bytes
;
7878 coding
->src_pos
= -chars
;
7879 coding
->src_pos_byte
= -bytes
;
7880 coding
->src_multibyte
= chars
< bytes
;
7881 coding
->dst_object
= coding
->src_object
;
7882 coding
->dst_pos
= PT
;
7883 coding
->dst_pos_byte
= PT_BYTE
;
7884 coding
->dst_multibyte
= ! NILP (BVAR (current_buffer
, enable_multibyte_characters
));
7886 coding
->head_ascii
= -1;
7887 coding
->detected_utf8_bytes
= coding
->detected_utf8_chars
= -1;
7888 coding
->eol_seen
= EOL_SEEN_NONE
;
7889 if (CODING_REQUIRE_DETECTION (coding
))
7890 detect_coding (coding
);
7891 attrs
= CODING_ID_ATTRS (coding
->id
);
7892 if (! disable_ascii_optimization
7893 && ! coding
->src_multibyte
7894 && ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
))
7895 && NILP (CODING_ATTR_POST_READ (attrs
))
7896 && NILP (get_translation_table (attrs
, 0, NULL
)))
7898 chars
= coding
->head_ascii
;
7900 chars
= check_ascii (coding
);
7903 /* There exists a non-ASCII byte. */
7904 if (EQ (CODING_ATTR_TYPE (attrs
), Qutf_8
)
7905 && coding
->detected_utf8_bytes
== coding
->src_bytes
)
7907 if (coding
->detected_utf8_chars
>= 0)
7908 chars
= coding
->detected_utf8_chars
;
7910 chars
= check_utf_8 (coding
);
7911 if (CODING_UTF_8_BOM (coding
) != utf_without_bom
7912 && coding
->head_ascii
== 0
7913 && coding
->source
[0] == UTF_8_BOM_1
7914 && coding
->source
[1] == UTF_8_BOM_2
7915 && coding
->source
[2] == UTF_8_BOM_3
)
7919 coding
->src_bytes
-= 3;
7927 Lisp_Object eol_type
;
7929 eol_type
= CODING_ID_EOL_TYPE (coding
->id
);
7930 if (VECTORP (eol_type
))
7932 if (coding
->eol_seen
!= EOL_SEEN_NONE
)
7933 eol_type
= adjust_coding_eol_type (coding
, coding
->eol_seen
);
7935 if (EQ (eol_type
, Qmac
))
7937 unsigned char *src_end
= GAP_END_ADDR
;
7938 unsigned char *src
= src_end
- coding
->src_bytes
;
7940 while (src
< src_end
)
7946 else if (EQ (eol_type
, Qdos
))
7948 unsigned char *src
= GAP_END_ADDR
;
7949 unsigned char *src_beg
= src
- coding
->src_bytes
;
7950 unsigned char *dst
= src
;
7953 while (src_beg
< src
)
7956 if (*src
== '\n' && src
> src_beg
&& src
[-1] == '\r')
7963 coding
->produced
= bytes
;
7964 coding
->produced_char
= chars
;
7965 insert_from_gap (chars
, bytes
, 1);
7969 code_conversion_save (0, 0);
7971 coding
->mode
|= CODING_MODE_LAST_BLOCK
;
7972 current_buffer
->text
->inhibit_shrinking
= 1;
7973 decode_coding (coding
);
7974 current_buffer
->text
->inhibit_shrinking
= 0;
7976 if (! NILP (CODING_ATTR_POST_READ (attrs
)))
7978 ptrdiff_t prev_Z
= Z
, prev_Z_BYTE
= Z_BYTE
;
7981 TEMP_SET_PT_BOTH (coding
->dst_pos
, coding
->dst_pos_byte
);
7982 val
= call1 (CODING_ATTR_POST_READ (attrs
),
7983 make_number (coding
->produced_char
));
7985 coding
->produced_char
+= Z
- prev_Z
;
7986 coding
->produced
+= Z_BYTE
- prev_Z_BYTE
;
7989 unbind_to (count
, Qnil
);
7993 /* Decode the text in the range FROM/FROM_BYTE and TO/TO_BYTE in
7994 SRC_OBJECT into DST_OBJECT by coding context CODING.
7996 SRC_OBJECT is a buffer, a string, or Qnil.
7998 If it is a buffer, the text is at point of the buffer. FROM and TO
7999 are positions in the buffer.
8001 If it is a string, the text is at the beginning of the string.
8002 FROM and TO are indices to the string.
8004 If it is nil, the text is at coding->source. FROM and TO are
8005 indices to coding->source.
8007 DST_OBJECT is a buffer, Qt, or Qnil.
8009 If it is a buffer, the decoded text is inserted at point of the
8010 buffer. If the buffer is the same as SRC_OBJECT, the source text
8013 If it is Qt, a string is made from the decoded text, and
8014 set in CODING->dst_object.
8016 If it is Qnil, the decoded text is stored at CODING->destination.
8017 The caller must allocate CODING->dst_bytes bytes at
8018 CODING->destination by xmalloc. If the decoded text is longer than
8019 CODING->dst_bytes, CODING->destination is relocated by xrealloc.
8023 decode_coding_object (struct coding_system
*coding
,
8024 Lisp_Object src_object
,
8025 ptrdiff_t from
, ptrdiff_t from_byte
,
8026 ptrdiff_t to
, ptrdiff_t to_byte
,
8027 Lisp_Object dst_object
)
8029 ptrdiff_t count
= SPECPDL_INDEX ();
8030 unsigned char *destination
IF_LINT (= NULL
);
8031 ptrdiff_t dst_bytes
IF_LINT (= 0);
8032 ptrdiff_t chars
= to
- from
;
8033 ptrdiff_t bytes
= to_byte
- from_byte
;
8035 ptrdiff_t saved_pt
= -1, saved_pt_byte
IF_LINT (= 0);
8036 bool need_marker_adjustment
= 0;
8037 Lisp_Object old_deactivate_mark
;
8039 old_deactivate_mark
= Vdeactivate_mark
;
8041 if (NILP (dst_object
))
8043 destination
= coding
->destination
;
8044 dst_bytes
= coding
->dst_bytes
;
8047 coding
->src_object
= src_object
;
8048 coding
->src_chars
= chars
;
8049 coding
->src_bytes
= bytes
;
8050 coding
->src_multibyte
= chars
< bytes
;
8052 if (STRINGP (src_object
))
8054 coding
->src_pos
= from
;
8055 coding
->src_pos_byte
= from_byte
;
8057 else if (BUFFERP (src_object
))
8059 set_buffer_internal (XBUFFER (src_object
));
8061 move_gap_both (from
, from_byte
);
8062 if (EQ (src_object
, dst_object
))
8064 struct Lisp_Marker
*tail
;
8066 for (tail
= BUF_MARKERS (current_buffer
); tail
; tail
= tail
->next
)
8068 tail
->need_adjustment
8069 = tail
->charpos
== (tail
->insertion_type
? from
: to
);
8070 need_marker_adjustment
|= tail
->need_adjustment
;
8072 saved_pt
= PT
, saved_pt_byte
= PT_BYTE
;
8073 TEMP_SET_PT_BOTH (from
, from_byte
);
8074 current_buffer
->text
->inhibit_shrinking
= 1;
8075 del_range_both (from
, from_byte
, to
, to_byte
, 1);
8076 coding
->src_pos
= -chars
;
8077 coding
->src_pos_byte
= -bytes
;
8081 coding
->src_pos
= from
;
8082 coding
->src_pos_byte
= from_byte
;
8086 if (CODING_REQUIRE_DETECTION (coding
))
8087 detect_coding (coding
);
8088 attrs
= CODING_ID_ATTRS (coding
->id
);
8090 if (EQ (dst_object
, Qt
)
8091 || (! NILP (CODING_ATTR_POST_READ (attrs
))
8092 && NILP (dst_object
)))
8094 coding
->dst_multibyte
= !CODING_FOR_UNIBYTE (coding
);
8095 coding
->dst_object
= code_conversion_save (1, coding
->dst_multibyte
);
8096 coding
->dst_pos
= BEG
;
8097 coding
->dst_pos_byte
= BEG_BYTE
;
8099 else if (BUFFERP (dst_object
))
8101 code_conversion_save (0, 0);
8102 coding
->dst_object
= dst_object
;
8103 coding
->dst_pos
= BUF_PT (XBUFFER (dst_object
));
8104 coding
->dst_pos_byte
= BUF_PT_BYTE (XBUFFER (dst_object
));
8105 coding
->dst_multibyte
8106 = ! NILP (BVAR (XBUFFER (dst_object
), enable_multibyte_characters
));
8110 code_conversion_save (0, 0);
8111 coding
->dst_object
= Qnil
;
8112 /* Most callers presume this will return a multibyte result, and they
8113 won't use `binary' or `raw-text' anyway, so let's not worry about
8114 CODING_FOR_UNIBYTE. */
8115 coding
->dst_multibyte
= 1;
8118 decode_coding (coding
);
8120 if (BUFFERP (coding
->dst_object
))
8121 set_buffer_internal (XBUFFER (coding
->dst_object
));
8123 if (! NILP (CODING_ATTR_POST_READ (attrs
)))
8125 ptrdiff_t prev_Z
= Z
, prev_Z_BYTE
= Z_BYTE
;
8128 TEMP_SET_PT_BOTH (coding
->dst_pos
, coding
->dst_pos_byte
);
8129 val
= safe_call1 (CODING_ATTR_POST_READ (attrs
),
8130 make_number (coding
->produced_char
));
8132 coding
->produced_char
+= Z
- prev_Z
;
8133 coding
->produced
+= Z_BYTE
- prev_Z_BYTE
;
8136 if (EQ (dst_object
, Qt
))
8138 coding
->dst_object
= Fbuffer_string ();
8140 else if (NILP (dst_object
) && BUFFERP (coding
->dst_object
))
8142 set_buffer_internal (XBUFFER (coding
->dst_object
));
8143 if (dst_bytes
< coding
->produced
)
8145 eassert (coding
->produced
> 0);
8146 destination
= xrealloc (destination
, coding
->produced
);
8147 if (BEGV
< GPT
&& GPT
< BEGV
+ coding
->produced_char
)
8148 move_gap_both (BEGV
, BEGV_BYTE
);
8149 memcpy (destination
, BEGV_ADDR
, coding
->produced
);
8150 coding
->destination
= destination
;
8156 /* This is the case of:
8157 (BUFFERP (src_object) && EQ (src_object, dst_object))
8158 As we have moved PT while replacing the original buffer
8159 contents, we must recover it now. */
8160 set_buffer_internal (XBUFFER (src_object
));
8161 current_buffer
->text
->inhibit_shrinking
= 0;
8162 if (saved_pt
< from
)
8163 TEMP_SET_PT_BOTH (saved_pt
, saved_pt_byte
);
8164 else if (saved_pt
< from
+ chars
)
8165 TEMP_SET_PT_BOTH (from
, from_byte
);
8166 else if (! NILP (BVAR (current_buffer
, enable_multibyte_characters
)))
8167 TEMP_SET_PT_BOTH (saved_pt
+ (coding
->produced_char
- chars
),
8168 saved_pt_byte
+ (coding
->produced
- bytes
));
8170 TEMP_SET_PT_BOTH (saved_pt
+ (coding
->produced
- bytes
),
8171 saved_pt_byte
+ (coding
->produced
- bytes
));
8173 if (need_marker_adjustment
)
8175 struct Lisp_Marker
*tail
;
8177 for (tail
= BUF_MARKERS (current_buffer
); tail
; tail
= tail
->next
)
8178 if (tail
->need_adjustment
)
8180 tail
->need_adjustment
= 0;
8181 if (tail
->insertion_type
)
8183 tail
->bytepos
= from_byte
;
8184 tail
->charpos
= from
;
8188 tail
->bytepos
= from_byte
+ coding
->produced
;
8190 = (NILP (BVAR (current_buffer
, enable_multibyte_characters
))
8191 ? tail
->bytepos
: from
+ coding
->produced_char
);
8197 Vdeactivate_mark
= old_deactivate_mark
;
8198 unbind_to (count
, coding
->dst_object
);
8203 encode_coding_object (struct coding_system
*coding
,
8204 Lisp_Object src_object
,
8205 ptrdiff_t from
, ptrdiff_t from_byte
,
8206 ptrdiff_t to
, ptrdiff_t to_byte
,
8207 Lisp_Object dst_object
)
8209 ptrdiff_t count
= SPECPDL_INDEX ();
8210 ptrdiff_t chars
= to
- from
;
8211 ptrdiff_t bytes
= to_byte
- from_byte
;
8213 ptrdiff_t saved_pt
= -1, saved_pt_byte
IF_LINT (= 0);
8214 bool need_marker_adjustment
= 0;
8215 bool kill_src_buffer
= 0;
8216 Lisp_Object old_deactivate_mark
;
8218 old_deactivate_mark
= Vdeactivate_mark
;
8220 coding
->src_object
= src_object
;
8221 coding
->src_chars
= chars
;
8222 coding
->src_bytes
= bytes
;
8223 coding
->src_multibyte
= chars
< bytes
;
8225 attrs
= CODING_ID_ATTRS (coding
->id
);
8227 if (EQ (src_object
, dst_object
))
8229 struct Lisp_Marker
*tail
;
8231 for (tail
= BUF_MARKERS (current_buffer
); tail
; tail
= tail
->next
)
8233 tail
->need_adjustment
8234 = tail
->charpos
== (tail
->insertion_type
? from
: to
);
8235 need_marker_adjustment
|= tail
->need_adjustment
;
8239 if (! NILP (CODING_ATTR_PRE_WRITE (attrs
)))
8241 coding
->src_object
= code_conversion_save (1, coding
->src_multibyte
);
8242 set_buffer_internal (XBUFFER (coding
->src_object
));
8243 if (STRINGP (src_object
))
8244 insert_from_string (src_object
, from
, from_byte
, chars
, bytes
, 0);
8245 else if (BUFFERP (src_object
))
8246 insert_from_buffer (XBUFFER (src_object
), from
, chars
, 0);
8248 insert_1_both ((char *) coding
->source
+ from
, chars
, bytes
, 0, 0, 0);
8250 if (EQ (src_object
, dst_object
))
8252 set_buffer_internal (XBUFFER (src_object
));
8253 saved_pt
= PT
, saved_pt_byte
= PT_BYTE
;
8254 del_range_both (from
, from_byte
, to
, to_byte
, 1);
8255 set_buffer_internal (XBUFFER (coding
->src_object
));
8258 safe_call2 (CODING_ATTR_PRE_WRITE (attrs
),
8259 make_number (BEG
), make_number (Z
));
8260 if (XBUFFER (coding
->src_object
) != current_buffer
)
8261 kill_src_buffer
= 1;
8262 coding
->src_object
= Fcurrent_buffer ();
8264 move_gap_both (BEG
, BEG_BYTE
);
8265 coding
->src_chars
= Z
- BEG
;
8266 coding
->src_bytes
= Z_BYTE
- BEG_BYTE
;
8267 coding
->src_pos
= BEG
;
8268 coding
->src_pos_byte
= BEG_BYTE
;
8269 coding
->src_multibyte
= Z
< Z_BYTE
;
8271 else if (STRINGP (src_object
))
8273 code_conversion_save (0, 0);
8274 coding
->src_pos
= from
;
8275 coding
->src_pos_byte
= from_byte
;
8277 else if (BUFFERP (src_object
))
8279 code_conversion_save (0, 0);
8280 set_buffer_internal (XBUFFER (src_object
));
8281 if (EQ (src_object
, dst_object
))
8283 saved_pt
= PT
, saved_pt_byte
= PT_BYTE
;
8284 coding
->src_object
= del_range_1 (from
, to
, 1, 1);
8285 coding
->src_pos
= 0;
8286 coding
->src_pos_byte
= 0;
8290 if (from
< GPT
&& to
>= GPT
)
8291 move_gap_both (from
, from_byte
);
8292 coding
->src_pos
= from
;
8293 coding
->src_pos_byte
= from_byte
;
8298 code_conversion_save (0, 0);
8299 coding
->src_pos
= from
;
8300 coding
->src_pos_byte
= from_byte
;
8303 if (BUFFERP (dst_object
))
8305 coding
->dst_object
= dst_object
;
8306 if (EQ (src_object
, dst_object
))
8308 coding
->dst_pos
= from
;
8309 coding
->dst_pos_byte
= from_byte
;
8313 struct buffer
*current
= current_buffer
;
8315 set_buffer_temp (XBUFFER (dst_object
));
8316 coding
->dst_pos
= PT
;
8317 coding
->dst_pos_byte
= PT_BYTE
;
8318 move_gap_both (coding
->dst_pos
, coding
->dst_pos_byte
);
8319 set_buffer_temp (current
);
8321 coding
->dst_multibyte
8322 = ! NILP (BVAR (XBUFFER (dst_object
), enable_multibyte_characters
));
8324 else if (EQ (dst_object
, Qt
))
8326 ptrdiff_t dst_bytes
= max (1, coding
->src_chars
);
8327 coding
->dst_object
= Qnil
;
8328 coding
->destination
= xmalloc (dst_bytes
);
8329 coding
->dst_bytes
= dst_bytes
;
8330 coding
->dst_multibyte
= 0;
8334 coding
->dst_object
= Qnil
;
8335 coding
->dst_multibyte
= 0;
8338 encode_coding (coding
);
8340 if (EQ (dst_object
, Qt
))
8342 if (BUFFERP (coding
->dst_object
))
8343 coding
->dst_object
= Fbuffer_string ();
8344 else if (coding
->raw_destination
)
8345 /* This is used to avoid creating huge Lisp string.
8346 NOTE: caller who sets `raw_destination' is also
8347 responsible for freeing `destination' buffer. */
8348 coding
->dst_object
= Qnil
;
8352 = make_unibyte_string ((char *) coding
->destination
,
8354 xfree (coding
->destination
);
8360 /* This is the case of:
8361 (BUFFERP (src_object) && EQ (src_object, dst_object))
8362 As we have moved PT while replacing the original buffer
8363 contents, we must recover it now. */
8364 set_buffer_internal (XBUFFER (src_object
));
8365 if (saved_pt
< from
)
8366 TEMP_SET_PT_BOTH (saved_pt
, saved_pt_byte
);
8367 else if (saved_pt
< from
+ chars
)
8368 TEMP_SET_PT_BOTH (from
, from_byte
);
8369 else if (! NILP (BVAR (current_buffer
, enable_multibyte_characters
)))
8370 TEMP_SET_PT_BOTH (saved_pt
+ (coding
->produced_char
- chars
),
8371 saved_pt_byte
+ (coding
->produced
- bytes
));
8373 TEMP_SET_PT_BOTH (saved_pt
+ (coding
->produced
- bytes
),
8374 saved_pt_byte
+ (coding
->produced
- bytes
));
8376 if (need_marker_adjustment
)
8378 struct Lisp_Marker
*tail
;
8380 for (tail
= BUF_MARKERS (current_buffer
); tail
; tail
= tail
->next
)
8381 if (tail
->need_adjustment
)
8383 tail
->need_adjustment
= 0;
8384 if (tail
->insertion_type
)
8386 tail
->bytepos
= from_byte
;
8387 tail
->charpos
= from
;
8391 tail
->bytepos
= from_byte
+ coding
->produced
;
8393 = (NILP (BVAR (current_buffer
, enable_multibyte_characters
))
8394 ? tail
->bytepos
: from
+ coding
->produced_char
);
8400 if (kill_src_buffer
)
8401 Fkill_buffer (coding
->src_object
);
8403 Vdeactivate_mark
= old_deactivate_mark
;
8404 unbind_to (count
, Qnil
);
8409 preferred_coding_system (void)
8411 int id
= coding_categories
[coding_priorities
[0]].id
;
8413 return CODING_ID_NAME (id
);
8416 #if defined (WINDOWSNT) || defined (CYGWIN)
8419 from_unicode (Lisp_Object str
)
8422 if (!STRING_MULTIBYTE (str
) &&
8425 str
= Fsubstring (str
, make_number (0), make_number (-1));
8428 return code_convert_string_norecord (str
, Qutf_16le
, 0);
8432 from_unicode_buffer (const wchar_t *wstr
)
8434 return from_unicode (
8435 make_unibyte_string (
8437 /* we get one of the two final 0 bytes for free. */
8438 1 + sizeof (wchar_t) * wcslen (wstr
)));
8442 to_unicode (Lisp_Object str
, Lisp_Object
*buf
)
8444 *buf
= code_convert_string_norecord (str
, Qutf_16le
, 1);
8445 /* We need to make another copy (in addition to the one made by
8446 code_convert_string_norecord) to ensure that the final string is
8447 _doubly_ zero terminated --- that is, that the string is
8448 terminated by two zero bytes and one utf-16le null character.
8449 Because strings are already terminated with a single zero byte,
8450 we just add one additional zero. */
8451 str
= make_uninit_string (SBYTES (*buf
) + 1);
8452 memcpy (SDATA (str
), SDATA (*buf
), SBYTES (*buf
));
8453 SDATA (str
) [SBYTES (*buf
)] = '\0';
8455 return WCSDATA (*buf
);
8458 #endif /* WINDOWSNT || CYGWIN */
8462 /*** 8. Emacs Lisp library functions ***/
8464 DEFUN ("coding-system-p", Fcoding_system_p
, Scoding_system_p
, 1, 1, 0,
8465 doc
: /* Return t if OBJECT is nil or a coding-system.
8466 See the documentation of `define-coding-system' for information
8467 about coding-system objects. */)
8468 (Lisp_Object object
)
8471 || CODING_SYSTEM_ID (object
) >= 0)
8473 if (! SYMBOLP (object
)
8474 || NILP (Fget (object
, Qcoding_system_define_form
)))
8479 DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system
,
8480 Sread_non_nil_coding_system
, 1, 1, 0,
8481 doc
: /* Read a coding system from the minibuffer, prompting with string PROMPT. */)
8482 (Lisp_Object prompt
)
8487 val
= Fcompleting_read (prompt
, Vcoding_system_alist
, Qnil
,
8488 Qt
, Qnil
, Qcoding_system_history
, Qnil
, Qnil
);
8490 while (SCHARS (val
) == 0);
8491 return (Fintern (val
, Qnil
));
8494 DEFUN ("read-coding-system", Fread_coding_system
, Sread_coding_system
, 1, 2, 0,
8495 doc
: /* Read a coding system from the minibuffer, prompting with string PROMPT.
8496 If the user enters null input, return second argument DEFAULT-CODING-SYSTEM.
8497 Ignores case when completing coding systems (all Emacs coding systems
8498 are lower-case). */)
8499 (Lisp_Object prompt
, Lisp_Object default_coding_system
)
8502 ptrdiff_t count
= SPECPDL_INDEX ();
8504 if (SYMBOLP (default_coding_system
))
8505 default_coding_system
= SYMBOL_NAME (default_coding_system
);
8506 specbind (Qcompletion_ignore_case
, Qt
);
8507 val
= Fcompleting_read (prompt
, Vcoding_system_alist
, Qnil
,
8508 Qt
, Qnil
, Qcoding_system_history
,
8509 default_coding_system
, Qnil
);
8510 unbind_to (count
, Qnil
);
8511 return (SCHARS (val
) == 0 ? Qnil
: Fintern (val
, Qnil
));
8514 DEFUN ("check-coding-system", Fcheck_coding_system
, Scheck_coding_system
,
8516 doc
: /* Check validity of CODING-SYSTEM.
8517 If valid, return CODING-SYSTEM, else signal a `coding-system-error' error.
8518 It is valid if it is nil or a symbol defined as a coding system by the
8519 function `define-coding-system'. */)
8520 (Lisp_Object coding_system
)
8522 Lisp_Object define_form
;
8524 define_form
= Fget (coding_system
, Qcoding_system_define_form
);
8525 if (! NILP (define_form
))
8527 Fput (coding_system
, Qcoding_system_define_form
, Qnil
);
8528 safe_eval (define_form
);
8530 if (!NILP (Fcoding_system_p (coding_system
)))
8531 return coding_system
;
8532 xsignal1 (Qcoding_system_error
, coding_system
);
8536 /* Detect how the bytes at SRC of length SRC_BYTES are encoded. If
8537 HIGHEST, return the coding system of the highest
8538 priority among the detected coding systems. Otherwise return a
8539 list of detected coding systems sorted by their priorities. If
8540 MULTIBYTEP, it is assumed that the bytes are in correct
8541 multibyte form but contains only ASCII and eight-bit chars.
8542 Otherwise, the bytes are raw bytes.
8544 CODING-SYSTEM controls the detection as below:
8546 If it is nil, detect both text-format and eol-format. If the
8547 text-format part of CODING-SYSTEM is already specified
8548 (e.g. `iso-latin-1'), detect only eol-format. If the eol-format
8549 part of CODING-SYSTEM is already specified (e.g. `undecided-unix'),
8550 detect only text-format. */
8553 detect_coding_system (const unsigned char *src
,
8554 ptrdiff_t src_chars
, ptrdiff_t src_bytes
,
8555 bool highest
, bool multibytep
,
8556 Lisp_Object coding_system
)
8558 const unsigned char *src_end
= src
+ src_bytes
;
8559 Lisp_Object attrs
, eol_type
;
8560 Lisp_Object val
= Qnil
;
8561 struct coding_system coding
;
8563 struct coding_detection_info detect_info
;
8564 enum coding_category base_category
;
8565 bool null_byte_found
= 0, eight_bit_found
= 0;
8567 if (NILP (coding_system
))
8568 coding_system
= Qundecided
;
8569 setup_coding_system (coding_system
, &coding
);
8570 attrs
= CODING_ID_ATTRS (coding
.id
);
8571 eol_type
= CODING_ID_EOL_TYPE (coding
.id
);
8572 coding_system
= CODING_ATTR_BASE_NAME (attrs
);
8574 coding
.source
= src
;
8575 coding
.src_chars
= src_chars
;
8576 coding
.src_bytes
= src_bytes
;
8577 coding
.src_multibyte
= multibytep
;
8578 coding
.consumed
= 0;
8579 coding
.mode
|= CODING_MODE_LAST_BLOCK
;
8580 coding
.head_ascii
= 0;
8582 detect_info
.checked
= detect_info
.found
= detect_info
.rejected
= 0;
8584 /* At first, detect text-format if necessary. */
8585 base_category
= XINT (CODING_ATTR_CATEGORY (attrs
));
8586 if (base_category
== coding_category_undecided
)
8588 enum coding_category category
IF_LINT (= 0);
8589 struct coding_system
*this IF_LINT (= NULL
);
8591 bool inhibit_nbd
= inhibit_flag (coding
.spec
.undecided
.inhibit_nbd
,
8592 inhibit_null_byte_detection
);
8593 bool inhibit_ied
= inhibit_flag (coding
.spec
.undecided
.inhibit_ied
,
8594 inhibit_iso_escape_detection
);
8595 bool prefer_utf_8
= coding
.spec
.undecided
.prefer_utf_8
;
8597 /* Skip all ASCII bytes except for a few ISO2022 controls. */
8598 for (; src
< src_end
; src
++)
8603 eight_bit_found
= 1;
8604 if (null_byte_found
)
8609 if ((c
== ISO_CODE_ESC
|| c
== ISO_CODE_SI
|| c
== ISO_CODE_SO
)
8611 && ! detect_info
.checked
)
8613 if (detect_coding_iso_2022 (&coding
, &detect_info
))
8615 /* We have scanned the whole data. */
8616 if (! (detect_info
.rejected
& CATEGORY_MASK_ISO_7_ELSE
))
8618 /* We didn't find an 8-bit code. We may
8619 have found a null-byte, but it's very
8620 rare that a binary file confirm to
8623 coding
.head_ascii
= src
- coding
.source
;
8625 detect_info
.rejected
|= ~CATEGORY_MASK_ISO_ESCAPE
;
8629 else if (! c
&& !inhibit_nbd
)
8631 null_byte_found
= 1;
8632 if (eight_bit_found
)
8635 if (! eight_bit_found
)
8636 coding
.head_ascii
++;
8638 else if (! eight_bit_found
)
8639 coding
.head_ascii
++;
8642 if (null_byte_found
|| eight_bit_found
8643 || coding
.head_ascii
< coding
.src_bytes
8644 || detect_info
.found
)
8646 if (coding
.head_ascii
== coding
.src_bytes
)
8647 /* As all bytes are 7-bit, we can ignore non-ISO-2022 codings. */
8648 for (i
= 0; i
< coding_category_raw_text
; i
++)
8650 category
= coding_priorities
[i
];
8651 this = coding_categories
+ category
;
8652 if (detect_info
.found
& (1 << category
))
8657 if (null_byte_found
)
8659 detect_info
.checked
|= ~CATEGORY_MASK_UTF_16
;
8660 detect_info
.rejected
|= ~CATEGORY_MASK_UTF_16
;
8662 else if (prefer_utf_8
8663 && detect_coding_utf_8 (&coding
, &detect_info
))
8665 detect_info
.checked
|= ~CATEGORY_MASK_UTF_8
;
8666 detect_info
.rejected
|= ~CATEGORY_MASK_UTF_8
;
8668 for (i
= 0; i
< coding_category_raw_text
; i
++)
8670 category
= coding_priorities
[i
];
8671 this = coding_categories
+ category
;
8675 /* No coding system of this category is defined. */
8676 detect_info
.rejected
|= (1 << category
);
8678 else if (category
>= coding_category_raw_text
)
8680 else if (detect_info
.checked
& (1 << category
))
8683 && (detect_info
.found
& (1 << category
)))
8686 else if ((*(this->detector
)) (&coding
, &detect_info
)
8688 && (detect_info
.found
& (1 << category
)))
8690 if (category
== coding_category_utf_16_auto
)
8692 if (detect_info
.found
& CATEGORY_MASK_UTF_16_LE
)
8693 category
= coding_category_utf_16_le
;
8695 category
= coding_category_utf_16_be
;
8703 if ((detect_info
.rejected
& CATEGORY_MASK_ANY
) == CATEGORY_MASK_ANY
8706 detect_info
.found
= CATEGORY_MASK_RAW_TEXT
;
8707 id
= CODING_SYSTEM_ID (Qno_conversion
);
8708 val
= list1 (make_number (id
));
8710 else if (! detect_info
.rejected
&& ! detect_info
.found
)
8712 detect_info
.found
= CATEGORY_MASK_ANY
;
8713 id
= coding_categories
[coding_category_undecided
].id
;
8714 val
= list1 (make_number (id
));
8718 if (detect_info
.found
)
8720 detect_info
.found
= 1 << category
;
8721 val
= list1 (make_number (this->id
));
8724 for (i
= 0; i
< coding_category_raw_text
; i
++)
8725 if (! (detect_info
.rejected
& (1 << coding_priorities
[i
])))
8727 detect_info
.found
= 1 << coding_priorities
[i
];
8728 id
= coding_categories
[coding_priorities
[i
]].id
;
8729 val
= list1 (make_number (id
));
8735 int mask
= detect_info
.rejected
| detect_info
.found
;
8738 for (i
= coding_category_raw_text
- 1; i
>= 0; i
--)
8740 category
= coding_priorities
[i
];
8741 if (! (mask
& (1 << category
)))
8743 found
|= 1 << category
;
8744 id
= coding_categories
[category
].id
;
8746 val
= list1 (make_number (id
));
8749 for (i
= coding_category_raw_text
- 1; i
>= 0; i
--)
8751 category
= coding_priorities
[i
];
8752 if (detect_info
.found
& (1 << category
))
8754 id
= coding_categories
[category
].id
;
8755 val
= Fcons (make_number (id
), val
);
8758 detect_info
.found
|= found
;
8761 else if (base_category
== coding_category_utf_8_auto
)
8763 if (detect_coding_utf_8 (&coding
, &detect_info
))
8765 struct coding_system
*this;
8767 if (detect_info
.found
& CATEGORY_MASK_UTF_8_SIG
)
8768 this = coding_categories
+ coding_category_utf_8_sig
;
8770 this = coding_categories
+ coding_category_utf_8_nosig
;
8771 val
= list1 (make_number (this->id
));
8774 else if (base_category
== coding_category_utf_16_auto
)
8776 if (detect_coding_utf_16 (&coding
, &detect_info
))
8778 struct coding_system
*this;
8780 if (detect_info
.found
& CATEGORY_MASK_UTF_16_LE
)
8781 this = coding_categories
+ coding_category_utf_16_le
;
8782 else if (detect_info
.found
& CATEGORY_MASK_UTF_16_BE
)
8783 this = coding_categories
+ coding_category_utf_16_be
;
8784 else if (detect_info
.rejected
& CATEGORY_MASK_UTF_16_LE_NOSIG
)
8785 this = coding_categories
+ coding_category_utf_16_be_nosig
;
8787 this = coding_categories
+ coding_category_utf_16_le_nosig
;
8788 val
= list1 (make_number (this->id
));
8793 detect_info
.found
= 1 << XINT (CODING_ATTR_CATEGORY (attrs
));
8794 val
= list1 (make_number (coding
.id
));
8797 /* Then, detect eol-format if necessary. */
8799 int normal_eol
= -1, utf_16_be_eol
= -1, utf_16_le_eol
= -1;
8802 if (VECTORP (eol_type
))
8804 if (detect_info
.found
& ~CATEGORY_MASK_UTF_16
)
8806 if (null_byte_found
)
8807 normal_eol
= EOL_SEEN_LF
;
8809 normal_eol
= detect_eol (coding
.source
, src_bytes
,
8810 coding_category_raw_text
);
8812 if (detect_info
.found
& (CATEGORY_MASK_UTF_16_BE
8813 | CATEGORY_MASK_UTF_16_BE_NOSIG
))
8814 utf_16_be_eol
= detect_eol (coding
.source
, src_bytes
,
8815 coding_category_utf_16_be
);
8816 if (detect_info
.found
& (CATEGORY_MASK_UTF_16_LE
8817 | CATEGORY_MASK_UTF_16_LE_NOSIG
))
8818 utf_16_le_eol
= detect_eol (coding
.source
, src_bytes
,
8819 coding_category_utf_16_le
);
8823 if (EQ (eol_type
, Qunix
))
8824 normal_eol
= utf_16_be_eol
= utf_16_le_eol
= EOL_SEEN_LF
;
8825 else if (EQ (eol_type
, Qdos
))
8826 normal_eol
= utf_16_be_eol
= utf_16_le_eol
= EOL_SEEN_CRLF
;
8828 normal_eol
= utf_16_be_eol
= utf_16_le_eol
= EOL_SEEN_CR
;
8831 for (tail
= val
; CONSP (tail
); tail
= XCDR (tail
))
8833 enum coding_category category
;
8836 id
= XINT (XCAR (tail
));
8837 attrs
= CODING_ID_ATTRS (id
);
8838 category
= XINT (CODING_ATTR_CATEGORY (attrs
));
8839 eol_type
= CODING_ID_EOL_TYPE (id
);
8840 if (VECTORP (eol_type
))
8842 if (category
== coding_category_utf_16_be
8843 || category
== coding_category_utf_16_be_nosig
)
8844 this_eol
= utf_16_be_eol
;
8845 else if (category
== coding_category_utf_16_le
8846 || category
== coding_category_utf_16_le_nosig
)
8847 this_eol
= utf_16_le_eol
;
8849 this_eol
= normal_eol
;
8851 if (this_eol
== EOL_SEEN_LF
)
8852 XSETCAR (tail
, AREF (eol_type
, 0));
8853 else if (this_eol
== EOL_SEEN_CRLF
)
8854 XSETCAR (tail
, AREF (eol_type
, 1));
8855 else if (this_eol
== EOL_SEEN_CR
)
8856 XSETCAR (tail
, AREF (eol_type
, 2));
8858 XSETCAR (tail
, CODING_ID_NAME (id
));
8861 XSETCAR (tail
, CODING_ID_NAME (id
));
8865 return (highest
? (CONSP (val
) ? XCAR (val
) : Qnil
) : val
);
8869 DEFUN ("detect-coding-region", Fdetect_coding_region
, Sdetect_coding_region
,
8871 doc
: /* Detect coding system of the text in the region between START and END.
8872 Return a list of possible coding systems ordered by priority.
8873 The coding systems to try and their priorities follows what
8874 the function `coding-system-priority-list' (which see) returns.
8876 If only ASCII characters are found (except for such ISO-2022 control
8877 characters as ESC), it returns a list of single element `undecided'
8878 or its subsidiary coding system according to a detected end-of-line
8881 If optional argument HIGHEST is non-nil, return the coding system of
8882 highest priority. */)
8883 (Lisp_Object start
, Lisp_Object end
, Lisp_Object highest
)
8886 ptrdiff_t from_byte
, to_byte
;
8888 validate_region (&start
, &end
);
8889 from
= XINT (start
), to
= XINT (end
);
8890 from_byte
= CHAR_TO_BYTE (from
);
8891 to_byte
= CHAR_TO_BYTE (to
);
8893 if (from
< GPT
&& to
>= GPT
)
8894 move_gap_both (to
, to_byte
);
8896 return detect_coding_system (BYTE_POS_ADDR (from_byte
),
8897 to
- from
, to_byte
- from_byte
,
8899 !NILP (BVAR (current_buffer
8900 , enable_multibyte_characters
)),
8904 DEFUN ("detect-coding-string", Fdetect_coding_string
, Sdetect_coding_string
,
8906 doc
: /* Detect coding system of the text in STRING.
8907 Return a list of possible coding systems ordered by priority.
8908 The coding systems to try and their priorities follows what
8909 the function `coding-system-priority-list' (which see) returns.
8911 If only ASCII characters are found (except for such ISO-2022 control
8912 characters as ESC), it returns a list of single element `undecided'
8913 or its subsidiary coding system according to a detected end-of-line
8916 If optional argument HIGHEST is non-nil, return the coding system of
8917 highest priority. */)
8918 (Lisp_Object string
, Lisp_Object highest
)
8920 CHECK_STRING (string
);
8922 return detect_coding_system (SDATA (string
),
8923 SCHARS (string
), SBYTES (string
),
8924 !NILP (highest
), STRING_MULTIBYTE (string
),
8930 char_encodable_p (int c
, Lisp_Object attrs
)
8933 struct charset
*charset
;
8934 Lisp_Object translation_table
;
8936 translation_table
= CODING_ATTR_TRANS_TBL (attrs
);
8937 if (! NILP (translation_table
))
8938 c
= translate_char (translation_table
, c
);
8939 for (tail
= CODING_ATTR_CHARSET_LIST (attrs
);
8940 CONSP (tail
); tail
= XCDR (tail
))
8942 charset
= CHARSET_FROM_ID (XINT (XCAR (tail
)));
8943 if (CHAR_CHARSET_P (c
, charset
))
8946 return (! NILP (tail
));
8950 /* Return a list of coding systems that safely encode the text between
8951 START and END. If EXCLUDE is non-nil, it is a list of coding
8952 systems not to check. The returned list doesn't contain any such
8953 coding systems. In any case, if the text contains only ASCII or is
8954 unibyte, return t. */
8956 DEFUN ("find-coding-systems-region-internal",
8957 Ffind_coding_systems_region_internal
,
8958 Sfind_coding_systems_region_internal
, 2, 3, 0,
8959 doc
: /* Internal use only. */)
8960 (Lisp_Object start
, Lisp_Object end
, Lisp_Object exclude
)
8962 Lisp_Object coding_attrs_list
, safe_codings
;
8963 ptrdiff_t start_byte
, end_byte
;
8964 const unsigned char *p
, *pbeg
, *pend
;
8966 Lisp_Object tail
, elt
, work_table
;
8968 if (STRINGP (start
))
8970 if (!STRING_MULTIBYTE (start
)
8971 || SCHARS (start
) == SBYTES (start
))
8974 end_byte
= SBYTES (start
);
8978 CHECK_NUMBER_COERCE_MARKER (start
);
8979 CHECK_NUMBER_COERCE_MARKER (end
);
8980 if (XINT (start
) < BEG
|| XINT (end
) > Z
|| XINT (start
) > XINT (end
))
8981 args_out_of_range (start
, end
);
8982 if (NILP (BVAR (current_buffer
, enable_multibyte_characters
)))
8984 start_byte
= CHAR_TO_BYTE (XINT (start
));
8985 end_byte
= CHAR_TO_BYTE (XINT (end
));
8986 if (XINT (end
) - XINT (start
) == end_byte
- start_byte
)
8989 if (XINT (start
) < GPT
&& XINT (end
) > GPT
)
8991 if ((GPT
- XINT (start
)) < (XINT (end
) - GPT
))
8992 move_gap_both (XINT (start
), start_byte
);
8994 move_gap_both (XINT (end
), end_byte
);
8998 coding_attrs_list
= Qnil
;
8999 for (tail
= Vcoding_system_list
; CONSP (tail
); tail
= XCDR (tail
))
9001 || NILP (Fmemq (XCAR (tail
), exclude
)))
9005 attrs
= AREF (CODING_SYSTEM_SPEC (XCAR (tail
)), 0);
9006 if (EQ (XCAR (tail
), CODING_ATTR_BASE_NAME (attrs
)))
9008 ASET (attrs
, coding_attr_trans_tbl
,
9009 get_translation_table (attrs
, 1, NULL
));
9010 coding_attrs_list
= Fcons (attrs
, coding_attrs_list
);
9014 if (STRINGP (start
))
9015 p
= pbeg
= SDATA (start
);
9017 p
= pbeg
= BYTE_POS_ADDR (start_byte
);
9018 pend
= p
+ (end_byte
- start_byte
);
9020 while (p
< pend
&& ASCII_CHAR_P (*p
)) p
++;
9021 while (p
< pend
&& ASCII_CHAR_P (*(pend
- 1))) pend
--;
9023 work_table
= Fmake_char_table (Qnil
, Qnil
);
9026 if (ASCII_CHAR_P (*p
))
9030 c
= STRING_CHAR_ADVANCE (p
);
9031 if (!NILP (char_table_ref (work_table
, c
)))
9032 /* This character was already checked. Ignore it. */
9035 charset_map_loaded
= 0;
9036 for (tail
= coding_attrs_list
; CONSP (tail
);)
9041 else if (char_encodable_p (c
, elt
))
9043 else if (CONSP (XCDR (tail
)))
9045 XSETCAR (tail
, XCAR (XCDR (tail
)));
9046 XSETCDR (tail
, XCDR (XCDR (tail
)));
9050 XSETCAR (tail
, Qnil
);
9054 if (charset_map_loaded
)
9056 ptrdiff_t p_offset
= p
- pbeg
, pend_offset
= pend
- pbeg
;
9058 if (STRINGP (start
))
9059 pbeg
= SDATA (start
);
9061 pbeg
= BYTE_POS_ADDR (start_byte
);
9062 p
= pbeg
+ p_offset
;
9063 pend
= pbeg
+ pend_offset
;
9065 char_table_set (work_table
, c
, Qt
);
9069 safe_codings
= list2 (Qraw_text
, Qno_conversion
);
9070 for (tail
= coding_attrs_list
; CONSP (tail
); tail
= XCDR (tail
))
9071 if (! NILP (XCAR (tail
)))
9072 safe_codings
= Fcons (CODING_ATTR_BASE_NAME (XCAR (tail
)), safe_codings
);
9074 return safe_codings
;
9078 DEFUN ("unencodable-char-position", Funencodable_char_position
,
9079 Sunencodable_char_position
, 3, 5, 0,
9080 doc
: /* Return position of first un-encodable character in a region.
9081 START and END specify the region and CODING-SYSTEM specifies the
9082 encoding to check. Return nil if CODING-SYSTEM does encode the region.
9084 If optional 4th argument COUNT is non-nil, it specifies at most how
9085 many un-encodable characters to search. In this case, the value is a
9088 If optional 5th argument STRING is non-nil, it is a string to search
9089 for un-encodable characters. In that case, START and END are indexes
9090 to the string and treated as in `substring'. */)
9091 (Lisp_Object start
, Lisp_Object end
, Lisp_Object coding_system
,
9092 Lisp_Object count
, Lisp_Object string
)
9095 struct coding_system coding
;
9096 Lisp_Object attrs
, charset_list
, translation_table
;
9097 Lisp_Object positions
;
9099 const unsigned char *p
, *stop
, *pend
;
9100 bool ascii_compatible
;
9102 setup_coding_system (Fcheck_coding_system (coding_system
), &coding
);
9103 attrs
= CODING_ID_ATTRS (coding
.id
);
9104 if (EQ (CODING_ATTR_TYPE (attrs
), Qraw_text
))
9106 ascii_compatible
= ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
));
9107 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
9108 translation_table
= get_translation_table (attrs
, 1, NULL
);
9112 validate_region (&start
, &end
);
9113 from
= XINT (start
);
9115 if (NILP (BVAR (current_buffer
, enable_multibyte_characters
))
9116 || (ascii_compatible
9117 && (to
- from
) == (CHAR_TO_BYTE (to
) - (CHAR_TO_BYTE (from
)))))
9119 p
= CHAR_POS_ADDR (from
);
9120 pend
= CHAR_POS_ADDR (to
);
9121 if (from
< GPT
&& to
>= GPT
)
9128 CHECK_STRING (string
);
9129 validate_subarray (string
, start
, end
, SCHARS (string
), &from
, &to
);
9130 if (! STRING_MULTIBYTE (string
))
9132 p
= SDATA (string
) + string_char_to_byte (string
, from
);
9133 stop
= pend
= SDATA (string
) + string_char_to_byte (string
, to
);
9134 if (ascii_compatible
&& (to
- from
) == (pend
- p
))
9142 CHECK_NATNUM (count
);
9147 charset_map_loaded
= 0;
9152 if (ascii_compatible
)
9153 while (p
< stop
&& ASCII_CHAR_P (*p
))
9163 c
= STRING_CHAR_ADVANCE (p
);
9164 if (! (ASCII_CHAR_P (c
) && ascii_compatible
)
9165 && ! char_charset (translate_char (translation_table
, c
),
9166 charset_list
, NULL
))
9168 positions
= Fcons (make_number (from
), positions
);
9175 if (charset_map_loaded
&& NILP (string
))
9177 p
= CHAR_POS_ADDR (from
);
9178 pend
= CHAR_POS_ADDR (to
);
9179 if (from
< GPT
&& to
>= GPT
)
9183 charset_map_loaded
= 0;
9187 return (NILP (count
) ? Fcar (positions
) : Fnreverse (positions
));
9191 DEFUN ("check-coding-systems-region", Fcheck_coding_systems_region
,
9192 Scheck_coding_systems_region
, 3, 3, 0,
9193 doc
: /* Check if the region is encodable by coding systems.
9195 START and END are buffer positions specifying the region.
9196 CODING-SYSTEM-LIST is a list of coding systems to check.
9198 The value is an alist ((CODING-SYSTEM POS0 POS1 ...) ...), where
9199 CODING-SYSTEM is a member of CODING-SYSTEM-LIST and can't encode the
9200 whole region, POS0, POS1, ... are buffer positions where non-encodable
9201 characters are found.
9203 If all coding systems in CODING-SYSTEM-LIST can encode the region, the
9206 START may be a string. In that case, check if the string is
9207 encodable, and the value contains indices to the string instead of
9208 buffer positions. END is ignored.
9210 If the current buffer (or START if it is a string) is unibyte, the value
9212 (Lisp_Object start
, Lisp_Object end
, Lisp_Object coding_system_list
)
9215 ptrdiff_t start_byte
, end_byte
;
9217 const unsigned char *p
, *pbeg
, *pend
;
9219 Lisp_Object tail
, elt
, attrs
;
9221 if (STRINGP (start
))
9223 if (!STRING_MULTIBYTE (start
)
9224 || SCHARS (start
) == SBYTES (start
))
9227 end_byte
= SBYTES (start
);
9232 CHECK_NUMBER_COERCE_MARKER (start
);
9233 CHECK_NUMBER_COERCE_MARKER (end
);
9234 if (XINT (start
) < BEG
|| XINT (end
) > Z
|| XINT (start
) > XINT (end
))
9235 args_out_of_range (start
, end
);
9236 if (NILP (BVAR (current_buffer
, enable_multibyte_characters
)))
9238 start_byte
= CHAR_TO_BYTE (XINT (start
));
9239 end_byte
= CHAR_TO_BYTE (XINT (end
));
9240 if (XINT (end
) - XINT (start
) == end_byte
- start_byte
)
9243 if (XINT (start
) < GPT
&& XINT (end
) > GPT
)
9245 if ((GPT
- XINT (start
)) < (XINT (end
) - GPT
))
9246 move_gap_both (XINT (start
), start_byte
);
9248 move_gap_both (XINT (end
), end_byte
);
9254 for (tail
= coding_system_list
; CONSP (tail
); tail
= XCDR (tail
))
9257 attrs
= AREF (CODING_SYSTEM_SPEC (elt
), 0);
9258 ASET (attrs
, coding_attr_trans_tbl
,
9259 get_translation_table (attrs
, 1, NULL
));
9260 list
= Fcons (list2 (elt
, attrs
), list
);
9263 if (STRINGP (start
))
9264 p
= pbeg
= SDATA (start
);
9266 p
= pbeg
= BYTE_POS_ADDR (start_byte
);
9267 pend
= p
+ (end_byte
- start_byte
);
9269 while (p
< pend
&& ASCII_CHAR_P (*p
)) p
++, pos
++;
9270 while (p
< pend
&& ASCII_CHAR_P (*(pend
- 1))) pend
--;
9274 if (ASCII_CHAR_P (*p
))
9278 c
= STRING_CHAR_ADVANCE (p
);
9280 charset_map_loaded
= 0;
9281 for (tail
= list
; CONSP (tail
); tail
= XCDR (tail
))
9283 elt
= XCDR (XCAR (tail
));
9284 if (! char_encodable_p (c
, XCAR (elt
)))
9285 XSETCDR (elt
, Fcons (make_number (pos
), XCDR (elt
)));
9287 if (charset_map_loaded
)
9289 ptrdiff_t p_offset
= p
- pbeg
, pend_offset
= pend
- pbeg
;
9291 if (STRINGP (start
))
9292 pbeg
= SDATA (start
);
9294 pbeg
= BYTE_POS_ADDR (start_byte
);
9295 p
= pbeg
+ p_offset
;
9296 pend
= pbeg
+ pend_offset
;
9304 for (; CONSP (tail
); tail
= XCDR (tail
))
9307 if (CONSP (XCDR (XCDR (elt
))))
9308 list
= Fcons (Fcons (XCAR (elt
), Fnreverse (XCDR (XCDR (elt
)))),
9317 code_convert_region (Lisp_Object start
, Lisp_Object end
,
9318 Lisp_Object coding_system
, Lisp_Object dst_object
,
9319 bool encodep
, bool norecord
)
9321 struct coding_system coding
;
9322 ptrdiff_t from
, from_byte
, to
, to_byte
;
9323 Lisp_Object src_object
;
9325 if (NILP (coding_system
))
9326 coding_system
= Qno_conversion
;
9328 CHECK_CODING_SYSTEM (coding_system
);
9329 src_object
= Fcurrent_buffer ();
9330 if (NILP (dst_object
))
9331 dst_object
= src_object
;
9332 else if (! EQ (dst_object
, Qt
))
9333 CHECK_BUFFER (dst_object
);
9335 validate_region (&start
, &end
);
9336 from
= XFASTINT (start
);
9337 from_byte
= CHAR_TO_BYTE (from
);
9338 to
= XFASTINT (end
);
9339 to_byte
= CHAR_TO_BYTE (to
);
9341 setup_coding_system (coding_system
, &coding
);
9342 coding
.mode
|= CODING_MODE_LAST_BLOCK
;
9344 if (BUFFERP (dst_object
) && !EQ (dst_object
, src_object
))
9346 struct buffer
*buf
= XBUFFER (dst_object
);
9347 ptrdiff_t buf_pt
= BUF_PT (buf
);
9349 invalidate_buffer_caches (buf
, buf_pt
, buf_pt
);
9353 encode_coding_object (&coding
, src_object
, from
, from_byte
, to
, to_byte
,
9356 decode_coding_object (&coding
, src_object
, from
, from_byte
, to
, to_byte
,
9359 Vlast_coding_system_used
= CODING_ID_NAME (coding
.id
);
9361 return (BUFFERP (dst_object
)
9362 ? make_number (coding
.produced_char
)
9363 : coding
.dst_object
);
9367 DEFUN ("decode-coding-region", Fdecode_coding_region
, Sdecode_coding_region
,
9368 3, 4, "r\nzCoding system: ",
9369 doc
: /* Decode the current region from the specified coding system.
9370 When called from a program, takes four arguments:
9371 START, END, CODING-SYSTEM, and DESTINATION.
9372 START and END are buffer positions.
9374 Optional 4th arguments DESTINATION specifies where the decoded text goes.
9375 If nil, the region between START and END is replaced by the decoded text.
9376 If buffer, the decoded text is inserted in that buffer after point (point
9378 In those cases, the length of the decoded text is returned.
9379 If DESTINATION is t, the decoded text is returned.
9381 This function sets `last-coding-system-used' to the precise coding system
9382 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
9383 not fully specified.) */)
9384 (Lisp_Object start
, Lisp_Object end
, Lisp_Object coding_system
, Lisp_Object destination
)
9386 return code_convert_region (start
, end
, coding_system
, destination
, 0, 0);
9389 DEFUN ("encode-coding-region", Fencode_coding_region
, Sencode_coding_region
,
9390 3, 4, "r\nzCoding system: ",
9391 doc
: /* Encode the current region by specified coding system.
9392 When called from a program, takes four arguments:
9393 START, END, CODING-SYSTEM and DESTINATION.
9394 START and END are buffer positions.
9396 Optional 4th arguments DESTINATION specifies where the encoded text goes.
9397 If nil, the region between START and END is replace by the encoded text.
9398 If buffer, the encoded text is inserted in that buffer after point (point
9400 In those cases, the length of the encoded text is returned.
9401 If DESTINATION is t, the encoded text is returned.
9403 This function sets `last-coding-system-used' to the precise coding system
9404 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
9405 not fully specified.) */)
9406 (Lisp_Object start
, Lisp_Object end
, Lisp_Object coding_system
, Lisp_Object destination
)
9408 return code_convert_region (start
, end
, coding_system
, destination
, 1, 0);
9412 code_convert_string (Lisp_Object string
, Lisp_Object coding_system
,
9413 Lisp_Object dst_object
, bool encodep
, bool nocopy
,
9416 struct coding_system coding
;
9417 ptrdiff_t chars
, bytes
;
9419 CHECK_STRING (string
);
9420 if (NILP (coding_system
))
9423 Vlast_coding_system_used
= Qno_conversion
;
9424 if (NILP (dst_object
))
9425 return (nocopy
? Fcopy_sequence (string
) : string
);
9428 if (NILP (coding_system
))
9429 coding_system
= Qno_conversion
;
9431 CHECK_CODING_SYSTEM (coding_system
);
9432 if (NILP (dst_object
))
9434 else if (! EQ (dst_object
, Qt
))
9435 CHECK_BUFFER (dst_object
);
9437 setup_coding_system (coding_system
, &coding
);
9438 coding
.mode
|= CODING_MODE_LAST_BLOCK
;
9439 chars
= SCHARS (string
);
9440 bytes
= SBYTES (string
);
9442 if (BUFFERP (dst_object
))
9444 struct buffer
*buf
= XBUFFER (dst_object
);
9445 ptrdiff_t buf_pt
= BUF_PT (buf
);
9447 invalidate_buffer_caches (buf
, buf_pt
, buf_pt
);
9451 encode_coding_object (&coding
, string
, 0, 0, chars
, bytes
, dst_object
);
9453 decode_coding_object (&coding
, string
, 0, 0, chars
, bytes
, dst_object
);
9455 Vlast_coding_system_used
= CODING_ID_NAME (coding
.id
);
9457 return (BUFFERP (dst_object
)
9458 ? make_number (coding
.produced_char
)
9459 : coding
.dst_object
);
9463 /* Encode or decode STRING according to CODING_SYSTEM.
9464 Do not set Vlast_coding_system_used.
9466 This function is called only from macros DECODE_FILE and
9467 ENCODE_FILE, thus we ignore character composition. */
9470 code_convert_string_norecord (Lisp_Object string
, Lisp_Object coding_system
,
9473 return code_convert_string (string
, coding_system
, Qt
, encodep
, 0, 1);
9476 /* Encode or decode a file name, to or from a unibyte string suitable
9477 for passing to C library functions. */
9479 decode_file_name (Lisp_Object fname
)
9482 /* The w32 build pretends to use UTF-8 for file-name encoding, and
9483 converts the file names either to UTF-16LE or to the system ANSI
9484 codepage internally, depending on the underlying OS; see w32.c. */
9485 if (! NILP (Fcoding_system_p (Qutf_8
)))
9486 return code_convert_string_norecord (fname
, Qutf_8
, 0);
9488 #else /* !WINDOWSNT */
9489 if (! NILP (Vfile_name_coding_system
))
9490 return code_convert_string_norecord (fname
, Vfile_name_coding_system
, 0);
9491 else if (! NILP (Vdefault_file_name_coding_system
))
9492 return code_convert_string_norecord (fname
,
9493 Vdefault_file_name_coding_system
, 0);
9500 encode_file_name (Lisp_Object fname
)
9502 /* This is especially important during bootstrap and dumping, when
9503 file-name encoding is not yet known, and therefore any non-ASCII
9504 file names are unibyte strings, and could only be thrashed if we
9505 try to encode them. */
9506 if (!STRING_MULTIBYTE (fname
))
9509 /* The w32 build pretends to use UTF-8 for file-name encoding, and
9510 converts the file names either to UTF-16LE or to the system ANSI
9511 codepage internally, depending on the underlying OS; see w32.c. */
9512 if (! NILP (Fcoding_system_p (Qutf_8
)))
9513 return code_convert_string_norecord (fname
, Qutf_8
, 1);
9515 #else /* !WINDOWSNT */
9516 if (! NILP (Vfile_name_coding_system
))
9517 return code_convert_string_norecord (fname
, Vfile_name_coding_system
, 1);
9518 else if (! NILP (Vdefault_file_name_coding_system
))
9519 return code_convert_string_norecord (fname
,
9520 Vdefault_file_name_coding_system
, 1);
9526 DEFUN ("decode-coding-string", Fdecode_coding_string
, Sdecode_coding_string
,
9528 doc
: /* Decode STRING which is encoded in CODING-SYSTEM, and return the result.
9530 Optional third arg NOCOPY non-nil means it is OK to return STRING itself
9531 if the decoding operation is trivial.
9533 Optional fourth arg BUFFER non-nil means that the decoded text is
9534 inserted in that buffer after point (point does not move). In this
9535 case, the return value is the length of the decoded text.
9537 This function sets `last-coding-system-used' to the precise coding system
9538 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
9539 not fully specified.) */)
9540 (Lisp_Object string
, Lisp_Object coding_system
, Lisp_Object nocopy
, Lisp_Object buffer
)
9542 return code_convert_string (string
, coding_system
, buffer
,
9543 0, ! NILP (nocopy
), 0);
9546 DEFUN ("encode-coding-string", Fencode_coding_string
, Sencode_coding_string
,
9548 doc
: /* Encode STRING to CODING-SYSTEM, and return the result.
9550 Optional third arg NOCOPY non-nil means it is OK to return STRING
9551 itself if the encoding operation is trivial.
9553 Optional fourth arg BUFFER non-nil means that the encoded text is
9554 inserted in that buffer after point (point does not move). In this
9555 case, the return value is the length of the encoded text.
9557 This function sets `last-coding-system-used' to the precise coding system
9558 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is
9559 not fully specified.) */)
9560 (Lisp_Object string
, Lisp_Object coding_system
, Lisp_Object nocopy
, Lisp_Object buffer
)
9562 return code_convert_string (string
, coding_system
, buffer
,
9563 1, ! NILP (nocopy
), 0);
9567 DEFUN ("decode-sjis-char", Fdecode_sjis_char
, Sdecode_sjis_char
, 1, 1, 0,
9568 doc
: /* Decode a Japanese character which has CODE in shift_jis encoding.
9569 Return the corresponding character. */)
9572 Lisp_Object spec
, attrs
, val
;
9573 struct charset
*charset_roman
, *charset_kanji
, *charset_kana
, *charset
;
9577 CHECK_NATNUM (code
);
9578 ch
= XFASTINT (code
);
9579 CHECK_CODING_SYSTEM_GET_SPEC (Vsjis_coding_system
, spec
);
9580 attrs
= AREF (spec
, 0);
9582 if (ASCII_CHAR_P (ch
)
9583 && ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
)))
9586 val
= CODING_ATTR_CHARSET_LIST (attrs
);
9587 charset_roman
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
9588 charset_kana
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
9589 charset_kanji
= CHARSET_FROM_ID (XINT (XCAR (val
)));
9594 charset
= charset_roman
;
9596 else if (ch
>= 0xA0 && ch
< 0xDF)
9599 charset
= charset_kana
;
9603 EMACS_INT c1
= ch
>> 8;
9606 if (c1
< 0x81 || (c1
> 0x9F && c1
< 0xE0) || c1
> 0xEF
9607 || c2
< 0x40 || c2
== 0x7F || c2
> 0xFC)
9608 error ("Invalid code: %"pI
"d", ch
);
9611 charset
= charset_kanji
;
9613 c
= DECODE_CHAR (charset
, c
);
9615 error ("Invalid code: %"pI
"d", ch
);
9616 return make_number (c
);
9620 DEFUN ("encode-sjis-char", Fencode_sjis_char
, Sencode_sjis_char
, 1, 1, 0,
9621 doc
: /* Encode a Japanese character CH to shift_jis encoding.
9622 Return the corresponding code in SJIS. */)
9625 Lisp_Object spec
, attrs
, charset_list
;
9627 struct charset
*charset
;
9630 CHECK_CHARACTER (ch
);
9632 CHECK_CODING_SYSTEM_GET_SPEC (Vsjis_coding_system
, spec
);
9633 attrs
= AREF (spec
, 0);
9635 if (ASCII_CHAR_P (c
)
9636 && ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
)))
9639 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
9640 charset
= char_charset (c
, charset_list
, &code
);
9641 if (code
== CHARSET_INVALID_CODE (charset
))
9642 error ("Can't encode by shift_jis encoding: %c", c
);
9645 return make_number (code
);
9648 DEFUN ("decode-big5-char", Fdecode_big5_char
, Sdecode_big5_char
, 1, 1, 0,
9649 doc
: /* Decode a Big5 character which has CODE in BIG5 coding system.
9650 Return the corresponding character. */)
9653 Lisp_Object spec
, attrs
, val
;
9654 struct charset
*charset_roman
, *charset_big5
, *charset
;
9658 CHECK_NATNUM (code
);
9659 ch
= XFASTINT (code
);
9660 CHECK_CODING_SYSTEM_GET_SPEC (Vbig5_coding_system
, spec
);
9661 attrs
= AREF (spec
, 0);
9663 if (ASCII_CHAR_P (ch
)
9664 && ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
)))
9667 val
= CODING_ATTR_CHARSET_LIST (attrs
);
9668 charset_roman
= CHARSET_FROM_ID (XINT (XCAR (val
))), val
= XCDR (val
);
9669 charset_big5
= CHARSET_FROM_ID (XINT (XCAR (val
)));
9674 charset
= charset_roman
;
9678 EMACS_INT b1
= ch
>> 8;
9680 if (b1
< 0xA1 || b1
> 0xFE
9681 || b2
< 0x40 || (b2
> 0x7E && b2
< 0xA1) || b2
> 0xFE)
9682 error ("Invalid code: %"pI
"d", ch
);
9684 charset
= charset_big5
;
9686 c
= DECODE_CHAR (charset
, c
);
9688 error ("Invalid code: %"pI
"d", ch
);
9689 return make_number (c
);
9692 DEFUN ("encode-big5-char", Fencode_big5_char
, Sencode_big5_char
, 1, 1, 0,
9693 doc
: /* Encode the Big5 character CH to BIG5 coding system.
9694 Return the corresponding character code in Big5. */)
9697 Lisp_Object spec
, attrs
, charset_list
;
9698 struct charset
*charset
;
9702 CHECK_CHARACTER (ch
);
9704 CHECK_CODING_SYSTEM_GET_SPEC (Vbig5_coding_system
, spec
);
9705 attrs
= AREF (spec
, 0);
9706 if (ASCII_CHAR_P (c
)
9707 && ! NILP (CODING_ATTR_ASCII_COMPAT (attrs
)))
9710 charset_list
= CODING_ATTR_CHARSET_LIST (attrs
);
9711 charset
= char_charset (c
, charset_list
, &code
);
9712 if (code
== CHARSET_INVALID_CODE (charset
))
9713 error ("Can't encode by Big5 encoding: %c", c
);
9715 return make_number (code
);
9719 DEFUN ("set-terminal-coding-system-internal", Fset_terminal_coding_system_internal
,
9720 Sset_terminal_coding_system_internal
, 1, 2, 0,
9721 doc
: /* Internal use only. */)
9722 (Lisp_Object coding_system
, Lisp_Object terminal
)
9724 struct terminal
*term
= decode_live_terminal (terminal
);
9725 struct coding_system
*terminal_coding
= TERMINAL_TERMINAL_CODING (term
);
9726 CHECK_SYMBOL (coding_system
);
9727 setup_coding_system (Fcheck_coding_system (coding_system
), terminal_coding
);
9728 /* We had better not send unsafe characters to terminal. */
9729 terminal_coding
->mode
|= CODING_MODE_SAFE_ENCODING
;
9730 /* Character composition should be disabled. */
9731 terminal_coding
->common_flags
&= ~CODING_ANNOTATE_COMPOSITION_MASK
;
9732 terminal_coding
->src_multibyte
= 1;
9733 terminal_coding
->dst_multibyte
= 0;
9735 (term
, (terminal_coding
->common_flags
& CODING_REQUIRE_ENCODING_MASK
9736 ? coding_charset_list (terminal_coding
)
9737 : list1 (make_number (charset_ascii
))));
9741 DEFUN ("set-safe-terminal-coding-system-internal",
9742 Fset_safe_terminal_coding_system_internal
,
9743 Sset_safe_terminal_coding_system_internal
, 1, 1, 0,
9744 doc
: /* Internal use only. */)
9745 (Lisp_Object coding_system
)
9747 CHECK_SYMBOL (coding_system
);
9748 setup_coding_system (Fcheck_coding_system (coding_system
),
9749 &safe_terminal_coding
);
9750 /* Character composition should be disabled. */
9751 safe_terminal_coding
.common_flags
&= ~CODING_ANNOTATE_COMPOSITION_MASK
;
9752 safe_terminal_coding
.src_multibyte
= 1;
9753 safe_terminal_coding
.dst_multibyte
= 0;
9757 DEFUN ("terminal-coding-system", Fterminal_coding_system
,
9758 Sterminal_coding_system
, 0, 1, 0,
9759 doc
: /* Return coding system specified for terminal output on the given terminal.
9760 TERMINAL may be a terminal object, a frame, or nil for the selected
9761 frame's terminal device. */)
9762 (Lisp_Object terminal
)
9764 struct coding_system
*terminal_coding
9765 = TERMINAL_TERMINAL_CODING (decode_live_terminal (terminal
));
9766 Lisp_Object coding_system
= CODING_ID_NAME (terminal_coding
->id
);
9768 /* For backward compatibility, return nil if it is `undecided'. */
9769 return (! EQ (coding_system
, Qundecided
) ? coding_system
: Qnil
);
9772 DEFUN ("set-keyboard-coding-system-internal", Fset_keyboard_coding_system_internal
,
9773 Sset_keyboard_coding_system_internal
, 1, 2, 0,
9774 doc
: /* Internal use only. */)
9775 (Lisp_Object coding_system
, Lisp_Object terminal
)
9777 struct terminal
*t
= decode_live_terminal (terminal
);
9778 CHECK_SYMBOL (coding_system
);
9779 if (NILP (coding_system
))
9780 coding_system
= Qno_conversion
;
9782 Fcheck_coding_system (coding_system
);
9783 setup_coding_system (coding_system
, TERMINAL_KEYBOARD_CODING (t
));
9784 /* Character composition should be disabled. */
9785 TERMINAL_KEYBOARD_CODING (t
)->common_flags
9786 &= ~CODING_ANNOTATE_COMPOSITION_MASK
;
9790 DEFUN ("keyboard-coding-system",
9791 Fkeyboard_coding_system
, Skeyboard_coding_system
, 0, 1, 0,
9792 doc
: /* Return coding system specified for decoding keyboard input. */)
9793 (Lisp_Object terminal
)
9795 return CODING_ID_NAME (TERMINAL_KEYBOARD_CODING
9796 (decode_live_terminal (terminal
))->id
);
9800 DEFUN ("find-operation-coding-system", Ffind_operation_coding_system
,
9801 Sfind_operation_coding_system
, 1, MANY
, 0,
9802 doc
: /* Choose a coding system for an operation based on the target name.
9803 The value names a pair of coding systems: (DECODING-SYSTEM . ENCODING-SYSTEM).
9804 DECODING-SYSTEM is the coding system to use for decoding
9805 (in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system
9806 for encoding (in case OPERATION does encoding).
9808 The first argument OPERATION specifies an I/O primitive:
9809 For file I/O, `insert-file-contents' or `write-region'.
9810 For process I/O, `call-process', `call-process-region', or `start-process'.
9811 For network I/O, `open-network-stream'.
9813 The remaining arguments should be the same arguments that were passed
9814 to the primitive. Depending on which primitive, one of those arguments
9815 is selected as the TARGET. For example, if OPERATION does file I/O,
9816 whichever argument specifies the file name is TARGET.
9818 TARGET has a meaning which depends on OPERATION:
9819 For file I/O, TARGET is a file name (except for the special case below).
9820 For process I/O, TARGET is a process name.
9821 For network I/O, TARGET is a service name or a port number.
9823 This function looks up what is specified for TARGET in
9824 `file-coding-system-alist', `process-coding-system-alist',
9825 or `network-coding-system-alist' depending on OPERATION.
9826 They may specify a coding system, a cons of coding systems,
9827 or a function symbol to call.
9828 In the last case, we call the function with one argument,
9829 which is a list of all the arguments given to this function.
9830 If the function can't decide a coding system, it can return
9831 `undecided' so that the normal code-detection is performed.
9833 If OPERATION is `insert-file-contents', the argument corresponding to
9834 TARGET may be a cons (FILENAME . BUFFER). In that case, FILENAME is a
9835 file name to look up, and BUFFER is a buffer that contains the file's
9836 contents (not yet decoded). If `file-coding-system-alist' specifies a
9837 function to call for FILENAME, that function should examine the
9838 contents of BUFFER instead of reading the file.
9840 usage: (find-operation-coding-system OPERATION ARGUMENTS...) */)
9841 (ptrdiff_t nargs
, Lisp_Object
*args
)
9843 Lisp_Object operation
, target_idx
, target
, val
;
9844 register Lisp_Object chain
;
9847 error ("Too few arguments");
9848 operation
= args
[0];
9849 if (!SYMBOLP (operation
)
9850 || (target_idx
= Fget (operation
, Qtarget_idx
), !NATNUMP (target_idx
)))
9851 error ("Invalid first argument");
9852 if (nargs
<= 1 + XFASTINT (target_idx
))
9853 error ("Too few arguments for operation `%s'",
9854 SDATA (SYMBOL_NAME (operation
)));
9855 target
= args
[XFASTINT (target_idx
) + 1];
9856 if (!(STRINGP (target
)
9857 || (EQ (operation
, Qinsert_file_contents
) && CONSP (target
)
9858 && STRINGP (XCAR (target
)) && BUFFERP (XCDR (target
)))
9859 || (EQ (operation
, Qopen_network_stream
) && INTEGERP (target
))))
9860 error ("Invalid argument %"pI
"d of operation `%s'",
9861 XFASTINT (target_idx
) + 1, SDATA (SYMBOL_NAME (operation
)));
9863 target
= XCAR (target
);
9865 chain
= ((EQ (operation
, Qinsert_file_contents
)
9866 || EQ (operation
, Qwrite_region
))
9867 ? Vfile_coding_system_alist
9868 : (EQ (operation
, Qopen_network_stream
)
9869 ? Vnetwork_coding_system_alist
9870 : Vprocess_coding_system_alist
));
9874 for (; CONSP (chain
); chain
= XCDR (chain
))
9880 && ((STRINGP (target
)
9881 && STRINGP (XCAR (elt
))
9882 && fast_string_match (XCAR (elt
), target
) >= 0)
9883 || (INTEGERP (target
) && EQ (target
, XCAR (elt
)))))
9886 /* Here, if VAL is both a valid coding system and a valid
9887 function symbol, we return VAL as a coding system. */
9890 if (! SYMBOLP (val
))
9892 if (! NILP (Fcoding_system_p (val
)))
9893 return Fcons (val
, val
);
9894 if (! NILP (Ffboundp (val
)))
9896 /* We use call1 rather than safe_call1
9897 so as to get bug reports about functions called here
9898 which don't handle the current interface. */
9899 val
= call1 (val
, Flist (nargs
, args
));
9902 if (SYMBOLP (val
) && ! NILP (Fcoding_system_p (val
)))
9903 return Fcons (val
, val
);
9911 DEFUN ("set-coding-system-priority", Fset_coding_system_priority
,
9912 Sset_coding_system_priority
, 0, MANY
, 0,
9913 doc
: /* Assign higher priority to the coding systems given as arguments.
9914 If multiple coding systems belong to the same category,
9915 all but the first one are ignored.
9917 usage: (set-coding-system-priority &rest coding-systems) */)
9918 (ptrdiff_t nargs
, Lisp_Object
*args
)
9921 bool changed
[coding_category_max
];
9922 enum coding_category priorities
[coding_category_max
];
9924 memset (changed
, 0, sizeof changed
);
9926 for (i
= j
= 0; i
< nargs
; i
++)
9928 enum coding_category category
;
9929 Lisp_Object spec
, attrs
;
9931 CHECK_CODING_SYSTEM_GET_SPEC (args
[i
], spec
);
9932 attrs
= AREF (spec
, 0);
9933 category
= XINT (CODING_ATTR_CATEGORY (attrs
));
9934 if (changed
[category
])
9935 /* Ignore this coding system because a coding system of the
9936 same category already had a higher priority. */
9938 changed
[category
] = 1;
9939 priorities
[j
++] = category
;
9940 if (coding_categories
[category
].id
>= 0
9941 && ! EQ (args
[i
], CODING_ID_NAME (coding_categories
[category
].id
)))
9942 setup_coding_system (args
[i
], &coding_categories
[category
]);
9943 Fset (AREF (Vcoding_category_table
, category
), args
[i
]);
9946 /* Now we have decided top J priorities. Reflect the order of the
9947 original priorities to the remaining priorities. */
9949 for (i
= j
, j
= 0; i
< coding_category_max
; i
++, j
++)
9951 while (j
< coding_category_max
9952 && changed
[coding_priorities
[j
]])
9954 if (j
== coding_category_max
)
9956 priorities
[i
] = coding_priorities
[j
];
9959 memcpy (coding_priorities
, priorities
, sizeof priorities
);
9961 /* Update `coding-category-list'. */
9962 Vcoding_category_list
= Qnil
;
9963 for (i
= coding_category_max
; i
-- > 0; )
9964 Vcoding_category_list
9965 = Fcons (AREF (Vcoding_category_table
, priorities
[i
]),
9966 Vcoding_category_list
);
9971 DEFUN ("coding-system-priority-list", Fcoding_system_priority_list
,
9972 Scoding_system_priority_list
, 0, 1, 0,
9973 doc
: /* Return a list of coding systems ordered by their priorities.
9974 The list contains a subset of coding systems; i.e. coding systems
9975 assigned to each coding category (see `coding-category-list').
9977 HIGHESTP non-nil means just return the highest priority one. */)
9978 (Lisp_Object highestp
)
9983 for (i
= 0, val
= Qnil
; i
< coding_category_max
; i
++)
9985 enum coding_category category
= coding_priorities
[i
];
9986 int id
= coding_categories
[category
].id
;
9991 attrs
= CODING_ID_ATTRS (id
);
9992 if (! NILP (highestp
))
9993 return CODING_ATTR_BASE_NAME (attrs
);
9994 val
= Fcons (CODING_ATTR_BASE_NAME (attrs
), val
);
9996 return Fnreverse (val
);
9999 static const char *const suffixes
[] = { "-unix", "-dos", "-mac" };
10002 make_subsidiaries (Lisp_Object base
)
10004 Lisp_Object subsidiaries
;
10005 ptrdiff_t base_name_len
= SBYTES (SYMBOL_NAME (base
));
10007 char *buf
= SAFE_ALLOCA (base_name_len
+ 6);
10010 memcpy (buf
, SDATA (SYMBOL_NAME (base
)), base_name_len
);
10011 subsidiaries
= make_uninit_vector (3);
10012 for (i
= 0; i
< 3; i
++)
10014 strcpy (buf
+ base_name_len
, suffixes
[i
]);
10015 ASET (subsidiaries
, i
, intern (buf
));
10018 return subsidiaries
;
10022 DEFUN ("define-coding-system-internal", Fdefine_coding_system_internal
,
10023 Sdefine_coding_system_internal
, coding_arg_max
, MANY
, 0,
10024 doc
: /* For internal use only.
10025 usage: (define-coding-system-internal ...) */)
10026 (ptrdiff_t nargs
, Lisp_Object
*args
)
10029 Lisp_Object spec_vec
; /* [ ATTRS ALIASE EOL_TYPE ] */
10030 Lisp_Object attrs
; /* Vector of attributes. */
10031 Lisp_Object eol_type
;
10032 Lisp_Object aliases
;
10033 Lisp_Object coding_type
, charset_list
, safe_charsets
;
10034 enum coding_category category
;
10035 Lisp_Object tail
, val
;
10036 int max_charset_id
= 0;
10039 if (nargs
< coding_arg_max
)
10042 attrs
= Fmake_vector (make_number (coding_attr_last_index
), Qnil
);
10044 name
= args
[coding_arg_name
];
10045 CHECK_SYMBOL (name
);
10046 ASET (attrs
, coding_attr_base_name
, name
);
10048 val
= args
[coding_arg_mnemonic
];
10049 if (! STRINGP (val
))
10050 CHECK_CHARACTER (val
);
10051 ASET (attrs
, coding_attr_mnemonic
, val
);
10053 coding_type
= args
[coding_arg_coding_type
];
10054 CHECK_SYMBOL (coding_type
);
10055 ASET (attrs
, coding_attr_type
, coding_type
);
10057 charset_list
= args
[coding_arg_charset_list
];
10058 if (SYMBOLP (charset_list
))
10060 if (EQ (charset_list
, Qiso_2022
))
10062 if (! EQ (coding_type
, Qiso_2022
))
10063 error ("Invalid charset-list");
10064 charset_list
= Viso_2022_charset_list
;
10066 else if (EQ (charset_list
, Qemacs_mule
))
10068 if (! EQ (coding_type
, Qemacs_mule
))
10069 error ("Invalid charset-list");
10070 charset_list
= Vemacs_mule_charset_list
;
10072 for (tail
= charset_list
; CONSP (tail
); tail
= XCDR (tail
))
10074 if (! RANGED_INTEGERP (0, XCAR (tail
), INT_MAX
- 1))
10075 error ("Invalid charset-list");
10076 if (max_charset_id
< XFASTINT (XCAR (tail
)))
10077 max_charset_id
= XFASTINT (XCAR (tail
));
10082 charset_list
= Fcopy_sequence (charset_list
);
10083 for (tail
= charset_list
; CONSP (tail
); tail
= XCDR (tail
))
10085 struct charset
*charset
;
10088 CHECK_CHARSET_GET_CHARSET (val
, charset
);
10089 if (EQ (coding_type
, Qiso_2022
)
10090 ? CHARSET_ISO_FINAL (charset
) < 0
10091 : EQ (coding_type
, Qemacs_mule
)
10092 ? CHARSET_EMACS_MULE_ID (charset
) < 0
10094 error ("Can't handle charset `%s'",
10095 SDATA (SYMBOL_NAME (CHARSET_NAME (charset
))));
10097 XSETCAR (tail
, make_number (charset
->id
));
10098 if (max_charset_id
< charset
->id
)
10099 max_charset_id
= charset
->id
;
10102 ASET (attrs
, coding_attr_charset_list
, charset_list
);
10104 safe_charsets
= make_uninit_string (max_charset_id
+ 1);
10105 memset (SDATA (safe_charsets
), 255, max_charset_id
+ 1);
10106 for (tail
= charset_list
; CONSP (tail
); tail
= XCDR (tail
))
10107 SSET (safe_charsets
, XFASTINT (XCAR (tail
)), 0);
10108 ASET (attrs
, coding_attr_safe_charsets
, safe_charsets
);
10110 ASET (attrs
, coding_attr_ascii_compat
, args
[coding_arg_ascii_compatible_p
]);
10112 val
= args
[coding_arg_decode_translation_table
];
10113 if (! CHAR_TABLE_P (val
) && ! CONSP (val
))
10114 CHECK_SYMBOL (val
);
10115 ASET (attrs
, coding_attr_decode_tbl
, val
);
10117 val
= args
[coding_arg_encode_translation_table
];
10118 if (! CHAR_TABLE_P (val
) && ! CONSP (val
))
10119 CHECK_SYMBOL (val
);
10120 ASET (attrs
, coding_attr_encode_tbl
, val
);
10122 val
= args
[coding_arg_post_read_conversion
];
10123 CHECK_SYMBOL (val
);
10124 ASET (attrs
, coding_attr_post_read
, val
);
10126 val
= args
[coding_arg_pre_write_conversion
];
10127 CHECK_SYMBOL (val
);
10128 ASET (attrs
, coding_attr_pre_write
, val
);
10130 val
= args
[coding_arg_default_char
];
10132 ASET (attrs
, coding_attr_default_char
, make_number (' '));
10135 CHECK_CHARACTER (val
);
10136 ASET (attrs
, coding_attr_default_char
, val
);
10139 val
= args
[coding_arg_for_unibyte
];
10140 ASET (attrs
, coding_attr_for_unibyte
, NILP (val
) ? Qnil
: Qt
);
10142 val
= args
[coding_arg_plist
];
10144 ASET (attrs
, coding_attr_plist
, val
);
10146 if (EQ (coding_type
, Qcharset
))
10148 /* Generate a lisp vector of 256 elements. Each element is nil,
10149 integer, or a list of charset IDs.
10151 If Nth element is nil, the byte code N is invalid in this
10154 If Nth element is a number NUM, N is the first byte of a
10155 charset whose ID is NUM.
10157 If Nth element is a list of charset IDs, N is the first byte
10158 of one of them. The list is sorted by dimensions of the
10159 charsets. A charset of smaller dimension comes first. */
10160 val
= Fmake_vector (make_number (256), Qnil
);
10162 for (tail
= charset_list
; CONSP (tail
); tail
= XCDR (tail
))
10164 struct charset
*charset
= CHARSET_FROM_ID (XFASTINT (XCAR (tail
)));
10165 int dim
= CHARSET_DIMENSION (charset
);
10166 int idx
= (dim
- 1) * 4;
10168 if (CHARSET_ASCII_COMPATIBLE_P (charset
))
10169 ASET (attrs
, coding_attr_ascii_compat
, Qt
);
10171 for (i
= charset
->code_space
[idx
];
10172 i
<= charset
->code_space
[idx
+ 1]; i
++)
10174 Lisp_Object tmp
, tmp2
;
10177 tmp
= AREF (val
, i
);
10180 else if (NUMBERP (tmp
))
10182 dim2
= CHARSET_DIMENSION (CHARSET_FROM_ID (XFASTINT (tmp
)));
10184 tmp
= list2 (XCAR (tail
), tmp
);
10186 tmp
= list2 (tmp
, XCAR (tail
));
10190 for (tmp2
= tmp
; CONSP (tmp2
); tmp2
= XCDR (tmp2
))
10192 dim2
= CHARSET_DIMENSION (CHARSET_FROM_ID (XFASTINT (XCAR (tmp2
))));
10197 tmp
= nconc2 (tmp
, list1 (XCAR (tail
)));
10200 XSETCDR (tmp2
, Fcons (XCAR (tmp2
), XCDR (tmp2
)));
10201 XSETCAR (tmp2
, XCAR (tail
));
10204 ASET (val
, i
, tmp
);
10207 ASET (attrs
, coding_attr_charset_valids
, val
);
10208 category
= coding_category_charset
;
10210 else if (EQ (coding_type
, Qccl
))
10212 Lisp_Object valids
;
10214 if (nargs
< coding_arg_ccl_max
)
10217 val
= args
[coding_arg_ccl_decoder
];
10218 CHECK_CCL_PROGRAM (val
);
10220 val
= Fcopy_sequence (val
);
10221 ASET (attrs
, coding_attr_ccl_decoder
, val
);
10223 val
= args
[coding_arg_ccl_encoder
];
10224 CHECK_CCL_PROGRAM (val
);
10226 val
= Fcopy_sequence (val
);
10227 ASET (attrs
, coding_attr_ccl_encoder
, val
);
10229 val
= args
[coding_arg_ccl_valids
];
10230 valids
= Fmake_string (make_number (256), make_number (0));
10231 for (tail
= val
; CONSP (tail
); tail
= XCDR (tail
))
10236 if (INTEGERP (val
))
10238 if (! (0 <= XINT (val
) && XINT (val
) <= 255))
10239 args_out_of_range_3 (val
, make_number (0), make_number (255));
10240 from
= to
= XINT (val
);
10245 CHECK_NATNUM_CAR (val
);
10246 CHECK_NUMBER_CDR (val
);
10247 if (XINT (XCAR (val
)) > 255)
10248 args_out_of_range_3 (XCAR (val
),
10249 make_number (0), make_number (255));
10250 from
= XINT (XCAR (val
));
10251 if (! (from
<= XINT (XCDR (val
)) && XINT (XCDR (val
)) <= 255))
10252 args_out_of_range_3 (XCDR (val
),
10253 XCAR (val
), make_number (255));
10254 to
= XINT (XCDR (val
));
10256 for (i
= from
; i
<= to
; i
++)
10257 SSET (valids
, i
, 1);
10259 ASET (attrs
, coding_attr_ccl_valids
, valids
);
10261 category
= coding_category_ccl
;
10263 else if (EQ (coding_type
, Qutf_16
))
10265 Lisp_Object bom
, endian
;
10267 ASET (attrs
, coding_attr_ascii_compat
, Qnil
);
10269 if (nargs
< coding_arg_utf16_max
)
10272 bom
= args
[coding_arg_utf16_bom
];
10273 if (! NILP (bom
) && ! EQ (bom
, Qt
))
10277 CHECK_CODING_SYSTEM (val
);
10279 CHECK_CODING_SYSTEM (val
);
10281 ASET (attrs
, coding_attr_utf_bom
, bom
);
10283 endian
= args
[coding_arg_utf16_endian
];
10284 CHECK_SYMBOL (endian
);
10287 else if (! EQ (endian
, Qbig
) && ! EQ (endian
, Qlittle
))
10288 error ("Invalid endian: %s", SDATA (SYMBOL_NAME (endian
)));
10289 ASET (attrs
, coding_attr_utf_16_endian
, endian
);
10291 category
= (CONSP (bom
)
10292 ? coding_category_utf_16_auto
10294 ? (EQ (endian
, Qbig
)
10295 ? coding_category_utf_16_be_nosig
10296 : coding_category_utf_16_le_nosig
)
10297 : (EQ (endian
, Qbig
)
10298 ? coding_category_utf_16_be
10299 : coding_category_utf_16_le
));
10301 else if (EQ (coding_type
, Qiso_2022
))
10303 Lisp_Object initial
, reg_usage
, request
, flags
;
10305 if (nargs
< coding_arg_iso2022_max
)
10308 initial
= Fcopy_sequence (args
[coding_arg_iso2022_initial
]);
10309 CHECK_VECTOR (initial
);
10310 for (i
= 0; i
< 4; i
++)
10312 val
= AREF (initial
, i
);
10315 struct charset
*charset
;
10317 CHECK_CHARSET_GET_CHARSET (val
, charset
);
10318 ASET (initial
, i
, make_number (CHARSET_ID (charset
)));
10319 if (i
== 0 && CHARSET_ASCII_COMPATIBLE_P (charset
))
10320 ASET (attrs
, coding_attr_ascii_compat
, Qt
);
10323 ASET (initial
, i
, make_number (-1));
10326 reg_usage
= args
[coding_arg_iso2022_reg_usage
];
10327 CHECK_CONS (reg_usage
);
10328 CHECK_NUMBER_CAR (reg_usage
);
10329 CHECK_NUMBER_CDR (reg_usage
);
10331 request
= Fcopy_sequence (args
[coding_arg_iso2022_request
]);
10332 for (tail
= request
; CONSP (tail
); tail
= XCDR (tail
))
10340 CHECK_CHARSET_GET_ID (tmp1
, id
);
10341 CHECK_NATNUM_CDR (val
);
10342 if (XINT (XCDR (val
)) >= 4)
10343 error ("Invalid graphic register number: %"pI
"d", XINT (XCDR (val
)));
10344 XSETCAR (val
, make_number (id
));
10347 flags
= args
[coding_arg_iso2022_flags
];
10348 CHECK_NATNUM (flags
);
10349 i
= XINT (flags
) & INT_MAX
;
10350 if (EQ (args
[coding_arg_charset_list
], Qiso_2022
))
10351 i
|= CODING_ISO_FLAG_FULL_SUPPORT
;
10352 flags
= make_number (i
);
10354 ASET (attrs
, coding_attr_iso_initial
, initial
);
10355 ASET (attrs
, coding_attr_iso_usage
, reg_usage
);
10356 ASET (attrs
, coding_attr_iso_request
, request
);
10357 ASET (attrs
, coding_attr_iso_flags
, flags
);
10358 setup_iso_safe_charsets (attrs
);
10360 if (i
& CODING_ISO_FLAG_SEVEN_BITS
)
10361 category
= ((i
& (CODING_ISO_FLAG_LOCKING_SHIFT
10362 | CODING_ISO_FLAG_SINGLE_SHIFT
))
10363 ? coding_category_iso_7_else
10364 : EQ (args
[coding_arg_charset_list
], Qiso_2022
)
10365 ? coding_category_iso_7
10366 : coding_category_iso_7_tight
);
10369 int id
= XINT (AREF (initial
, 1));
10371 category
= (((i
& CODING_ISO_FLAG_LOCKING_SHIFT
)
10372 || EQ (args
[coding_arg_charset_list
], Qiso_2022
)
10374 ? coding_category_iso_8_else
10375 : (CHARSET_DIMENSION (CHARSET_FROM_ID (id
)) == 1)
10376 ? coding_category_iso_8_1
10377 : coding_category_iso_8_2
);
10379 if (category
!= coding_category_iso_8_1
10380 && category
!= coding_category_iso_8_2
)
10381 ASET (attrs
, coding_attr_ascii_compat
, Qnil
);
10383 else if (EQ (coding_type
, Qemacs_mule
))
10385 if (EQ (args
[coding_arg_charset_list
], Qemacs_mule
))
10386 ASET (attrs
, coding_attr_emacs_mule_full
, Qt
);
10387 ASET (attrs
, coding_attr_ascii_compat
, Qt
);
10388 category
= coding_category_emacs_mule
;
10390 else if (EQ (coding_type
, Qshift_jis
))
10393 struct charset
*charset
;
10395 if (XINT (Flength (charset_list
)) != 3
10396 && XINT (Flength (charset_list
)) != 4)
10397 error ("There should be three or four charsets");
10399 charset
= CHARSET_FROM_ID (XINT (XCAR (charset_list
)));
10400 if (CHARSET_DIMENSION (charset
) != 1)
10401 error ("Dimension of charset %s is not one",
10402 SDATA (SYMBOL_NAME (CHARSET_NAME (charset
))));
10403 if (CHARSET_ASCII_COMPATIBLE_P (charset
))
10404 ASET (attrs
, coding_attr_ascii_compat
, Qt
);
10406 charset_list
= XCDR (charset_list
);
10407 charset
= CHARSET_FROM_ID (XINT (XCAR (charset_list
)));
10408 if (CHARSET_DIMENSION (charset
) != 1)
10409 error ("Dimension of charset %s is not one",
10410 SDATA (SYMBOL_NAME (CHARSET_NAME (charset
))));
10412 charset_list
= XCDR (charset_list
);
10413 charset
= CHARSET_FROM_ID (XINT (XCAR (charset_list
)));
10414 if (CHARSET_DIMENSION (charset
) != 2)
10415 error ("Dimension of charset %s is not two",
10416 SDATA (SYMBOL_NAME (CHARSET_NAME (charset
))));
10418 charset_list
= XCDR (charset_list
);
10419 if (! NILP (charset_list
))
10421 charset
= CHARSET_FROM_ID (XINT (XCAR (charset_list
)));
10422 if (CHARSET_DIMENSION (charset
) != 2)
10423 error ("Dimension of charset %s is not two",
10424 SDATA (SYMBOL_NAME (CHARSET_NAME (charset
))));
10427 category
= coding_category_sjis
;
10428 Vsjis_coding_system
= name
;
10430 else if (EQ (coding_type
, Qbig5
))
10432 struct charset
*charset
;
10434 if (XINT (Flength (charset_list
)) != 2)
10435 error ("There should be just two charsets");
10437 charset
= CHARSET_FROM_ID (XINT (XCAR (charset_list
)));
10438 if (CHARSET_DIMENSION (charset
) != 1)
10439 error ("Dimension of charset %s is not one",
10440 SDATA (SYMBOL_NAME (CHARSET_NAME (charset
))));
10441 if (CHARSET_ASCII_COMPATIBLE_P (charset
))
10442 ASET (attrs
, coding_attr_ascii_compat
, Qt
);
10444 charset_list
= XCDR (charset_list
);
10445 charset
= CHARSET_FROM_ID (XINT (XCAR (charset_list
)));
10446 if (CHARSET_DIMENSION (charset
) != 2)
10447 error ("Dimension of charset %s is not two",
10448 SDATA (SYMBOL_NAME (CHARSET_NAME (charset
))));
10450 category
= coding_category_big5
;
10451 Vbig5_coding_system
= name
;
10453 else if (EQ (coding_type
, Qraw_text
))
10455 category
= coding_category_raw_text
;
10456 ASET (attrs
, coding_attr_ascii_compat
, Qt
);
10458 else if (EQ (coding_type
, Qutf_8
))
10462 if (nargs
< coding_arg_utf8_max
)
10465 bom
= args
[coding_arg_utf8_bom
];
10466 if (! NILP (bom
) && ! EQ (bom
, Qt
))
10470 CHECK_CODING_SYSTEM (val
);
10472 CHECK_CODING_SYSTEM (val
);
10474 ASET (attrs
, coding_attr_utf_bom
, bom
);
10476 ASET (attrs
, coding_attr_ascii_compat
, Qt
);
10478 category
= (CONSP (bom
) ? coding_category_utf_8_auto
10479 : NILP (bom
) ? coding_category_utf_8_nosig
10480 : coding_category_utf_8_sig
);
10482 else if (EQ (coding_type
, Qundecided
))
10484 if (nargs
< coding_arg_undecided_max
)
10486 ASET (attrs
, coding_attr_undecided_inhibit_null_byte_detection
,
10487 args
[coding_arg_undecided_inhibit_null_byte_detection
]);
10488 ASET (attrs
, coding_attr_undecided_inhibit_iso_escape_detection
,
10489 args
[coding_arg_undecided_inhibit_iso_escape_detection
]);
10490 ASET (attrs
, coding_attr_undecided_prefer_utf_8
,
10491 args
[coding_arg_undecided_prefer_utf_8
]);
10492 category
= coding_category_undecided
;
10495 error ("Invalid coding system type: %s",
10496 SDATA (SYMBOL_NAME (coding_type
)));
10498 ASET (attrs
, coding_attr_category
, make_number (category
));
10499 ASET (attrs
, coding_attr_plist
,
10501 Fcons (AREF (Vcoding_category_table
, category
),
10502 CODING_ATTR_PLIST (attrs
))));
10503 ASET (attrs
, coding_attr_plist
,
10504 Fcons (QCascii_compatible_p
,
10505 Fcons (CODING_ATTR_ASCII_COMPAT (attrs
),
10506 CODING_ATTR_PLIST (attrs
))));
10508 eol_type
= args
[coding_arg_eol_type
];
10509 if (! NILP (eol_type
)
10510 && ! EQ (eol_type
, Qunix
)
10511 && ! EQ (eol_type
, Qdos
)
10512 && ! EQ (eol_type
, Qmac
))
10513 error ("Invalid eol-type");
10515 aliases
= list1 (name
);
10517 if (NILP (eol_type
))
10519 eol_type
= make_subsidiaries (name
);
10520 for (i
= 0; i
< 3; i
++)
10522 Lisp_Object this_spec
, this_name
, this_aliases
, this_eol_type
;
10524 this_name
= AREF (eol_type
, i
);
10525 this_aliases
= list1 (this_name
);
10526 this_eol_type
= (i
== 0 ? Qunix
: i
== 1 ? Qdos
: Qmac
);
10527 this_spec
= make_uninit_vector (3);
10528 ASET (this_spec
, 0, attrs
);
10529 ASET (this_spec
, 1, this_aliases
);
10530 ASET (this_spec
, 2, this_eol_type
);
10531 Fputhash (this_name
, this_spec
, Vcoding_system_hash_table
);
10532 Vcoding_system_list
= Fcons (this_name
, Vcoding_system_list
);
10533 val
= Fassoc (Fsymbol_name (this_name
), Vcoding_system_alist
);
10535 Vcoding_system_alist
10536 = Fcons (Fcons (Fsymbol_name (this_name
), Qnil
),
10537 Vcoding_system_alist
);
10541 spec_vec
= make_uninit_vector (3);
10542 ASET (spec_vec
, 0, attrs
);
10543 ASET (spec_vec
, 1, aliases
);
10544 ASET (spec_vec
, 2, eol_type
);
10546 Fputhash (name
, spec_vec
, Vcoding_system_hash_table
);
10547 Vcoding_system_list
= Fcons (name
, Vcoding_system_list
);
10548 val
= Fassoc (Fsymbol_name (name
), Vcoding_system_alist
);
10550 Vcoding_system_alist
= Fcons (Fcons (Fsymbol_name (name
), Qnil
),
10551 Vcoding_system_alist
);
10554 int id
= coding_categories
[category
].id
;
10556 if (id
< 0 || EQ (name
, CODING_ID_NAME (id
)))
10557 setup_coding_system (name
, &coding_categories
[category
]);
10563 return Fsignal (Qwrong_number_of_arguments
,
10564 Fcons (intern ("define-coding-system-internal"),
10565 make_number (nargs
)));
10569 DEFUN ("coding-system-put", Fcoding_system_put
, Scoding_system_put
,
10571 doc
: /* Change value in CODING-SYSTEM's property list PROP to VAL. */)
10572 (Lisp_Object coding_system
, Lisp_Object prop
, Lisp_Object val
)
10574 Lisp_Object spec
, attrs
;
10576 CHECK_CODING_SYSTEM_GET_SPEC (coding_system
, spec
);
10577 attrs
= AREF (spec
, 0);
10578 if (EQ (prop
, QCmnemonic
))
10580 if (! STRINGP (val
))
10581 CHECK_CHARACTER (val
);
10582 ASET (attrs
, coding_attr_mnemonic
, val
);
10584 else if (EQ (prop
, QCdefault_char
))
10587 val
= make_number (' ');
10589 CHECK_CHARACTER (val
);
10590 ASET (attrs
, coding_attr_default_char
, val
);
10592 else if (EQ (prop
, QCdecode_translation_table
))
10594 if (! CHAR_TABLE_P (val
) && ! CONSP (val
))
10595 CHECK_SYMBOL (val
);
10596 ASET (attrs
, coding_attr_decode_tbl
, val
);
10598 else if (EQ (prop
, QCencode_translation_table
))
10600 if (! CHAR_TABLE_P (val
) && ! CONSP (val
))
10601 CHECK_SYMBOL (val
);
10602 ASET (attrs
, coding_attr_encode_tbl
, val
);
10604 else if (EQ (prop
, QCpost_read_conversion
))
10606 CHECK_SYMBOL (val
);
10607 ASET (attrs
, coding_attr_post_read
, val
);
10609 else if (EQ (prop
, QCpre_write_conversion
))
10611 CHECK_SYMBOL (val
);
10612 ASET (attrs
, coding_attr_pre_write
, val
);
10614 else if (EQ (prop
, QCascii_compatible_p
))
10616 ASET (attrs
, coding_attr_ascii_compat
, val
);
10619 ASET (attrs
, coding_attr_plist
,
10620 Fplist_put (CODING_ATTR_PLIST (attrs
), prop
, val
));
10625 DEFUN ("define-coding-system-alias", Fdefine_coding_system_alias
,
10626 Sdefine_coding_system_alias
, 2, 2, 0,
10627 doc
: /* Define ALIAS as an alias for CODING-SYSTEM. */)
10628 (Lisp_Object alias
, Lisp_Object coding_system
)
10630 Lisp_Object spec
, aliases
, eol_type
, val
;
10632 CHECK_SYMBOL (alias
);
10633 CHECK_CODING_SYSTEM_GET_SPEC (coding_system
, spec
);
10634 aliases
= AREF (spec
, 1);
10635 /* ALIASES should be a list of length more than zero, and the first
10636 element is a base coding system. Append ALIAS at the tail of the
10638 while (!NILP (XCDR (aliases
)))
10639 aliases
= XCDR (aliases
);
10640 XSETCDR (aliases
, list1 (alias
));
10642 eol_type
= AREF (spec
, 2);
10643 if (VECTORP (eol_type
))
10645 Lisp_Object subsidiaries
;
10648 subsidiaries
= make_subsidiaries (alias
);
10649 for (i
= 0; i
< 3; i
++)
10650 Fdefine_coding_system_alias (AREF (subsidiaries
, i
),
10651 AREF (eol_type
, i
));
10654 Fputhash (alias
, spec
, Vcoding_system_hash_table
);
10655 Vcoding_system_list
= Fcons (alias
, Vcoding_system_list
);
10656 val
= Fassoc (Fsymbol_name (alias
), Vcoding_system_alist
);
10658 Vcoding_system_alist
= Fcons (Fcons (Fsymbol_name (alias
), Qnil
),
10659 Vcoding_system_alist
);
10664 DEFUN ("coding-system-base", Fcoding_system_base
, Scoding_system_base
,
10666 doc
: /* Return the base of CODING-SYSTEM.
10667 Any alias or subsidiary coding system is not a base coding system. */)
10668 (Lisp_Object coding_system
)
10670 Lisp_Object spec
, attrs
;
10672 if (NILP (coding_system
))
10673 return (Qno_conversion
);
10674 CHECK_CODING_SYSTEM_GET_SPEC (coding_system
, spec
);
10675 attrs
= AREF (spec
, 0);
10676 return CODING_ATTR_BASE_NAME (attrs
);
10679 DEFUN ("coding-system-plist", Fcoding_system_plist
, Scoding_system_plist
,
10681 doc
: /* Return the property list of CODING-SYSTEM. */)
10682 (Lisp_Object coding_system
)
10684 Lisp_Object spec
, attrs
;
10686 if (NILP (coding_system
))
10687 coding_system
= Qno_conversion
;
10688 CHECK_CODING_SYSTEM_GET_SPEC (coding_system
, spec
);
10689 attrs
= AREF (spec
, 0);
10690 return CODING_ATTR_PLIST (attrs
);
10694 DEFUN ("coding-system-aliases", Fcoding_system_aliases
, Scoding_system_aliases
,
10696 doc
: /* Return the list of aliases of CODING-SYSTEM. */)
10697 (Lisp_Object coding_system
)
10701 if (NILP (coding_system
))
10702 coding_system
= Qno_conversion
;
10703 CHECK_CODING_SYSTEM_GET_SPEC (coding_system
, spec
);
10704 return AREF (spec
, 1);
10707 DEFUN ("coding-system-eol-type", Fcoding_system_eol_type
,
10708 Scoding_system_eol_type
, 1, 1, 0,
10709 doc
: /* Return eol-type of CODING-SYSTEM.
10710 An eol-type is an integer 0, 1, 2, or a vector of coding systems.
10712 Integer values 0, 1, and 2 indicate a format of end-of-line; LF, CRLF,
10713 and CR respectively.
10715 A vector value indicates that a format of end-of-line should be
10716 detected automatically. Nth element of the vector is the subsidiary
10717 coding system whose eol-type is N. */)
10718 (Lisp_Object coding_system
)
10720 Lisp_Object spec
, eol_type
;
10723 if (NILP (coding_system
))
10724 coding_system
= Qno_conversion
;
10725 if (! CODING_SYSTEM_P (coding_system
))
10727 spec
= CODING_SYSTEM_SPEC (coding_system
);
10728 eol_type
= AREF (spec
, 2);
10729 if (VECTORP (eol_type
))
10730 return Fcopy_sequence (eol_type
);
10731 n
= EQ (eol_type
, Qunix
) ? 0 : EQ (eol_type
, Qdos
) ? 1 : 2;
10732 return make_number (n
);
10738 /*** 9. Post-amble ***/
10741 init_coding_once (void)
10745 for (i
= 0; i
< coding_category_max
; i
++)
10747 coding_categories
[i
].id
= -1;
10748 coding_priorities
[i
] = i
;
10751 /* ISO2022 specific initialize routine. */
10752 for (i
= 0; i
< 0x20; i
++)
10753 iso_code_class
[i
] = ISO_control_0
;
10754 for (i
= 0x21; i
< 0x7F; i
++)
10755 iso_code_class
[i
] = ISO_graphic_plane_0
;
10756 for (i
= 0x80; i
< 0xA0; i
++)
10757 iso_code_class
[i
] = ISO_control_1
;
10758 for (i
= 0xA1; i
< 0xFF; i
++)
10759 iso_code_class
[i
] = ISO_graphic_plane_1
;
10760 iso_code_class
[0x20] = iso_code_class
[0x7F] = ISO_0x20_or_0x7F
;
10761 iso_code_class
[0xA0] = iso_code_class
[0xFF] = ISO_0xA0_or_0xFF
;
10762 iso_code_class
[ISO_CODE_SO
] = ISO_shift_out
;
10763 iso_code_class
[ISO_CODE_SI
] = ISO_shift_in
;
10764 iso_code_class
[ISO_CODE_SS2_7
] = ISO_single_shift_2_7
;
10765 iso_code_class
[ISO_CODE_ESC
] = ISO_escape
;
10766 iso_code_class
[ISO_CODE_SS2
] = ISO_single_shift_2
;
10767 iso_code_class
[ISO_CODE_SS3
] = ISO_single_shift_3
;
10768 iso_code_class
[ISO_CODE_CSI
] = ISO_control_sequence_introducer
;
10770 for (i
= 0; i
< 256; i
++)
10772 emacs_mule_bytes
[i
] = 1;
10774 emacs_mule_bytes
[EMACS_MULE_LEADING_CODE_PRIVATE_11
] = 3;
10775 emacs_mule_bytes
[EMACS_MULE_LEADING_CODE_PRIVATE_12
] = 3;
10776 emacs_mule_bytes
[EMACS_MULE_LEADING_CODE_PRIVATE_21
] = 4;
10777 emacs_mule_bytes
[EMACS_MULE_LEADING_CODE_PRIVATE_22
] = 4;
10783 syms_of_coding (void)
10785 staticpro (&Vcoding_system_hash_table
);
10786 Vcoding_system_hash_table
= CALLN (Fmake_hash_table
, QCtest
, Qeq
);
10788 staticpro (&Vsjis_coding_system
);
10789 Vsjis_coding_system
= Qnil
;
10791 staticpro (&Vbig5_coding_system
);
10792 Vbig5_coding_system
= Qnil
;
10794 staticpro (&Vcode_conversion_reused_workbuf
);
10795 Vcode_conversion_reused_workbuf
= Qnil
;
10797 staticpro (&Vcode_conversion_workbuf_name
);
10798 Vcode_conversion_workbuf_name
= build_pure_c_string (" *code-conversion-work*");
10800 reused_workbuf_in_use
= 0;
10802 DEFSYM (Qcharset
, "charset");
10803 DEFSYM (Qtarget_idx
, "target-idx");
10804 DEFSYM (Qcoding_system_history
, "coding-system-history");
10805 Fset (Qcoding_system_history
, Qnil
);
10807 /* Target FILENAME is the first argument. */
10808 Fput (Qinsert_file_contents
, Qtarget_idx
, make_number (0));
10809 /* Target FILENAME is the third argument. */
10810 Fput (Qwrite_region
, Qtarget_idx
, make_number (2));
10812 DEFSYM (Qcall_process
, "call-process");
10813 /* Target PROGRAM is the first argument. */
10814 Fput (Qcall_process
, Qtarget_idx
, make_number (0));
10816 DEFSYM (Qcall_process_region
, "call-process-region");
10817 /* Target PROGRAM is the third argument. */
10818 Fput (Qcall_process_region
, Qtarget_idx
, make_number (2));
10820 DEFSYM (Qstart_process
, "start-process");
10821 /* Target PROGRAM is the third argument. */
10822 Fput (Qstart_process
, Qtarget_idx
, make_number (2));
10824 DEFSYM (Qopen_network_stream
, "open-network-stream");
10825 /* Target SERVICE is the fourth argument. */
10826 Fput (Qopen_network_stream
, Qtarget_idx
, make_number (3));
10828 DEFSYM (Qunix
, "unix");
10829 DEFSYM (Qdos
, "dos");
10830 DEFSYM (Qmac
, "mac");
10832 DEFSYM (Qbuffer_file_coding_system
, "buffer-file-coding-system");
10833 DEFSYM (Qundecided
, "undecided");
10834 DEFSYM (Qno_conversion
, "no-conversion");
10835 DEFSYM (Qraw_text
, "raw-text");
10837 DEFSYM (Qiso_2022
, "iso-2022");
10839 DEFSYM (Qutf_8
, "utf-8");
10840 DEFSYM (Qutf_8_emacs
, "utf-8-emacs");
10842 #if defined (WINDOWSNT) || defined (CYGWIN)
10843 /* No, not utf-16-le: that one has a BOM. */
10844 DEFSYM (Qutf_16le
, "utf-16le");
10847 DEFSYM (Qutf_16
, "utf-16");
10848 DEFSYM (Qbig
, "big");
10849 DEFSYM (Qlittle
, "little");
10851 DEFSYM (Qshift_jis
, "shift-jis");
10852 DEFSYM (Qbig5
, "big5");
10854 DEFSYM (Qcoding_system_p
, "coding-system-p");
10856 /* Error signaled when there's a problem with detecting a coding system. */
10857 DEFSYM (Qcoding_system_error
, "coding-system-error");
10858 Fput (Qcoding_system_error
, Qerror_conditions
,
10859 listn (CONSTYPE_PURE
, 2, Qcoding_system_error
, Qerror
));
10860 Fput (Qcoding_system_error
, Qerror_message
,
10861 build_pure_c_string ("Invalid coding system"));
10863 DEFSYM (Qtranslation_table
, "translation-table");
10864 Fput (Qtranslation_table
, Qchar_table_extra_slots
, make_number (2));
10865 DEFSYM (Qtranslation_table_id
, "translation-table-id");
10867 /* Coding system emacs-mule and raw-text are for converting only
10868 end-of-line format. */
10869 DEFSYM (Qemacs_mule
, "emacs-mule");
10871 DEFSYM (QCcategory
, ":category");
10872 DEFSYM (QCmnemonic
, ":mnemonic");
10873 DEFSYM (QCdefault_char
, ":default-char");
10874 DEFSYM (QCdecode_translation_table
, ":decode-translation-table");
10875 DEFSYM (QCencode_translation_table
, ":encode-translation-table");
10876 DEFSYM (QCpost_read_conversion
, ":post-read-conversion");
10877 DEFSYM (QCpre_write_conversion
, ":pre-write-conversion");
10878 DEFSYM (QCascii_compatible_p
, ":ascii-compatible-p");
10880 Vcoding_category_table
10881 = Fmake_vector (make_number (coding_category_max
), Qnil
);
10882 staticpro (&Vcoding_category_table
);
10883 /* Followings are target of code detection. */
10884 ASET (Vcoding_category_table
, coding_category_iso_7
,
10885 intern_c_string ("coding-category-iso-7"));
10886 ASET (Vcoding_category_table
, coding_category_iso_7_tight
,
10887 intern_c_string ("coding-category-iso-7-tight"));
10888 ASET (Vcoding_category_table
, coding_category_iso_8_1
,
10889 intern_c_string ("coding-category-iso-8-1"));
10890 ASET (Vcoding_category_table
, coding_category_iso_8_2
,
10891 intern_c_string ("coding-category-iso-8-2"));
10892 ASET (Vcoding_category_table
, coding_category_iso_7_else
,
10893 intern_c_string ("coding-category-iso-7-else"));
10894 ASET (Vcoding_category_table
, coding_category_iso_8_else
,
10895 intern_c_string ("coding-category-iso-8-else"));
10896 ASET (Vcoding_category_table
, coding_category_utf_8_auto
,
10897 intern_c_string ("coding-category-utf-8-auto"));
10898 ASET (Vcoding_category_table
, coding_category_utf_8_nosig
,
10899 intern_c_string ("coding-category-utf-8"));
10900 ASET (Vcoding_category_table
, coding_category_utf_8_sig
,
10901 intern_c_string ("coding-category-utf-8-sig"));
10902 ASET (Vcoding_category_table
, coding_category_utf_16_be
,
10903 intern_c_string ("coding-category-utf-16-be"));
10904 ASET (Vcoding_category_table
, coding_category_utf_16_auto
,
10905 intern_c_string ("coding-category-utf-16-auto"));
10906 ASET (Vcoding_category_table
, coding_category_utf_16_le
,
10907 intern_c_string ("coding-category-utf-16-le"));
10908 ASET (Vcoding_category_table
, coding_category_utf_16_be_nosig
,
10909 intern_c_string ("coding-category-utf-16-be-nosig"));
10910 ASET (Vcoding_category_table
, coding_category_utf_16_le_nosig
,
10911 intern_c_string ("coding-category-utf-16-le-nosig"));
10912 ASET (Vcoding_category_table
, coding_category_charset
,
10913 intern_c_string ("coding-category-charset"));
10914 ASET (Vcoding_category_table
, coding_category_sjis
,
10915 intern_c_string ("coding-category-sjis"));
10916 ASET (Vcoding_category_table
, coding_category_big5
,
10917 intern_c_string ("coding-category-big5"));
10918 ASET (Vcoding_category_table
, coding_category_ccl
,
10919 intern_c_string ("coding-category-ccl"));
10920 ASET (Vcoding_category_table
, coding_category_emacs_mule
,
10921 intern_c_string ("coding-category-emacs-mule"));
10922 /* Followings are NOT target of code detection. */
10923 ASET (Vcoding_category_table
, coding_category_raw_text
,
10924 intern_c_string ("coding-category-raw-text"));
10925 ASET (Vcoding_category_table
, coding_category_undecided
,
10926 intern_c_string ("coding-category-undecided"));
10928 DEFSYM (Qinsufficient_source
, "insufficient-source");
10929 DEFSYM (Qinvalid_source
, "invalid-source");
10930 DEFSYM (Qinterrupted
, "interrupted");
10932 /* If a symbol has this property, evaluate the value to define the
10933 symbol as a coding system. */
10934 DEFSYM (Qcoding_system_define_form
, "coding-system-define-form");
10936 defsubr (&Scoding_system_p
);
10937 defsubr (&Sread_coding_system
);
10938 defsubr (&Sread_non_nil_coding_system
);
10939 defsubr (&Scheck_coding_system
);
10940 defsubr (&Sdetect_coding_region
);
10941 defsubr (&Sdetect_coding_string
);
10942 defsubr (&Sfind_coding_systems_region_internal
);
10943 defsubr (&Sunencodable_char_position
);
10944 defsubr (&Scheck_coding_systems_region
);
10945 defsubr (&Sdecode_coding_region
);
10946 defsubr (&Sencode_coding_region
);
10947 defsubr (&Sdecode_coding_string
);
10948 defsubr (&Sencode_coding_string
);
10949 defsubr (&Sdecode_sjis_char
);
10950 defsubr (&Sencode_sjis_char
);
10951 defsubr (&Sdecode_big5_char
);
10952 defsubr (&Sencode_big5_char
);
10953 defsubr (&Sset_terminal_coding_system_internal
);
10954 defsubr (&Sset_safe_terminal_coding_system_internal
);
10955 defsubr (&Sterminal_coding_system
);
10956 defsubr (&Sset_keyboard_coding_system_internal
);
10957 defsubr (&Skeyboard_coding_system
);
10958 defsubr (&Sfind_operation_coding_system
);
10959 defsubr (&Sset_coding_system_priority
);
10960 defsubr (&Sdefine_coding_system_internal
);
10961 defsubr (&Sdefine_coding_system_alias
);
10962 defsubr (&Scoding_system_put
);
10963 defsubr (&Scoding_system_base
);
10964 defsubr (&Scoding_system_plist
);
10965 defsubr (&Scoding_system_aliases
);
10966 defsubr (&Scoding_system_eol_type
);
10967 defsubr (&Scoding_system_priority_list
);
10969 DEFVAR_LISP ("coding-system-list", Vcoding_system_list
,
10970 doc
: /* List of coding systems.
10972 Do not alter the value of this variable manually. This variable should be
10973 updated by the functions `define-coding-system' and
10974 `define-coding-system-alias'. */);
10975 Vcoding_system_list
= Qnil
;
10977 DEFVAR_LISP ("coding-system-alist", Vcoding_system_alist
,
10978 doc
: /* Alist of coding system names.
10979 Each element is one element list of coding system name.
10980 This variable is given to `completing-read' as COLLECTION argument.
10982 Do not alter the value of this variable manually. This variable should be
10983 updated by the functions `make-coding-system' and
10984 `define-coding-system-alias'. */);
10985 Vcoding_system_alist
= Qnil
;
10987 DEFVAR_LISP ("coding-category-list", Vcoding_category_list
,
10988 doc
: /* List of coding-categories (symbols) ordered by priority.
10990 On detecting a coding system, Emacs tries code detection algorithms
10991 associated with each coding-category one by one in this order. When
10992 one algorithm agrees with a byte sequence of source text, the coding
10993 system bound to the corresponding coding-category is selected.
10995 Don't modify this variable directly, but use `set-coding-system-priority'. */);
10999 Vcoding_category_list
= Qnil
;
11000 for (i
= coding_category_max
- 1; i
>= 0; i
--)
11001 Vcoding_category_list
11002 = Fcons (AREF (Vcoding_category_table
, i
),
11003 Vcoding_category_list
);
11006 DEFVAR_LISP ("coding-system-for-read", Vcoding_system_for_read
,
11007 doc
: /* Specify the coding system for read operations.
11008 It is useful to bind this variable with `let', but do not set it globally.
11009 If the value is a coding system, it is used for decoding on read operation.
11010 If not, an appropriate element is used from one of the coding system alists.
11011 There are three such tables: `file-coding-system-alist',
11012 `process-coding-system-alist', and `network-coding-system-alist'. */);
11013 Vcoding_system_for_read
= Qnil
;
11015 DEFVAR_LISP ("coding-system-for-write", Vcoding_system_for_write
,
11016 doc
: /* Specify the coding system for write operations.
11017 Programs bind this variable with `let', but you should not set it globally.
11018 If the value is a coding system, it is used for encoding of output,
11019 when writing it to a file and when sending it to a file or subprocess.
11021 If this does not specify a coding system, an appropriate element
11022 is used from one of the coding system alists.
11023 There are three such tables: `file-coding-system-alist',
11024 `process-coding-system-alist', and `network-coding-system-alist'.
11025 For output to files, if the above procedure does not specify a coding system,
11026 the value of `buffer-file-coding-system' is used. */);
11027 Vcoding_system_for_write
= Qnil
;
11029 DEFVAR_LISP ("last-coding-system-used", Vlast_coding_system_used
,
11031 Coding system used in the latest file or process I/O. */);
11032 Vlast_coding_system_used
= Qnil
;
11034 DEFVAR_LISP ("last-code-conversion-error", Vlast_code_conversion_error
,
11036 Error status of the last code conversion.
11038 When an error was detected in the last code conversion, this variable
11039 is set to one of the following symbols.
11040 `insufficient-source'
11044 `insufficient-memory'
11045 When no error was detected, the value doesn't change. So, to check
11046 the error status of a code conversion by this variable, you must
11047 explicitly set this variable to nil before performing code
11049 Vlast_code_conversion_error
= Qnil
;
11051 DEFVAR_BOOL ("inhibit-eol-conversion", inhibit_eol_conversion
,
11053 Non-nil means always inhibit code conversion of end-of-line format.
11054 See info node `Coding Systems' and info node `Text and Binary' concerning
11055 such conversion. */);
11056 inhibit_eol_conversion
= 0;
11058 DEFVAR_BOOL ("inherit-process-coding-system", inherit_process_coding_system
,
11060 Non-nil means process buffer inherits coding system of process output.
11061 Bind it to t if the process output is to be treated as if it were a file
11062 read from some filesystem. */);
11063 inherit_process_coding_system
= 0;
11065 DEFVAR_LISP ("file-coding-system-alist", Vfile_coding_system_alist
,
11067 Alist to decide a coding system to use for a file I/O operation.
11068 The format is ((PATTERN . VAL) ...),
11069 where PATTERN is a regular expression matching a file name,
11070 VAL is a coding system, a cons of coding systems, or a function symbol.
11071 If VAL is a coding system, it is used for both decoding and encoding
11073 If VAL is a cons of coding systems, the car part is used for decoding,
11074 and the cdr part is used for encoding.
11075 If VAL is a function symbol, the function must return a coding system
11076 or a cons of coding systems which are used as above. The function is
11077 called with an argument that is a list of the arguments with which
11078 `find-operation-coding-system' was called. If the function can't decide
11079 a coding system, it can return `undecided' so that the normal
11080 code-detection is performed.
11082 See also the function `find-operation-coding-system'
11083 and the variable `auto-coding-alist'. */);
11084 Vfile_coding_system_alist
= Qnil
;
11086 DEFVAR_LISP ("process-coding-system-alist", Vprocess_coding_system_alist
,
11088 Alist to decide a coding system to use for a process I/O operation.
11089 The format is ((PATTERN . VAL) ...),
11090 where PATTERN is a regular expression matching a program name,
11091 VAL is a coding system, a cons of coding systems, or a function symbol.
11092 If VAL is a coding system, it is used for both decoding what received
11093 from the program and encoding what sent to the program.
11094 If VAL is a cons of coding systems, the car part is used for decoding,
11095 and the cdr part is used for encoding.
11096 If VAL is a function symbol, the function must return a coding system
11097 or a cons of coding systems which are used as above.
11099 See also the function `find-operation-coding-system'. */);
11100 Vprocess_coding_system_alist
= Qnil
;
11102 DEFVAR_LISP ("network-coding-system-alist", Vnetwork_coding_system_alist
,
11104 Alist to decide a coding system to use for a network I/O operation.
11105 The format is ((PATTERN . VAL) ...),
11106 where PATTERN is a regular expression matching a network service name
11107 or is a port number to connect to,
11108 VAL is a coding system, a cons of coding systems, or a function symbol.
11109 If VAL is a coding system, it is used for both decoding what received
11110 from the network stream and encoding what sent to the network stream.
11111 If VAL is a cons of coding systems, the car part is used for decoding,
11112 and the cdr part is used for encoding.
11113 If VAL is a function symbol, the function must return a coding system
11114 or a cons of coding systems which are used as above.
11116 See also the function `find-operation-coding-system'. */);
11117 Vnetwork_coding_system_alist
= Qnil
;
11119 DEFVAR_LISP ("locale-coding-system", Vlocale_coding_system
,
11120 doc
: /* Coding system to use with system messages.
11121 Also used for decoding keyboard input on X Window system, and for
11122 encoding standard output and error streams. */);
11123 Vlocale_coding_system
= Qnil
;
11125 /* The eol mnemonics are reset in startup.el system-dependently. */
11126 DEFVAR_LISP ("eol-mnemonic-unix", eol_mnemonic_unix
,
11128 String displayed in mode line for UNIX-like (LF) end-of-line format. */);
11129 eol_mnemonic_unix
= build_pure_c_string (":");
11131 DEFVAR_LISP ("eol-mnemonic-dos", eol_mnemonic_dos
,
11133 String displayed in mode line for DOS-like (CRLF) end-of-line format. */);
11134 eol_mnemonic_dos
= build_pure_c_string ("\\");
11136 DEFVAR_LISP ("eol-mnemonic-mac", eol_mnemonic_mac
,
11138 String displayed in mode line for MAC-like (CR) end-of-line format. */);
11139 eol_mnemonic_mac
= build_pure_c_string ("/");
11141 DEFVAR_LISP ("eol-mnemonic-undecided", eol_mnemonic_undecided
,
11143 String displayed in mode line when end-of-line format is not yet determined. */);
11144 eol_mnemonic_undecided
= build_pure_c_string (":");
11146 DEFVAR_LISP ("enable-character-translation", Venable_character_translation
,
11148 Non-nil enables character translation while encoding and decoding. */);
11149 Venable_character_translation
= Qt
;
11151 DEFVAR_LISP ("standard-translation-table-for-decode",
11152 Vstandard_translation_table_for_decode
,
11153 doc
: /* Table for translating characters while decoding. */);
11154 Vstandard_translation_table_for_decode
= Qnil
;
11156 DEFVAR_LISP ("standard-translation-table-for-encode",
11157 Vstandard_translation_table_for_encode
,
11158 doc
: /* Table for translating characters while encoding. */);
11159 Vstandard_translation_table_for_encode
= Qnil
;
11161 DEFVAR_LISP ("charset-revision-table", Vcharset_revision_table
,
11162 doc
: /* Alist of charsets vs revision numbers.
11163 While encoding, if a charset (car part of an element) is found,
11164 designate it with the escape sequence identifying revision (cdr part
11165 of the element). */);
11166 Vcharset_revision_table
= Qnil
;
11168 DEFVAR_LISP ("default-process-coding-system",
11169 Vdefault_process_coding_system
,
11170 doc
: /* Cons of coding systems used for process I/O by default.
11171 The car part is used for decoding a process output,
11172 the cdr part is used for encoding a text to be sent to a process. */);
11173 Vdefault_process_coding_system
= Qnil
;
11175 DEFVAR_LISP ("latin-extra-code-table", Vlatin_extra_code_table
,
11177 Table of extra Latin codes in the range 128..159 (inclusive).
11178 This is a vector of length 256.
11179 If Nth element is non-nil, the existence of code N in a file
11180 (or output of subprocess) doesn't prevent it to be detected as
11181 a coding system of ISO 2022 variant which has a flag
11182 `accept-latin-extra-code' t (e.g. iso-latin-1) on reading a file
11183 or reading output of a subprocess.
11184 Only 128th through 159th elements have a meaning. */);
11185 Vlatin_extra_code_table
= Fmake_vector (make_number (256), Qnil
);
11187 DEFVAR_LISP ("select-safe-coding-system-function",
11188 Vselect_safe_coding_system_function
,
11190 Function to call to select safe coding system for encoding a text.
11192 If set, this function is called to force a user to select a proper
11193 coding system which can encode the text in the case that a default
11194 coding system used in each operation can't encode the text. The
11195 function should take care that the buffer is not modified while
11196 the coding system is being selected.
11198 The default value is `select-safe-coding-system' (which see). */);
11199 Vselect_safe_coding_system_function
= Qnil
;
11201 DEFVAR_BOOL ("coding-system-require-warning",
11202 coding_system_require_warning
,
11203 doc
: /* Internal use only.
11204 If non-nil, on writing a file, `select-safe-coding-system-function' is
11205 called even if `coding-system-for-write' is non-nil. The command
11206 `universal-coding-system-argument' binds this variable to t temporarily. */);
11207 coding_system_require_warning
= 0;
11210 DEFVAR_BOOL ("inhibit-iso-escape-detection",
11211 inhibit_iso_escape_detection
,
11213 If non-nil, Emacs ignores ISO-2022 escape sequences during code detection.
11215 When Emacs reads text, it tries to detect how the text is encoded.
11216 This code detection is sensitive to escape sequences. If Emacs sees
11217 a valid ISO-2022 escape sequence, it assumes the text is encoded in one
11218 of the ISO2022 encodings, and decodes text by the corresponding coding
11219 system (e.g. `iso-2022-7bit').
11221 However, there may be a case that you want to read escape sequences in
11222 a file as is. In such a case, you can set this variable to non-nil.
11223 Then the code detection will ignore any escape sequences, and no text is
11224 detected as encoded in some ISO-2022 encoding. The result is that all
11225 escape sequences become visible in a buffer.
11227 The default value is nil, and it is strongly recommended not to change
11228 it. That is because many Emacs Lisp source files that contain
11229 non-ASCII characters are encoded by the coding system `iso-2022-7bit'
11230 in Emacs's distribution, and they won't be decoded correctly on
11231 reading if you suppress escape sequence detection.
11233 The other way to read escape sequences in a file without decoding is
11234 to explicitly specify some coding system that doesn't use ISO-2022
11235 escape sequence (e.g., `latin-1') on reading by \\[universal-coding-system-argument]. */);
11236 inhibit_iso_escape_detection
= 0;
11238 DEFVAR_BOOL ("inhibit-null-byte-detection",
11239 inhibit_null_byte_detection
,
11240 doc
: /* If non-nil, Emacs ignores null bytes on code detection.
11241 By default, Emacs treats it as binary data, and does not attempt to
11242 decode it. The effect is as if you specified `no-conversion' for
11245 Set this to non-nil when a regular text happens to include null bytes.
11246 Examples are Index nodes of Info files and null-byte delimited output
11247 from GNU Find and GNU Grep. Emacs will then ignore the null bytes and
11248 decode text as usual. */);
11249 inhibit_null_byte_detection
= 0;
11251 DEFVAR_BOOL ("disable-ascii-optimization", disable_ascii_optimization
,
11252 doc
: /* If non-nil, Emacs does not optimize code decoder for ASCII files.
11253 Internal use only. Remove after the experimental optimizer becomes stable. */);
11254 disable_ascii_optimization
= 0;
11256 DEFVAR_LISP ("translation-table-for-input", Vtranslation_table_for_input
,
11257 doc
: /* Char table for translating self-inserting characters.
11258 This is applied to the result of input methods, not their input.
11259 See also `keyboard-translate-table'.
11261 Use of this variable for character code unification was rendered
11262 obsolete in Emacs 23.1 and later, since Unicode is now the basis of
11263 internal character representation. */);
11264 Vtranslation_table_for_input
= Qnil
;
11266 Lisp_Object args
[coding_arg_undecided_max
];
11267 memclear (args
, sizeof args
);
11269 Lisp_Object plist
[] =
11272 args
[coding_arg_name
] = Qno_conversion
,
11274 args
[coding_arg_mnemonic
] = make_number ('='),
11275 intern_c_string (":coding-type"),
11276 args
[coding_arg_coding_type
] = Qraw_text
,
11277 QCascii_compatible_p
,
11278 args
[coding_arg_ascii_compatible_p
] = Qt
,
11280 args
[coding_arg_default_char
] = make_number (0),
11281 intern_c_string (":for-unibyte"),
11282 args
[coding_arg_for_unibyte
] = Qt
,
11283 intern_c_string (":docstring"),
11284 (build_pure_c_string
11285 ("Do no conversion.\n"
11287 "When you visit a file with this coding, the file is read into a\n"
11288 "unibyte buffer as is, thus each byte of a file is treated as a\n"
11290 intern_c_string (":eol-type"),
11291 args
[coding_arg_eol_type
] = Qunix
,
11293 args
[coding_arg_plist
] = CALLMANY (Flist
, plist
);
11294 Fdefine_coding_system_internal (coding_arg_max
, args
);
11296 plist
[1] = args
[coding_arg_name
] = Qundecided
;
11297 plist
[3] = args
[coding_arg_mnemonic
] = make_number ('-');
11298 plist
[5] = args
[coding_arg_coding_type
] = Qundecided
;
11299 /* This is already set.
11300 plist[7] = args[coding_arg_ascii_compatible_p] = Qt; */
11301 plist
[8] = intern_c_string (":charset-list");
11302 plist
[9] = args
[coding_arg_charset_list
] = Fcons (Qascii
, Qnil
);
11303 plist
[11] = args
[coding_arg_for_unibyte
] = Qnil
;
11304 plist
[13] = build_pure_c_string ("No conversion on encoding, "
11305 "automatic conversion on decoding.");
11306 plist
[15] = args
[coding_arg_eol_type
] = Qnil
;
11307 args
[coding_arg_plist
] = CALLMANY (Flist
, plist
);
11308 args
[coding_arg_undecided_inhibit_null_byte_detection
] = make_number (0);
11309 args
[coding_arg_undecided_inhibit_iso_escape_detection
] = make_number (0);
11310 Fdefine_coding_system_internal (coding_arg_undecided_max
, args
);
11312 setup_coding_system (Qno_conversion
, &safe_terminal_coding
);
11314 for (int i
= 0; i
< coding_category_max
; i
++)
11315 Fset (AREF (Vcoding_category_table
, i
), Qno_conversion
);
11317 #if defined (DOS_NT)
11318 system_eol_type
= Qdos
;
11320 system_eol_type
= Qunix
;
11322 staticpro (&system_eol_type
);
11326 emacs_strerror (int error_number
)
11330 synchronize_system_messages_locale ();
11331 str
= strerror (error_number
);
11333 if (! NILP (Vlocale_coding_system
))
11335 Lisp_Object dec
= code_convert_string_norecord (build_string (str
),
11336 Vlocale_coding_system
,
11338 str
= SSDATA (dec
);