1 /* Coding system handler (conversion, detection, and etc).
2 Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN.
3 Licensed to the Free Software Foundation.
5 This file is part of GNU Emacs.
7 GNU Emacs is free software; you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation; either version 2, or (at your option)
12 GNU Emacs is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
17 You should have received a copy of the GNU General Public License
18 along with GNU Emacs; see the file COPYING. If not, write to
19 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
20 Boston, MA 02111-1307, USA. */
22 /*** TABLE OF CONTENTS ***
25 2. Emacs' internal format (emacs-mule) handlers
27 4. Shift-JIS and BIG5 handlers
28 5. End-of-line handlers
29 6. C library functions
30 7. Emacs Lisp library functions
35 /*** GENERAL NOTE on CODING SYSTEM ***
37 Coding system is an encoding mechanism of one or more character
38 sets. Here's a list of coding systems which Emacs can handle. When
39 we say "decode", it means converting some other coding system to
40 Emacs' internal format (emacs-internal), and when we say "encode",
41 it means converting the coding system emacs-mule to some other
44 0. Emacs' internal format (emacs-mule)
46 Emacs itself holds a multi-lingual character in a buffer and a string
47 in a special format. Details are described in section 2.
51 The most famous coding system for multiple character sets. X's
52 Compound Text, various EUCs (Extended Unix Code), and coding
53 systems used in Internet communication such as ISO-2022-JP are
54 all variants of ISO2022. Details are described in section 3.
56 2. SJIS (or Shift-JIS or MS-Kanji-Code)
58 A coding system to encode character sets: ASCII, JISX0201, and
59 JISX0208. Widely used for PC's in Japan. Details are described in
64 A coding system to encode character sets: ASCII and Big5. Widely
65 used by Chinese (mainly in Taiwan and Hong Kong). Details are
66 described in section 4. In this file, when we write "BIG5"
67 (all uppercase), we mean the coding system, and when we write
68 "Big5" (capitalized), we mean the character set.
72 A coding system for a text containing random 8-bit code. Emacs does
73 no code conversion on such a text except for end-of-line format.
77 If a user wants to read/write a text encoded in a coding system not
78 listed above, he can supply a decoder and an encoder for it in CCL
79 (Code Conversion Language) programs. Emacs executes the CCL program
80 while reading/writing.
82 Emacs represents a coding system by a Lisp symbol that has a property
83 `coding-system'. But, before actually using the coding system, the
84 information about it is set in a structure of type `struct
85 coding_system' for rapid processing. See section 6 for more details.
89 /*** GENERAL NOTES on END-OF-LINE FORMAT ***
91 How end-of-line of a text is encoded depends on a system. For
92 instance, Unix's format is just one byte of `line-feed' code,
93 whereas DOS's format is two-byte sequence of `carriage-return' and
94 `line-feed' codes. MacOS's format is usually one byte of
97 Since text characters encoding and end-of-line encoding are
98 independent, any coding system described above can take
99 any format of end-of-line. So, Emacs has information of format of
100 end-of-line in each coding-system. See section 6 for more details.
104 /*** GENERAL NOTES on `detect_coding_XXX ()' functions ***
106 These functions check if a text between SRC and SRC_END is encoded
107 in the coding system category XXX. Each returns an integer value in
108 which appropriate flag bits for the category XXX is set. The flag
109 bits are defined in macros CODING_CATEGORY_MASK_XXX. Below is the
110 template of these functions. */
113 detect_coding_emacs_mule (src
, src_end
)
114 unsigned char *src
, *src_end
;
120 /*** GENERAL NOTES on `decode_coding_XXX ()' functions ***
122 These functions decode SRC_BYTES length text at SOURCE encoded in
123 CODING to Emacs' internal format (emacs-mule). The resulting text
124 goes to a place pointed to by DESTINATION, the length of which
125 should not exceed DST_BYTES. These functions set the information of
126 original and decoded texts in the members produced, produced_char,
127 consumed, and consumed_char of the structure *CODING.
129 The return value is an integer (CODING_FINISH_XXX) indicating how
130 the decoding finished.
132 DST_BYTES zero means that source area and destination area are
133 overlapped, which means that we can produce a decoded text until it
134 reaches at the head of not-yet-decoded source text.
136 Below is a template of these functions. */
138 decode_coding_XXX (coding
, source
, destination
, src_bytes
, dst_bytes
)
139 struct coding_system
*coding
;
140 unsigned char *source
, *destination
;
141 int src_bytes
, dst_bytes
;
147 /*** GENERAL NOTES on `encode_coding_XXX ()' functions ***
149 These functions encode SRC_BYTES length text at SOURCE of Emacs'
150 internal format (emacs-mule) to CODING. The resulting text goes to
151 a place pointed to by DESTINATION, the length of which should not
152 exceed DST_BYTES. These functions set the information of
153 original and encoded texts in the members produced, produced_char,
154 consumed, and consumed_char of the structure *CODING.
156 The return value is an integer (CODING_FINISH_XXX) indicating how
157 the encoding finished.
159 DST_BYTES zero means that source area and destination area are
160 overlapped, which means that we can produce a decoded text until it
161 reaches at the head of not-yet-decoded source text.
163 Below is a template of these functions. */
165 encode_coding_XXX (coding
, source
, destination
, src_bytes
, dst_bytes
)
166 struct coding_system
*coding
;
167 unsigned char *source
, *destination
;
168 int src_bytes
, dst_bytes
;
174 /*** COMMONLY USED MACROS ***/
176 /* The following three macros ONE_MORE_BYTE, TWO_MORE_BYTES, and
177 THREE_MORE_BYTES safely get one, two, and three bytes from the
178 source text respectively. If there are not enough bytes in the
179 source, they jump to `label_end_of_loop'. The caller should set
180 variables `src' and `src_end' to appropriate areas in advance. */
182 #define ONE_MORE_BYTE(c1) \
187 goto label_end_of_loop; \
190 #define TWO_MORE_BYTES(c1, c2) \
192 if (src + 1 < src_end) \
193 c1 = *src++, c2 = *src++; \
195 goto label_end_of_loop; \
198 #define THREE_MORE_BYTES(c1, c2, c3) \
200 if (src + 2 < src_end) \
201 c1 = *src++, c2 = *src++, c3 = *src++; \
203 goto label_end_of_loop; \
206 /* The following three macros DECODE_CHARACTER_ASCII,
207 DECODE_CHARACTER_DIMENSION1, and DECODE_CHARACTER_DIMENSION2 put
208 the multi-byte form of a character of each class at the place
209 pointed by `dst'. The caller should set the variable `dst' to
210 point to an appropriate area and the variable `coding' to point to
211 the coding-system of the currently decoding text in advance. */
213 /* Decode one ASCII character C. */
215 #define DECODE_CHARACTER_ASCII(c) \
217 if (COMPOSING_P (coding->composing)) \
218 *dst++ = 0xA0, *dst++ = (c) | 0x80; \
222 coding->produced_char++; \
226 /* Decode one DIMENSION1 character whose charset is CHARSET and whose
227 position-code is C. */
229 #define DECODE_CHARACTER_DIMENSION1(charset, c) \
231 unsigned char leading_code = CHARSET_LEADING_CODE_BASE (charset); \
232 if (COMPOSING_P (coding->composing)) \
233 *dst++ = leading_code + 0x20; \
236 *dst++ = leading_code; \
237 coding->produced_char++; \
239 if (leading_code = CHARSET_LEADING_CODE_EXT (charset)) \
240 *dst++ = leading_code; \
241 *dst++ = (c) | 0x80; \
244 /* Decode one DIMENSION2 character whose charset is CHARSET and whose
245 position-codes are C1 and C2. */
247 #define DECODE_CHARACTER_DIMENSION2(charset, c1, c2) \
249 DECODE_CHARACTER_DIMENSION1 (charset, c1); \
250 *dst++ = (c2) | 0x80; \
254 /*** 1. Preamble ***/
268 #else /* not emacs */
272 #endif /* not emacs */
274 Lisp_Object Qcoding_system
, Qeol_type
;
275 Lisp_Object Qbuffer_file_coding_system
;
276 Lisp_Object Qpost_read_conversion
, Qpre_write_conversion
;
277 Lisp_Object Qno_conversion
, Qundecided
;
278 Lisp_Object Qcoding_system_history
;
279 Lisp_Object Qsafe_charsets
;
281 extern Lisp_Object Qinsert_file_contents
, Qwrite_region
;
282 Lisp_Object Qcall_process
, Qcall_process_region
, Qprocess_argument
;
283 Lisp_Object Qstart_process
, Qopen_network_stream
;
284 Lisp_Object Qtarget_idx
;
286 Lisp_Object Vselect_safe_coding_system_function
;
288 /* Mnemonic character of each format of end-of-line. */
289 int eol_mnemonic_unix
, eol_mnemonic_dos
, eol_mnemonic_mac
;
290 /* Mnemonic character to indicate format of end-of-line is not yet
292 int eol_mnemonic_undecided
;
294 /* Format of end-of-line decided by system. This is CODING_EOL_LF on
295 Unix, CODING_EOL_CRLF on DOS/Windows, and CODING_EOL_CR on Mac. */
300 Lisp_Object Vcoding_system_list
, Vcoding_system_alist
;
302 Lisp_Object Qcoding_system_p
, Qcoding_system_error
;
304 /* Coding system emacs-mule and raw-text are for converting only
305 end-of-line format. */
306 Lisp_Object Qemacs_mule
, Qraw_text
;
308 /* Coding-systems are handed between Emacs Lisp programs and C internal
309 routines by the following three variables. */
310 /* Coding-system for reading files and receiving data from process. */
311 Lisp_Object Vcoding_system_for_read
;
312 /* Coding-system for writing files and sending data to process. */
313 Lisp_Object Vcoding_system_for_write
;
314 /* Coding-system actually used in the latest I/O. */
315 Lisp_Object Vlast_coding_system_used
;
317 /* A vector of length 256 which contains information about special
318 Latin codes (especially for dealing with Microsoft codes). */
319 Lisp_Object Vlatin_extra_code_table
;
321 /* Flag to inhibit code conversion of end-of-line format. */
322 int inhibit_eol_conversion
;
324 /* Flag to make buffer-file-coding-system inherit from process-coding. */
325 int inherit_process_coding_system
;
327 /* Coding system to be used to encode text for terminal display. */
328 struct coding_system terminal_coding
;
330 /* Coding system to be used to encode text for terminal display when
331 terminal coding system is nil. */
332 struct coding_system safe_terminal_coding
;
334 /* Coding system of what is sent from terminal keyboard. */
335 struct coding_system keyboard_coding
;
337 Lisp_Object Vfile_coding_system_alist
;
338 Lisp_Object Vprocess_coding_system_alist
;
339 Lisp_Object Vnetwork_coding_system_alist
;
343 Lisp_Object Qcoding_category
, Qcoding_category_index
;
345 /* List of symbols `coding-category-xxx' ordered by priority. */
346 Lisp_Object Vcoding_category_list
;
348 /* Table of coding categories (Lisp symbols). */
349 Lisp_Object Vcoding_category_table
;
351 /* Table of names of symbol for each coding-category. */
352 char *coding_category_name
[CODING_CATEGORY_IDX_MAX
] = {
353 "coding-category-emacs-mule",
354 "coding-category-sjis",
355 "coding-category-iso-7",
356 "coding-category-iso-7-tight",
357 "coding-category-iso-8-1",
358 "coding-category-iso-8-2",
359 "coding-category-iso-7-else",
360 "coding-category-iso-8-else",
361 "coding-category-big5",
362 "coding-category-raw-text",
363 "coding-category-binary"
366 /* Table of pointers to coding systems corresponding to each coding
368 struct coding_system
*coding_system_table
[CODING_CATEGORY_IDX_MAX
];
370 /* Table of coding category masks. Nth element is a mask for a coding
371 cateogry of which priority is Nth. */
373 int coding_priorities
[CODING_CATEGORY_IDX_MAX
];
375 /* Flag to tell if we look up translation table on character code
377 Lisp_Object Venable_character_translation
;
378 /* Standard translation table to look up on decoding (reading). */
379 Lisp_Object Vstandard_translation_table_for_decode
;
380 /* Standard translation table to look up on encoding (writing). */
381 Lisp_Object Vstandard_translation_table_for_encode
;
383 Lisp_Object Qtranslation_table
;
384 Lisp_Object Qtranslation_table_id
;
385 Lisp_Object Qtranslation_table_for_decode
;
386 Lisp_Object Qtranslation_table_for_encode
;
388 /* Alist of charsets vs revision number. */
389 Lisp_Object Vcharset_revision_alist
;
391 /* Default coding systems used for process I/O. */
392 Lisp_Object Vdefault_process_coding_system
;
395 /*** 2. Emacs internal format (emacs-mule) handlers ***/
397 /* Emacs' internal format for encoding multiple character sets is a
398 kind of multi-byte encoding, i.e. characters are encoded by
399 variable-length sequences of one-byte codes. ASCII characters
400 and control characters (e.g. `tab', `newline') are represented by
401 one-byte sequences which are their ASCII codes, in the range 0x00
402 through 0x7F. The other characters are represented by a sequence
403 of `base leading-code', optional `extended leading-code', and one
404 or two `position-code's. The length of the sequence is determined
405 by the base leading-code. Leading-code takes the range 0x80
406 through 0x9F, whereas extended leading-code and position-code take
407 the range 0xA0 through 0xFF. See `charset.h' for more details
408 about leading-code and position-code.
410 There's one exception to this rule. Special leading-code
411 `leading-code-composition' denotes that the following several
412 characters should be composed into one character. Leading-codes of
413 components (except for ASCII) are added 0x20. An ASCII character
414 component is represented by a 2-byte sequence of `0xA0' and
415 `ASCII-code + 0x80'. See also the comments in `charset.h' for the
416 details of composite character. Hence, we can summarize the code
419 --- CODE RANGE of Emacs' internal format ---
420 (character set) (range)
422 ELSE (1st byte) 0x80 .. 0x9F
423 (rest bytes) 0xA0 .. 0xFF
424 ---------------------------------------------
428 enum emacs_code_class_type emacs_code_class
[256];
430 /* Go to the next statement only if *SRC is accessible and the code is
431 greater than 0xA0. */
432 #define CHECK_CODE_RANGE_A0_FF \
434 if (src >= src_end) \
435 goto label_end_of_switch; \
436 else if (*src++ < 0xA0) \
440 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
441 Check if a text is encoded in Emacs' internal format. If it is,
442 return CODING_CATEGORY_MASK_EMACS_MULE, else return 0. */
445 detect_coding_emacs_mule (src
, src_end
)
446 unsigned char *src
, *src_end
;
451 while (src
< src_end
)
463 switch (emacs_code_class
[c
])
465 case EMACS_ascii_code
:
466 case EMACS_linefeed_code
:
469 case EMACS_control_code
:
470 if (c
== ISO_CODE_ESC
|| c
== ISO_CODE_SI
|| c
== ISO_CODE_SO
)
474 case EMACS_invalid_code
:
477 case EMACS_leading_code_composition
: /* c == 0x80 */
479 CHECK_CODE_RANGE_A0_FF
;
484 case EMACS_leading_code_4
:
485 CHECK_CODE_RANGE_A0_FF
;
486 /* fall down to check it two more times ... */
488 case EMACS_leading_code_3
:
489 CHECK_CODE_RANGE_A0_FF
;
490 /* fall down to check it one more time ... */
492 case EMACS_leading_code_2
:
493 CHECK_CODE_RANGE_A0_FF
;
501 return CODING_CATEGORY_MASK_EMACS_MULE
;
505 /*** 3. ISO2022 handlers ***/
507 /* The following note describes the coding system ISO2022 briefly.
508 Since the intention of this note is to help in understanding of
509 the programs in this file, some parts are NOT ACCURATE or OVERLY
510 SIMPLIFIED. For the thorough understanding, please refer to the
511 original document of ISO2022.
513 ISO2022 provides many mechanisms to encode several character sets
514 in 7-bit and 8-bit environment. If one chooses 7-bite environment,
515 all text is encoded by codes of less than 128. This may make the
516 encoded text a little bit longer, but the text gets more stability
517 to pass through several gateways (some of them strip off the MSB).
519 There are two kinds of character set: control character set and
520 graphic character set. The former contains control characters such
521 as `newline' and `escape' to provide control functions (control
522 functions are provided also by escape sequences). The latter
523 contains graphic characters such as ' A' and '-'. Emacs recognizes
524 two control character sets and many graphic character sets.
526 Graphic character sets are classified into one of the following
527 four classes, DIMENSION1_CHARS94, DIMENSION1_CHARS96,
528 DIMENSION2_CHARS94, DIMENSION2_CHARS96 according to the number of
529 bytes (DIMENSION) and the number of characters in one dimension
530 (CHARS) of the set. In addition, each character set is assigned an
531 identification tag (called "final character" and denoted as <F>
532 here after) which is unique in each class. <F> of each character
533 set is decided by ECMA(*) when it is registered in ISO. Code range
534 of <F> is 0x30..0x7F (0x30..0x3F are for private use only).
536 Note (*): ECMA = European Computer Manufacturers Association
538 Here are examples of graphic character set [NAME(<F>)]:
539 o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ...
540 o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ...
541 o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ...
542 o DIMENSION2_CHARS96 -- none for the moment
544 A code area (1byte=8bits) is divided into 4 areas, C0, GL, C1, and GR.
545 C0 [0x00..0x1F] -- control character plane 0
546 GL [0x20..0x7F] -- graphic character plane 0
547 C1 [0x80..0x9F] -- control character plane 1
548 GR [0xA0..0xFF] -- graphic character plane 1
550 A control character set is directly designated and invoked to C0 or
551 C1 by an escape sequence. The most common case is that ISO646's
552 control character set is designated/invoked to C0 and ISO6429's
553 control character set is designated/invoked to C1, and usually
554 these designations/invocations are omitted in a coded text. With
555 7-bit environment, only C0 can be used, and a control character for
556 C1 is encoded by an appropriate escape sequence to fit in the
557 environment. All control characters for C1 are defined the
558 corresponding escape sequences.
560 A graphic character set is at first designated to one of four
561 graphic registers (G0 through G3), then these graphic registers are
562 invoked to GL or GR. These designations and invocations can be
563 done independently. The most common case is that G0 is invoked to
564 GL, G1 is invoked to GR, and ASCII is designated to G0, and usually
565 these invocations and designations are omitted in a coded text.
566 With 7-bit environment, only GL can be used.
568 When a graphic character set of CHARS94 is invoked to GL, code 0x20
569 and 0x7F of GL area work as control characters SPACE and DEL
570 respectively, and code 0xA0 and 0xFF of GR area should not be used.
572 There are two ways of invocation: locking-shift and single-shift.
573 With locking-shift, the invocation lasts until the next different
574 invocation, whereas with single-shift, the invocation works only
575 for the following character and doesn't affect locking-shift.
576 Invocations are done by the following control characters or escape
579 ----------------------------------------------------------------------
580 function control char escape sequence description
581 ----------------------------------------------------------------------
582 SI (shift-in) 0x0F none invoke G0 to GL
583 SO (shift-out) 0x0E none invoke G1 to GL
584 LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL
585 LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL
586 SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 into GL
587 SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 into GL
588 ----------------------------------------------------------------------
589 The first four are for locking-shift. Control characters for these
590 functions are defined by macros ISO_CODE_XXX in `coding.h'.
592 Designations are done by the following escape sequences.
593 ----------------------------------------------------------------------
594 escape sequence description
595 ----------------------------------------------------------------------
596 ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0
597 ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1
598 ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2
599 ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3
600 ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*)
601 ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1
602 ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2
603 ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3
604 ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**)
605 ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1
606 ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2
607 ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3
608 ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*)
609 ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1
610 ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2
611 ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3
612 ----------------------------------------------------------------------
614 In this list, "DIMENSION1_CHARS94<F>" means a graphic character set
615 of dimension 1, chars 94, and final character <F>, and etc.
617 Note (*): Although these designations are not allowed in ISO2022,
618 Emacs accepts them on decoding, and produces them on encoding
619 CHARS96 character set in a coding system which is characterized as
620 7-bit environment, non-locking-shift, and non-single-shift.
622 Note (**): If <F> is '@', 'A', or 'B', the intermediate character
623 '(' can be omitted. We call this as "short-form" here after.
625 Now you may notice that there are a lot of ways for encoding the
626 same multilingual text in ISO2022. Actually, there exists many
627 coding systems such as Compound Text (used in X's inter client
628 communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR
629 (used in Korean Internet), EUC (Extended UNIX Code, used in Asian
630 localized platforms), and all of these are variants of ISO2022.
632 In addition to the above, Emacs handles two more kinds of escape
633 sequences: ISO6429's direction specification and Emacs' private
634 sequence for specifying character composition.
636 ISO6429's direction specification takes the following format:
637 o CSI ']' -- end of the current direction
638 o CSI '0' ']' -- end of the current direction
639 o CSI '1' ']' -- start of left-to-right text
640 o CSI '2' ']' -- start of right-to-left text
641 The control character CSI (0x9B: control sequence introducer) is
642 abbreviated to the escape sequence ESC '[' in 7-bit environment.
644 Character composition specification takes the following format:
645 o ESC '0' -- start character composition
646 o ESC '1' -- end character composition
647 Since these are not standard escape sequences of any ISO, the use
648 of them for these meaning is restricted to Emacs only. */
650 enum iso_code_class_type iso_code_class
[256];
652 #define CHARSET_OK(idx, charset) \
653 (coding_system_table[idx]->safe_charsets[charset] \
654 || (CODING_SPEC_ISO_REQUESTED_DESIGNATION \
655 (coding_system_table[idx], charset) \
656 != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION))
658 #define SHIFT_OUT_OK(idx) \
659 (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding_system_table[idx], 1) >= 0)
661 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
662 Check if a text is encoded in ISO2022. If it is, returns an
663 integer in which appropriate flag bits any of:
664 CODING_CATEGORY_MASK_ISO_7
665 CODING_CATEGORY_MASK_ISO_7_TIGHT
666 CODING_CATEGORY_MASK_ISO_8_1
667 CODING_CATEGORY_MASK_ISO_8_2
668 CODING_CATEGORY_MASK_ISO_7_ELSE
669 CODING_CATEGORY_MASK_ISO_8_ELSE
670 are set. If a code which should never appear in ISO2022 is found,
674 detect_coding_iso2022 (src
, src_end
)
675 unsigned char *src
, *src_end
;
677 int mask
= CODING_CATEGORY_MASK_ISO
;
679 int reg
[4], shift_out
= 0;
680 int c
, c1
, i
, charset
;
682 reg
[0] = CHARSET_ASCII
, reg
[1] = reg
[2] = reg
[3] = -1;
683 while (mask
&& src
< src_end
)
692 if (c
>= '(' && c
<= '/')
694 /* Designation sequence for a charset of dimension 1. */
698 if (c1
< ' ' || c1
>= 0x80
699 || (charset
= iso_charset_table
[0][c
>= ','][c1
]) < 0)
700 /* Invalid designation sequence. Just ignore. */
702 reg
[(c
- '(') % 4] = charset
;
706 /* Designation sequence for a charset of dimension 2. */
710 if (c
>= '@' && c
<= 'B')
711 /* Designation for JISX0208.1978, GB2312, or JISX0208. */
712 reg
[0] = charset
= iso_charset_table
[1][0][c
];
713 else if (c
>= '(' && c
<= '/')
718 if (c1
< ' ' || c1
>= 0x80
719 || (charset
= iso_charset_table
[1][c
>= ','][c1
]) < 0)
720 /* Invalid designation sequence. Just ignore. */
722 reg
[(c
- '(') % 4] = charset
;
725 /* Invalid designation sequence. Just ignore. */
728 else if (c
== 'N' || c
== 'n')
732 || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE
)
733 || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE
)))
735 /* Locking shift out. */
736 mask
&= ~CODING_CATEGORY_MASK_ISO_7BIT
;
737 mask_found
|= CODING_CATEGORY_MASK_ISO_SHIFT
;
742 else if (c
== 'O' || c
== 'o')
746 /* Locking shift in. */
747 mask
&= ~CODING_CATEGORY_MASK_ISO_7BIT
;
748 mask_found
|= CODING_CATEGORY_MASK_ISO_SHIFT
;
753 else if (c
== '0' || c
== '1' || c
== '2')
754 /* Start/end composition. Just ignore. */
757 /* Invalid escape sequence. Just ignore. */
760 /* We found a valid designation sequence for CHARSET. */
761 mask
&= ~CODING_CATEGORY_MASK_ISO_8BIT
;
762 if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7
, charset
))
763 mask_found
|= CODING_CATEGORY_MASK_ISO_7
;
765 mask
&= ~CODING_CATEGORY_MASK_ISO_7
;
766 if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT
, charset
))
767 mask_found
|= CODING_CATEGORY_MASK_ISO_7_TIGHT
;
769 mask
&= ~CODING_CATEGORY_MASK_ISO_7_TIGHT
;
770 if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE
, charset
))
771 mask
&= ~CODING_CATEGORY_MASK_ISO_7_ELSE
;
772 if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE
, charset
))
773 mask
&= ~CODING_CATEGORY_MASK_ISO_8_ELSE
;
779 || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE
)
780 || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE
)))
782 /* Locking shift out. */
783 mask
&= ~CODING_CATEGORY_MASK_ISO_7BIT
;
784 mask_found
|= CODING_CATEGORY_MASK_ISO_SHIFT
;
791 /* Locking shift in. */
792 mask
&= ~CODING_CATEGORY_MASK_ISO_7BIT
;
793 mask_found
|= CODING_CATEGORY_MASK_ISO_SHIFT
;
801 int newmask
= CODING_CATEGORY_MASK_ISO_8_ELSE
;
803 if (c
!= ISO_CODE_CSI
)
805 if (coding_system_table
[CODING_CATEGORY_IDX_ISO_8_1
]->flags
806 & CODING_FLAG_ISO_SINGLE_SHIFT
)
807 newmask
|= CODING_CATEGORY_MASK_ISO_8_1
;
808 if (coding_system_table
[CODING_CATEGORY_IDX_ISO_8_2
]->flags
809 & CODING_FLAG_ISO_SINGLE_SHIFT
)
810 newmask
|= CODING_CATEGORY_MASK_ISO_8_2
;
812 if (VECTORP (Vlatin_extra_code_table
)
813 && !NILP (XVECTOR (Vlatin_extra_code_table
)->contents
[c
]))
815 if (coding_system_table
[CODING_CATEGORY_IDX_ISO_8_1
]->flags
816 & CODING_FLAG_ISO_LATIN_EXTRA
)
817 newmask
|= CODING_CATEGORY_MASK_ISO_8_1
;
818 if (coding_system_table
[CODING_CATEGORY_IDX_ISO_8_2
]->flags
819 & CODING_FLAG_ISO_LATIN_EXTRA
)
820 newmask
|= CODING_CATEGORY_MASK_ISO_8_2
;
823 mask_found
|= newmask
;
832 if (VECTORP (Vlatin_extra_code_table
)
833 && !NILP (XVECTOR (Vlatin_extra_code_table
)->contents
[c
]))
837 if (coding_system_table
[CODING_CATEGORY_IDX_ISO_8_1
]->flags
838 & CODING_FLAG_ISO_LATIN_EXTRA
)
839 newmask
|= CODING_CATEGORY_MASK_ISO_8_1
;
840 if (coding_system_table
[CODING_CATEGORY_IDX_ISO_8_2
]->flags
841 & CODING_FLAG_ISO_LATIN_EXTRA
)
842 newmask
|= CODING_CATEGORY_MASK_ISO_8_2
;
844 mask_found
|= newmask
;
851 unsigned char *src_begin
= src
;
853 mask
&= ~(CODING_CATEGORY_MASK_ISO_7BIT
854 | CODING_CATEGORY_MASK_ISO_7_ELSE
);
855 mask_found
|= CODING_CATEGORY_MASK_ISO_8_1
;
856 while (src
< src_end
&& *src
>= 0xA0)
858 if ((src
- src_begin
- 1) & 1 && src
< src_end
)
859 mask
&= ~CODING_CATEGORY_MASK_ISO_8_2
;
861 mask_found
|= CODING_CATEGORY_MASK_ISO_8_2
;
867 return (mask
& mask_found
);
870 /* Decode a character of which charset is CHARSET and the 1st position
871 code is C1. If dimension of CHARSET is 2, the 2nd position code is
872 fetched from SRC and set to C2. If CHARSET is negative, it means
873 that we are decoding ill formed text, and what we can do is just to
876 #define DECODE_ISO_CHARACTER(charset, c1) \
878 int c_alt, charset_alt = (charset); \
879 if (COMPOSING_HEAD_P (coding->composing)) \
881 *dst++ = LEADING_CODE_COMPOSITION; \
882 if (COMPOSING_WITH_RULE_P (coding->composing)) \
883 /* To tell composition rules are embeded. */ \
885 coding->composing += 2; \
887 if ((charset) >= 0) \
889 if (CHARSET_DIMENSION (charset) == 2) \
891 ONE_MORE_BYTE (c2); \
892 if (iso_code_class[(c2) & 0x7F] != ISO_0x20_or_0x7F \
893 && iso_code_class[(c2) & 0x7F] != ISO_graphic_plane_0) \
899 if (!NILP (translation_table) \
900 && ((c_alt = translate_char (translation_table, \
901 -1, (charset), c1, c2)) >= 0)) \
902 SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
904 if (charset_alt == CHARSET_ASCII || charset_alt < 0) \
905 DECODE_CHARACTER_ASCII (c1); \
906 else if (CHARSET_DIMENSION (charset_alt) == 1) \
907 DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \
909 DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \
910 if (COMPOSING_WITH_RULE_P (coding->composing)) \
911 /* To tell a composition rule follows. */ \
912 coding->composing = COMPOSING_WITH_RULE_RULE; \
915 /* Set designation state into CODING. */
916 #define DECODE_DESIGNATION(reg, dimension, chars, final_char) \
918 int charset = ISO_CHARSET_TABLE (make_number (dimension), \
919 make_number (chars), \
920 make_number (final_char)); \
922 && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) == reg \
923 || coding->safe_charsets[charset])) \
925 if (coding->spec.iso2022.last_invalid_designation_register == 0 \
927 && charset == CHARSET_ASCII) \
929 /* We should insert this designation sequence as is so \
930 that it is surely written back to a file. */ \
931 coding->spec.iso2022.last_invalid_designation_register = -1; \
932 goto label_invalid_code; \
934 coding->spec.iso2022.last_invalid_designation_register = -1; \
935 if ((coding->mode & CODING_MODE_DIRECTION) \
936 && CHARSET_REVERSE_CHARSET (charset) >= 0) \
937 charset = CHARSET_REVERSE_CHARSET (charset); \
938 CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \
942 coding->spec.iso2022.last_invalid_designation_register = reg; \
943 goto label_invalid_code; \
947 /* Check if the current composing sequence contains only valid codes.
948 If the composing sequence doesn't end before SRC_END, return -1.
949 Else, if it contains only valid codes, return 0.
950 Else return the length of the composing sequence. */
953 check_composing_code (coding
, src
, src_end
)
954 struct coding_system
*coding
;
955 unsigned char *src
, *src_end
;
957 unsigned char *src_start
= src
;
958 int invalid_code_found
= 0;
959 int charset
, c
, c1
, dim
;
961 while (src
< src_end
)
963 if (*src
++ != ISO_CODE_ESC
) continue;
964 if (src
>= src_end
) break;
965 if ((c
= *src
++) == '1') /* end of compsition */
966 return (invalid_code_found
? src
- src_start
: 0);
967 if (src
+ 2 >= src_end
) break;
968 if (!coding
->flags
& CODING_FLAG_ISO_DESIGNATION
)
969 invalid_code_found
= 1;
976 c
= (*src
>= '@' && *src
<= 'B') ? '(' : *src
++;
978 if (c
>= '(' && c
<= '/')
981 if ((c1
< ' ' || c1
>= 0x80)
982 || (charset
= iso_charset_table
[dim
][c
>= ','][c1
]) < 0
983 || ! coding
->safe_charsets
[charset
]
984 || (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
)
985 == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION
))
986 invalid_code_found
= 1;
989 invalid_code_found
= 1;
992 return (invalid_code_found
994 : (coding
->mode
& CODING_MODE_LAST_BLOCK
? 0 : -1));
997 /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
1000 decode_coding_iso2022 (coding
, source
, destination
, src_bytes
, dst_bytes
)
1001 struct coding_system
*coding
;
1002 unsigned char *source
, *destination
;
1003 int src_bytes
, dst_bytes
;
1005 unsigned char *src
= source
;
1006 unsigned char *src_end
= source
+ src_bytes
;
1007 unsigned char *dst
= destination
;
1008 unsigned char *dst_end
= destination
+ dst_bytes
;
1009 /* Since the maximum bytes produced by each loop is 7, we subtract 6
1010 from DST_END to assure that overflow checking is necessary only
1011 at the head of loop. */
1012 unsigned char *adjusted_dst_end
= dst_end
- 6;
1014 /* Charsets invoked to graphic plane 0 and 1 respectively. */
1015 int charset0
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 0);
1016 int charset1
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 1);
1017 Lisp_Object translation_table
1018 = coding
->translation_table_for_decode
;
1019 int result
= CODING_FINISH_NORMAL
;
1021 if (!NILP (Venable_character_translation
) && NILP (translation_table
))
1022 translation_table
= Vstandard_translation_table_for_decode
;
1024 coding
->produced_char
= 0;
1025 coding
->fake_multibyte
= 0;
1026 while (src
< src_end
&& (dst_bytes
1027 ? (dst
< adjusted_dst_end
)
1030 /* SRC_BASE remembers the start position in source in each loop.
1031 The loop will be exited when there's not enough source text
1032 to analyze long escape sequence or 2-byte code (within macros
1033 ONE_MORE_BYTE or TWO_MORE_BYTES). In that case, SRC is reset
1034 to SRC_BASE before exiting. */
1035 unsigned char *src_base
= src
;
1036 int c1
= *src
++, c2
;
1038 switch (iso_code_class
[c1
])
1040 case ISO_0x20_or_0x7F
:
1041 if (!coding
->composing
1042 && (charset0
< 0 || CHARSET_CHARS (charset0
) == 94))
1044 /* This is SPACE or DEL. */
1046 coding
->produced_char
++;
1049 /* This is a graphic character, we fall down ... */
1051 case ISO_graphic_plane_0
:
1052 if (coding
->composing
== COMPOSING_WITH_RULE_RULE
)
1054 /* This is a composition rule. */
1056 coding
->composing
= COMPOSING_WITH_RULE_TAIL
;
1059 DECODE_ISO_CHARACTER (charset0
, c1
);
1062 case ISO_0xA0_or_0xFF
:
1063 if (charset1
< 0 || CHARSET_CHARS (charset1
) == 94
1064 || coding
->flags
& CODING_FLAG_ISO_SEVEN_BITS
)
1065 goto label_invalid_code
;
1066 /* This is a graphic character, we fall down ... */
1068 case ISO_graphic_plane_1
:
1069 if (coding
->flags
& CODING_FLAG_ISO_SEVEN_BITS
)
1070 goto label_invalid_code
;
1072 DECODE_ISO_CHARACTER (charset1
, c1
);
1075 case ISO_control_code
:
1076 /* All ISO2022 control characters in this class have the
1077 same representation in Emacs internal format. */
1079 && (coding
->mode
& CODING_MODE_INHIBIT_INCONSISTENT_EOL
)
1080 && (coding
->eol_type
== CODING_EOL_CR
1081 || coding
->eol_type
== CODING_EOL_CRLF
))
1083 result
= CODING_FINISH_INCONSISTENT_EOL
;
1084 goto label_end_of_loop_2
;
1087 coding
->produced_char
++;
1090 case ISO_carriage_return
:
1091 if (coding
->eol_type
== CODING_EOL_CR
)
1093 else if (coding
->eol_type
== CODING_EOL_CRLF
)
1096 if (c1
== ISO_CODE_LF
)
1100 if (coding
->mode
& CODING_MODE_INHIBIT_INCONSISTENT_EOL
)
1102 result
= CODING_FINISH_INCONSISTENT_EOL
;
1103 goto label_end_of_loop_2
;
1111 coding
->produced_char
++;
1115 if (! (coding
->flags
& CODING_FLAG_ISO_LOCKING_SHIFT
)
1116 || CODING_SPEC_ISO_DESIGNATION (coding
, 1) < 0)
1117 goto label_invalid_code
;
1118 CODING_SPEC_ISO_INVOCATION (coding
, 0) = 1;
1119 charset0
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 0);
1123 if (! (coding
->flags
& CODING_FLAG_ISO_LOCKING_SHIFT
))
1124 goto label_invalid_code
;
1125 CODING_SPEC_ISO_INVOCATION (coding
, 0) = 0;
1126 charset0
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 0);
1129 case ISO_single_shift_2_7
:
1130 case ISO_single_shift_2
:
1131 if (! (coding
->flags
& CODING_FLAG_ISO_SINGLE_SHIFT
))
1132 goto label_invalid_code
;
1133 /* SS2 is handled as an escape sequence of ESC 'N' */
1135 goto label_escape_sequence
;
1137 case ISO_single_shift_3
:
1138 if (! (coding
->flags
& CODING_FLAG_ISO_SINGLE_SHIFT
))
1139 goto label_invalid_code
;
1140 /* SS2 is handled as an escape sequence of ESC 'O' */
1142 goto label_escape_sequence
;
1144 case ISO_control_sequence_introducer
:
1145 /* CSI is handled as an escape sequence of ESC '[' ... */
1147 goto label_escape_sequence
;
1151 label_escape_sequence
:
1152 /* Escape sequences handled by Emacs are invocation,
1153 designation, direction specification, and character
1154 composition specification. */
1157 case '&': /* revision of following character set */
1159 if (!(c1
>= '@' && c1
<= '~'))
1160 goto label_invalid_code
;
1162 if (c1
!= ISO_CODE_ESC
)
1163 goto label_invalid_code
;
1165 goto label_escape_sequence
;
1167 case '$': /* designation of 2-byte character set */
1168 if (! (coding
->flags
& CODING_FLAG_ISO_DESIGNATION
))
1169 goto label_invalid_code
;
1171 if (c1
>= '@' && c1
<= 'B')
1172 { /* designation of JISX0208.1978, GB2312.1980,
1174 DECODE_DESIGNATION (0, 2, 94, c1
);
1176 else if (c1
>= 0x28 && c1
<= 0x2B)
1177 { /* designation of DIMENSION2_CHARS94 character set */
1179 DECODE_DESIGNATION (c1
- 0x28, 2, 94, c2
);
1181 else if (c1
>= 0x2C && c1
<= 0x2F)
1182 { /* designation of DIMENSION2_CHARS96 character set */
1184 DECODE_DESIGNATION (c1
- 0x2C, 2, 96, c2
);
1187 goto label_invalid_code
;
1190 case 'n': /* invocation of locking-shift-2 */
1191 if (! (coding
->flags
& CODING_FLAG_ISO_LOCKING_SHIFT
)
1192 || CODING_SPEC_ISO_DESIGNATION (coding
, 2) < 0)
1193 goto label_invalid_code
;
1194 CODING_SPEC_ISO_INVOCATION (coding
, 0) = 2;
1195 charset0
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 0);
1198 case 'o': /* invocation of locking-shift-3 */
1199 if (! (coding
->flags
& CODING_FLAG_ISO_LOCKING_SHIFT
)
1200 || CODING_SPEC_ISO_DESIGNATION (coding
, 3) < 0)
1201 goto label_invalid_code
;
1202 CODING_SPEC_ISO_INVOCATION (coding
, 0) = 3;
1203 charset0
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 0);
1206 case 'N': /* invocation of single-shift-2 */
1207 if (! (coding
->flags
& CODING_FLAG_ISO_SINGLE_SHIFT
)
1208 || CODING_SPEC_ISO_DESIGNATION (coding
, 2) < 0)
1209 goto label_invalid_code
;
1211 charset
= CODING_SPEC_ISO_DESIGNATION (coding
, 2);
1212 DECODE_ISO_CHARACTER (charset
, c1
);
1215 case 'O': /* invocation of single-shift-3 */
1216 if (! (coding
->flags
& CODING_FLAG_ISO_SINGLE_SHIFT
)
1217 || CODING_SPEC_ISO_DESIGNATION (coding
, 3) < 0)
1218 goto label_invalid_code
;
1220 charset
= CODING_SPEC_ISO_DESIGNATION (coding
, 3);
1221 DECODE_ISO_CHARACTER (charset
, c1
);
1224 case '0': case '2': /* start composing */
1225 /* Before processing composing, we must be sure that all
1226 characters being composed are supported by CODING.
1227 If not, we must give up composing and insert the
1228 bunch of codes for composing as is without decoding. */
1232 result1
= check_composing_code (coding
, src
, src_end
);
1235 coding
->composing
= (c1
== '0'
1236 ? COMPOSING_NO_RULE_HEAD
1237 : COMPOSING_WITH_RULE_HEAD
);
1238 coding
->produced_char
++;
1240 else if (result1
> 0)
1242 if (result1
+ 2 < (dst_bytes
? dst_end
: src_base
) - dst
)
1244 bcopy (src_base
, dst
, result1
+ 2);
1247 coding
->produced_char
+= result1
+ 2;
1251 result
= CODING_FINISH_INSUFFICIENT_DST
;
1252 goto label_end_of_loop_2
;
1256 goto label_end_of_loop
;
1260 case '1': /* end composing */
1261 coding
->composing
= COMPOSING_NO
;
1264 case '[': /* specification of direction */
1265 if (coding
->flags
& CODING_FLAG_ISO_NO_DIRECTION
)
1266 goto label_invalid_code
;
1267 /* For the moment, nested direction is not supported.
1268 So, `coding->mode & CODING_MODE_DIRECTION' zero means
1269 left-to-right, and nozero means right-to-left. */
1273 case ']': /* end of the current direction */
1274 coding
->mode
&= ~CODING_MODE_DIRECTION
;
1276 case '0': /* end of the current direction */
1277 case '1': /* start of left-to-right direction */
1280 coding
->mode
&= ~CODING_MODE_DIRECTION
;
1282 goto label_invalid_code
;
1285 case '2': /* start of right-to-left direction */
1288 coding
->mode
|= CODING_MODE_DIRECTION
;
1290 goto label_invalid_code
;
1294 goto label_invalid_code
;
1299 if (! (coding
->flags
& CODING_FLAG_ISO_DESIGNATION
))
1300 goto label_invalid_code
;
1301 if (c1
>= 0x28 && c1
<= 0x2B)
1302 { /* designation of DIMENSION1_CHARS94 character set */
1304 DECODE_DESIGNATION (c1
- 0x28, 1, 94, c2
);
1306 else if (c1
>= 0x2C && c1
<= 0x2F)
1307 { /* designation of DIMENSION1_CHARS96 character set */
1309 DECODE_DESIGNATION (c1
- 0x2C, 1, 96, c2
);
1313 goto label_invalid_code
;
1316 /* We must update these variables now. */
1317 charset0
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 0);
1318 charset1
= CODING_SPEC_ISO_PLANE_CHARSET (coding
, 1);
1322 while (src_base
< src
)
1323 *dst
++ = *src_base
++;
1324 coding
->fake_multibyte
= 1;
1329 result
= CODING_FINISH_INSUFFICIENT_SRC
;
1330 label_end_of_loop_2
:
1337 if (result
== CODING_FINISH_NORMAL
)
1338 result
= CODING_FINISH_INSUFFICIENT_DST
;
1339 else if (result
!= CODING_FINISH_INCONSISTENT_EOL
1340 && coding
->mode
& CODING_MODE_LAST_BLOCK
)
1342 /* This is the last block of the text to be decoded. We had
1343 better just flush out all remaining codes in the text
1344 although they are not valid characters. */
1345 src_bytes
= src_end
- src
;
1346 if (dst_bytes
&& (dst_end
- dst
< src_bytes
))
1347 src_bytes
= dst_end
- dst
;
1348 bcopy (src
, dst
, src_bytes
);
1351 coding
->fake_multibyte
= 1;
1355 coding
->consumed
= coding
->consumed_char
= src
- source
;
1356 coding
->produced
= dst
- destination
;
1360 /* ISO2022 encoding stuff. */
1363 It is not enough to say just "ISO2022" on encoding, we have to
1364 specify more details. In Emacs, each coding system of ISO2022
1365 variant has the following specifications:
1366 1. Initial designation to G0 thru G3.
1367 2. Allows short-form designation?
1368 3. ASCII should be designated to G0 before control characters?
1369 4. ASCII should be designated to G0 at end of line?
1370 5. 7-bit environment or 8-bit environment?
1371 6. Use locking-shift?
1372 7. Use Single-shift?
1373 And the following two are only for Japanese:
1374 8. Use ASCII in place of JIS0201-1976-Roman?
1375 9. Use JISX0208-1983 in place of JISX0208-1978?
1376 These specifications are encoded in `coding->flags' as flag bits
1377 defined by macros CODING_FLAG_ISO_XXX. See `coding.h' for more
1381 /* Produce codes (escape sequence) for designating CHARSET to graphic
1382 register REG. If <final-char> of CHARSET is '@', 'A', or 'B' and
1383 the coding system CODING allows, produce designation sequence of
1386 #define ENCODE_DESIGNATION(charset, reg, coding) \
1388 unsigned char final_char = CHARSET_ISO_FINAL_CHAR (charset); \
1389 char *intermediate_char_94 = "()*+"; \
1390 char *intermediate_char_96 = ",-./"; \
1391 int revision = CODING_SPEC_ISO_REVISION_NUMBER(coding, charset); \
1392 if (revision < 255) \
1394 *dst++ = ISO_CODE_ESC; \
1396 *dst++ = '@' + revision; \
1398 *dst++ = ISO_CODE_ESC; \
1399 if (CHARSET_DIMENSION (charset) == 1) \
1401 if (CHARSET_CHARS (charset) == 94) \
1402 *dst++ = (unsigned char) (intermediate_char_94[reg]); \
1404 *dst++ = (unsigned char) (intermediate_char_96[reg]); \
1409 if (CHARSET_CHARS (charset) == 94) \
1411 if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \
1413 || final_char < '@' || final_char > 'B') \
1414 *dst++ = (unsigned char) (intermediate_char_94[reg]); \
1417 *dst++ = (unsigned char) (intermediate_char_96[reg]); \
1419 *dst++ = final_char; \
1420 CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \
1423 /* The following two macros produce codes (control character or escape
1424 sequence) for ISO2022 single-shift functions (single-shift-2 and
1427 #define ENCODE_SINGLE_SHIFT_2 \
1429 if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \
1430 *dst++ = ISO_CODE_ESC, *dst++ = 'N'; \
1433 *dst++ = ISO_CODE_SS2; \
1434 coding->fake_multibyte = 1; \
1436 CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \
1439 #define ENCODE_SINGLE_SHIFT_3 \
1441 if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \
1442 *dst++ = ISO_CODE_ESC, *dst++ = 'O'; \
1445 *dst++ = ISO_CODE_SS3; \
1446 coding->fake_multibyte = 1; \
1448 CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \
1451 /* The following four macros produce codes (control character or
1452 escape sequence) for ISO2022 locking-shift functions (shift-in,
1453 shift-out, locking-shift-2, and locking-shift-3). */
1455 #define ENCODE_SHIFT_IN \
1457 *dst++ = ISO_CODE_SI; \
1458 CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; \
1461 #define ENCODE_SHIFT_OUT \
1463 *dst++ = ISO_CODE_SO; \
1464 CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; \
1467 #define ENCODE_LOCKING_SHIFT_2 \
1469 *dst++ = ISO_CODE_ESC, *dst++ = 'n'; \
1470 CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; \
1473 #define ENCODE_LOCKING_SHIFT_3 \
1475 *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \
1476 CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; \
1479 /* Produce codes for a DIMENSION1 character whose character set is
1480 CHARSET and whose position-code is C1. Designation and invocation
1481 sequences are also produced in advance if necessary. */
1484 #define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \
1486 if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \
1488 if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \
1489 *dst++ = c1 & 0x7F; \
1491 *dst++ = c1 | 0x80; \
1492 CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \
1495 else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \
1497 *dst++ = c1 & 0x7F; \
1500 else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \
1502 *dst++ = c1 | 0x80; \
1505 else if (coding->flags & CODING_FLAG_ISO_SAFE \
1506 && !coding->safe_charsets[charset]) \
1508 /* We should not encode this character, instead produce one or \
1510 *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
1511 if (CHARSET_WIDTH (charset) == 2) \
1512 *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
1516 /* Since CHARSET is not yet invoked to any graphic planes, we \
1517 must invoke it, or, at first, designate it to some graphic \
1518 register. Then repeat the loop to actually produce the \
1520 dst = encode_invocation_designation (charset, coding, dst); \
1523 /* Produce codes for a DIMENSION2 character whose character set is
1524 CHARSET and whose position-codes are C1 and C2. Designation and
1525 invocation codes are also produced in advance if necessary. */
1527 #define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2) \
1529 if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \
1531 if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \
1532 *dst++ = c1 & 0x7F, *dst++ = c2 & 0x7F; \
1534 *dst++ = c1 | 0x80, *dst++ = c2 | 0x80; \
1535 CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \
1538 else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \
1540 *dst++ = c1 & 0x7F, *dst++= c2 & 0x7F; \
1543 else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \
1545 *dst++ = c1 | 0x80, *dst++= c2 | 0x80; \
1548 else if (coding->flags & CODING_FLAG_ISO_SAFE \
1549 && !coding->safe_charsets[charset]) \
1551 /* We should not encode this character, instead produce one or \
1553 *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
1554 if (CHARSET_WIDTH (charset) == 2) \
1555 *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
1559 /* Since CHARSET is not yet invoked to any graphic planes, we \
1560 must invoke it, or, at first, designate it to some graphic \
1561 register. Then repeat the loop to actually produce the \
1563 dst = encode_invocation_designation (charset, coding, dst); \
1566 #define ENCODE_ISO_CHARACTER(charset, c1, c2) \
1568 int c_alt, charset_alt; \
1569 if (!NILP (translation_table) \
1570 && ((c_alt = translate_char (translation_table, -1, \
1573 SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
1575 charset_alt = charset; \
1576 if (CHARSET_DIMENSION (charset_alt) == 1) \
1578 if (charset == CHARSET_ASCII \
1579 && coding->flags & CODING_FLAG_ISO_USE_ROMAN) \
1580 charset_alt = charset_latin_jisx0201; \
1581 ENCODE_ISO_CHARACTER_DIMENSION1 (charset_alt, c1); \
1585 if (charset == charset_jisx0208 \
1586 && coding->flags & CODING_FLAG_ISO_USE_OLDJIS) \
1587 charset_alt = charset_jisx0208_1978; \
1588 ENCODE_ISO_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \
1590 if (! COMPOSING_P (coding->composing)) \
1591 coding->consumed_char++; \
1594 /* Produce designation and invocation codes at a place pointed by DST
1595 to use CHARSET. The element `spec.iso2022' of *CODING is updated.
1599 encode_invocation_designation (charset
, coding
, dst
)
1601 struct coding_system
*coding
;
1604 int reg
; /* graphic register number */
1606 /* At first, check designations. */
1607 for (reg
= 0; reg
< 4; reg
++)
1608 if (charset
== CODING_SPEC_ISO_DESIGNATION (coding
, reg
))
1613 /* CHARSET is not yet designated to any graphic registers. */
1614 /* At first check the requested designation. */
1615 reg
= CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
);
1616 if (reg
== CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION
)
1617 /* Since CHARSET requests no special designation, designate it
1618 to graphic register 0. */
1621 ENCODE_DESIGNATION (charset
, reg
, coding
);
1624 if (CODING_SPEC_ISO_INVOCATION (coding
, 0) != reg
1625 && CODING_SPEC_ISO_INVOCATION (coding
, 1) != reg
)
1627 /* Since the graphic register REG is not invoked to any graphic
1628 planes, invoke it to graphic plane 0. */
1631 case 0: /* graphic register 0 */
1635 case 1: /* graphic register 1 */
1639 case 2: /* graphic register 2 */
1640 if (coding
->flags
& CODING_FLAG_ISO_SINGLE_SHIFT
)
1641 ENCODE_SINGLE_SHIFT_2
;
1643 ENCODE_LOCKING_SHIFT_2
;
1646 case 3: /* graphic register 3 */
1647 if (coding
->flags
& CODING_FLAG_ISO_SINGLE_SHIFT
)
1648 ENCODE_SINGLE_SHIFT_3
;
1650 ENCODE_LOCKING_SHIFT_3
;
1657 /* The following two macros produce codes for indicating composition. */
1658 #define ENCODE_COMPOSITION_NO_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '0'
1659 #define ENCODE_COMPOSITION_WITH_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '2'
1660 #define ENCODE_COMPOSITION_END *dst++ = ISO_CODE_ESC, *dst++ = '1'
1662 /* The following three macros produce codes for indicating direction
1664 #define ENCODE_CONTROL_SEQUENCE_INTRODUCER \
1666 if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS) \
1667 *dst++ = ISO_CODE_ESC, *dst++ = '['; \
1669 *dst++ = ISO_CODE_CSI; \
1672 #define ENCODE_DIRECTION_R2L \
1673 ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '2', *dst++ = ']'
1675 #define ENCODE_DIRECTION_L2R \
1676 ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '0', *dst++ = ']'
1678 /* Produce codes for designation and invocation to reset the graphic
1679 planes and registers to initial state. */
1680 #define ENCODE_RESET_PLANE_AND_REGISTER \
1683 if (CODING_SPEC_ISO_INVOCATION (coding, 0) != 0) \
1685 for (reg = 0; reg < 4; reg++) \
1686 if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg) >= 0 \
1687 && (CODING_SPEC_ISO_DESIGNATION (coding, reg) \
1688 != CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg))) \
1689 ENCODE_DESIGNATION \
1690 (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg), reg, coding); \
1693 /* Produce designation sequences of charsets in the line started from
1694 SRC to a place pointed by *DSTP, and update DSTP.
1696 If the current block ends before any end-of-line, we may fail to
1697 find all the necessary designations. */
1700 encode_designation_at_bol (coding
, table
, src
, src_end
, dstp
)
1701 struct coding_system
*coding
;
1703 unsigned char *src
, *src_end
, **dstp
;
1705 int charset
, c
, found
= 0, reg
;
1706 /* Table of charsets to be designated to each graphic register. */
1708 unsigned char *dst
= *dstp
;
1710 for (reg
= 0; reg
< 4; reg
++)
1713 while (src
< src_end
&& *src
!= '\n' && found
< 4)
1715 int bytes
= BYTES_BY_CHAR_HEAD (*src
);
1718 charset
= CHARSET_AT (src
);
1722 unsigned char c1
, c2
;
1724 SPLIT_STRING(src
, bytes
, charset
, c1
, c2
);
1725 if ((c_alt
= translate_char (table
, -1, charset
, c1
, c2
)) >= 0)
1726 charset
= CHAR_CHARSET (c_alt
);
1729 reg
= CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
);
1730 if (reg
!= CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION
&& r
[reg
] < 0)
1741 for (reg
= 0; reg
< 4; reg
++)
1743 && CODING_SPEC_ISO_DESIGNATION (coding
, reg
) != r
[reg
])
1744 ENCODE_DESIGNATION (r
[reg
], reg
, coding
);
1749 /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */
1752 encode_coding_iso2022 (coding
, source
, destination
, src_bytes
, dst_bytes
)
1753 struct coding_system
*coding
;
1754 unsigned char *source
, *destination
;
1755 int src_bytes
, dst_bytes
;
1757 unsigned char *src
= source
;
1758 unsigned char *src_end
= source
+ src_bytes
;
1759 unsigned char *dst
= destination
;
1760 unsigned char *dst_end
= destination
+ dst_bytes
;
1761 /* Since the maximum bytes produced by each loop is 20, we subtract 19
1762 from DST_END to assure overflow checking is necessary only at the
1764 unsigned char *adjusted_dst_end
= dst_end
- 19;
1765 Lisp_Object translation_table
1766 = coding
->translation_table_for_encode
;
1767 int result
= CODING_FINISH_NORMAL
;
1769 if (!NILP (Venable_character_translation
) && NILP (translation_table
))
1770 translation_table
= Vstandard_translation_table_for_encode
;
1772 coding
->consumed_char
= 0;
1773 coding
->fake_multibyte
= 0;
1774 while (src
< src_end
&& (dst_bytes
1775 ? (dst
< adjusted_dst_end
)
1776 : (dst
< src
- 19)))
1778 /* SRC_BASE remembers the start position in source in each loop.
1779 The loop will be exited when there's not enough source text
1780 to analyze multi-byte codes (within macros ONE_MORE_BYTE,
1781 TWO_MORE_BYTES, and THREE_MORE_BYTES). In that case, SRC is
1782 reset to SRC_BASE before exiting. */
1783 unsigned char *src_base
= src
;
1784 int charset
, c1
, c2
, c3
, c4
;
1786 if (coding
->flags
& CODING_FLAG_ISO_DESIGNATE_AT_BOL
1787 && CODING_SPEC_ISO_BOL (coding
))
1789 /* We have to produce designation sequences if any now. */
1790 encode_designation_at_bol (coding
, translation_table
,
1791 src
, src_end
, &dst
);
1792 CODING_SPEC_ISO_BOL (coding
) = 0;
1796 /* If we are seeing a component of a composite character, we are
1797 seeing a leading-code encoded irregularly for composition, or
1798 a composition rule if composing with rule. We must set C1 to
1799 a normal leading-code or an ASCII code. If we are not seeing
1800 a composite character, we must reset composition,
1801 designation, and invocation states. */
1802 if (COMPOSING_P (coding
->composing
))
1806 /* We are not in a composite character any longer. */
1807 coding
->composing
= COMPOSING_NO
;
1808 ENCODE_RESET_PLANE_AND_REGISTER
;
1809 ENCODE_COMPOSITION_END
;
1813 if (coding
->composing
== COMPOSING_WITH_RULE_RULE
)
1816 coding
->composing
= COMPOSING_WITH_RULE_HEAD
;
1819 else if (coding
->composing
== COMPOSING_WITH_RULE_HEAD
)
1820 coding
->composing
= COMPOSING_WITH_RULE_RULE
;
1823 /* This is an ASCII component. */
1828 /* This is a leading-code of non ASCII component. */
1833 /* Now encode one character. C1 is a control character, an
1834 ASCII character, or a leading-code of multi-byte character. */
1835 switch (emacs_code_class
[c1
])
1837 case EMACS_ascii_code
:
1838 ENCODE_ISO_CHARACTER (CHARSET_ASCII
, c1
, /* dummy */ c2
);
1841 case EMACS_control_code
:
1842 if (coding
->flags
& CODING_FLAG_ISO_RESET_AT_CNTL
)
1843 ENCODE_RESET_PLANE_AND_REGISTER
;
1845 coding
->consumed_char
++;
1848 case EMACS_carriage_return_code
:
1849 if (! (coding
->mode
& CODING_MODE_SELECTIVE_DISPLAY
))
1851 if (coding
->flags
& CODING_FLAG_ISO_RESET_AT_CNTL
)
1852 ENCODE_RESET_PLANE_AND_REGISTER
;
1854 coding
->consumed_char
++;
1857 /* fall down to treat '\r' as '\n' ... */
1859 case EMACS_linefeed_code
:
1860 if (coding
->flags
& CODING_FLAG_ISO_RESET_AT_EOL
)
1861 ENCODE_RESET_PLANE_AND_REGISTER
;
1862 if (coding
->flags
& CODING_FLAG_ISO_INIT_AT_BOL
)
1863 bcopy (coding
->spec
.iso2022
.initial_designation
,
1864 coding
->spec
.iso2022
.current_designation
,
1865 sizeof coding
->spec
.iso2022
.initial_designation
);
1866 if (coding
->eol_type
== CODING_EOL_LF
1867 || coding
->eol_type
== CODING_EOL_UNDECIDED
)
1868 *dst
++ = ISO_CODE_LF
;
1869 else if (coding
->eol_type
== CODING_EOL_CRLF
)
1870 *dst
++ = ISO_CODE_CR
, *dst
++ = ISO_CODE_LF
;
1872 *dst
++ = ISO_CODE_CR
;
1873 CODING_SPEC_ISO_BOL (coding
) = 1;
1874 coding
->consumed_char
++;
1877 case EMACS_leading_code_2
:
1881 /* invalid sequence */
1884 coding
->consumed_char
++;
1887 ENCODE_ISO_CHARACTER (c1
, c2
, /* dummy */ c3
);
1890 case EMACS_leading_code_3
:
1891 TWO_MORE_BYTES (c2
, c3
);
1892 if (c2
< 0xA0 || c3
< 0xA0)
1894 /* invalid sequence */
1897 coding
->consumed_char
++;
1899 else if (c1
< LEADING_CODE_PRIVATE_11
)
1900 ENCODE_ISO_CHARACTER (c1
, c2
, c3
);
1902 ENCODE_ISO_CHARACTER (c2
, c3
, /* dummy */ c4
);
1905 case EMACS_leading_code_4
:
1906 THREE_MORE_BYTES (c2
, c3
, c4
);
1907 if (c2
< 0xA0 || c3
< 0xA0 || c4
< 0xA0)
1909 /* invalid sequence */
1912 coding
->consumed_char
++;
1915 ENCODE_ISO_CHARACTER (c2
, c3
, c4
);
1918 case EMACS_leading_code_composition
:
1922 /* invalid sequence */
1925 coding
->consumed_char
++;
1927 else if (c2
== 0xFF)
1929 ENCODE_RESET_PLANE_AND_REGISTER
;
1930 coding
->composing
= COMPOSING_WITH_RULE_HEAD
;
1931 ENCODE_COMPOSITION_WITH_RULE_START
;
1932 coding
->consumed_char
++;
1936 ENCODE_RESET_PLANE_AND_REGISTER
;
1937 /* Rewind one byte because it is a character code of
1938 composition elements. */
1940 coding
->composing
= COMPOSING_NO_RULE_HEAD
;
1941 ENCODE_COMPOSITION_NO_RULE_START
;
1942 coding
->consumed_char
++;
1946 case EMACS_invalid_code
:
1948 coding
->consumed_char
++;
1953 result
= CODING_FINISH_INSUFFICIENT_SRC
;
1958 if (src
< src_end
&& result
== CODING_FINISH_NORMAL
)
1959 result
= CODING_FINISH_INSUFFICIENT_DST
;
1961 /* If this is the last block of the text to be encoded, we must
1962 reset graphic planes and registers to the initial state, and
1963 flush out the carryover if any. */
1964 if (coding
->mode
& CODING_MODE_LAST_BLOCK
)
1966 ENCODE_RESET_PLANE_AND_REGISTER
;
1967 if (COMPOSING_P (coding
->composing
))
1968 ENCODE_COMPOSITION_END
;
1970 coding
->consumed
= src
- source
;
1971 coding
->produced
= coding
->produced_char
= dst
- destination
;
1976 /*** 4. SJIS and BIG5 handlers ***/
1978 /* Although SJIS and BIG5 are not ISO's coding system, they are used
1979 quite widely. So, for the moment, Emacs supports them in the bare
1980 C code. But, in the future, they may be supported only by CCL. */
1982 /* SJIS is a coding system encoding three character sets: ASCII, right
1983 half of JISX0201-Kana, and JISX0208. An ASCII character is encoded
1984 as is. A character of charset katakana-jisx0201 is encoded by
1985 "position-code + 0x80". A character of charset japanese-jisx0208
1986 is encoded in 2-byte but two position-codes are divided and shifted
1987 so that it fit in the range below.
1989 --- CODE RANGE of SJIS ---
1990 (character set) (range)
1992 KATAKANA-JISX0201 0xA0 .. 0xDF
1993 JISX0208 (1st byte) 0x80 .. 0x9F and 0xE0 .. 0xFF
1994 (2nd byte) 0x40 .. 0xFF
1995 -------------------------------
1999 /* BIG5 is a coding system encoding two character sets: ASCII and
2000 Big5. An ASCII character is encoded as is. Big5 is a two-byte
2001 character set and is encoded in two-byte.
2003 --- CODE RANGE of BIG5 ---
2004 (character set) (range)
2006 Big5 (1st byte) 0xA1 .. 0xFE
2007 (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE
2008 --------------------------
2010 Since the number of characters in Big5 is larger than maximum
2011 characters in Emacs' charset (96x96), it can't be handled as one
2012 charset. So, in Emacs, Big5 is divided into two: `charset-big5-1'
2013 and `charset-big5-2'. Both are DIMENSION2 and CHARS94. The former
2014 contains frequently used characters and the latter contains less
2015 frequently used characters. */
2017 /* Macros to decode or encode a character of Big5 in BIG5. B1 and B2
2018 are the 1st and 2nd position-codes of Big5 in BIG5 coding system.
2019 C1 and C2 are the 1st and 2nd position-codes of of Emacs' internal
2020 format. CHARSET is `charset_big5_1' or `charset_big5_2'. */
2022 /* Number of Big5 characters which have the same code in 1st byte. */
2023 #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40)
2025 #define DECODE_BIG5(b1, b2, charset, c1, c2) \
2028 = (b1 - 0xA1) * BIG5_SAME_ROW + b2 - (b2 < 0x7F ? 0x40 : 0x62); \
2030 charset = charset_big5_1; \
2033 charset = charset_big5_2; \
2034 temp -= (0xC9 - 0xA1) * BIG5_SAME_ROW; \
2036 c1 = temp / (0xFF - 0xA1) + 0x21; \
2037 c2 = temp % (0xFF - 0xA1) + 0x21; \
2040 #define ENCODE_BIG5(charset, c1, c2, b1, b2) \
2042 unsigned int temp = (c1 - 0x21) * (0xFF - 0xA1) + (c2 - 0x21); \
2043 if (charset == charset_big5_2) \
2044 temp += BIG5_SAME_ROW * (0xC9 - 0xA1); \
2045 b1 = temp / BIG5_SAME_ROW + 0xA1; \
2046 b2 = temp % BIG5_SAME_ROW; \
2047 b2 += b2 < 0x3F ? 0x40 : 0x62; \
2050 #define DECODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \
2052 int c_alt, charset_alt = (charset); \
2053 if (!NILP (translation_table) \
2054 && ((c_alt = translate_char (translation_table, \
2055 -1, (charset), c1, c2)) >= 0)) \
2056 SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
2057 if (charset_alt == CHARSET_ASCII || charset_alt < 0) \
2058 DECODE_CHARACTER_ASCII (c1); \
2059 else if (CHARSET_DIMENSION (charset_alt) == 1) \
2060 DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \
2062 DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \
2065 #define ENCODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \
2067 int c_alt, charset_alt; \
2068 if (!NILP (translation_table) \
2069 && ((c_alt = translate_char (translation_table, -1, \
2072 SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
2074 charset_alt = charset; \
2075 if (charset_alt == charset_ascii) \
2077 else if (CHARSET_DIMENSION (charset_alt) == 1) \
2079 if (sjis_p && charset_alt == charset_katakana_jisx0201) \
2083 *dst++ = charset_alt, *dst++ = c1; \
2084 coding->fake_multibyte = 1; \
2089 c1 &= 0x7F, c2 &= 0x7F; \
2090 if (sjis_p && charset_alt == charset_jisx0208) \
2092 unsigned char s1, s2; \
2094 ENCODE_SJIS (c1, c2, s1, s2); \
2095 *dst++ = s1, *dst++ = s2; \
2096 coding->fake_multibyte = 1; \
2099 && (charset_alt == charset_big5_1 \
2100 || charset_alt == charset_big5_2)) \
2102 unsigned char b1, b2; \
2104 ENCODE_BIG5 (charset_alt, c1, c2, b1, b2); \
2105 *dst++ = b1, *dst++ = b2; \
2109 *dst++ = charset_alt, *dst++ = c1, *dst++ = c2; \
2110 coding->fake_multibyte = 1; \
2113 coding->consumed_char++; \
2116 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
2117 Check if a text is encoded in SJIS. If it is, return
2118 CODING_CATEGORY_MASK_SJIS, else return 0. */
2121 detect_coding_sjis (src
, src_end
)
2122 unsigned char *src
, *src_end
;
2126 while (src
< src_end
)
2129 if ((c
>= 0x80 && c
< 0xA0) || c
>= 0xE0)
2131 if (src
< src_end
&& *src
++ < 0x40)
2135 return CODING_CATEGORY_MASK_SJIS
;
2138 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
2139 Check if a text is encoded in BIG5. If it is, return
2140 CODING_CATEGORY_MASK_BIG5, else return 0. */
2143 detect_coding_big5 (src
, src_end
)
2144 unsigned char *src
, *src_end
;
2148 while (src
< src_end
)
2156 if (c
< 0x40 || (c
>= 0x7F && c
<= 0xA0))
2160 return CODING_CATEGORY_MASK_BIG5
;
2163 /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".
2164 If SJIS_P is 1, decode SJIS text, else decode BIG5 test. */
2167 decode_coding_sjis_big5 (coding
, source
, destination
,
2168 src_bytes
, dst_bytes
, sjis_p
)
2169 struct coding_system
*coding
;
2170 unsigned char *source
, *destination
;
2171 int src_bytes
, dst_bytes
;
2174 unsigned char *src
= source
;
2175 unsigned char *src_end
= source
+ src_bytes
;
2176 unsigned char *dst
= destination
;
2177 unsigned char *dst_end
= destination
+ dst_bytes
;
2178 /* Since the maximum bytes produced by each loop is 4, we subtract 3
2179 from DST_END to assure overflow checking is necessary only at the
2181 unsigned char *adjusted_dst_end
= dst_end
- 3;
2182 Lisp_Object translation_table
2183 = coding
->translation_table_for_decode
;
2184 int result
= CODING_FINISH_NORMAL
;
2186 if (!NILP (Venable_character_translation
) && NILP (translation_table
))
2187 translation_table
= Vstandard_translation_table_for_decode
;
2189 coding
->produced_char
= 0;
2190 coding
->fake_multibyte
= 0;
2191 while (src
< src_end
&& (dst_bytes
2192 ? (dst
< adjusted_dst_end
)
2195 /* SRC_BASE remembers the start position in source in each loop.
2196 The loop will be exited when there's not enough source text
2197 to analyze two-byte character (within macro ONE_MORE_BYTE).
2198 In that case, SRC is reset to SRC_BASE before exiting. */
2199 unsigned char *src_base
= src
;
2200 unsigned char c1
= *src
++, c2
, c3
, c4
;
2206 if (coding
->eol_type
== CODING_EOL_CRLF
)
2211 else if (coding
->mode
& CODING_MODE_INHIBIT_INCONSISTENT_EOL
)
2213 result
= CODING_FINISH_INCONSISTENT_EOL
;
2214 goto label_end_of_loop_2
;
2217 /* To process C2 again, SRC is subtracted by 1. */
2220 else if (coding
->eol_type
== CODING_EOL_CR
)
2226 && (coding
->mode
& CODING_MODE_INHIBIT_INCONSISTENT_EOL
)
2227 && (coding
->eol_type
== CODING_EOL_CR
2228 || coding
->eol_type
== CODING_EOL_CRLF
))
2230 result
= CODING_FINISH_INCONSISTENT_EOL
;
2231 goto label_end_of_loop_2
;
2235 coding
->produced_char
++;
2238 DECODE_SJIS_BIG5_CHARACTER (charset_ascii
, c1
, /* dummy */ c2
);
2241 /* SJIS -> JISX0208 */
2247 DECODE_SJIS (c1
, c2
, c3
, c4
);
2248 DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208
, c3
, c4
);
2251 goto label_invalid_code_2
;
2254 goto label_invalid_code_1
;
2258 /* SJIS -> JISX0201-Kana, BIG5 -> Big5 */
2260 DECODE_SJIS_BIG5_CHARACTER (charset_katakana_jisx0201
, c1
,
2267 if ((c2
>= 0x40 && c2
<= 0x7E) || (c2
>= 0xA1 && c2
<= 0xFE))
2269 DECODE_BIG5 (c1
, c2
, charset
, c3
, c4
);
2270 DECODE_SJIS_BIG5_CHARACTER (charset
, c3
, c4
);
2273 goto label_invalid_code_2
;
2276 else /* C1 >= 0xE0 */
2278 /* SJIS -> JISX0208, BIG5 -> Big5 */
2284 DECODE_SJIS (c1
, c2
, c3
, c4
);
2285 DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208
, c3
, c4
);
2288 goto label_invalid_code_2
;
2295 if ((c2
>= 0x40 && c2
<= 0x7E) || (c2
>= 0xA1 && c2
<= 0xFE))
2297 DECODE_BIG5 (c1
, c2
, charset
, c3
, c4
);
2298 DECODE_SJIS_BIG5_CHARACTER (charset
, c3
, c4
);
2301 goto label_invalid_code_2
;
2306 label_invalid_code_1
:
2308 coding
->produced_char
++;
2309 coding
->fake_multibyte
= 1;
2312 label_invalid_code_2
:
2313 *dst
++ = c1
; *dst
++= c2
;
2314 coding
->produced_char
+= 2;
2315 coding
->fake_multibyte
= 1;
2319 result
= CODING_FINISH_INSUFFICIENT_SRC
;
2320 label_end_of_loop_2
:
2327 if (result
== CODING_FINISH_NORMAL
)
2328 result
= CODING_FINISH_INSUFFICIENT_DST
;
2329 else if (result
!= CODING_FINISH_INCONSISTENT_EOL
2330 && coding
->mode
& CODING_MODE_LAST_BLOCK
)
2332 src_bytes
= src_end
- src
;
2333 if (dst_bytes
&& (dst_end
- dst
< src_bytes
))
2334 src_bytes
= dst_end
- dst
;
2335 bcopy (dst
, src
, src_bytes
);
2338 coding
->fake_multibyte
= 1;
2342 coding
->consumed
= coding
->consumed_char
= src
- source
;
2343 coding
->produced
= dst
- destination
;
2347 /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions".
2348 This function can encode `charset_ascii', `charset_katakana_jisx0201',
2349 `charset_jisx0208', `charset_big5_1', and `charset_big5-2'. We are
2350 sure that all these charsets are registered as official charset
2351 (i.e. do not have extended leading-codes). Characters of other
2352 charsets are produced without any encoding. If SJIS_P is 1, encode
2353 SJIS text, else encode BIG5 text. */
2356 encode_coding_sjis_big5 (coding
, source
, destination
,
2357 src_bytes
, dst_bytes
, sjis_p
)
2358 struct coding_system
*coding
;
2359 unsigned char *source
, *destination
;
2360 int src_bytes
, dst_bytes
;
2363 unsigned char *src
= source
;
2364 unsigned char *src_end
= source
+ src_bytes
;
2365 unsigned char *dst
= destination
;
2366 unsigned char *dst_end
= destination
+ dst_bytes
;
2367 /* Since the maximum bytes produced by each loop is 2, we subtract 1
2368 from DST_END to assure overflow checking is necessary only at the
2370 unsigned char *adjusted_dst_end
= dst_end
- 1;
2371 Lisp_Object translation_table
2372 = coding
->translation_table_for_encode
;
2373 int result
= CODING_FINISH_NORMAL
;
2375 if (!NILP (Venable_character_translation
) && NILP (translation_table
))
2376 translation_table
= Vstandard_translation_table_for_encode
;
2378 coding
->consumed_char
= 0;
2379 coding
->fake_multibyte
= 0;
2380 while (src
< src_end
&& (dst_bytes
2381 ? (dst
< adjusted_dst_end
)
2384 /* SRC_BASE remembers the start position in source in each loop.
2385 The loop will be exited when there's not enough source text
2386 to analyze multi-byte codes (within macros ONE_MORE_BYTE and
2387 TWO_MORE_BYTES). In that case, SRC is reset to SRC_BASE
2389 unsigned char *src_base
= src
;
2390 unsigned char c1
= *src
++, c2
, c3
, c4
;
2392 if (coding
->composing
)
2399 else if (c1
>= 0xA0)
2402 coding
->composing
= 0;
2405 switch (emacs_code_class
[c1
])
2407 case EMACS_ascii_code
:
2408 ENCODE_SJIS_BIG5_CHARACTER (charset_ascii
, c1
, /* dummy */ c2
);
2411 case EMACS_control_code
:
2413 coding
->consumed_char
++;
2416 case EMACS_carriage_return_code
:
2417 if (! (coding
->mode
& CODING_MODE_SELECTIVE_DISPLAY
))
2420 coding
->consumed_char
++;
2423 /* fall down to treat '\r' as '\n' ... */
2425 case EMACS_linefeed_code
:
2426 if (coding
->eol_type
== CODING_EOL_LF
2427 || coding
->eol_type
== CODING_EOL_UNDECIDED
)
2429 else if (coding
->eol_type
== CODING_EOL_CRLF
)
2430 *dst
++ = '\r', *dst
++ = '\n';
2433 coding
->consumed_char
++;
2436 case EMACS_leading_code_2
:
2438 ENCODE_SJIS_BIG5_CHARACTER (c1
, c2
, /* dummy */ c3
);
2441 case EMACS_leading_code_3
:
2442 TWO_MORE_BYTES (c2
, c3
);
2443 ENCODE_SJIS_BIG5_CHARACTER (c1
, c2
, c3
);
2446 case EMACS_leading_code_4
:
2447 THREE_MORE_BYTES (c2
, c3
, c4
);
2448 ENCODE_SJIS_BIG5_CHARACTER (c2
, c3
, c4
);
2451 case EMACS_leading_code_composition
:
2452 coding
->composing
= 1;
2455 default: /* i.e. case EMACS_invalid_code: */
2457 coding
->consumed_char
++;
2462 result
= CODING_FINISH_INSUFFICIENT_SRC
;
2467 if (result
== CODING_FINISH_NORMAL
2469 result
= CODING_FINISH_INSUFFICIENT_DST
;
2470 coding
->consumed
= src
- source
;
2471 coding
->produced
= coding
->produced_char
= dst
- destination
;
2476 /*** 5. End-of-line handlers ***/
2478 /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".
2479 This function is called only when `coding->eol_type' is
2480 CODING_EOL_CRLF or CODING_EOL_CR. */
2483 decode_eol (coding
, source
, destination
, src_bytes
, dst_bytes
)
2484 struct coding_system
*coding
;
2485 unsigned char *source
, *destination
;
2486 int src_bytes
, dst_bytes
;
2488 unsigned char *src
= source
;
2489 unsigned char *src_end
= source
+ src_bytes
;
2490 unsigned char *dst
= destination
;
2491 unsigned char *dst_end
= destination
+ dst_bytes
;
2493 int result
= CODING_FINISH_NORMAL
;
2495 coding
->fake_multibyte
= 0;
2500 switch (coding
->eol_type
)
2502 case CODING_EOL_CRLF
:
2504 /* Since the maximum bytes produced by each loop is 2, we
2505 subtract 1 from DST_END to assure overflow checking is
2506 necessary only at the head of loop. */
2507 unsigned char *adjusted_dst_end
= dst_end
- 1;
2509 while (src
< src_end
&& (dst_bytes
2510 ? (dst
< adjusted_dst_end
)
2513 unsigned char *src_base
= src
;
2521 if (coding
->mode
& CODING_MODE_INHIBIT_INCONSISTENT_EOL
)
2523 result
= CODING_FINISH_INCONSISTENT_EOL
;
2524 goto label_end_of_loop_2
;
2527 if (BASE_LEADING_CODE_P (c
))
2528 coding
->fake_multibyte
= 1;
2533 && (coding
->mode
& CODING_MODE_INHIBIT_INCONSISTENT_EOL
))
2535 result
= CODING_FINISH_INCONSISTENT_EOL
;
2536 goto label_end_of_loop_2
;
2541 if (BASE_LEADING_CODE_P (c
))
2542 coding
->fake_multibyte
= 1;
2547 result
= CODING_FINISH_INSUFFICIENT_SRC
;
2548 label_end_of_loop_2
:
2552 if (result
== CODING_FINISH_NORMAL
2554 result
= CODING_FINISH_INSUFFICIENT_DST
;
2559 if (coding
->mode
& CODING_MODE_INHIBIT_INCONSISTENT_EOL
)
2561 while (src
< src_end
)
2563 if ((c
= *src
++) == '\n')
2565 if (BASE_LEADING_CODE_P (c
))
2566 coding
->fake_multibyte
= 1;
2570 src_bytes
= src
- source
;
2571 result
= CODING_FINISH_INCONSISTENT_EOL
;
2574 if (dst_bytes
&& src_bytes
> dst_bytes
)
2576 result
= CODING_FINISH_INSUFFICIENT_DST
;
2577 src_bytes
= dst_bytes
;
2580 bcopy (source
, destination
, src_bytes
);
2582 safe_bcopy (source
, destination
, src_bytes
);
2583 src
= source
+ src_bytes
;
2584 while (src_bytes
--) if (*dst
++ == '\r') dst
[-1] = '\n';
2587 default: /* i.e. case: CODING_EOL_LF */
2588 if (dst_bytes
&& src_bytes
> dst_bytes
)
2590 result
= CODING_FINISH_INSUFFICIENT_DST
;
2591 src_bytes
= dst_bytes
;
2594 bcopy (source
, destination
, src_bytes
);
2596 safe_bcopy (source
, destination
, src_bytes
);
2599 coding
->fake_multibyte
= 1;
2603 coding
->consumed
= coding
->consumed_char
= src
- source
;
2604 coding
->produced
= coding
->produced_char
= dst
- destination
;
2608 /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". Encode
2609 format of end-of-line according to `coding->eol_type'. If
2610 `coding->mode & CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code
2611 '\r' in source text also means end-of-line. */
2614 encode_eol (coding
, source
, destination
, src_bytes
, dst_bytes
)
2615 struct coding_system
*coding
;
2616 unsigned char *source
, *destination
;
2617 int src_bytes
, dst_bytes
;
2619 unsigned char *src
= source
;
2620 unsigned char *dst
= destination
;
2621 int result
= CODING_FINISH_NORMAL
;
2623 coding
->fake_multibyte
= 0;
2625 if (coding
->eol_type
== CODING_EOL_CRLF
)
2628 unsigned char *src_end
= source
+ src_bytes
;
2629 unsigned char *dst_end
= destination
+ dst_bytes
;
2630 /* Since the maximum bytes produced by each loop is 2, we
2631 subtract 1 from DST_END to assure overflow checking is
2632 necessary only at the head of loop. */
2633 unsigned char *adjusted_dst_end
= dst_end
- 1;
2635 while (src
< src_end
&& (dst_bytes
2636 ? (dst
< adjusted_dst_end
)
2641 || (c
== '\r' && (coding
->mode
& CODING_MODE_SELECTIVE_DISPLAY
)))
2642 *dst
++ = '\r', *dst
++ = '\n';
2646 if (BASE_LEADING_CODE_P (c
))
2647 coding
->fake_multibyte
= 1;
2651 result
= CODING_FINISH_INSUFFICIENT_DST
;
2657 if (dst_bytes
&& src_bytes
> dst_bytes
)
2659 src_bytes
= dst_bytes
;
2660 result
= CODING_FINISH_INSUFFICIENT_DST
;
2663 bcopy (source
, destination
, src_bytes
);
2665 safe_bcopy (source
, destination
, src_bytes
);
2666 dst_bytes
= src_bytes
;
2667 if (coding
->eol_type
== CODING_EOL_CR
)
2671 if ((c
= *dst
++) == '\n')
2673 else if (BASE_LEADING_CODE_P (c
))
2674 coding
->fake_multibyte
= 1;
2679 if (coding
->mode
& CODING_MODE_SELECTIVE_DISPLAY
)
2682 if (*dst
++ == '\r') dst
[-1] = '\n';
2684 coding
->fake_multibyte
= 1;
2686 src
= source
+ dst_bytes
;
2687 dst
= destination
+ dst_bytes
;
2690 coding
->consumed
= coding
->consumed_char
= src
- source
;
2691 coding
->produced
= coding
->produced_char
= dst
- destination
;
2696 /*** 6. C library functions ***/
2698 /* In Emacs Lisp, coding system is represented by a Lisp symbol which
2699 has a property `coding-system'. The value of this property is a
2700 vector of length 5 (called as coding-vector). Among elements of
2701 this vector, the first (element[0]) and the fifth (element[4])
2702 carry important information for decoding/encoding. Before
2703 decoding/encoding, this information should be set in fields of a
2704 structure of type `coding_system'.
2706 A value of property `coding-system' can be a symbol of another
2707 subsidiary coding-system. In that case, Emacs gets coding-vector
2710 `element[0]' contains information to be set in `coding->type'. The
2711 value and its meaning is as follows:
2713 0 -- coding_type_emacs_mule
2714 1 -- coding_type_sjis
2715 2 -- coding_type_iso2022
2716 3 -- coding_type_big5
2717 4 -- coding_type_ccl encoder/decoder written in CCL
2718 nil -- coding_type_no_conversion
2719 t -- coding_type_undecided (automatic conversion on decoding,
2720 no-conversion on encoding)
2722 `element[4]' contains information to be set in `coding->flags' and
2723 `coding->spec'. The meaning varies by `coding->type'.
2725 If `coding->type' is `coding_type_iso2022', element[4] is a vector
2726 of length 32 (of which the first 13 sub-elements are used now).
2727 Meanings of these sub-elements are:
2729 sub-element[N] where N is 0 through 3: to be set in `coding->spec.iso2022'
2730 If the value is an integer of valid charset, the charset is
2731 assumed to be designated to graphic register N initially.
2733 If the value is minus, it is a minus value of charset which
2734 reserves graphic register N, which means that the charset is
2735 not designated initially but should be designated to graphic
2736 register N just before encoding a character in that charset.
2738 If the value is nil, graphic register N is never used on
2741 sub-element[N] where N is 4 through 11: to be set in `coding->flags'
2742 Each value takes t or nil. See the section ISO2022 of
2743 `coding.h' for more information.
2745 If `coding->type' is `coding_type_big5', element[4] is t to denote
2746 BIG5-ETen or nil to denote BIG5-HKU.
2748 If `coding->type' takes the other value, element[4] is ignored.
2750 Emacs Lisp's coding system also carries information about format of
2751 end-of-line in a value of property `eol-type'. If the value is
2752 integer, 0 means CODING_EOL_LF, 1 means CODING_EOL_CRLF, and 2
2753 means CODING_EOL_CR. If it is not integer, it should be a vector
2754 of subsidiary coding systems of which property `eol-type' has one
2759 /* Extract information for decoding/encoding from CODING_SYSTEM_SYMBOL
2760 and set it in CODING. If CODING_SYSTEM_SYMBOL is invalid, CODING
2761 is setup so that no conversion is necessary and return -1, else
2765 setup_coding_system (coding_system
, coding
)
2766 Lisp_Object coding_system
;
2767 struct coding_system
*coding
;
2769 Lisp_Object coding_spec
, coding_type
, eol_type
, plist
;
2773 /* Initialize some fields required for all kinds of coding systems. */
2774 coding
->symbol
= coding_system
;
2775 coding
->common_flags
= 0;
2777 coding
->heading_ascii
= -1;
2778 coding
->post_read_conversion
= coding
->pre_write_conversion
= Qnil
;
2779 coding_spec
= Fget (coding_system
, Qcoding_system
);
2780 if (!VECTORP (coding_spec
)
2781 || XVECTOR (coding_spec
)->size
!= 5
2782 || !CONSP (XVECTOR (coding_spec
)->contents
[3]))
2783 goto label_invalid_coding_system
;
2785 eol_type
= inhibit_eol_conversion
? Qnil
: Fget (coding_system
, Qeol_type
);
2786 if (VECTORP (eol_type
))
2788 coding
->eol_type
= CODING_EOL_UNDECIDED
;
2789 coding
->common_flags
= CODING_REQUIRE_DETECTION_MASK
;
2791 else if (XFASTINT (eol_type
) == 1)
2793 coding
->eol_type
= CODING_EOL_CRLF
;
2794 coding
->common_flags
2795 = CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
;
2797 else if (XFASTINT (eol_type
) == 2)
2799 coding
->eol_type
= CODING_EOL_CR
;
2800 coding
->common_flags
2801 = CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
;
2804 coding
->eol_type
= CODING_EOL_LF
;
2806 coding_type
= XVECTOR (coding_spec
)->contents
[0];
2807 /* Try short cut. */
2808 if (SYMBOLP (coding_type
))
2810 if (EQ (coding_type
, Qt
))
2812 coding
->type
= coding_type_undecided
;
2813 coding
->common_flags
|= CODING_REQUIRE_DETECTION_MASK
;
2816 coding
->type
= coding_type_no_conversion
;
2820 /* Initialize remaining fields. */
2821 coding
->composing
= 0;
2822 coding
->translation_table_for_decode
= Qnil
;
2823 coding
->translation_table_for_encode
= Qnil
;
2825 /* Get values of coding system properties:
2826 `post-read-conversion', `pre-write-conversion',
2827 `translation-table-for-decode', `translation-table-for-encode'. */
2828 plist
= XVECTOR (coding_spec
)->contents
[3];
2829 coding
->post_read_conversion
= Fplist_get (plist
, Qpost_read_conversion
);
2830 coding
->pre_write_conversion
= Fplist_get (plist
, Qpre_write_conversion
);
2831 val
= Fplist_get (plist
, Qtranslation_table_for_decode
);
2833 val
= Fget (val
, Qtranslation_table_for_decode
);
2834 coding
->translation_table_for_decode
= CHAR_TABLE_P (val
) ? val
: Qnil
;
2835 val
= Fplist_get (plist
, Qtranslation_table_for_encode
);
2837 val
= Fget (val
, Qtranslation_table_for_encode
);
2838 coding
->translation_table_for_encode
= CHAR_TABLE_P (val
) ? val
: Qnil
;
2839 val
= Fplist_get (plist
, Qcoding_category
);
2842 val
= Fget (val
, Qcoding_category_index
);
2844 coding
->category_idx
= XINT (val
);
2846 goto label_invalid_coding_system
;
2849 goto label_invalid_coding_system
;
2851 val
= Fplist_get (plist
, Qsafe_charsets
);
2854 for (i
= 0; i
<= MAX_CHARSET
; i
++)
2855 coding
->safe_charsets
[i
] = 1;
2859 bzero (coding
->safe_charsets
, MAX_CHARSET
+ 1);
2862 if ((i
= get_charset_id (XCONS (val
)->car
)) >= 0)
2863 coding
->safe_charsets
[i
] = 1;
2864 val
= XCONS (val
)->cdr
;
2868 switch (XFASTINT (coding_type
))
2871 coding
->type
= coding_type_emacs_mule
;
2872 if (!NILP (coding
->post_read_conversion
))
2873 coding
->common_flags
|= CODING_REQUIRE_DECODING_MASK
;
2874 if (!NILP (coding
->pre_write_conversion
))
2875 coding
->common_flags
|= CODING_REQUIRE_ENCODING_MASK
;
2879 coding
->type
= coding_type_sjis
;
2880 coding
->common_flags
2881 |= CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
;
2885 coding
->type
= coding_type_iso2022
;
2886 coding
->common_flags
2887 |= CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
;
2889 Lisp_Object val
, temp
;
2891 int i
, charset
, reg_bits
= 0;
2893 val
= XVECTOR (coding_spec
)->contents
[4];
2895 if (!VECTORP (val
) || XVECTOR (val
)->size
!= 32)
2896 goto label_invalid_coding_system
;
2898 flags
= XVECTOR (val
)->contents
;
2900 = ((NILP (flags
[4]) ? 0 : CODING_FLAG_ISO_SHORT_FORM
)
2901 | (NILP (flags
[5]) ? 0 : CODING_FLAG_ISO_RESET_AT_EOL
)
2902 | (NILP (flags
[6]) ? 0 : CODING_FLAG_ISO_RESET_AT_CNTL
)
2903 | (NILP (flags
[7]) ? 0 : CODING_FLAG_ISO_SEVEN_BITS
)
2904 | (NILP (flags
[8]) ? 0 : CODING_FLAG_ISO_LOCKING_SHIFT
)
2905 | (NILP (flags
[9]) ? 0 : CODING_FLAG_ISO_SINGLE_SHIFT
)
2906 | (NILP (flags
[10]) ? 0 : CODING_FLAG_ISO_USE_ROMAN
)
2907 | (NILP (flags
[11]) ? 0 : CODING_FLAG_ISO_USE_OLDJIS
)
2908 | (NILP (flags
[12]) ? 0 : CODING_FLAG_ISO_NO_DIRECTION
)
2909 | (NILP (flags
[13]) ? 0 : CODING_FLAG_ISO_INIT_AT_BOL
)
2910 | (NILP (flags
[14]) ? 0 : CODING_FLAG_ISO_DESIGNATE_AT_BOL
)
2911 | (NILP (flags
[15]) ? 0 : CODING_FLAG_ISO_SAFE
)
2912 | (NILP (flags
[16]) ? 0 : CODING_FLAG_ISO_LATIN_EXTRA
)
2915 /* Invoke graphic register 0 to plane 0. */
2916 CODING_SPEC_ISO_INVOCATION (coding
, 0) = 0;
2917 /* Invoke graphic register 1 to plane 1 if we can use full 8-bit. */
2918 CODING_SPEC_ISO_INVOCATION (coding
, 1)
2919 = (coding
->flags
& CODING_FLAG_ISO_SEVEN_BITS
? -1 : 1);
2920 /* Not single shifting at first. */
2921 CODING_SPEC_ISO_SINGLE_SHIFTING (coding
) = 0;
2922 /* Beginning of buffer should also be regarded as bol. */
2923 CODING_SPEC_ISO_BOL (coding
) = 1;
2925 for (charset
= 0; charset
<= MAX_CHARSET
; charset
++)
2926 CODING_SPEC_ISO_REVISION_NUMBER (coding
, charset
) = 255;
2927 val
= Vcharset_revision_alist
;
2930 charset
= get_charset_id (Fcar_safe (XCONS (val
)->car
));
2932 && (temp
= Fcdr_safe (XCONS (val
)->car
), INTEGERP (temp
))
2933 && (i
= XINT (temp
), (i
>= 0 && (i
+ '@') < 128)))
2934 CODING_SPEC_ISO_REVISION_NUMBER (coding
, charset
) = i
;
2935 val
= XCONS (val
)->cdr
;
2938 /* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations.
2939 FLAGS[REG] can be one of below:
2940 integer CHARSET: CHARSET occupies register I,
2941 t: designate nothing to REG initially, but can be used
2943 list of integer, nil, or t: designate the first
2944 element (if integer) to REG initially, the remaining
2945 elements (if integer) is designated to REG on request,
2946 if an element is t, REG can be used by any charsets,
2947 nil: REG is never used. */
2948 for (charset
= 0; charset
<= MAX_CHARSET
; charset
++)
2949 CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
)
2950 = CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION
;
2951 for (i
= 0; i
< 4; i
++)
2953 if (INTEGERP (flags
[i
])
2954 && (charset
= XINT (flags
[i
]), CHARSET_VALID_P (charset
))
2955 || (charset
= get_charset_id (flags
[i
])) >= 0)
2957 CODING_SPEC_ISO_INITIAL_DESIGNATION (coding
, i
) = charset
;
2958 CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
) = i
;
2960 else if (EQ (flags
[i
], Qt
))
2962 CODING_SPEC_ISO_INITIAL_DESIGNATION (coding
, i
) = -1;
2964 coding
->flags
|= CODING_FLAG_ISO_DESIGNATION
;
2966 else if (CONSP (flags
[i
]))
2968 Lisp_Object tail
= flags
[i
];
2970 coding
->flags
|= CODING_FLAG_ISO_DESIGNATION
;
2971 if (INTEGERP (XCONS (tail
)->car
)
2972 && (charset
= XINT (XCONS (tail
)->car
),
2973 CHARSET_VALID_P (charset
))
2974 || (charset
= get_charset_id (XCONS (tail
)->car
)) >= 0)
2976 CODING_SPEC_ISO_INITIAL_DESIGNATION (coding
, i
) = charset
;
2977 CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
) =i
;
2980 CODING_SPEC_ISO_INITIAL_DESIGNATION (coding
, i
) = -1;
2981 tail
= XCONS (tail
)->cdr
;
2982 while (CONSP (tail
))
2984 if (INTEGERP (XCONS (tail
)->car
)
2985 && (charset
= XINT (XCONS (tail
)->car
),
2986 CHARSET_VALID_P (charset
))
2987 || (charset
= get_charset_id (XCONS (tail
)->car
)) >= 0)
2988 CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
)
2990 else if (EQ (XCONS (tail
)->car
, Qt
))
2992 tail
= XCONS (tail
)->cdr
;
2996 CODING_SPEC_ISO_INITIAL_DESIGNATION (coding
, i
) = -1;
2998 CODING_SPEC_ISO_DESIGNATION (coding
, i
)
2999 = CODING_SPEC_ISO_INITIAL_DESIGNATION (coding
, i
);
3002 if (reg_bits
&& ! (coding
->flags
& CODING_FLAG_ISO_LOCKING_SHIFT
))
3004 /* REG 1 can be used only by locking shift in 7-bit env. */
3005 if (coding
->flags
& CODING_FLAG_ISO_SEVEN_BITS
)
3007 if (! (coding
->flags
& CODING_FLAG_ISO_SINGLE_SHIFT
))
3008 /* Without any shifting, only REG 0 and 1 can be used. */
3013 for (charset
= 0; charset
<= MAX_CHARSET
; charset
++)
3015 if (CHARSET_VALID_P (charset
))
3017 /* There exist some default graphic registers to be
3020 /* We had better avoid designating a charset of
3021 CHARS96 to REG 0 as far as possible. */
3022 if (CHARSET_CHARS (charset
) == 96)
3023 CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
)
3025 ? 1 : (reg_bits
& 4 ? 2 : (reg_bits
& 8 ? 3 : 0)));
3027 CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding
, charset
)
3029 ? 0 : (reg_bits
& 2 ? 1 : (reg_bits
& 4 ? 2 : 3)));
3033 coding
->common_flags
|= CODING_REQUIRE_FLUSHING_MASK
;
3034 coding
->spec
.iso2022
.last_invalid_designation_register
= -1;
3038 coding
->type
= coding_type_big5
;
3039 coding
->common_flags
3040 |= CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
;
3042 = (NILP (XVECTOR (coding_spec
)->contents
[4])
3043 ? CODING_FLAG_BIG5_HKU
3044 : CODING_FLAG_BIG5_ETEN
);
3048 coding
->type
= coding_type_ccl
;
3049 coding
->common_flags
3050 |= CODING_REQUIRE_DECODING_MASK
| CODING_REQUIRE_ENCODING_MASK
;
3052 Lisp_Object val
= XVECTOR (coding_spec
)->contents
[4];
3053 Lisp_Object decoder
, encoder
;
3056 && SYMBOLP (XCONS (val
)->car
)
3057 && !NILP (decoder
= Fget (XCONS (val
)->car
, Qccl_program_idx
))
3058 && !NILP (decoder
= Fcdr (Faref (Vccl_program_table
, decoder
)))
3059 && SYMBOLP (XCONS (val
)->cdr
)
3060 && !NILP (encoder
= Fget (XCONS (val
)->cdr
, Qccl_program_idx
))
3061 && !NILP (encoder
= Fcdr (Faref (Vccl_program_table
, encoder
))))
3063 setup_ccl_program (&(coding
->spec
.ccl
.decoder
), decoder
);
3064 setup_ccl_program (&(coding
->spec
.ccl
.encoder
), encoder
);
3067 goto label_invalid_coding_system
;
3069 coding
->common_flags
|= CODING_REQUIRE_FLUSHING_MASK
;
3073 coding
->type
= coding_type_raw_text
;
3077 goto label_invalid_coding_system
;
3081 label_invalid_coding_system
:
3082 coding
->type
= coding_type_no_conversion
;
3083 coding
->category_idx
= CODING_CATEGORY_IDX_BINARY
;
3084 coding
->common_flags
= 0;
3085 coding
->eol_type
= CODING_EOL_LF
;
3086 coding
->pre_write_conversion
= coding
->post_read_conversion
= Qnil
;
3090 /* Emacs has a mechanism to automatically detect a coding system if it
3091 is one of Emacs' internal format, ISO2022, SJIS, and BIG5. But,
3092 it's impossible to distinguish some coding systems accurately
3093 because they use the same range of codes. So, at first, coding
3094 systems are categorized into 7, those are:
3096 o coding-category-emacs-mule
3098 The category for a coding system which has the same code range
3099 as Emacs' internal format. Assigned the coding-system (Lisp
3100 symbol) `emacs-mule' by default.
3102 o coding-category-sjis
3104 The category for a coding system which has the same code range
3105 as SJIS. Assigned the coding-system (Lisp
3106 symbol) `japanese-shift-jis' by default.
3108 o coding-category-iso-7
3110 The category for a coding system which has the same code range
3111 as ISO2022 of 7-bit environment. This doesn't use any locking
3112 shift and single shift functions. This can encode/decode all
3113 charsets. Assigned the coding-system (Lisp symbol)
3114 `iso-2022-7bit' by default.
3116 o coding-category-iso-7-tight
3118 Same as coding-category-iso-7 except that this can
3119 encode/decode only the specified charsets.
3121 o coding-category-iso-8-1
3123 The category for a coding system which has the same code range
3124 as ISO2022 of 8-bit environment and graphic plane 1 used only
3125 for DIMENSION1 charset. This doesn't use any locking shift
3126 and single shift functions. Assigned the coding-system (Lisp
3127 symbol) `iso-latin-1' by default.
3129 o coding-category-iso-8-2
3131 The category for a coding system which has the same code range
3132 as ISO2022 of 8-bit environment and graphic plane 1 used only
3133 for DIMENSION2 charset. This doesn't use any locking shift
3134 and single shift functions. Assigned the coding-system (Lisp
3135 symbol) `japanese-iso-8bit' by default.
3137 o coding-category-iso-7-else
3139 The category for a coding system which has the same code range
3140 as ISO2022 of 7-bit environemnt but uses locking shift or
3141 single shift functions. Assigned the coding-system (Lisp
3142 symbol) `iso-2022-7bit-lock' by default.
3144 o coding-category-iso-8-else
3146 The category for a coding system which has the same code range
3147 as ISO2022 of 8-bit environemnt but uses locking shift or
3148 single shift functions. Assigned the coding-system (Lisp
3149 symbol) `iso-2022-8bit-ss2' by default.
3151 o coding-category-big5
3153 The category for a coding system which has the same code range
3154 as BIG5. Assigned the coding-system (Lisp symbol)
3155 `cn-big5' by default.
3157 o coding-category-binary
3159 The category for a coding system not categorized in any of the
3160 above. Assigned the coding-system (Lisp symbol)
3161 `no-conversion' by default.
3163 Each of them is a Lisp symbol and the value is an actual
3164 `coding-system's (this is also a Lisp symbol) assigned by a user.
3165 What Emacs does actually is to detect a category of coding system.
3166 Then, it uses a `coding-system' assigned to it. If Emacs can't
3167 decide only one possible category, it selects a category of the
3168 highest priority. Priorities of categories are also specified by a
3169 user in a Lisp variable `coding-category-list'.
3174 int ascii_skip_code
[256];
3176 /* Detect how a text of length SRC_BYTES pointed by SOURCE is encoded.
3177 If it detects possible coding systems, return an integer in which
3178 appropriate flag bits are set. Flag bits are defined by macros
3179 CODING_CATEGORY_MASK_XXX in `coding.h'.
3181 How many ASCII characters are at the head is returned as *SKIP. */
3184 detect_coding_mask (source
, src_bytes
, priorities
, skip
)
3185 unsigned char *source
;
3186 int src_bytes
, *priorities
, *skip
;
3188 register unsigned char c
;
3189 unsigned char *src
= source
, *src_end
= source
+ src_bytes
;
3193 /* At first, skip all ASCII characters and control characters except
3194 for three ISO2022 specific control characters. */
3195 ascii_skip_code
[ISO_CODE_SO
] = 0;
3196 ascii_skip_code
[ISO_CODE_SI
] = 0;
3197 ascii_skip_code
[ISO_CODE_ESC
] = 0;
3199 label_loop_detect_coding
:
3200 while (src
< src_end
&& ascii_skip_code
[*src
]) src
++;
3201 *skip
= src
- source
;
3204 /* We found nothing other than ASCII. There's nothing to do. */
3208 /* The text seems to be encoded in some multilingual coding system.
3209 Now, try to find in which coding system the text is encoded. */
3212 /* i.e. (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) */
3213 /* C is an ISO2022 specific control code of C0. */
3214 mask
= detect_coding_iso2022 (src
, src_end
);
3217 /* No valid ISO2022 code follows C. Try again. */
3219 if (c
== ISO_CODE_ESC
)
3220 ascii_skip_code
[ISO_CODE_ESC
] = 1;
3222 ascii_skip_code
[ISO_CODE_SO
] = ascii_skip_code
[ISO_CODE_SI
] = 1;
3223 goto label_loop_detect_coding
;
3226 goto label_return_highest_only
;
3234 /* C is the first byte of SJIS character code,
3235 or a leading-code of Emacs' internal format (emacs-mule). */
3236 try = CODING_CATEGORY_MASK_SJIS
| CODING_CATEGORY_MASK_EMACS_MULE
;
3238 /* Or, if C is a special latin extra code,
3239 or is an ISO2022 specific control code of C1 (SS2 or SS3),
3240 or is an ISO2022 control-sequence-introducer (CSI),
3241 we should also consider the possibility of ISO2022 codings. */
3242 if ((VECTORP (Vlatin_extra_code_table
)
3243 && !NILP (XVECTOR (Vlatin_extra_code_table
)->contents
[c
]))
3244 || (c
== ISO_CODE_SS2
|| c
== ISO_CODE_SS3
)
3245 || (c
== ISO_CODE_CSI
3248 || ((*src
== '0' || *src
== '1' || *src
== '2')
3249 && src
+ 1 < src_end
3250 && src
[1] == ']')))))
3251 try |= (CODING_CATEGORY_MASK_ISO_8_ELSE
3252 | CODING_CATEGORY_MASK_ISO_8BIT
);
3255 /* C is a character of ISO2022 in graphic plane right,
3256 or a SJIS's 1-byte character code (i.e. JISX0201),
3257 or the first byte of BIG5's 2-byte code. */
3258 try = (CODING_CATEGORY_MASK_ISO_8_ELSE
3259 | CODING_CATEGORY_MASK_ISO_8BIT
3260 | CODING_CATEGORY_MASK_SJIS
3261 | CODING_CATEGORY_MASK_BIG5
);
3266 for (i
= 0; i
< CODING_CATEGORY_IDX_MAX
; i
++)
3268 if (priorities
[i
] & try & CODING_CATEGORY_MASK_ISO
)
3269 mask
= detect_coding_iso2022 (src
, src_end
);
3270 else if (priorities
[i
] & try & CODING_CATEGORY_MASK_SJIS
)
3271 mask
= detect_coding_sjis (src
, src_end
);
3272 else if (priorities
[i
] & try & CODING_CATEGORY_MASK_BIG5
)
3273 mask
= detect_coding_big5 (src
, src_end
);
3274 else if (priorities
[i
] & try & CODING_CATEGORY_MASK_EMACS_MULE
)
3275 mask
= detect_coding_emacs_mule (src
, src_end
);
3276 else if (priorities
[i
] & CODING_CATEGORY_MASK_RAW_TEXT
)
3277 mask
= CODING_CATEGORY_MASK_RAW_TEXT
;
3278 else if (priorities
[i
] & CODING_CATEGORY_MASK_BINARY
)
3279 mask
= CODING_CATEGORY_MASK_BINARY
;
3281 goto label_return_highest_only
;
3283 return CODING_CATEGORY_MASK_RAW_TEXT
;
3285 if (try & CODING_CATEGORY_MASK_ISO
)
3286 mask
|= detect_coding_iso2022 (src
, src_end
);
3287 if (try & CODING_CATEGORY_MASK_SJIS
)
3288 mask
|= detect_coding_sjis (src
, src_end
);
3289 if (try & CODING_CATEGORY_MASK_BIG5
)
3290 mask
|= detect_coding_big5 (src
, src_end
);
3291 if (try & CODING_CATEGORY_MASK_EMACS_MULE
)
3292 mask
|= detect_coding_emacs_mule (src
, src_end
);
3294 return (mask
| CODING_CATEGORY_MASK_RAW_TEXT
| CODING_CATEGORY_MASK_BINARY
);
3296 label_return_highest_only
:
3297 for (i
= 0; i
< CODING_CATEGORY_IDX_MAX
; i
++)
3299 if (mask
& priorities
[i
])
3300 return priorities
[i
];
3302 return CODING_CATEGORY_MASK_RAW_TEXT
;
3305 /* Detect how a text of length SRC_BYTES pointed by SRC is encoded.
3306 The information of the detected coding system is set in CODING. */
3309 detect_coding (coding
, src
, src_bytes
)
3310 struct coding_system
*coding
;
3316 Lisp_Object val
= Vcoding_category_list
;
3318 mask
= detect_coding_mask (src
, src_bytes
, coding_priorities
, &skip
);
3319 coding
->heading_ascii
= skip
;
3323 /* We found a single coding system of the highest priority in MASK. */
3325 while (mask
&& ! (mask
& 1)) mask
>>= 1, idx
++;
3327 idx
= CODING_CATEGORY_IDX_RAW_TEXT
;
3329 val
= XSYMBOL (XVECTOR (Vcoding_category_table
)->contents
[idx
])->value
;
3331 if (coding
->eol_type
!= CODING_EOL_UNDECIDED
)
3333 Lisp_Object tmp
= Fget (val
, Qeol_type
);
3336 val
= XVECTOR (tmp
)->contents
[coding
->eol_type
];
3338 setup_coding_system (val
, coding
);
3339 /* Set this again because setup_coding_system reset this member. */
3340 coding
->heading_ascii
= skip
;
3343 /* Detect how end-of-line of a text of length SRC_BYTES pointed by
3344 SOURCE is encoded. Return one of CODING_EOL_LF, CODING_EOL_CRLF,
3345 CODING_EOL_CR, and CODING_EOL_UNDECIDED.
3347 How many non-eol characters are at the head is returned as *SKIP. */
3349 #define MAX_EOL_CHECK_COUNT 3
3352 detect_eol_type (source
, src_bytes
, skip
)
3353 unsigned char *source
;
3354 int src_bytes
, *skip
;
3356 unsigned char *src
= source
, *src_end
= src
+ src_bytes
;
3358 int total
= 0; /* How many end-of-lines are found so far. */
3359 int eol_type
= CODING_EOL_UNDECIDED
;
3364 while (src
< src_end
&& total
< MAX_EOL_CHECK_COUNT
)
3367 if (c
== '\n' || c
== '\r')
3370 *skip
= src
- 1 - source
;
3373 this_eol_type
= CODING_EOL_LF
;
3374 else if (src
>= src_end
|| *src
!= '\n')
3375 this_eol_type
= CODING_EOL_CR
;
3377 this_eol_type
= CODING_EOL_CRLF
, src
++;
3379 if (eol_type
== CODING_EOL_UNDECIDED
)
3380 /* This is the first end-of-line. */
3381 eol_type
= this_eol_type
;
3382 else if (eol_type
!= this_eol_type
)
3384 /* The found type is different from what found before. */
3385 eol_type
= CODING_EOL_INCONSISTENT
;
3392 *skip
= src_end
- source
;
3396 /* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC
3397 is encoded. If it detects an appropriate format of end-of-line, it
3398 sets the information in *CODING. */
3401 detect_eol (coding
, src
, src_bytes
)
3402 struct coding_system
*coding
;
3408 int eol_type
= detect_eol_type (src
, src_bytes
, &skip
);
3410 if (coding
->heading_ascii
> skip
)
3411 coding
->heading_ascii
= skip
;
3413 skip
= coding
->heading_ascii
;
3415 if (eol_type
== CODING_EOL_UNDECIDED
)
3417 if (eol_type
== CODING_EOL_INCONSISTENT
)
3420 /* This code is suppressed until we find a better way to
3421 distinguish raw text file and binary file. */
3423 /* If we have already detected that the coding is raw-text, the
3424 coding should actually be no-conversion. */
3425 if (coding
->type
== coding_type_raw_text
)
3427 setup_coding_system (Qno_conversion
, coding
);
3430 /* Else, let's decode only text code anyway. */
3432 eol_type
= CODING_EOL_LF
;
3435 val
= Fget (coding
->symbol
, Qeol_type
);
3436 if (VECTORP (val
) && XVECTOR (val
)->size
== 3)
3438 setup_coding_system (XVECTOR (val
)->contents
[eol_type
], coding
);
3439 coding
->heading_ascii
= skip
;
3443 #define CONVERSION_BUFFER_EXTRA_ROOM 256
3445 #define DECODING_BUFFER_MAG(coding) \
3446 (coding->type == coding_type_iso2022 \
3448 : ((coding->type == coding_type_sjis || coding->type == coding_type_big5) \
3450 : (coding->type == coding_type_raw_text \
3452 : (coding->type == coding_type_ccl \
3453 ? coding->spec.ccl.decoder.buf_magnification \
3456 /* Return maximum size (bytes) of a buffer enough for decoding
3457 SRC_BYTES of text encoded in CODING. */
3460 decoding_buffer_size (coding
, src_bytes
)
3461 struct coding_system
*coding
;
3464 return (src_bytes
* DECODING_BUFFER_MAG (coding
)
3465 + CONVERSION_BUFFER_EXTRA_ROOM
);
3468 /* Return maximum size (bytes) of a buffer enough for encoding
3469 SRC_BYTES of text to CODING. */
3472 encoding_buffer_size (coding
, src_bytes
)
3473 struct coding_system
*coding
;
3478 if (coding
->type
== coding_type_ccl
)
3479 magnification
= coding
->spec
.ccl
.encoder
.buf_magnification
;
3483 return (src_bytes
* magnification
+ CONVERSION_BUFFER_EXTRA_ROOM
);
3486 #ifndef MINIMUM_CONVERSION_BUFFER_SIZE
3487 #define MINIMUM_CONVERSION_BUFFER_SIZE 1024
3490 char *conversion_buffer
;
3491 int conversion_buffer_size
;
3493 /* Return a pointer to a SIZE bytes of buffer to be used for encoding
3494 or decoding. Sufficient memory is allocated automatically. If we
3495 run out of memory, return NULL. */
3498 get_conversion_buffer (size
)
3501 if (size
> conversion_buffer_size
)
3504 int real_size
= conversion_buffer_size
* 2;
3506 while (real_size
< size
) real_size
*= 2;
3507 buf
= (char *) xmalloc (real_size
);
3508 xfree (conversion_buffer
);
3509 conversion_buffer
= buf
;
3510 conversion_buffer_size
= real_size
;
3512 return conversion_buffer
;
3516 ccl_coding_driver (coding
, source
, destination
, src_bytes
, dst_bytes
, encodep
)
3517 struct coding_system
*coding
;
3518 unsigned char *source
, *destination
;
3519 int src_bytes
, dst_bytes
, encodep
;
3521 struct ccl_program
*ccl
3522 = encodep
? &coding
->spec
.ccl
.encoder
: &coding
->spec
.ccl
.decoder
;
3525 coding
->produced
= ccl_driver (ccl
, source
, destination
,
3526 src_bytes
, dst_bytes
, &(coding
->consumed
));
3529 coding
->produced_char
= coding
->produced
;
3530 coding
->consumed_char
3531 = multibyte_chars_in_text (source
, coding
->consumed
);
3535 coding
->produced_char
3536 = multibyte_chars_in_text (destination
, coding
->produced
);
3537 coding
->consumed_char
= coding
->consumed
;
3539 switch (ccl
->status
)
3541 case CCL_STAT_SUSPEND_BY_SRC
:
3542 result
= CODING_FINISH_INSUFFICIENT_SRC
;
3544 case CCL_STAT_SUSPEND_BY_DST
:
3545 result
= CODING_FINISH_INSUFFICIENT_DST
;
3548 result
= CODING_FINISH_NORMAL
;
3554 /* See "GENERAL NOTES about `decode_coding_XXX ()' functions". Before
3555 decoding, it may detect coding system and format of end-of-line if
3556 those are not yet decided. */
3559 decode_coding (coding
, source
, destination
, src_bytes
, dst_bytes
)
3560 struct coding_system
*coding
;
3561 unsigned char *source
, *destination
;
3562 int src_bytes
, dst_bytes
;
3568 coding
->produced
= coding
->produced_char
= 0;
3569 coding
->consumed
= coding
->consumed_char
= 0;
3570 coding
->fake_multibyte
= 0;
3571 return CODING_FINISH_NORMAL
;
3574 if (coding
->type
== coding_type_undecided
)
3575 detect_coding (coding
, source
, src_bytes
);
3577 if (coding
->eol_type
== CODING_EOL_UNDECIDED
)
3578 detect_eol (coding
, source
, src_bytes
);
3580 switch (coding
->type
)
3582 case coding_type_emacs_mule
:
3583 case coding_type_undecided
:
3584 case coding_type_raw_text
:
3585 if (coding
->eol_type
== CODING_EOL_LF
3586 || coding
->eol_type
== CODING_EOL_UNDECIDED
)
3587 goto label_no_conversion
;
3588 result
= decode_eol (coding
, source
, destination
, src_bytes
, dst_bytes
);
3591 case coding_type_sjis
:
3592 result
= decode_coding_sjis_big5 (coding
, source
, destination
,
3593 src_bytes
, dst_bytes
, 1);
3596 case coding_type_iso2022
:
3597 result
= decode_coding_iso2022 (coding
, source
, destination
,
3598 src_bytes
, dst_bytes
);
3601 case coding_type_big5
:
3602 result
= decode_coding_sjis_big5 (coding
, source
, destination
,
3603 src_bytes
, dst_bytes
, 0);
3606 case coding_type_ccl
:
3607 result
= ccl_coding_driver (coding
, source
, destination
,
3608 src_bytes
, dst_bytes
, 0);
3611 default: /* i.e. case coding_type_no_conversion: */
3612 label_no_conversion
:
3613 if (dst_bytes
&& src_bytes
> dst_bytes
)
3615 coding
->produced
= dst_bytes
;
3616 result
= CODING_FINISH_INSUFFICIENT_DST
;
3620 coding
->produced
= src_bytes
;
3621 result
= CODING_FINISH_NORMAL
;
3624 bcopy (source
, destination
, coding
->produced
);
3626 safe_bcopy (source
, destination
, coding
->produced
);
3627 coding
->fake_multibyte
= 1;
3629 = coding
->consumed_char
= coding
->produced_char
= coding
->produced
;
3636 /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". */
3639 encode_coding (coding
, source
, destination
, src_bytes
, dst_bytes
)
3640 struct coding_system
*coding
;
3641 unsigned char *source
, *destination
;
3642 int src_bytes
, dst_bytes
;
3648 coding
->produced
= coding
->produced_char
= 0;
3649 coding
->consumed
= coding
->consumed_char
= 0;
3650 coding
->fake_multibyte
= 0;
3651 return CODING_FINISH_NORMAL
;
3654 switch (coding
->type
)
3656 case coding_type_emacs_mule
:
3657 case coding_type_undecided
:
3658 case coding_type_raw_text
:
3659 if (coding
->eol_type
== CODING_EOL_LF
3660 || coding
->eol_type
== CODING_EOL_UNDECIDED
)
3661 goto label_no_conversion
;
3662 result
= encode_eol (coding
, source
, destination
, src_bytes
, dst_bytes
);
3665 case coding_type_sjis
:
3666 result
= encode_coding_sjis_big5 (coding
, source
, destination
,
3667 src_bytes
, dst_bytes
, 1);
3670 case coding_type_iso2022
:
3671 result
= encode_coding_iso2022 (coding
, source
, destination
,
3672 src_bytes
, dst_bytes
);
3675 case coding_type_big5
:
3676 result
= encode_coding_sjis_big5 (coding
, source
, destination
,
3677 src_bytes
, dst_bytes
, 0);
3680 case coding_type_ccl
:
3681 result
= ccl_coding_driver (coding
, source
, destination
,
3682 src_bytes
, dst_bytes
, 1);
3685 default: /* i.e. case coding_type_no_conversion: */
3686 label_no_conversion
:
3687 if (dst_bytes
&& src_bytes
> dst_bytes
)
3689 coding
->produced
= dst_bytes
;
3690 result
= CODING_FINISH_INSUFFICIENT_DST
;
3694 coding
->produced
= src_bytes
;
3695 result
= CODING_FINISH_NORMAL
;
3698 bcopy (source
, destination
, coding
->produced
);
3700 safe_bcopy (source
, destination
, coding
->produced
);
3701 if (coding
->mode
& CODING_MODE_SELECTIVE_DISPLAY
)
3703 unsigned char *p
= destination
, *pend
= p
+ coding
->produced
;
3705 if (*p
++ == '\015') p
[-1] = '\n';
3707 coding
->fake_multibyte
= 1;
3709 = coding
->consumed_char
= coding
->produced_char
= coding
->produced
;
3716 /* Scan text in the region between *BEG and *END (byte positions),
3717 skip characters which we don't have to decode by coding system
3718 CODING at the head and tail, then set *BEG and *END to the region
3719 of the text we actually have to convert. The caller should move
3720 the gap out of the region in advance.
3722 If STR is not NULL, *BEG and *END are indices into STR. */
3725 shrink_decoding_region (beg
, end
, coding
, str
)
3727 struct coding_system
*coding
;
3730 unsigned char *begp_orig
, *begp
, *endp_orig
, *endp
, c
;
3733 if (coding
->type
== coding_type_ccl
3734 || coding
->type
== coding_type_undecided
3735 || !NILP (coding
->post_read_conversion
))
3737 /* We can't skip any data. */
3740 else if (coding
->type
== coding_type_no_conversion
)
3742 /* We need no conversion, but don't have to skip any data here.
3743 Decoding routine handles them effectively anyway. */
3747 eol_conversion
= (coding
->eol_type
!= CODING_EOL_LF
);
3749 if ((! eol_conversion
) && (coding
->heading_ascii
>= 0))
3750 /* Detection routine has already found how much we can skip at the
3752 *beg
+= coding
->heading_ascii
;
3756 begp_orig
= begp
= str
+ *beg
;
3757 endp_orig
= endp
= str
+ *end
;
3761 begp_orig
= begp
= BYTE_POS_ADDR (*beg
);
3762 endp_orig
= endp
= begp
+ *end
- *beg
;
3765 switch (coding
->type
)
3767 case coding_type_emacs_mule
:
3768 case coding_type_raw_text
:
3771 if (coding
->heading_ascii
< 0)
3772 while (begp
< endp
&& *begp
!= '\r' && *begp
< 0x80) begp
++;
3773 while (begp
< endp
&& endp
[-1] != '\r' && endp
[-1] < 0x80)
3775 /* Do not consider LF as ascii if preceded by CR, since that
3776 confuses eol decoding. */
3777 if (begp
< endp
&& endp
< endp_orig
&& endp
[-1] == '\r' && endp
[0] == '\n')
3784 case coding_type_sjis
:
3785 case coding_type_big5
:
3786 /* We can skip all ASCII characters at the head. */
3787 if (coding
->heading_ascii
< 0)
3790 while (begp
< endp
&& *begp
< 0x80 && *begp
!= '\r') begp
++;
3792 while (begp
< endp
&& *begp
< 0x80) begp
++;
3794 /* We can skip all ASCII characters at the tail except for the
3795 second byte of SJIS or BIG5 code. */
3797 while (begp
< endp
&& endp
[-1] < 0x80 && endp
[-1] != '\r') endp
--;
3799 while (begp
< endp
&& endp
[-1] < 0x80) endp
--;
3800 /* Do not consider LF as ascii if preceded by CR, since that
3801 confuses eol decoding. */
3802 if (begp
< endp
&& endp
< endp_orig
&& endp
[-1] == '\r' && endp
[0] == '\n')
3804 if (begp
< endp
&& endp
< endp_orig
&& endp
[-1] >= 0x80)
3808 default: /* i.e. case coding_type_iso2022: */
3809 if (coding
->heading_ascii
< 0)
3811 /* We can skip all ASCII characters at the head except for a
3812 few control codes. */
3813 while (begp
< endp
&& (c
= *begp
) < 0x80
3814 && c
!= ISO_CODE_CR
&& c
!= ISO_CODE_SO
3815 && c
!= ISO_CODE_SI
&& c
!= ISO_CODE_ESC
3816 && (!eol_conversion
|| c
!= ISO_CODE_LF
))
3819 switch (coding
->category_idx
)
3821 case CODING_CATEGORY_IDX_ISO_8_1
:
3822 case CODING_CATEGORY_IDX_ISO_8_2
:
3823 /* We can skip all ASCII characters at the tail. */
3825 while (begp
< endp
&& (c
= endp
[-1]) < 0x80 && c
!= '\r') endp
--;
3827 while (begp
< endp
&& endp
[-1] < 0x80) endp
--;
3828 /* Do not consider LF as ascii if preceded by CR, since that
3829 confuses eol decoding. */
3830 if (begp
< endp
&& endp
< endp_orig
&& endp
[-1] == '\r' && endp
[0] == '\n')
3834 case CODING_CATEGORY_IDX_ISO_7
:
3835 case CODING_CATEGORY_IDX_ISO_7_TIGHT
:
3836 /* We can skip all charactes at the tail except for ESC and
3837 the following 2-byte at the tail. */
3840 && (c
= endp
[-1]) < 0x80 && c
!= ISO_CODE_ESC
&& c
!= '\r')
3844 && (c
= endp
[-1]) < 0x80 && c
!= ISO_CODE_ESC
)
3846 /* Do not consider LF as ascii if preceded by CR, since that
3847 confuses eol decoding. */
3848 if (begp
< endp
&& endp
< endp_orig
&& endp
[-1] == '\r' && endp
[0] == '\n')
3850 if (begp
< endp
&& endp
[-1] == ISO_CODE_ESC
)
3852 if (endp
+ 1 < endp_orig
&& end
[0] == '(' && end
[1] == 'B')
3853 /* This is an ASCII designation sequence. We can
3854 surely skip the tail. */
3857 /* Hmmm, we can't skip the tail. */
3862 *beg
+= begp
- begp_orig
;
3863 *end
+= endp
- endp_orig
;
3867 /* Like shrink_decoding_region but for encoding. */
3870 shrink_encoding_region (beg
, end
, coding
, str
)
3872 struct coding_system
*coding
;
3875 unsigned char *begp_orig
, *begp
, *endp_orig
, *endp
;
3878 if (coding
->type
== coding_type_ccl
)
3879 /* We can't skip any data. */
3881 else if (coding
->type
== coding_type_no_conversion
)
3883 /* We need no conversion. */
3890 begp_orig
= begp
= str
+ *beg
;
3891 endp_orig
= endp
= str
+ *end
;
3895 begp_orig
= begp
= BYTE_POS_ADDR (*beg
);
3896 endp_orig
= endp
= begp
+ *end
- *beg
;
3899 eol_conversion
= (coding
->eol_type
== CODING_EOL_CR
3900 || coding
->eol_type
== CODING_EOL_CRLF
);
3902 /* Here, we don't have to check coding->pre_write_conversion because
3903 the caller is expected to have handled it already. */
3904 switch (coding
->type
)
3906 case coding_type_undecided
:
3907 case coding_type_emacs_mule
:
3908 case coding_type_raw_text
:
3911 while (begp
< endp
&& *begp
!= '\n') begp
++;
3912 while (begp
< endp
&& endp
[-1] != '\n') endp
--;
3918 case coding_type_iso2022
:
3919 if (coding
->flags
& CODING_FLAG_ISO_DESIGNATE_AT_BOL
)
3921 unsigned char *bol
= begp
;
3922 while (begp
< endp
&& *begp
< 0x80)
3925 if (begp
[-1] == '\n')
3929 goto label_skip_tail
;
3934 /* We can skip all ASCII characters at the head and tail. */
3936 while (begp
< endp
&& *begp
< 0x80 && *begp
!= '\n') begp
++;
3938 while (begp
< endp
&& *begp
< 0x80) begp
++;
3941 while (begp
< endp
&& endp
[-1] < 0x80 && endp
[-1] != '\n') endp
--;
3943 while (begp
< endp
&& *(endp
- 1) < 0x80) endp
--;
3947 *beg
+= begp
- begp_orig
;
3948 *end
+= endp
- endp_orig
;
3952 /* Decode (if ENCODEP is zero) or encode (if ENCODEP is nonzero) the
3953 text from FROM to TO (byte positions are FROM_BYTE and TO_BYTE) by
3954 coding system CODING, and return the status code of code conversion
3955 (currently, this value has no meaning).
3957 How many characters (and bytes) are converted to how many
3958 characters (and bytes) are recorded in members of the structure
3961 If REPLACE is nonzero, we do various things as if the original text
3962 is deleted and a new text is inserted. See the comments in
3963 replace_range (insdel.c) to know what we are doing. */
3966 code_convert_region (from
, from_byte
, to
, to_byte
, coding
, encodep
, replace
)
3967 int from
, from_byte
, to
, to_byte
, encodep
, replace
;
3968 struct coding_system
*coding
;
3970 int len
= to
- from
, len_byte
= to_byte
- from_byte
;
3971 int require
, inserted
, inserted_byte
;
3972 int head_skip
, tail_skip
, total_skip
;
3973 Lisp_Object saved_coding_symbol
= Qnil
;
3974 int multibyte
= !NILP (current_buffer
->enable_multibyte_characters
);
3976 int fake_multibyte
= 0;
3977 unsigned char *src
, *dst
;
3978 Lisp_Object deletion
= Qnil
;
3980 if (from
< PT
&& PT
< to
)
3981 SET_PT_BOTH (from
, from_byte
);
3985 int saved_from
= from
;
3987 prepare_to_modify_buffer (from
, to
, &from
);
3988 if (saved_from
!= from
)
3992 from_byte
= CHAR_TO_BYTE (from
), to_byte
= CHAR_TO_BYTE (to
);
3994 from_byte
= from
, to_byte
= to
;
3995 len_byte
= to_byte
- from_byte
;
3999 if (! encodep
&& CODING_REQUIRE_DETECTION (coding
))
4001 /* We must detect encoding of text and eol format. */
4003 if (from
< GPT
&& to
> GPT
)
4004 move_gap_both (from
, from_byte
);
4005 if (coding
->type
== coding_type_undecided
)
4007 detect_coding (coding
, BYTE_POS_ADDR (from_byte
), len_byte
);
4008 if (coding
->type
== coding_type_undecided
)
4009 /* It seems that the text contains only ASCII, but we
4010 should not left it undecided because the deeper
4011 decoding routine (decode_coding) tries to detect the
4012 encodings again in vain. */
4013 coding
->type
= coding_type_emacs_mule
;
4015 if (coding
->eol_type
== CODING_EOL_UNDECIDED
)
4017 saved_coding_symbol
= coding
->symbol
;
4018 detect_eol (coding
, BYTE_POS_ADDR (from_byte
), len_byte
);
4019 if (coding
->eol_type
== CODING_EOL_UNDECIDED
)
4020 coding
->eol_type
= CODING_EOL_LF
;
4021 /* We had better recover the original eol format if we
4022 encounter an inconsitent eol format while decoding. */
4023 coding
->mode
|= CODING_MODE_INHIBIT_INCONSISTENT_EOL
;
4027 coding
->consumed_char
= len
, coding
->consumed
= len_byte
;
4030 ? ! CODING_REQUIRE_ENCODING (coding
)
4031 : ! CODING_REQUIRE_DECODING (coding
))
4033 coding
->produced
= len_byte
;
4036 /* See the comment of the member heading_ascii in coding.h. */
4037 && coding
->heading_ascii
< len_byte
)
4039 /* We still may have to combine byte at the head and the
4040 tail of the text in the region. */
4041 if (from
< GPT
&& GPT
< to
)
4042 move_gap_both (to
, to_byte
);
4043 len
= multibyte_chars_in_text (BYTE_POS_ADDR (from_byte
), len_byte
);
4044 adjust_after_insert (from
, from_byte
, to
, to_byte
, len
);
4045 coding
->produced_char
= len
;
4050 adjust_after_insert (from
, from_byte
, to
, to_byte
, len_byte
);
4051 coding
->produced_char
= len_byte
;
4056 /* Now we convert the text. */
4058 /* For encoding, we must process pre-write-conversion in advance. */
4060 && ! NILP (coding
->pre_write_conversion
)
4061 && SYMBOLP (coding
->pre_write_conversion
)
4062 && ! NILP (Ffboundp (coding
->pre_write_conversion
)))
4064 /* The function in pre-write-conversion may put a new text in a
4066 struct buffer
*prev
= current_buffer
, *new;
4068 call2 (coding
->pre_write_conversion
,
4069 make_number (from
), make_number (to
));
4070 if (current_buffer
!= prev
)
4073 new = current_buffer
;
4074 set_buffer_internal_1 (prev
);
4075 del_range_2 (from
, from_byte
, to
, to_byte
);
4076 insert_from_buffer (new, BEG
, len
, 0);
4078 to_byte
= multibyte
? CHAR_TO_BYTE (to
) : to
;
4079 len_byte
= to_byte
- from_byte
;
4084 deletion
= make_buffer_string_both (from
, from_byte
, to
, to_byte
, 1);
4086 /* Try to skip the heading and tailing ASCIIs. */
4088 int from_byte_orig
= from_byte
, to_byte_orig
= to_byte
;
4090 if (from
< GPT
&& GPT
< to
)
4091 move_gap_both (from
, from_byte
);
4093 shrink_encoding_region (&from_byte
, &to_byte
, coding
, NULL
);
4095 shrink_decoding_region (&from_byte
, &to_byte
, coding
, NULL
);
4096 if (from_byte
== to_byte
)
4098 coding
->produced
= len_byte
;
4099 coding
->produced_char
= multibyte
? len
: len_byte
;
4101 /* We must record and adjust for this new text now. */
4102 adjust_after_insert (from
, from_byte_orig
, to
, to_byte_orig
, len
);
4106 head_skip
= from_byte
- from_byte_orig
;
4107 tail_skip
= to_byte_orig
- to_byte
;
4108 total_skip
= head_skip
+ tail_skip
;
4111 len
-= total_skip
; len_byte
-= total_skip
;
4114 /* For converion, we must put the gap before the text in addition to
4115 making the gap larger for efficient decoding. The required gap
4116 size starts from 2000 which is the magic number used in make_gap.
4117 But, after one batch of conversion, it will be incremented if we
4118 find that it is not enough . */
4121 if (GAP_SIZE
< require
)
4122 make_gap (require
- GAP_SIZE
);
4123 move_gap_both (from
, from_byte
);
4125 if (GPT
- BEG
< beg_unchanged
)
4126 beg_unchanged
= GPT
- BEG
;
4127 if (Z
- GPT
< end_unchanged
)
4128 end_unchanged
= Z
- GPT
;
4130 inserted
= inserted_byte
= 0;
4131 src
= GAP_END_ADDR
, dst
= GPT_ADDR
;
4133 GAP_SIZE
+= len_byte
;
4136 ZV_BYTE
-= len_byte
;
4143 /* The buffer memory is changed from:
4144 +--------+converted-text+---------+-------original-text------+---+
4145 |<-from->|<--inserted-->|---------|<-----------len---------->|---|
4146 |<------------------- GAP_SIZE -------------------->| */
4148 result
= encode_coding (coding
, src
, dst
, len_byte
, 0);
4150 result
= decode_coding (coding
, src
, dst
, len_byte
, 0);
4152 +--------+-------converted-text--------+--+---original-text--+---+
4153 |<-from->|<--inserted-->|<--produced-->|--|<-(len-consumed)->|---|
4154 |<------------------- GAP_SIZE -------------------->| */
4155 if (coding
->fake_multibyte
)
4158 if (!encodep
&& !multibyte
)
4159 coding
->produced_char
= coding
->produced
;
4160 inserted
+= coding
->produced_char
;
4161 inserted_byte
+= coding
->produced
;
4162 len_byte
-= coding
->consumed
;
4163 src
+= coding
->consumed
;
4164 dst
+= inserted_byte
;
4166 if (! encodep
&& result
== CODING_FINISH_INCONSISTENT_EOL
)
4168 unsigned char *pend
= dst
, *p
= pend
- inserted_byte
;
4170 /* Encode LFs back to the original eol format (CR or CRLF). */
4171 if (coding
->eol_type
== CODING_EOL_CR
)
4173 while (p
< pend
) if (*p
++ == '\n') p
[-1] = '\r';
4179 while (p
< pend
) if (*p
++ == '\n') count
++;
4180 if (src
- dst
< count
)
4182 /* We don't have sufficient room for putting LFs
4183 back to CRLF. We must record converted and
4184 not-yet-converted text back to the buffer
4185 content, enlarge the gap, then record them out of
4186 the buffer contents again. */
4187 int add
= len_byte
+ inserted_byte
;
4190 ZV
+= add
; Z
+= add
; ZV_BYTE
+= add
; Z_BYTE
+= add
;
4191 GPT
+= inserted_byte
; GPT_BYTE
+= inserted_byte
;
4192 make_gap (count
- GAP_SIZE
);
4194 ZV
-= add
; Z
-= add
; ZV_BYTE
-= add
; Z_BYTE
-= add
;
4195 GPT
-= inserted_byte
; GPT_BYTE
-= inserted_byte
;
4196 /* Don't forget to update SRC, DST, and PEND. */
4197 src
= GAP_END_ADDR
- len_byte
;
4198 dst
= GPT_ADDR
+ inserted_byte
;
4202 inserted_byte
+= count
;
4203 coding
->produced
+= count
;
4204 p
= dst
= pend
+ count
;
4208 if (*p
== '\n') count
--, *--p
= '\r';
4212 /* Suppress eol-format conversion in the further conversion. */
4213 coding
->eol_type
= CODING_EOL_LF
;
4215 /* Restore the original symbol. */
4216 coding
->symbol
= saved_coding_symbol
;
4222 if (result
== CODING_FINISH_INSUFFICIENT_SRC
)
4224 /* The source text ends in invalid codes. Let's just
4225 make them valid buffer contents, and finish conversion. */
4226 inserted
+= len_byte
;
4227 inserted_byte
+= len_byte
;
4235 /* We have just done the first batch of conversion which was
4236 stoped because of insufficient gap. Let's reconsider the
4237 required gap size (i.e. SRT - DST) now.
4239 We have converted ORIG bytes (== coding->consumed) into
4240 NEW bytes (coding->produced). To convert the remaining
4241 LEN bytes, we may need REQUIRE bytes of gap, where:
4242 REQUIRE + LEN_BYTE = LEN_BYTE * (NEW / ORIG)
4243 REQUIRE = LEN_BYTE * (NEW - ORIG) / ORIG
4244 Here, we are sure that NEW >= ORIG. */
4245 float ratio
= coding
->produced
- coding
->consumed
;
4246 ratio
/= coding
->consumed
;
4247 require
= len_byte
* ratio
;
4250 if ((src
- dst
) < (require
+ 2000))
4252 /* See the comment above the previous call of make_gap. */
4253 int add
= len_byte
+ inserted_byte
;
4256 ZV
+= add
; Z
+= add
; ZV_BYTE
+= add
; Z_BYTE
+= add
;
4257 GPT
+= inserted_byte
; GPT_BYTE
+= inserted_byte
;
4258 make_gap (require
+ 2000);
4260 ZV
-= add
; Z
-= add
; ZV_BYTE
-= add
; Z_BYTE
-= add
;
4261 GPT
-= inserted_byte
; GPT_BYTE
-= inserted_byte
;
4262 /* Don't forget to update SRC, DST. */
4263 src
= GAP_END_ADDR
- len_byte
;
4264 dst
= GPT_ADDR
+ inserted_byte
;
4267 if (src
- dst
> 0) *dst
= 0; /* Put an anchor. */
4271 || !encodep
&& (to
- from
) != (to_byte
- from_byte
)))
4272 inserted
= multibyte_chars_in_text (GPT_ADDR
, inserted_byte
);
4274 /* If we have shrinked the conversion area, adjust it now. */
4278 safe_bcopy (GAP_END_ADDR
, GPT_ADDR
+ inserted_byte
, tail_skip
);
4279 inserted
+= total_skip
; inserted_byte
+= total_skip
;
4280 GAP_SIZE
+= total_skip
;
4281 GPT
-= head_skip
; GPT_BYTE
-= head_skip
;
4282 ZV
-= total_skip
; ZV_BYTE
-= total_skip
;
4283 Z
-= total_skip
; Z_BYTE
-= total_skip
;
4284 from
-= head_skip
; from_byte
-= head_skip
;
4285 to
+= tail_skip
; to_byte
+= tail_skip
;
4288 adjust_after_replace (from
, from_byte
, deletion
, inserted
, inserted_byte
);
4290 if (! encodep
&& ! NILP (coding
->post_read_conversion
))
4293 int orig_inserted
= inserted
, pos
= PT
;
4296 temp_set_point_both (current_buffer
, from
, from_byte
);
4297 val
= call1 (coding
->post_read_conversion
, make_number (inserted
));
4300 CHECK_NUMBER (val
, 0);
4301 inserted
= XFASTINT (val
);
4303 if (pos
>= from
+ orig_inserted
)
4304 temp_set_point (current_buffer
, pos
+ (inserted
- orig_inserted
));
4307 signal_after_change (from
, to
- from
, inserted
);
4310 coding
->consumed
= to_byte
- from_byte
;
4311 coding
->consumed_char
= to
- from
;
4312 coding
->produced
= inserted_byte
;
4313 coding
->produced_char
= inserted
;
4320 code_convert_string (str
, coding
, encodep
, nocopy
)
4322 struct coding_system
*coding
;
4323 int encodep
, nocopy
;
4327 int from
= 0, to
= XSTRING (str
)->size
;
4328 int to_byte
= STRING_BYTES (XSTRING (str
));
4329 struct gcpro gcpro1
;
4330 Lisp_Object saved_coding_symbol
= Qnil
;
4333 if (encodep
&& !NILP (coding
->pre_write_conversion
)
4334 || !encodep
&& !NILP (coding
->post_read_conversion
))
4336 /* Since we have to call Lisp functions which assume target text
4337 is in a buffer, after setting a temporary buffer, call
4338 code_convert_region. */
4339 int count
= specpdl_ptr
- specpdl
;
4340 struct buffer
*prev
= current_buffer
;
4342 record_unwind_protect (Fset_buffer
, Fcurrent_buffer ());
4343 temp_output_buffer_setup (" *code-converting-work*");
4344 set_buffer_internal (XBUFFER (Vstandard_output
));
4346 insert_from_string (str
, 0, 0, to
, to_byte
, 0);
4349 /* We must insert the contents of STR as is without
4350 unibyte<->multibyte conversion. */
4351 current_buffer
->enable_multibyte_characters
= Qnil
;
4352 insert_from_string (str
, 0, 0, to_byte
, to_byte
, 0);
4353 current_buffer
->enable_multibyte_characters
= Qt
;
4355 code_convert_region (BEGV
, BEGV_BYTE
, ZV
, ZV_BYTE
, coding
, encodep
, 1);
4357 /* We must return the buffer contents as unibyte string. */
4358 current_buffer
->enable_multibyte_characters
= Qnil
;
4359 str
= make_buffer_string (BEGV
, ZV
, 0);
4360 set_buffer_internal (prev
);
4361 return unbind_to (count
, str
);
4364 if (! encodep
&& CODING_REQUIRE_DETECTION (coding
))
4366 /* See the comments in code_convert_region. */
4367 if (coding
->type
== coding_type_undecided
)
4369 detect_coding (coding
, XSTRING (str
)->data
, to_byte
);
4370 if (coding
->type
== coding_type_undecided
)
4371 coding
->type
= coding_type_emacs_mule
;
4373 if (coding
->eol_type
== CODING_EOL_UNDECIDED
)
4375 saved_coding_symbol
= coding
->symbol
;
4376 detect_eol (coding
, XSTRING (str
)->data
, to_byte
);
4377 if (coding
->eol_type
== CODING_EOL_UNDECIDED
)
4378 coding
->eol_type
= CODING_EOL_LF
;
4379 /* We had better recover the original eol format if we
4380 encounter an inconsitent eol format while decoding. */
4381 coding
->mode
|= CODING_MODE_INHIBIT_INCONSISTENT_EOL
;
4386 ? ! CODING_REQUIRE_ENCODING (coding
)
4387 : ! CODING_REQUIRE_DECODING (coding
))
4391 /* Try to skip the heading and tailing ASCIIs. */
4393 shrink_encoding_region (&from
, &to_byte
, coding
, XSTRING (str
)->data
);
4395 shrink_decoding_region (&from
, &to_byte
, coding
, XSTRING (str
)->data
);
4397 if (from
== to_byte
)
4398 return (nocopy
? str
: Fcopy_sequence (str
));
4401 len
= encoding_buffer_size (coding
, to_byte
- from
);
4403 len
= decoding_buffer_size (coding
, to_byte
- from
);
4404 len
+= from
+ STRING_BYTES (XSTRING (str
)) - to_byte
;
4406 buf
= get_conversion_buffer (len
);
4410 bcopy (XSTRING (str
)->data
, buf
, from
);
4412 ? encode_coding (coding
, XSTRING (str
)->data
+ from
,
4413 buf
+ from
, to_byte
- from
, len
)
4414 : decode_coding (coding
, XSTRING (str
)->data
+ from
,
4415 buf
+ from
, to_byte
- from
, len
));
4416 if (! encodep
&& result
== CODING_FINISH_INCONSISTENT_EOL
)
4418 /* We simple try to decode the whole string again but without
4419 eol-conversion this time. */
4420 coding
->eol_type
= CODING_EOL_LF
;
4421 coding
->symbol
= saved_coding_symbol
;
4422 return code_convert_string (str
, coding
, encodep
, nocopy
);
4425 bcopy (XSTRING (str
)->data
+ to_byte
, buf
+ from
+ coding
->produced
,
4426 STRING_BYTES (XSTRING (str
)) - to_byte
);
4428 len
= from
+ STRING_BYTES (XSTRING (str
)) - to_byte
;
4430 str
= make_unibyte_string (buf
, len
+ coding
->produced
);
4432 str
= make_string_from_bytes (buf
, len
+ coding
->produced_char
,
4433 len
+ coding
->produced
);
4439 /*** 7. Emacs Lisp library functions ***/
4441 DEFUN ("coding-system-p", Fcoding_system_p
, Scoding_system_p
, 1, 1, 0,
4442 "Return t if OBJECT is nil or a coding-system.\n\
4443 See the documentation of `make-coding-system' for information\n\
4444 about coding-system objects.")
4452 /* Get coding-spec vector for OBJ. */
4453 obj
= Fget (obj
, Qcoding_system
);
4454 return ((VECTORP (obj
) && XVECTOR (obj
)->size
== 5)
4458 DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system
,
4459 Sread_non_nil_coding_system
, 1, 1, 0,
4460 "Read a coding system from the minibuffer, prompting with string PROMPT.")
4467 val
= Fcompleting_read (prompt
, Vcoding_system_alist
, Qnil
,
4468 Qt
, Qnil
, Qcoding_system_history
, Qnil
, Qnil
);
4470 while (XSTRING (val
)->size
== 0);
4471 return (Fintern (val
, Qnil
));
4474 DEFUN ("read-coding-system", Fread_coding_system
, Sread_coding_system
, 1, 2, 0,
4475 "Read a coding system from the minibuffer, prompting with string PROMPT.\n\
4476 If the user enters null input, return second argument DEFAULT-CODING-SYSTEM.")
4477 (prompt
, default_coding_system
)
4478 Lisp_Object prompt
, default_coding_system
;
4481 if (SYMBOLP (default_coding_system
))
4482 XSETSTRING (default_coding_system
, XSYMBOL (default_coding_system
)->name
);
4483 val
= Fcompleting_read (prompt
, Vcoding_system_alist
, Qnil
,
4484 Qt
, Qnil
, Qcoding_system_history
,
4485 default_coding_system
, Qnil
);
4486 return (XSTRING (val
)->size
== 0 ? Qnil
: Fintern (val
, Qnil
));
4489 DEFUN ("check-coding-system", Fcheck_coding_system
, Scheck_coding_system
,
4491 "Check validity of CODING-SYSTEM.\n\
4492 If valid, return CODING-SYSTEM, else signal a `coding-system-error' error.\n\
4493 It is valid if it is a symbol with a non-nil `coding-system' property.\n\
4494 The value of property should be a vector of length 5.")
4496 Lisp_Object coding_system
;
4498 CHECK_SYMBOL (coding_system
, 0);
4499 if (!NILP (Fcoding_system_p (coding_system
)))
4500 return coding_system
;
4502 Fsignal (Qcoding_system_error
, Fcons (coding_system
, Qnil
));
4506 detect_coding_system (src
, src_bytes
, highest
)
4508 int src_bytes
, highest
;
4510 int coding_mask
, eol_type
;
4511 Lisp_Object val
, tmp
;
4514 coding_mask
= detect_coding_mask (src
, src_bytes
, NULL
, &dummy
);
4515 eol_type
= detect_eol_type (src
, src_bytes
, &dummy
);
4516 if (eol_type
== CODING_EOL_INCONSISTENT
)
4517 eol_type
== CODING_EOL_UNDECIDED
;
4522 if (eol_type
!= CODING_EOL_UNDECIDED
)
4525 val2
= Fget (Qundecided
, Qeol_type
);
4527 val
= XVECTOR (val2
)->contents
[eol_type
];
4529 return (highest
? val
: Fcons (val
, Qnil
));
4532 /* At first, gather possible coding systems in VAL. */
4534 for (tmp
= Vcoding_category_list
; !NILP (tmp
); tmp
= XCONS (tmp
)->cdr
)
4537 = XFASTINT (Fget (XCONS (tmp
)->car
, Qcoding_category_index
));
4538 if (coding_mask
& (1 << idx
))
4540 val
= Fcons (Fsymbol_value (XCONS (tmp
)->car
), val
);
4546 val
= Fnreverse (val
);
4548 /* Then, replace the elements with subsidiary coding systems. */
4549 for (tmp
= val
; !NILP (tmp
); tmp
= XCONS (tmp
)->cdr
)
4551 if (eol_type
!= CODING_EOL_UNDECIDED
4552 && eol_type
!= CODING_EOL_INCONSISTENT
)
4555 eol
= Fget (XCONS (tmp
)->car
, Qeol_type
);
4557 XCONS (tmp
)->car
= XVECTOR (eol
)->contents
[eol_type
];
4560 return (highest
? XCONS (val
)->car
: val
);
4563 DEFUN ("detect-coding-region", Fdetect_coding_region
, Sdetect_coding_region
,
4565 "Detect coding system of the text in the region between START and END.\n\
4566 Return a list of possible coding systems ordered by priority.\n\
4568 If only ASCII characters are found, it returns a list of single element\n\
4569 `undecided' or its subsidiary coding system according to a detected\n\
4570 end-of-line format.\n\
4572 If optional argument HIGHEST is non-nil, return the coding system of\n\
4574 (start
, end
, highest
)
4575 Lisp_Object start
, end
, highest
;
4578 int from_byte
, to_byte
;
4580 CHECK_NUMBER_COERCE_MARKER (start
, 0);
4581 CHECK_NUMBER_COERCE_MARKER (end
, 1);
4583 validate_region (&start
, &end
);
4584 from
= XINT (start
), to
= XINT (end
);
4585 from_byte
= CHAR_TO_BYTE (from
);
4586 to_byte
= CHAR_TO_BYTE (to
);
4588 if (from
< GPT
&& to
>= GPT
)
4589 move_gap_both (to
, to_byte
);
4591 return detect_coding_system (BYTE_POS_ADDR (from_byte
),
4592 to_byte
- from_byte
,
4596 DEFUN ("detect-coding-string", Fdetect_coding_string
, Sdetect_coding_string
,
4598 "Detect coding system of the text in STRING.\n\
4599 Return a list of possible coding systems ordered by priority.\n\
4601 If only ASCII characters are found, it returns a list of single element\n\
4602 `undecided' or its subsidiary coding system according to a detected\n\
4603 end-of-line format.\n\
4605 If optional argument HIGHEST is non-nil, return the coding system of\n\
4608 Lisp_Object string
, highest
;
4610 CHECK_STRING (string
, 0);
4612 return detect_coding_system (XSTRING (string
)->data
,
4613 STRING_BYTES (XSTRING (string
)),
4618 code_convert_region1 (start
, end
, coding_system
, encodep
)
4619 Lisp_Object start
, end
, coding_system
;
4622 struct coding_system coding
;
4625 CHECK_NUMBER_COERCE_MARKER (start
, 0);
4626 CHECK_NUMBER_COERCE_MARKER (end
, 1);
4627 CHECK_SYMBOL (coding_system
, 2);
4629 validate_region (&start
, &end
);
4630 from
= XFASTINT (start
);
4631 to
= XFASTINT (end
);
4633 if (NILP (coding_system
))
4634 return make_number (to
- from
);
4636 if (setup_coding_system (Fcheck_coding_system (coding_system
), &coding
) < 0)
4637 error ("Invalid coding system: %s", XSYMBOL (coding_system
)->name
->data
);
4639 coding
.mode
|= CODING_MODE_LAST_BLOCK
;
4640 code_convert_region (from
, CHAR_TO_BYTE (from
), to
, CHAR_TO_BYTE (to
),
4641 &coding
, encodep
, 1);
4642 Vlast_coding_system_used
= coding
.symbol
;
4643 return make_number (coding
.produced_char
);
4646 DEFUN ("decode-coding-region", Fdecode_coding_region
, Sdecode_coding_region
,
4647 3, 3, "r\nzCoding system: ",
4648 "Decode the current region by specified coding system.\n\
4649 When called from a program, takes three arguments:\n\
4650 START, END, and CODING-SYSTEM. START and END are buffer positions.\n\
4651 This function sets `last-coding-system-used' to the precise coding system\n\
4652 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\
4653 not fully specified.)\n\
4654 It returns the length of the decoded text.")
4655 (start
, end
, coding_system
)
4656 Lisp_Object start
, end
, coding_system
;
4658 return code_convert_region1 (start
, end
, coding_system
, 0);
4661 DEFUN ("encode-coding-region", Fencode_coding_region
, Sencode_coding_region
,
4662 3, 3, "r\nzCoding system: ",
4663 "Encode the current region by specified coding system.\n\
4664 When called from a program, takes three arguments:\n\
4665 START, END, and CODING-SYSTEM. START and END are buffer positions.\n\
4666 This function sets `last-coding-system-used' to the precise coding system\n\
4667 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\
4668 not fully specified.)\n\
4669 It returns the length of the encoded text.")
4670 (start
, end
, coding_system
)
4671 Lisp_Object start
, end
, coding_system
;
4673 return code_convert_region1 (start
, end
, coding_system
, 1);
4677 code_convert_string1 (string
, coding_system
, nocopy
, encodep
)
4678 Lisp_Object string
, coding_system
, nocopy
;
4681 struct coding_system coding
;
4683 CHECK_STRING (string
, 0);
4684 CHECK_SYMBOL (coding_system
, 1);
4686 if (NILP (coding_system
))
4687 return (NILP (nocopy
) ? Fcopy_sequence (string
) : string
);
4689 if (setup_coding_system (Fcheck_coding_system (coding_system
), &coding
) < 0)
4690 error ("Invalid coding system: %s", XSYMBOL (coding_system
)->name
->data
);
4692 coding
.mode
|= CODING_MODE_LAST_BLOCK
;
4693 Vlast_coding_system_used
= coding
.symbol
;
4694 return code_convert_string (string
, &coding
, encodep
, !NILP (nocopy
));
4697 DEFUN ("decode-coding-string", Fdecode_coding_string
, Sdecode_coding_string
,
4699 "Decode STRING which is encoded in CODING-SYSTEM, and return the result.\n\
4700 Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\
4701 if the decoding operation is trivial.\n\
4702 This function sets `last-coding-system-used' to the precise coding system\n\
4703 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\
4704 not fully specified.)")
4705 (string
, coding_system
, nocopy
)
4706 Lisp_Object string
, coding_system
, nocopy
;
4708 return code_convert_string1 (string
, coding_system
, nocopy
, 0);
4711 DEFUN ("encode-coding-string", Fencode_coding_string
, Sencode_coding_string
,
4713 "Encode STRING to CODING-SYSTEM, and return the result.\n\
4714 Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\
4715 if the encoding operation is trivial.\n\
4716 This function sets `last-coding-system-used' to the precise coding system\n\
4717 used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\
4718 not fully specified.)")
4719 (string
, coding_system
, nocopy
)
4720 Lisp_Object string
, coding_system
, nocopy
;
4722 return code_convert_string1 (string
, coding_system
, nocopy
, 1);
4725 /* Encode or decode STRING according to CODING_SYSTEM.
4726 Do not set Vlast_coding_system_used. */
4729 code_convert_string_norecord (string
, coding_system
, encodep
)
4730 Lisp_Object string
, coding_system
;
4733 struct coding_system coding
;
4735 CHECK_STRING (string
, 0);
4736 CHECK_SYMBOL (coding_system
, 1);
4738 if (NILP (coding_system
))
4741 if (setup_coding_system (Fcheck_coding_system (coding_system
), &coding
) < 0)
4742 error ("Invalid coding system: %s", XSYMBOL (coding_system
)->name
->data
);
4744 coding
.mode
|= CODING_MODE_LAST_BLOCK
;
4745 return code_convert_string (string
, &coding
, encodep
, Qt
);
4748 DEFUN ("decode-sjis-char", Fdecode_sjis_char
, Sdecode_sjis_char
, 1, 1, 0,
4749 "Decode a JISX0208 character of shift-jis encoding.\n\
4750 CODE is the character code in SJIS.\n\
4751 Return the corresponding character.")
4755 unsigned char c1
, c2
, s1
, s2
;
4758 CHECK_NUMBER (code
, 0);
4759 s1
= (XFASTINT (code
)) >> 8, s2
= (XFASTINT (code
)) & 0xFF;
4760 DECODE_SJIS (s1
, s2
, c1
, c2
);
4761 XSETFASTINT (val
, MAKE_NON_ASCII_CHAR (charset_jisx0208
, c1
, c2
));
4765 DEFUN ("encode-sjis-char", Fencode_sjis_char
, Sencode_sjis_char
, 1, 1, 0,
4766 "Encode a JISX0208 character CHAR to SJIS coding system.\n\
4767 Return the corresponding character code in SJIS.")
4771 int charset
, c1
, c2
, s1
, s2
;
4774 CHECK_NUMBER (ch
, 0);
4775 SPLIT_CHAR (XFASTINT (ch
), charset
, c1
, c2
);
4776 if (charset
== charset_jisx0208
)
4778 ENCODE_SJIS (c1
, c2
, s1
, s2
);
4779 XSETFASTINT (val
, (s1
<< 8) | s2
);
4782 XSETFASTINT (val
, 0);
4786 DEFUN ("decode-big5-char", Fdecode_big5_char
, Sdecode_big5_char
, 1, 1, 0,
4787 "Decode a Big5 character CODE of BIG5 coding system.\n\
4788 CODE is the character code in BIG5.\n\
4789 Return the corresponding character.")
4794 unsigned char b1
, b2
, c1
, c2
;
4797 CHECK_NUMBER (code
, 0);
4798 b1
= (XFASTINT (code
)) >> 8, b2
= (XFASTINT (code
)) & 0xFF;
4799 DECODE_BIG5 (b1
, b2
, charset
, c1
, c2
);
4800 XSETFASTINT (val
, MAKE_NON_ASCII_CHAR (charset
, c1
, c2
));
4804 DEFUN ("encode-big5-char", Fencode_big5_char
, Sencode_big5_char
, 1, 1, 0,
4805 "Encode the Big5 character CHAR to BIG5 coding system.\n\
4806 Return the corresponding character code in Big5.")
4810 int charset
, c1
, c2
, b1
, b2
;
4813 CHECK_NUMBER (ch
, 0);
4814 SPLIT_CHAR (XFASTINT (ch
), charset
, c1
, c2
);
4815 if (charset
== charset_big5_1
|| charset
== charset_big5_2
)
4817 ENCODE_BIG5 (charset
, c1
, c2
, b1
, b2
);
4818 XSETFASTINT (val
, (b1
<< 8) | b2
);
4821 XSETFASTINT (val
, 0);
4825 DEFUN ("set-terminal-coding-system-internal",
4826 Fset_terminal_coding_system_internal
,
4827 Sset_terminal_coding_system_internal
, 1, 1, 0, "")
4829 Lisp_Object coding_system
;
4831 CHECK_SYMBOL (coding_system
, 0);
4832 setup_coding_system (Fcheck_coding_system (coding_system
), &terminal_coding
);
4833 /* We had better not send unsafe characters to terminal. */
4834 terminal_coding
.flags
|= CODING_FLAG_ISO_SAFE
;
4839 DEFUN ("set-safe-terminal-coding-system-internal",
4840 Fset_safe_terminal_coding_system_internal
,
4841 Sset_safe_terminal_coding_system_internal
, 1, 1, 0, "")
4843 Lisp_Object coding_system
;
4845 CHECK_SYMBOL (coding_system
, 0);
4846 setup_coding_system (Fcheck_coding_system (coding_system
),
4847 &safe_terminal_coding
);
4851 DEFUN ("terminal-coding-system",
4852 Fterminal_coding_system
, Sterminal_coding_system
, 0, 0, 0,
4853 "Return coding system specified for terminal output.")
4856 return terminal_coding
.symbol
;
4859 DEFUN ("set-keyboard-coding-system-internal",
4860 Fset_keyboard_coding_system_internal
,
4861 Sset_keyboard_coding_system_internal
, 1, 1, 0, "")
4863 Lisp_Object coding_system
;
4865 CHECK_SYMBOL (coding_system
, 0);
4866 setup_coding_system (Fcheck_coding_system (coding_system
), &keyboard_coding
);
4870 DEFUN ("keyboard-coding-system",
4871 Fkeyboard_coding_system
, Skeyboard_coding_system
, 0, 0, 0,
4872 "Return coding system specified for decoding keyboard input.")
4875 return keyboard_coding
.symbol
;
4879 DEFUN ("find-operation-coding-system", Ffind_operation_coding_system
,
4880 Sfind_operation_coding_system
, 1, MANY
, 0,
4881 "Choose a coding system for an operation based on the target name.\n\
4882 The value names a pair of coding systems: (DECODING-SYSTEM ENCODING-SYSTEM).\n\
4883 DECODING-SYSTEM is the coding system to use for decoding\n\
4884 \(in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system\n\
4885 for encoding (in case OPERATION does encoding).\n\
4887 The first argument OPERATION specifies an I/O primitive:\n\
4888 For file I/O, `insert-file-contents' or `write-region'.\n\
4889 For process I/O, `call-process', `call-process-region', or `start-process'.\n\
4890 For network I/O, `open-network-stream'.\n\
4892 The remaining arguments should be the same arguments that were passed\n\
4893 to the primitive. Depending on which primitive, one of those arguments\n\
4894 is selected as the TARGET. For example, if OPERATION does file I/O,\n\
4895 whichever argument specifies the file name is TARGET.\n\
4897 TARGET has a meaning which depends on OPERATION:\n\
4898 For file I/O, TARGET is a file name.\n\
4899 For process I/O, TARGET is a process name.\n\
4900 For network I/O, TARGET is a service name or a port number\n\
4902 This function looks up what specified for TARGET in,\n\
4903 `file-coding-system-alist', `process-coding-system-alist',\n\
4904 or `network-coding-system-alist' depending on OPERATION.\n\
4905 They may specify a coding system, a cons of coding systems,\n\
4906 or a function symbol to call.\n\
4907 In the last case, we call the function with one argument,\n\
4908 which is a list of all the arguments given to this function.")
4913 Lisp_Object operation
, target_idx
, target
, val
;
4914 register Lisp_Object chain
;
4917 error ("Too few arguments");
4918 operation
= args
[0];
4919 if (!SYMBOLP (operation
)
4920 || !INTEGERP (target_idx
= Fget (operation
, Qtarget_idx
)))
4921 error ("Invalid first arguement");
4922 if (nargs
< 1 + XINT (target_idx
))
4923 error ("Too few arguments for operation: %s",
4924 XSYMBOL (operation
)->name
->data
);
4925 target
= args
[XINT (target_idx
) + 1];
4926 if (!(STRINGP (target
)
4927 || (EQ (operation
, Qopen_network_stream
) && INTEGERP (target
))))
4928 error ("Invalid %dth argument", XINT (target_idx
) + 1);
4930 chain
= ((EQ (operation
, Qinsert_file_contents
)
4931 || EQ (operation
, Qwrite_region
))
4932 ? Vfile_coding_system_alist
4933 : (EQ (operation
, Qopen_network_stream
)
4934 ? Vnetwork_coding_system_alist
4935 : Vprocess_coding_system_alist
));
4939 for (; CONSP (chain
); chain
= XCONS (chain
)->cdr
)
4942 elt
= XCONS (chain
)->car
;
4945 && ((STRINGP (target
)
4946 && STRINGP (XCONS (elt
)->car
)
4947 && fast_string_match (XCONS (elt
)->car
, target
) >= 0)
4948 || (INTEGERP (target
) && EQ (target
, XCONS (elt
)->car
))))
4950 val
= XCONS (elt
)->cdr
;
4951 /* Here, if VAL is both a valid coding system and a valid
4952 function symbol, we return VAL as a coding system. */
4955 if (! SYMBOLP (val
))
4957 if (! NILP (Fcoding_system_p (val
)))
4958 return Fcons (val
, val
);
4959 if (! NILP (Ffboundp (val
)))
4961 val
= call1 (val
, Flist (nargs
, args
));
4964 if (SYMBOLP (val
) && ! NILP (Fcoding_system_p (val
)))
4965 return Fcons (val
, val
);
4973 DEFUN ("update-iso-coding-systems", Fupdate_iso_coding_systems
,
4974 Supdate_iso_coding_systems
, 0, 0, 0,
4975 "Update internal database for ISO2022 based coding systems.\n\
4976 When values of the following coding categories are changed, you must\n\
4977 call this function:\n\
4978 coding-category-iso-7, coding-category-iso-7-tight,\n\
4979 coding-category-iso-8-1, coding-category-iso-8-2,\n\
4980 coding-category-iso-7-else, coding-category-iso-8-else")
4985 for (i
= CODING_CATEGORY_IDX_ISO_7
; i
<= CODING_CATEGORY_IDX_ISO_8_ELSE
;
4988 if (! coding_system_table
[i
])
4989 coding_system_table
[i
]
4990 = (struct coding_system
*) xmalloc (sizeof (struct coding_system
));
4992 (XSYMBOL (XVECTOR (Vcoding_category_table
)->contents
[i
])->value
,
4993 coding_system_table
[i
]);
4998 DEFUN ("set-coding-priority-internal", Fset_coding_priority_internal
,
4999 Sset_coding_priority_internal
, 0, 0, 0,
5000 "Update internal database for the current value of `coding-category-list'.\n\
5001 This function is internal use only.")
5005 Lisp_Object val
= Vcoding_category_list
;
5007 while (CONSP (val
) && i
< CODING_CATEGORY_IDX_MAX
)
5009 if (! SYMBOLP (XCONS (val
)->car
))
5011 idx
= XFASTINT (Fget (XCONS (val
)->car
, Qcoding_category_index
));
5012 if (idx
>= CODING_CATEGORY_IDX_MAX
)
5014 coding_priorities
[i
++] = (1 << idx
);
5015 val
= XCONS (val
)->cdr
;
5017 /* If coding-category-list is valid and contains all coding
5018 categories, `i' should be CODING_CATEGORY_IDX_MAX now. If not,
5019 the following code saves Emacs from craching. */
5020 while (i
< CODING_CATEGORY_IDX_MAX
)
5021 coding_priorities
[i
++] = CODING_CATEGORY_MASK_RAW_TEXT
;
5029 /*** 8. Post-amble ***/
5036 /* Emacs' internal format specific initialize routine. */
5037 for (i
= 0; i
<= 0x20; i
++)
5038 emacs_code_class
[i
] = EMACS_control_code
;
5039 emacs_code_class
[0x0A] = EMACS_linefeed_code
;
5040 emacs_code_class
[0x0D] = EMACS_carriage_return_code
;
5041 for (i
= 0x21 ; i
< 0x7F; i
++)
5042 emacs_code_class
[i
] = EMACS_ascii_code
;
5043 emacs_code_class
[0x7F] = EMACS_control_code
;
5044 emacs_code_class
[0x80] = EMACS_leading_code_composition
;
5045 for (i
= 0x81; i
< 0xFF; i
++)
5046 emacs_code_class
[i
] = EMACS_invalid_code
;
5047 emacs_code_class
[LEADING_CODE_PRIVATE_11
] = EMACS_leading_code_3
;
5048 emacs_code_class
[LEADING_CODE_PRIVATE_12
] = EMACS_leading_code_3
;
5049 emacs_code_class
[LEADING_CODE_PRIVATE_21
] = EMACS_leading_code_4
;
5050 emacs_code_class
[LEADING_CODE_PRIVATE_22
] = EMACS_leading_code_4
;
5052 /* ISO2022 specific initialize routine. */
5053 for (i
= 0; i
< 0x20; i
++)
5054 iso_code_class
[i
] = ISO_control_code
;
5055 for (i
= 0x21; i
< 0x7F; i
++)
5056 iso_code_class
[i
] = ISO_graphic_plane_0
;
5057 for (i
= 0x80; i
< 0xA0; i
++)
5058 iso_code_class
[i
] = ISO_control_code
;
5059 for (i
= 0xA1; i
< 0xFF; i
++)
5060 iso_code_class
[i
] = ISO_graphic_plane_1
;
5061 iso_code_class
[0x20] = iso_code_class
[0x7F] = ISO_0x20_or_0x7F
;
5062 iso_code_class
[0xA0] = iso_code_class
[0xFF] = ISO_0xA0_or_0xFF
;
5063 iso_code_class
[ISO_CODE_CR
] = ISO_carriage_return
;
5064 iso_code_class
[ISO_CODE_SO
] = ISO_shift_out
;
5065 iso_code_class
[ISO_CODE_SI
] = ISO_shift_in
;
5066 iso_code_class
[ISO_CODE_SS2_7
] = ISO_single_shift_2_7
;
5067 iso_code_class
[ISO_CODE_ESC
] = ISO_escape
;
5068 iso_code_class
[ISO_CODE_SS2
] = ISO_single_shift_2
;
5069 iso_code_class
[ISO_CODE_SS3
] = ISO_single_shift_3
;
5070 iso_code_class
[ISO_CODE_CSI
] = ISO_control_sequence_introducer
;
5072 conversion_buffer_size
= MINIMUM_CONVERSION_BUFFER_SIZE
;
5073 conversion_buffer
= (char *) xmalloc (MINIMUM_CONVERSION_BUFFER_SIZE
);
5075 setup_coding_system (Qnil
, &keyboard_coding
);
5076 setup_coding_system (Qnil
, &terminal_coding
);
5077 setup_coding_system (Qnil
, &safe_terminal_coding
);
5079 bzero (coding_system_table
, sizeof coding_system_table
);
5081 bzero (ascii_skip_code
, sizeof ascii_skip_code
);
5082 for (i
= 0; i
< 128; i
++)
5083 ascii_skip_code
[i
] = 1;
5085 #if defined (MSDOS) || defined (WINDOWSNT)
5086 system_eol_type
= CODING_EOL_CRLF
;
5088 system_eol_type
= CODING_EOL_LF
;
5097 Qtarget_idx
= intern ("target-idx");
5098 staticpro (&Qtarget_idx
);
5100 Qcoding_system_history
= intern ("coding-system-history");
5101 staticpro (&Qcoding_system_history
);
5102 Fset (Qcoding_system_history
, Qnil
);
5104 /* Target FILENAME is the first argument. */
5105 Fput (Qinsert_file_contents
, Qtarget_idx
, make_number (0));
5106 /* Target FILENAME is the third argument. */
5107 Fput (Qwrite_region
, Qtarget_idx
, make_number (2));
5109 Qcall_process
= intern ("call-process");
5110 staticpro (&Qcall_process
);
5111 /* Target PROGRAM is the first argument. */
5112 Fput (Qcall_process
, Qtarget_idx
, make_number (0));
5114 Qcall_process_region
= intern ("call-process-region");
5115 staticpro (&Qcall_process_region
);
5116 /* Target PROGRAM is the third argument. */
5117 Fput (Qcall_process_region
, Qtarget_idx
, make_number (2));
5119 Qstart_process
= intern ("start-process");
5120 staticpro (&Qstart_process
);
5121 /* Target PROGRAM is the third argument. */
5122 Fput (Qstart_process
, Qtarget_idx
, make_number (2));
5124 Qopen_network_stream
= intern ("open-network-stream");
5125 staticpro (&Qopen_network_stream
);
5126 /* Target SERVICE is the fourth argument. */
5127 Fput (Qopen_network_stream
, Qtarget_idx
, make_number (3));
5129 Qcoding_system
= intern ("coding-system");
5130 staticpro (&Qcoding_system
);
5132 Qeol_type
= intern ("eol-type");
5133 staticpro (&Qeol_type
);
5135 Qbuffer_file_coding_system
= intern ("buffer-file-coding-system");
5136 staticpro (&Qbuffer_file_coding_system
);
5138 Qpost_read_conversion
= intern ("post-read-conversion");
5139 staticpro (&Qpost_read_conversion
);
5141 Qpre_write_conversion
= intern ("pre-write-conversion");
5142 staticpro (&Qpre_write_conversion
);
5144 Qno_conversion
= intern ("no-conversion");
5145 staticpro (&Qno_conversion
);
5147 Qundecided
= intern ("undecided");
5148 staticpro (&Qundecided
);
5150 Qcoding_system_p
= intern ("coding-system-p");
5151 staticpro (&Qcoding_system_p
);
5153 Qcoding_system_error
= intern ("coding-system-error");
5154 staticpro (&Qcoding_system_error
);
5156 Fput (Qcoding_system_error
, Qerror_conditions
,
5157 Fcons (Qcoding_system_error
, Fcons (Qerror
, Qnil
)));
5158 Fput (Qcoding_system_error
, Qerror_message
,
5159 build_string ("Invalid coding system"));
5161 Qcoding_category
= intern ("coding-category");
5162 staticpro (&Qcoding_category
);
5163 Qcoding_category_index
= intern ("coding-category-index");
5164 staticpro (&Qcoding_category_index
);
5166 Vcoding_category_table
5167 = Fmake_vector (make_number (CODING_CATEGORY_IDX_MAX
), Qnil
);
5168 staticpro (&Vcoding_category_table
);
5171 for (i
= 0; i
< CODING_CATEGORY_IDX_MAX
; i
++)
5173 XVECTOR (Vcoding_category_table
)->contents
[i
]
5174 = intern (coding_category_name
[i
]);
5175 Fput (XVECTOR (Vcoding_category_table
)->contents
[i
],
5176 Qcoding_category_index
, make_number (i
));
5180 Qtranslation_table
= intern ("translation-table");
5181 staticpro (&Qtranslation_table
);
5182 Fput (Qtranslation_table
, Qchar_table_extra_slots
, make_number (0));
5184 Qtranslation_table_id
= intern ("translation-table-id");
5185 staticpro (&Qtranslation_table_id
);
5187 Qtranslation_table_for_decode
= intern ("translation-table-for-decode");
5188 staticpro (&Qtranslation_table_for_decode
);
5190 Qtranslation_table_for_encode
= intern ("translation-table-for-encode");
5191 staticpro (&Qtranslation_table_for_encode
);
5193 Qsafe_charsets
= intern ("safe-charsets");
5194 staticpro (&Qsafe_charsets
);
5196 Qemacs_mule
= intern ("emacs-mule");
5197 staticpro (&Qemacs_mule
);
5199 Qraw_text
= intern ("raw-text");
5200 staticpro (&Qraw_text
);
5202 defsubr (&Scoding_system_p
);
5203 defsubr (&Sread_coding_system
);
5204 defsubr (&Sread_non_nil_coding_system
);
5205 defsubr (&Scheck_coding_system
);
5206 defsubr (&Sdetect_coding_region
);
5207 defsubr (&Sdetect_coding_string
);
5208 defsubr (&Sdecode_coding_region
);
5209 defsubr (&Sencode_coding_region
);
5210 defsubr (&Sdecode_coding_string
);
5211 defsubr (&Sencode_coding_string
);
5212 defsubr (&Sdecode_sjis_char
);
5213 defsubr (&Sencode_sjis_char
);
5214 defsubr (&Sdecode_big5_char
);
5215 defsubr (&Sencode_big5_char
);
5216 defsubr (&Sset_terminal_coding_system_internal
);
5217 defsubr (&Sset_safe_terminal_coding_system_internal
);
5218 defsubr (&Sterminal_coding_system
);
5219 defsubr (&Sset_keyboard_coding_system_internal
);
5220 defsubr (&Skeyboard_coding_system
);
5221 defsubr (&Sfind_operation_coding_system
);
5222 defsubr (&Supdate_iso_coding_systems
);
5223 defsubr (&Sset_coding_priority_internal
);
5225 DEFVAR_LISP ("coding-system-list", &Vcoding_system_list
,
5226 "List of coding systems.\n\
5228 Do not alter the value of this variable manually. This variable should be\n\
5229 updated by the functions `make-coding-system' and\n\
5230 `define-coding-system-alias'.");
5231 Vcoding_system_list
= Qnil
;
5233 DEFVAR_LISP ("coding-system-alist", &Vcoding_system_alist
,
5234 "Alist of coding system names.\n\
5235 Each element is one element list of coding system name.\n\
5236 This variable is given to `completing-read' as TABLE argument.\n\
5238 Do not alter the value of this variable manually. This variable should be\n\
5239 updated by the functions `make-coding-system' and\n\
5240 `define-coding-system-alias'.");
5241 Vcoding_system_alist
= Qnil
;
5243 DEFVAR_LISP ("coding-category-list", &Vcoding_category_list
,
5244 "List of coding-categories (symbols) ordered by priority.");
5248 Vcoding_category_list
= Qnil
;
5249 for (i
= CODING_CATEGORY_IDX_MAX
- 1; i
>= 0; i
--)
5250 Vcoding_category_list
5251 = Fcons (XVECTOR (Vcoding_category_table
)->contents
[i
],
5252 Vcoding_category_list
);
5255 DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read
,
5256 "Specify the coding system for read operations.\n\
5257 It is useful to bind this variable with `let', but do not set it globally.\n\
5258 If the value is a coding system, it is used for decoding on read operation.\n\
5259 If not, an appropriate element is used from one of the coding system alists:\n\
5260 There are three such tables, `file-coding-system-alist',\n\
5261 `process-coding-system-alist', and `network-coding-system-alist'.");
5262 Vcoding_system_for_read
= Qnil
;
5264 DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write
,
5265 "Specify the coding system for write operations.\n\
5266 It is useful to bind this variable with `let', but do not set it globally.\n\
5267 If the value is a coding system, it is used for encoding on write operation.\n\
5268 If not, an appropriate element is used from one of the coding system alists:\n\
5269 There are three such tables, `file-coding-system-alist',\n\
5270 `process-coding-system-alist', and `network-coding-system-alist'.");
5271 Vcoding_system_for_write
= Qnil
;
5273 DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used
,
5274 "Coding system used in the latest file or process I/O.");
5275 Vlast_coding_system_used
= Qnil
;
5277 DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion
,
5278 "*Non-nil inhibit code conversion of end-of-line format in any cases.");
5279 inhibit_eol_conversion
= 0;
5281 DEFVAR_BOOL ("inherit-process-coding-system", &inherit_process_coding_system
,
5282 "Non-nil means process buffer inherits coding system of process output.\n\
5283 Bind it to t if the process output is to be treated as if it were a file\n\
5284 read from some filesystem.");
5285 inherit_process_coding_system
= 0;
5287 DEFVAR_LISP ("file-coding-system-alist", &Vfile_coding_system_alist
,
5288 "Alist to decide a coding system to use for a file I/O operation.\n\
5289 The format is ((PATTERN . VAL) ...),\n\
5290 where PATTERN is a regular expression matching a file name,\n\
5291 VAL is a coding system, a cons of coding systems, or a function symbol.\n\
5292 If VAL is a coding system, it is used for both decoding and encoding\n\
5293 the file contents.\n\
5294 If VAL is a cons of coding systems, the car part is used for decoding,\n\
5295 and the cdr part is used for encoding.\n\
5296 If VAL is a function symbol, the function must return a coding system\n\
5297 or a cons of coding systems which are used as above.\n\
5299 See also the function `find-operation-coding-system'.");
5300 Vfile_coding_system_alist
= Qnil
;
5302 DEFVAR_LISP ("process-coding-system-alist", &Vprocess_coding_system_alist
,
5303 "Alist to decide a coding system to use for a process I/O operation.\n\
5304 The format is ((PATTERN . VAL) ...),\n\
5305 where PATTERN is a regular expression matching a program name,\n\
5306 VAL is a coding system, a cons of coding systems, or a function symbol.\n\
5307 If VAL is a coding system, it is used for both decoding what received\n\
5308 from the program and encoding what sent to the program.\n\
5309 If VAL is a cons of coding systems, the car part is used for decoding,\n\
5310 and the cdr part is used for encoding.\n\
5311 If VAL is a function symbol, the function must return a coding system\n\
5312 or a cons of coding systems which are used as above.\n\
5314 See also the function `find-operation-coding-system'.");
5315 Vprocess_coding_system_alist
= Qnil
;
5317 DEFVAR_LISP ("network-coding-system-alist", &Vnetwork_coding_system_alist
,
5318 "Alist to decide a coding system to use for a network I/O operation.\n\
5319 The format is ((PATTERN . VAL) ...),\n\
5320 where PATTERN is a regular expression matching a network service name\n\
5321 or is a port number to connect to,\n\
5322 VAL is a coding system, a cons of coding systems, or a function symbol.\n\
5323 If VAL is a coding system, it is used for both decoding what received\n\
5324 from the network stream and encoding what sent to the network stream.\n\
5325 If VAL is a cons of coding systems, the car part is used for decoding,\n\
5326 and the cdr part is used for encoding.\n\
5327 If VAL is a function symbol, the function must return a coding system\n\
5328 or a cons of coding systems which are used as above.\n\
5330 See also the function `find-operation-coding-system'.");
5331 Vnetwork_coding_system_alist
= Qnil
;
5333 DEFVAR_INT ("eol-mnemonic-unix", &eol_mnemonic_unix
,
5334 "Mnemonic character indicating UNIX-like end-of-line format (i.e. LF) .");
5335 eol_mnemonic_unix
= ':';
5337 DEFVAR_INT ("eol-mnemonic-dos", &eol_mnemonic_dos
,
5338 "Mnemonic character indicating DOS-like end-of-line format (i.e. CRLF).");
5339 eol_mnemonic_dos
= '\\';
5341 DEFVAR_INT ("eol-mnemonic-mac", &eol_mnemonic_mac
,
5342 "Mnemonic character indicating MAC-like end-of-line format (i.e. CR).");
5343 eol_mnemonic_mac
= '/';
5345 DEFVAR_INT ("eol-mnemonic-undecided", &eol_mnemonic_undecided
,
5346 "Mnemonic character indicating end-of-line format is not yet decided.");
5347 eol_mnemonic_undecided
= ':';
5349 DEFVAR_LISP ("enable-character-translation", &Venable_character_translation
,
5350 "*Non-nil enables character translation while encoding and decoding.");
5351 Venable_character_translation
= Qt
;
5353 DEFVAR_LISP ("standard-translation-table-for-decode",
5354 &Vstandard_translation_table_for_decode
,
5355 "Table for translating characters while decoding.");
5356 Vstandard_translation_table_for_decode
= Qnil
;
5358 DEFVAR_LISP ("standard-translation-table-for-encode",
5359 &Vstandard_translation_table_for_encode
,
5360 "Table for translationg characters while encoding.");
5361 Vstandard_translation_table_for_encode
= Qnil
;
5363 DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist
,
5364 "Alist of charsets vs revision numbers.\n\
5365 While encoding, if a charset (car part of an element) is found,\n\
5366 designate it with the escape sequence identifing revision (cdr part of the element).");
5367 Vcharset_revision_alist
= Qnil
;
5369 DEFVAR_LISP ("default-process-coding-system",
5370 &Vdefault_process_coding_system
,
5371 "Cons of coding systems used for process I/O by default.\n\
5372 The car part is used for decoding a process output,\n\
5373 the cdr part is used for encoding a text to be sent to a process.");
5374 Vdefault_process_coding_system
= Qnil
;
5376 DEFVAR_LISP ("latin-extra-code-table", &Vlatin_extra_code_table
,
5377 "Table of extra Latin codes in the range 128..159 (inclusive).\n\
5378 This is a vector of length 256.\n\
5379 If Nth element is non-nil, the existence of code N in a file\n\
5380 \(or output of subprocess) doesn't prevent it to be detected as\n\
5381 a coding system of ISO 2022 variant which has a flag\n\
5382 `accept-latin-extra-code' t (e.g. iso-latin-1) on reading a file\n\
5383 or reading output of a subprocess.\n\
5384 Only 128th through 159th elements has a meaning.");
5385 Vlatin_extra_code_table
= Fmake_vector (make_number (256), Qnil
);
5387 DEFVAR_LISP ("select-safe-coding-system-function",
5388 &Vselect_safe_coding_system_function
,
5389 "Function to call to select safe coding system for encoding a text.\n\
5391 If set, this function is called to force a user to select a proper\n\
5392 coding system which can encode the text in the case that a default\n\
5393 coding system used in each operation can't encode the text.\n\
5395 The default value is `select-safe-codign-system' (which see).");
5396 Vselect_safe_coding_system_function
= Qnil
;