Commit | Line | Data |
---|---|---|
4ed46869 | 1 | /* Coding system handler (conversion, detection, and etc). |
4a2f9c6a | 2 | Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN. |
203cb916 | 3 | Licensed to the Free Software Foundation. |
4ed46869 | 4 | |
369314dc KH |
5 | This file is part of GNU Emacs. |
6 | ||
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) | |
10 | any later version. | |
4ed46869 | 11 | |
369314dc KH |
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. | |
4ed46869 | 16 | |
369314dc KH |
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. */ | |
4ed46869 KH |
21 | |
22 | /*** TABLE OF CONTENTS *** | |
23 | ||
24 | 1. Preamble | |
0ef69138 | 25 | 2. Emacs' internal format (emacs-mule) handlers |
4ed46869 KH |
26 | 3. ISO2022 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 | |
31 | 8. Post-amble | |
32 | ||
33 | */ | |
34 | ||
35 | /*** GENERAL NOTE on CODING SYSTEM *** | |
36 | ||
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 | |
0ef69138 KH |
40 | Emacs' internal format (emacs-internal), and when we say "encode", |
41 | it means converting the coding system emacs-mule to some other | |
42 | coding system. | |
4ed46869 | 43 | |
0ef69138 | 44 | 0. Emacs' internal format (emacs-mule) |
4ed46869 KH |
45 | |
46 | Emacs itself holds a multi-lingual character in a buffer and a string | |
f4dee582 | 47 | in a special format. Details are described in section 2. |
4ed46869 KH |
48 | |
49 | 1. ISO2022 | |
50 | ||
51 | The most famous coding system for multiple character sets. X's | |
f4dee582 RS |
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. | |
4ed46869 KH |
55 | |
56 | 2. SJIS (or Shift-JIS or MS-Kanji-Code) | |
57 | ||
58 | A coding system to encode character sets: ASCII, JISX0201, and | |
59 | JISX0208. Widely used for PC's in Japan. Details are described in | |
f4dee582 | 60 | section 4. |
4ed46869 KH |
61 | |
62 | 3. BIG5 | |
63 | ||
64 | A coding system to encode character sets: ASCII and Big5. Widely | |
65 | used by Chinese (mainly in Taiwan and Hong Kong). Details are | |
f4dee582 RS |
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. | |
4ed46869 | 69 | |
27901516 KH |
70 | 4. Raw text |
71 | ||
4608c386 KH |
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. | |
27901516 KH |
74 | |
75 | 5. Other | |
4ed46869 | 76 | |
f4dee582 | 77 | If a user wants to read/write a text encoded in a coding system not |
4ed46869 KH |
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. | |
81 | ||
d46c5b12 KH |
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 | |
4ed46869 | 84 | information about it is set in a structure of type `struct |
f4dee582 | 85 | coding_system' for rapid processing. See section 6 for more details. |
4ed46869 KH |
86 | |
87 | */ | |
88 | ||
89 | /*** GENERAL NOTES on END-OF-LINE FORMAT *** | |
90 | ||
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, | |
f4dee582 | 93 | whereas DOS's format is two-byte sequence of `carriage-return' and |
d46c5b12 KH |
94 | `line-feed' codes. MacOS's format is usually one byte of |
95 | `carriage-return'. | |
4ed46869 | 96 | |
f4dee582 RS |
97 | Since text characters encoding and end-of-line encoding are |
98 | independent, any coding system described above can take | |
4ed46869 | 99 | any format of end-of-line. So, Emacs has information of format of |
f4dee582 | 100 | end-of-line in each coding-system. See section 6 for more details. |
4ed46869 KH |
101 | |
102 | */ | |
103 | ||
104 | /*** GENERAL NOTES on `detect_coding_XXX ()' functions *** | |
105 | ||
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. */ | |
111 | #if 0 | |
112 | int | |
0ef69138 | 113 | detect_coding_emacs_mule (src, src_end) |
4ed46869 KH |
114 | unsigned char *src, *src_end; |
115 | { | |
116 | ... | |
117 | } | |
118 | #endif | |
119 | ||
120 | /*** GENERAL NOTES on `decode_coding_XXX ()' functions *** | |
121 | ||
122 | These functions decode SRC_BYTES length text at SOURCE encoded in | |
0ef69138 | 123 | CODING to Emacs' internal format (emacs-mule). The resulting text |
d46c5b12 KH |
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. | |
128 | ||
129 | The return value is an integer (CODING_FINISH_XXX) indicating how | |
130 | the decoding finished. | |
131 | ||
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. | |
135 | ||
136 | Below is a template of these functions. */ | |
4ed46869 | 137 | #if 0 |
d46c5b12 | 138 | decode_coding_XXX (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
139 | struct coding_system *coding; |
140 | unsigned char *source, *destination; | |
141 | int src_bytes, dst_bytes; | |
4ed46869 KH |
142 | { |
143 | ... | |
144 | } | |
145 | #endif | |
146 | ||
147 | /*** GENERAL NOTES on `encode_coding_XXX ()' functions *** | |
148 | ||
0ef69138 KH |
149 | These functions encode SRC_BYTES length text at SOURCE of Emacs' |
150 | internal format (emacs-mule) to CODING. The resulting text goes to | |
f4dee582 | 151 | a place pointed to by DESTINATION, the length of which should not |
d46c5b12 KH |
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. | |
155 | ||
156 | The return value is an integer (CODING_FINISH_XXX) indicating how | |
157 | the encoding finished. | |
158 | ||
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. | |
162 | ||
163 | Below is a template of these functions. */ | |
4ed46869 | 164 | #if 0 |
d46c5b12 | 165 | encode_coding_XXX (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
166 | struct coding_system *coding; |
167 | unsigned char *source, *destination; | |
168 | int src_bytes, dst_bytes; | |
4ed46869 KH |
169 | { |
170 | ... | |
171 | } | |
172 | #endif | |
173 | ||
174 | /*** COMMONLY USED MACROS ***/ | |
175 | ||
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. */ | |
181 | ||
182 | #define ONE_MORE_BYTE(c1) \ | |
183 | do { \ | |
184 | if (src < src_end) \ | |
185 | c1 = *src++; \ | |
186 | else \ | |
187 | goto label_end_of_loop; \ | |
188 | } while (0) | |
189 | ||
190 | #define TWO_MORE_BYTES(c1, c2) \ | |
191 | do { \ | |
192 | if (src + 1 < src_end) \ | |
193 | c1 = *src++, c2 = *src++; \ | |
194 | else \ | |
195 | goto label_end_of_loop; \ | |
196 | } while (0) | |
197 | ||
198 | #define THREE_MORE_BYTES(c1, c2, c3) \ | |
199 | do { \ | |
200 | if (src + 2 < src_end) \ | |
201 | c1 = *src++, c2 = *src++, c3 = *src++; \ | |
202 | else \ | |
203 | goto label_end_of_loop; \ | |
204 | } while (0) | |
205 | ||
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. */ | |
212 | ||
213 | /* Decode one ASCII character C. */ | |
214 | ||
215 | #define DECODE_CHARACTER_ASCII(c) \ | |
216 | do { \ | |
217 | if (COMPOSING_P (coding->composing)) \ | |
218 | *dst++ = 0xA0, *dst++ = (c) | 0x80; \ | |
219 | else \ | |
d46c5b12 KH |
220 | { \ |
221 | *dst++ = (c); \ | |
222 | coding->produced_char++; \ | |
223 | } \ | |
4ed46869 KH |
224 | } while (0) |
225 | ||
f4dee582 | 226 | /* Decode one DIMENSION1 character whose charset is CHARSET and whose |
4ed46869 KH |
227 | position-code is C. */ |
228 | ||
229 | #define DECODE_CHARACTER_DIMENSION1(charset, c) \ | |
230 | do { \ | |
231 | unsigned char leading_code = CHARSET_LEADING_CODE_BASE (charset); \ | |
232 | if (COMPOSING_P (coding->composing)) \ | |
233 | *dst++ = leading_code + 0x20; \ | |
234 | else \ | |
d46c5b12 KH |
235 | { \ |
236 | *dst++ = leading_code; \ | |
237 | coding->produced_char++; \ | |
238 | } \ | |
4ed46869 KH |
239 | if (leading_code = CHARSET_LEADING_CODE_EXT (charset)) \ |
240 | *dst++ = leading_code; \ | |
241 | *dst++ = (c) | 0x80; \ | |
242 | } while (0) | |
243 | ||
f4dee582 | 244 | /* Decode one DIMENSION2 character whose charset is CHARSET and whose |
4ed46869 KH |
245 | position-codes are C1 and C2. */ |
246 | ||
247 | #define DECODE_CHARACTER_DIMENSION2(charset, c1, c2) \ | |
248 | do { \ | |
249 | DECODE_CHARACTER_DIMENSION1 (charset, c1); \ | |
250 | *dst++ = (c2) | 0x80; \ | |
251 | } while (0) | |
252 | ||
253 | \f | |
254 | /*** 1. Preamble ***/ | |
255 | ||
256 | #include <stdio.h> | |
257 | ||
258 | #ifdef emacs | |
259 | ||
260 | #include <config.h> | |
261 | #include "lisp.h" | |
262 | #include "buffer.h" | |
263 | #include "charset.h" | |
264 | #include "ccl.h" | |
265 | #include "coding.h" | |
266 | #include "window.h" | |
267 | ||
268 | #else /* not emacs */ | |
269 | ||
270 | #include "mulelib.h" | |
271 | ||
272 | #endif /* not emacs */ | |
273 | ||
274 | Lisp_Object Qcoding_system, Qeol_type; | |
275 | Lisp_Object Qbuffer_file_coding_system; | |
276 | Lisp_Object Qpost_read_conversion, Qpre_write_conversion; | |
27901516 | 277 | Lisp_Object Qno_conversion, Qundecided; |
bb0115a2 | 278 | Lisp_Object Qcoding_system_history; |
70c22245 | 279 | Lisp_Object Qsafe_charsets; |
4ed46869 KH |
280 | |
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; | |
285 | ||
d46c5b12 KH |
286 | Lisp_Object Vselect_safe_coding_system_function; |
287 | ||
4ed46869 KH |
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 | |
291 | decided. */ | |
292 | int eol_mnemonic_undecided; | |
293 | ||
9ce27fde KH |
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. */ | |
296 | int system_eol_type; | |
297 | ||
4ed46869 KH |
298 | #ifdef emacs |
299 | ||
4608c386 KH |
300 | Lisp_Object Vcoding_system_list, Vcoding_system_alist; |
301 | ||
302 | Lisp_Object Qcoding_system_p, Qcoding_system_error; | |
4ed46869 | 303 | |
d46c5b12 KH |
304 | /* Coding system emacs-mule and raw-text are for converting only |
305 | end-of-line format. */ | |
306 | Lisp_Object Qemacs_mule, Qraw_text; | |
9ce27fde | 307 | |
4ed46869 KH |
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; | |
316 | ||
c4825358 | 317 | /* A vector of length 256 which contains information about special |
3f003981 KH |
318 | Latin codes (espepcially for dealing with Microsoft code). */ |
319 | Lisp_Object Vlatin_extra_code_table; | |
c4825358 | 320 | |
9ce27fde KH |
321 | /* Flag to inhibit code conversion of end-of-line format. */ |
322 | int inhibit_eol_conversion; | |
323 | ||
c4825358 | 324 | /* Coding system to be used to encode text for terminal display. */ |
4ed46869 KH |
325 | struct coding_system terminal_coding; |
326 | ||
c4825358 KH |
327 | /* Coding system to be used to encode text for terminal display when |
328 | terminal coding system is nil. */ | |
329 | struct coding_system safe_terminal_coding; | |
330 | ||
331 | /* Coding system of what is sent from terminal keyboard. */ | |
4ed46869 KH |
332 | struct coding_system keyboard_coding; |
333 | ||
02ba4723 KH |
334 | Lisp_Object Vfile_coding_system_alist; |
335 | Lisp_Object Vprocess_coding_system_alist; | |
336 | Lisp_Object Vnetwork_coding_system_alist; | |
4ed46869 KH |
337 | |
338 | #endif /* emacs */ | |
339 | ||
d46c5b12 | 340 | Lisp_Object Qcoding_category, Qcoding_category_index; |
4ed46869 KH |
341 | |
342 | /* List of symbols `coding-category-xxx' ordered by priority. */ | |
343 | Lisp_Object Vcoding_category_list; | |
344 | ||
d46c5b12 KH |
345 | /* Table of coding categories (Lisp symbols). */ |
346 | Lisp_Object Vcoding_category_table; | |
4ed46869 KH |
347 | |
348 | /* Table of names of symbol for each coding-category. */ | |
349 | char *coding_category_name[CODING_CATEGORY_IDX_MAX] = { | |
0ef69138 | 350 | "coding-category-emacs-mule", |
4ed46869 KH |
351 | "coding-category-sjis", |
352 | "coding-category-iso-7", | |
d46c5b12 | 353 | "coding-category-iso-7-tight", |
4ed46869 KH |
354 | "coding-category-iso-8-1", |
355 | "coding-category-iso-8-2", | |
7717c392 KH |
356 | "coding-category-iso-7-else", |
357 | "coding-category-iso-8-else", | |
4ed46869 | 358 | "coding-category-big5", |
27901516 | 359 | "coding-category-raw-text", |
4ed46869 KH |
360 | "coding-category-binary" |
361 | }; | |
362 | ||
d46c5b12 KH |
363 | /* Table pointers to coding systems corresponding to each coding |
364 | categories. */ | |
365 | struct coding_system *coding_system_table[CODING_CATEGORY_IDX_MAX]; | |
366 | ||
bdd9fb48 KH |
367 | /* Flag to tell if we look up unification table on character code |
368 | conversion. */ | |
369 | Lisp_Object Venable_character_unification; | |
a5d301df KH |
370 | /* Standard unification table to look up on decoding (reading). */ |
371 | Lisp_Object Vstandard_character_unification_table_for_decode; | |
372 | /* Standard unification table to look up on encoding (writing). */ | |
373 | Lisp_Object Vstandard_character_unification_table_for_encode; | |
bdd9fb48 KH |
374 | |
375 | Lisp_Object Qcharacter_unification_table; | |
a5d301df KH |
376 | Lisp_Object Qcharacter_unification_table_for_decode; |
377 | Lisp_Object Qcharacter_unification_table_for_encode; | |
4ed46869 KH |
378 | |
379 | /* Alist of charsets vs revision number. */ | |
380 | Lisp_Object Vcharset_revision_alist; | |
381 | ||
02ba4723 KH |
382 | /* Default coding systems used for process I/O. */ |
383 | Lisp_Object Vdefault_process_coding_system; | |
384 | ||
4ed46869 | 385 | \f |
0ef69138 | 386 | /*** 2. Emacs internal format (emacs-mule) handlers ***/ |
4ed46869 KH |
387 | |
388 | /* Emacs' internal format for encoding multiple character sets is a | |
f4dee582 RS |
389 | kind of multi-byte encoding, i.e. characters are encoded by |
390 | variable-length sequences of one-byte codes. ASCII characters | |
391 | and control characters (e.g. `tab', `newline') are represented by | |
392 | one-byte sequences which are their ASCII codes, in the range 0x00 | |
393 | through 0x7F. The other characters are represented by a sequence | |
394 | of `base leading-code', optional `extended leading-code', and one | |
395 | or two `position-code's. The length of the sequence is determined | |
396 | by the base leading-code. Leading-code takes the range 0x80 | |
397 | through 0x9F, whereas extended leading-code and position-code take | |
398 | the range 0xA0 through 0xFF. See `charset.h' for more details | |
399 | about leading-code and position-code. | |
400 | ||
401 | There's one exception to this rule. Special leading-code | |
4ed46869 KH |
402 | `leading-code-composition' denotes that the following several |
403 | characters should be composed into one character. Leading-codes of | |
404 | components (except for ASCII) are added 0x20. An ASCII character | |
405 | component is represented by a 2-byte sequence of `0xA0' and | |
f4dee582 RS |
406 | `ASCII-code + 0x80'. See also the comments in `charset.h' for the |
407 | details of composite character. Hence, we can summarize the code | |
4ed46869 KH |
408 | range as follows: |
409 | ||
410 | --- CODE RANGE of Emacs' internal format --- | |
411 | (character set) (range) | |
412 | ASCII 0x00 .. 0x7F | |
413 | ELSE (1st byte) 0x80 .. 0x9F | |
414 | (rest bytes) 0xA0 .. 0xFF | |
415 | --------------------------------------------- | |
416 | ||
417 | */ | |
418 | ||
419 | enum emacs_code_class_type emacs_code_class[256]; | |
420 | ||
421 | /* Go to the next statement only if *SRC is accessible and the code is | |
422 | greater than 0xA0. */ | |
423 | #define CHECK_CODE_RANGE_A0_FF \ | |
424 | do { \ | |
425 | if (src >= src_end) \ | |
426 | goto label_end_of_switch; \ | |
427 | else if (*src++ < 0xA0) \ | |
428 | return 0; \ | |
429 | } while (0) | |
430 | ||
431 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". | |
432 | Check if a text is encoded in Emacs' internal format. If it is, | |
d46c5b12 | 433 | return CODING_CATEGORY_MASK_EMACS_MULE, else return 0. */ |
4ed46869 KH |
434 | |
435 | int | |
0ef69138 | 436 | detect_coding_emacs_mule (src, src_end) |
4ed46869 KH |
437 | unsigned char *src, *src_end; |
438 | { | |
439 | unsigned char c; | |
440 | int composing = 0; | |
441 | ||
442 | while (src < src_end) | |
443 | { | |
444 | c = *src++; | |
445 | ||
446 | if (composing) | |
447 | { | |
448 | if (c < 0xA0) | |
449 | composing = 0; | |
450 | else | |
451 | c -= 0x20; | |
452 | } | |
453 | ||
454 | switch (emacs_code_class[c]) | |
455 | { | |
456 | case EMACS_ascii_code: | |
457 | case EMACS_linefeed_code: | |
458 | break; | |
459 | ||
460 | case EMACS_control_code: | |
461 | if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) | |
462 | return 0; | |
463 | break; | |
464 | ||
465 | case EMACS_invalid_code: | |
466 | return 0; | |
467 | ||
468 | case EMACS_leading_code_composition: /* c == 0x80 */ | |
469 | if (composing) | |
470 | CHECK_CODE_RANGE_A0_FF; | |
471 | else | |
472 | composing = 1; | |
473 | break; | |
474 | ||
475 | case EMACS_leading_code_4: | |
476 | CHECK_CODE_RANGE_A0_FF; | |
477 | /* fall down to check it two more times ... */ | |
478 | ||
479 | case EMACS_leading_code_3: | |
480 | CHECK_CODE_RANGE_A0_FF; | |
481 | /* fall down to check it one more time ... */ | |
482 | ||
483 | case EMACS_leading_code_2: | |
484 | CHECK_CODE_RANGE_A0_FF; | |
485 | break; | |
486 | ||
487 | default: | |
488 | label_end_of_switch: | |
489 | break; | |
490 | } | |
491 | } | |
0ef69138 | 492 | return CODING_CATEGORY_MASK_EMACS_MULE; |
4ed46869 KH |
493 | } |
494 | ||
495 | \f | |
496 | /*** 3. ISO2022 handlers ***/ | |
497 | ||
498 | /* The following note describes the coding system ISO2022 briefly. | |
f4dee582 RS |
499 | Since the intention of this note is to help in understanding of |
500 | the programs in this file, some parts are NOT ACCURATE or OVERLY | |
4ed46869 KH |
501 | SIMPLIFIED. For the thorough understanding, please refer to the |
502 | original document of ISO2022. | |
503 | ||
504 | ISO2022 provides many mechanisms to encode several character sets | |
f4dee582 | 505 | in 7-bit and 8-bit environment. If one chooses 7-bite environment, |
4ed46869 | 506 | all text is encoded by codes of less than 128. This may make the |
f4dee582 RS |
507 | encoded text a little bit longer, but the text gets more stability |
508 | to pass through several gateways (some of them strip off the MSB). | |
4ed46869 | 509 | |
f4dee582 | 510 | There are two kinds of character set: control character set and |
4ed46869 KH |
511 | graphic character set. The former contains control characters such |
512 | as `newline' and `escape' to provide control functions (control | |
f4dee582 | 513 | functions are provided also by escape sequences). The latter |
4ed46869 KH |
514 | contains graphic characters such as ' A' and '-'. Emacs recognizes |
515 | two control character sets and many graphic character sets. | |
516 | ||
517 | Graphic character sets are classified into one of the following | |
518 | four classes, DIMENSION1_CHARS94, DIMENSION1_CHARS96, | |
519 | DIMENSION2_CHARS94, DIMENSION2_CHARS96 according to the number of | |
520 | bytes (DIMENSION) and the number of characters in one dimension | |
521 | (CHARS) of the set. In addition, each character set is assigned an | |
522 | identification tag (called "final character" and denoted as <F> | |
523 | here after) which is unique in each class. <F> of each character | |
524 | set is decided by ECMA(*) when it is registered in ISO. Code range | |
525 | of <F> is 0x30..0x7F (0x30..0x3F are for private use only). | |
526 | ||
527 | Note (*): ECMA = European Computer Manufacturers Association | |
528 | ||
529 | Here are examples of graphic character set [NAME(<F>)]: | |
530 | o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ... | |
531 | o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ... | |
532 | o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ... | |
533 | o DIMENSION2_CHARS96 -- none for the moment | |
534 | ||
535 | A code area (1byte=8bits) is divided into 4 areas, C0, GL, C1, and GR. | |
536 | C0 [0x00..0x1F] -- control character plane 0 | |
537 | GL [0x20..0x7F] -- graphic character plane 0 | |
538 | C1 [0x80..0x9F] -- control character plane 1 | |
539 | GR [0xA0..0xFF] -- graphic character plane 1 | |
540 | ||
541 | A control character set is directly designated and invoked to C0 or | |
542 | C1 by an escape sequence. The most common case is that ISO646's | |
543 | control character set is designated/invoked to C0 and ISO6429's | |
544 | control character set is designated/invoked to C1, and usually | |
545 | these designations/invocations are omitted in a coded text. With | |
546 | 7-bit environment, only C0 can be used, and a control character for | |
547 | C1 is encoded by an appropriate escape sequence to fit in the | |
548 | environment. All control characters for C1 are defined the | |
549 | corresponding escape sequences. | |
550 | ||
551 | A graphic character set is at first designated to one of four | |
552 | graphic registers (G0 through G3), then these graphic registers are | |
553 | invoked to GL or GR. These designations and invocations can be | |
554 | done independently. The most common case is that G0 is invoked to | |
555 | GL, G1 is invoked to GR, and ASCII is designated to G0, and usually | |
556 | these invocations and designations are omitted in a coded text. | |
557 | With 7-bit environment, only GL can be used. | |
558 | ||
559 | When a graphic character set of CHARS94 is invoked to GL, code 0x20 | |
560 | and 0x7F of GL area work as control characters SPACE and DEL | |
561 | respectively, and code 0xA0 and 0xFF of GR area should not be used. | |
562 | ||
563 | There are two ways of invocation: locking-shift and single-shift. | |
564 | With locking-shift, the invocation lasts until the next different | |
565 | invocation, whereas with single-shift, the invocation works only | |
566 | for the following character and doesn't affect locking-shift. | |
567 | Invocations are done by the following control characters or escape | |
568 | sequences. | |
569 | ||
570 | ---------------------------------------------------------------------- | |
571 | function control char escape sequence description | |
572 | ---------------------------------------------------------------------- | |
573 | SI (shift-in) 0x0F none invoke G0 to GL | |
10bff6f1 | 574 | SO (shift-out) 0x0E none invoke G1 to GL |
4ed46869 KH |
575 | LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL |
576 | LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL | |
577 | SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 into GL | |
578 | SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 into GL | |
579 | ---------------------------------------------------------------------- | |
580 | The first four are for locking-shift. Control characters for these | |
581 | functions are defined by macros ISO_CODE_XXX in `coding.h'. | |
582 | ||
583 | Designations are done by the following escape sequences. | |
584 | ---------------------------------------------------------------------- | |
585 | escape sequence description | |
586 | ---------------------------------------------------------------------- | |
587 | ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0 | |
588 | ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1 | |
589 | ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2 | |
590 | ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3 | |
591 | ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*) | |
592 | ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1 | |
593 | ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2 | |
594 | ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3 | |
595 | ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**) | |
596 | ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1 | |
597 | ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2 | |
598 | ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3 | |
599 | ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*) | |
600 | ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1 | |
601 | ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2 | |
602 | ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3 | |
603 | ---------------------------------------------------------------------- | |
604 | ||
605 | In this list, "DIMENSION1_CHARS94<F>" means a graphic character set | |
606 | of dimension 1, chars 94, and final character <F>, and etc. | |
607 | ||
608 | Note (*): Although these designations are not allowed in ISO2022, | |
609 | Emacs accepts them on decoding, and produces them on encoding | |
610 | CHARS96 character set in a coding system which is characterized as | |
611 | 7-bit environment, non-locking-shift, and non-single-shift. | |
612 | ||
613 | Note (**): If <F> is '@', 'A', or 'B', the intermediate character | |
614 | '(' can be omitted. We call this as "short-form" here after. | |
615 | ||
616 | Now you may notice that there are a lot of ways for encoding the | |
f4dee582 | 617 | same multilingual text in ISO2022. Actually, there exists many |
4ed46869 KH |
618 | coding systems such as Compound Text (used in X's inter client |
619 | communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR | |
620 | (used in Korean Internet), EUC (Extended UNIX Code, used in Asian | |
621 | localized platforms), and all of these are variants of ISO2022. | |
622 | ||
623 | In addition to the above, Emacs handles two more kinds of escape | |
624 | sequences: ISO6429's direction specification and Emacs' private | |
625 | sequence for specifying character composition. | |
626 | ||
627 | ISO6429's direction specification takes the following format: | |
628 | o CSI ']' -- end of the current direction | |
629 | o CSI '0' ']' -- end of the current direction | |
630 | o CSI '1' ']' -- start of left-to-right text | |
631 | o CSI '2' ']' -- start of right-to-left text | |
632 | The control character CSI (0x9B: control sequence introducer) is | |
633 | abbreviated to the escape sequence ESC '[' in 7-bit environment. | |
634 | ||
635 | Character composition specification takes the following format: | |
636 | o ESC '0' -- start character composition | |
637 | o ESC '1' -- end character composition | |
638 | Since these are not standard escape sequences of any ISO, the use | |
639 | of them for these meaning is restricted to Emacs only. */ | |
640 | ||
641 | enum iso_code_class_type iso_code_class[256]; | |
642 | ||
d46c5b12 KH |
643 | #define CHARSET_OK(idx, charset) \ |
644 | (CODING_SPEC_ISO_REQUESTED_DESIGNATION \ | |
645 | (coding_system_table[idx], charset) \ | |
646 | != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION) | |
647 | ||
648 | #define SHIFT_OUT_OK(idx) \ | |
649 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding_system_table[idx], 1) >= 0) | |
650 | ||
4ed46869 KH |
651 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
652 | Check if a text is encoded in ISO2022. If it is, returns an | |
653 | integer in which appropriate flag bits any of: | |
654 | CODING_CATEGORY_MASK_ISO_7 | |
d46c5b12 | 655 | CODING_CATEGORY_MASK_ISO_7_TIGHT |
4ed46869 KH |
656 | CODING_CATEGORY_MASK_ISO_8_1 |
657 | CODING_CATEGORY_MASK_ISO_8_2 | |
7717c392 KH |
658 | CODING_CATEGORY_MASK_ISO_7_ELSE |
659 | CODING_CATEGORY_MASK_ISO_8_ELSE | |
4ed46869 KH |
660 | are set. If a code which should never appear in ISO2022 is found, |
661 | returns 0. */ | |
662 | ||
663 | int | |
664 | detect_coding_iso2022 (src, src_end) | |
665 | unsigned char *src, *src_end; | |
666 | { | |
d46c5b12 KH |
667 | int mask = CODING_CATEGORY_MASK_ISO; |
668 | int mask_found = 0; | |
669 | int reg[4], shift_out = 0; | |
670 | int c, c1, i, charset; | |
3f003981 | 671 | |
d46c5b12 | 672 | reg[0] = CHARSET_ASCII, reg[1] = reg[2] = reg[3] = -1; |
3f003981 | 673 | while (mask && src < src_end) |
4ed46869 KH |
674 | { |
675 | c = *src++; | |
676 | switch (c) | |
677 | { | |
678 | case ISO_CODE_ESC: | |
e0e989f6 | 679 | if (src >= src_end) |
4ed46869 KH |
680 | break; |
681 | c = *src++; | |
d46c5b12 | 682 | if (c >= '(' && c <= '/') |
4ed46869 | 683 | { |
bf9cdd4e KH |
684 | /* Designation sequence for a charset of dimension 1. */ |
685 | if (src >= src_end) | |
686 | break; | |
d46c5b12 KH |
687 | c1 = *src++; |
688 | if (c1 < ' ' || c1 >= 0x80 | |
689 | || (charset = iso_charset_table[0][c >= ','][c1]) < 0) | |
690 | /* Invalid designation sequence. Just ignore. */ | |
691 | break; | |
692 | reg[(c - '(') % 4] = charset; | |
bf9cdd4e KH |
693 | } |
694 | else if (c == '$') | |
695 | { | |
696 | /* Designation sequence for a charset of dimension 2. */ | |
697 | if (src >= src_end) | |
698 | break; | |
699 | c = *src++; | |
700 | if (c >= '@' && c <= 'B') | |
701 | /* Designation for JISX0208.1978, GB2312, or JISX0208. */ | |
d46c5b12 | 702 | reg[0] = charset = iso_charset_table[1][0][c]; |
bf9cdd4e | 703 | else if (c >= '(' && c <= '/') |
bcf26d6a | 704 | { |
bf9cdd4e KH |
705 | if (src >= src_end) |
706 | break; | |
d46c5b12 KH |
707 | c1 = *src++; |
708 | if (c1 < ' ' || c1 >= 0x80 | |
709 | || (charset = iso_charset_table[1][c >= ','][c1]) < 0) | |
710 | /* Invalid designation sequence. Just ignore. */ | |
711 | break; | |
712 | reg[(c - '(') % 4] = charset; | |
bcf26d6a | 713 | } |
bf9cdd4e | 714 | else |
d46c5b12 KH |
715 | /* Invalid designation sequence. Just ignore. */ |
716 | break; | |
717 | } | |
718 | else if (c == 'N' || c == 'n') | |
719 | { | |
720 | if (shift_out == 0 | |
721 | && (reg[1] >= 0 | |
722 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE) | |
723 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))) | |
724 | { | |
725 | /* Locking shift out. */ | |
726 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
727 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
728 | shift_out = 1; | |
729 | } | |
730 | break; | |
731 | } | |
732 | else if (c == 'O' || c == 'o') | |
733 | { | |
734 | if (shift_out == 1) | |
735 | { | |
736 | /* Locking shift in. */ | |
737 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
738 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
739 | shift_out = 0; | |
740 | } | |
741 | break; | |
4ed46869 | 742 | } |
bf9cdd4e | 743 | else if (c == '0' || c == '1' || c == '2') |
d46c5b12 KH |
744 | /* Start/end composition. Just ignore. */ |
745 | break; | |
bf9cdd4e | 746 | else |
d46c5b12 KH |
747 | /* Invalid escape sequence. Just ignore. */ |
748 | break; | |
749 | ||
750 | /* We found a valid designation sequence for CHARSET. */ | |
751 | mask &= ~CODING_CATEGORY_MASK_ISO_8BIT; | |
752 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7, charset)) | |
753 | mask_found |= CODING_CATEGORY_MASK_ISO_7; | |
754 | else | |
755 | mask &= ~CODING_CATEGORY_MASK_ISO_7; | |
756 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT, charset)) | |
757 | mask_found |= CODING_CATEGORY_MASK_ISO_7_TIGHT; | |
758 | else | |
759 | mask &= ~CODING_CATEGORY_MASK_ISO_7_TIGHT; | |
760 | if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE, charset)) | |
761 | mask &= ~CODING_CATEGORY_MASK_ISO_7_ELSE; | |
762 | if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE, charset)) | |
763 | mask &= ~CODING_CATEGORY_MASK_ISO_8_ELSE; | |
4ed46869 KH |
764 | break; |
765 | ||
4ed46869 | 766 | case ISO_CODE_SO: |
d46c5b12 KH |
767 | if (shift_out == 0 |
768 | && (reg[1] >= 0 | |
769 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE) | |
770 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))) | |
771 | { | |
772 | /* Locking shift out. */ | |
773 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
774 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
775 | } | |
e0e989f6 KH |
776 | break; |
777 | ||
d46c5b12 KH |
778 | case ISO_CODE_SI: |
779 | if (shift_out == 1) | |
780 | { | |
781 | /* Locking shift in. */ | |
782 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
783 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
784 | } | |
785 | break; | |
786 | ||
4ed46869 KH |
787 | case ISO_CODE_CSI: |
788 | case ISO_CODE_SS2: | |
789 | case ISO_CODE_SS3: | |
3f003981 KH |
790 | { |
791 | int newmask = CODING_CATEGORY_MASK_ISO_8_ELSE; | |
792 | ||
70c22245 KH |
793 | if (c != ISO_CODE_CSI) |
794 | { | |
d46c5b12 KH |
795 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
796 | & CODING_FLAG_ISO_SINGLE_SHIFT) | |
70c22245 | 797 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
d46c5b12 KH |
798 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
799 | & CODING_FLAG_ISO_SINGLE_SHIFT) | |
70c22245 KH |
800 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
801 | } | |
3f003981 KH |
802 | if (VECTORP (Vlatin_extra_code_table) |
803 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) | |
804 | { | |
d46c5b12 KH |
805 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
806 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 | 807 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
d46c5b12 KH |
808 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
809 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 KH |
810 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
811 | } | |
812 | mask &= newmask; | |
d46c5b12 | 813 | mask_found |= newmask; |
3f003981 KH |
814 | } |
815 | break; | |
4ed46869 KH |
816 | |
817 | default: | |
818 | if (c < 0x80) | |
819 | break; | |
820 | else if (c < 0xA0) | |
c4825358 | 821 | { |
3f003981 KH |
822 | if (VECTORP (Vlatin_extra_code_table) |
823 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) | |
c4825358 | 824 | { |
3f003981 KH |
825 | int newmask = 0; |
826 | ||
d46c5b12 KH |
827 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
828 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 | 829 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
d46c5b12 KH |
830 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
831 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 KH |
832 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
833 | mask &= newmask; | |
d46c5b12 | 834 | mask_found |= newmask; |
c4825358 | 835 | } |
3f003981 KH |
836 | else |
837 | return 0; | |
c4825358 | 838 | } |
4ed46869 KH |
839 | else |
840 | { | |
7717c392 | 841 | unsigned char *src_begin = src; |
4ed46869 | 842 | |
d46c5b12 | 843 | mask &= ~(CODING_CATEGORY_MASK_ISO_7BIT |
7717c392 | 844 | | CODING_CATEGORY_MASK_ISO_7_ELSE); |
d46c5b12 | 845 | mask_found |= CODING_CATEGORY_MASK_ISO_8_1; |
e0e989f6 | 846 | while (src < src_end && *src >= 0xA0) |
7717c392 KH |
847 | src++; |
848 | if ((src - src_begin - 1) & 1 && src < src_end) | |
4ed46869 | 849 | mask &= ~CODING_CATEGORY_MASK_ISO_8_2; |
d46c5b12 KH |
850 | else |
851 | mask_found |= CODING_CATEGORY_MASK_ISO_8_2; | |
4ed46869 KH |
852 | } |
853 | break; | |
854 | } | |
855 | } | |
856 | ||
d46c5b12 | 857 | return (mask & mask_found); |
4ed46869 KH |
858 | } |
859 | ||
860 | /* Decode a character of which charset is CHARSET and the 1st position | |
bdd9fb48 | 861 | code is C1. If dimension of CHARSET is 2, the 2nd position code is |
4ed46869 KH |
862 | fetched from SRC and set to C2. If CHARSET is negative, it means |
863 | that we are decoding ill formed text, and what we can do is just to | |
864 | read C1 as is. */ | |
865 | ||
bdd9fb48 KH |
866 | #define DECODE_ISO_CHARACTER(charset, c1) \ |
867 | do { \ | |
868 | int c_alt, charset_alt = (charset); \ | |
869 | if (COMPOSING_HEAD_P (coding->composing)) \ | |
870 | { \ | |
871 | *dst++ = LEADING_CODE_COMPOSITION; \ | |
872 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ | |
873 | /* To tell composition rules are embeded. */ \ | |
874 | *dst++ = 0xFF; \ | |
875 | coding->composing += 2; \ | |
876 | } \ | |
877 | if ((charset) >= 0) \ | |
878 | { \ | |
879 | if (CHARSET_DIMENSION (charset) == 2) \ | |
70c22245 KH |
880 | { \ |
881 | ONE_MORE_BYTE (c2); \ | |
882 | if (iso_code_class[(c2) & 0x7F] != ISO_0x20_or_0x7F \ | |
883 | && iso_code_class[(c2) & 0x7F] != ISO_graphic_plane_0) \ | |
884 | { \ | |
885 | src--; \ | |
886 | c2 = ' '; \ | |
887 | } \ | |
888 | } \ | |
bdd9fb48 KH |
889 | if (!NILP (unification_table) \ |
890 | && ((c_alt = unify_char (unification_table, \ | |
891 | -1, (charset), c1, c2)) >= 0)) \ | |
892 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
893 | } \ | |
894 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ | |
895 | DECODE_CHARACTER_ASCII (c1); \ | |
896 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
897 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
898 | else \ | |
899 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
900 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ | |
901 | /* To tell a composition rule follows. */ \ | |
902 | coding->composing = COMPOSING_WITH_RULE_RULE; \ | |
4ed46869 KH |
903 | } while (0) |
904 | ||
905 | /* Set designation state into CODING. */ | |
d46c5b12 KH |
906 | #define DECODE_DESIGNATION(reg, dimension, chars, final_char) \ |
907 | do { \ | |
908 | int charset = ISO_CHARSET_TABLE (make_number (dimension), \ | |
909 | make_number (chars), \ | |
910 | make_number (final_char)); \ | |
911 | if (charset >= 0 \ | |
912 | && CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) == reg) \ | |
913 | { \ | |
914 | if (coding->spec.iso2022.last_invalid_designation_register == 0 \ | |
915 | && reg == 0 \ | |
916 | && charset == CHARSET_ASCII) \ | |
917 | { \ | |
918 | /* We should insert this designation sequence as is so \ | |
919 | that it is surely written back to a file. */ \ | |
920 | coding->spec.iso2022.last_invalid_designation_register = -1; \ | |
921 | goto label_invalid_code; \ | |
922 | } \ | |
923 | coding->spec.iso2022.last_invalid_designation_register = -1; \ | |
924 | if ((coding->mode & CODING_MODE_DIRECTION) \ | |
925 | && CHARSET_REVERSE_CHARSET (charset) >= 0) \ | |
926 | charset = CHARSET_REVERSE_CHARSET (charset); \ | |
927 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ | |
928 | } \ | |
929 | else \ | |
930 | { \ | |
931 | coding->spec.iso2022.last_invalid_designation_register = reg; \ | |
932 | goto label_invalid_code; \ | |
933 | } \ | |
4ed46869 KH |
934 | } while (0) |
935 | ||
d46c5b12 KH |
936 | /* Check if the current composing sequence contains only valid codes. |
937 | If the composing sequence doesn't end before SRC_END, return -1. | |
938 | Else, if it contains only valid codes, return 0. | |
939 | Else return the length of the composing sequence. */ | |
940 | ||
941 | int check_composing_code (coding, src, src_end) | |
942 | struct coding_system *coding; | |
943 | unsigned char *src, *src_end; | |
944 | { | |
945 | unsigned char *src_start = src; | |
946 | int invalid_code_found = 0; | |
947 | int charset, c, c1, dim; | |
948 | ||
949 | while (src < src_end) | |
950 | { | |
951 | if (*src++ != ISO_CODE_ESC) continue; | |
952 | if (src >= src_end) break; | |
953 | if ((c = *src++) == '1') /* end of compsition */ | |
954 | return (invalid_code_found ? src - src_start : 0); | |
955 | if (src + 2 >= src_end) break; | |
956 | if (!coding->flags & CODING_FLAG_ISO_DESIGNATION) | |
957 | invalid_code_found = 1; | |
958 | else | |
959 | { | |
960 | dim = 0; | |
961 | if (c == '$') | |
962 | { | |
963 | dim = 1; | |
964 | c = (*src >= '@' && *src <= 'B') ? '(' : *src++; | |
965 | } | |
966 | if (c >= '(' && c <= '/') | |
967 | { | |
968 | c1 = *src++; | |
969 | if ((c1 < ' ' || c1 >= 0x80) | |
970 | || (charset = iso_charset_table[dim][c >= ','][c1]) < 0 | |
971 | || (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) | |
972 | == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)) | |
973 | invalid_code_found = 1; | |
974 | } | |
975 | else | |
976 | invalid_code_found = 1; | |
977 | } | |
978 | } | |
979 | return ((coding->mode & CODING_MODE_LAST_BLOCK) ? src_end - src_start : -1); | |
980 | } | |
981 | ||
4ed46869 KH |
982 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */ |
983 | ||
984 | int | |
d46c5b12 | 985 | decode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
986 | struct coding_system *coding; |
987 | unsigned char *source, *destination; | |
988 | int src_bytes, dst_bytes; | |
4ed46869 KH |
989 | { |
990 | unsigned char *src = source; | |
991 | unsigned char *src_end = source + src_bytes; | |
992 | unsigned char *dst = destination; | |
993 | unsigned char *dst_end = destination + dst_bytes; | |
994 | /* Since the maximum bytes produced by each loop is 7, we subtract 6 | |
995 | from DST_END to assure that overflow checking is necessary only | |
996 | at the head of loop. */ | |
997 | unsigned char *adjusted_dst_end = dst_end - 6; | |
998 | int charset; | |
999 | /* Charsets invoked to graphic plane 0 and 1 respectively. */ | |
1000 | int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1001 | int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); | |
a5d301df | 1002 | Lisp_Object unification_table |
d46c5b12 KH |
1003 | = coding->character_unification_table_for_decode; |
1004 | int result = CODING_FINISH_NORMAL; | |
bdd9fb48 KH |
1005 | |
1006 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
a5d301df | 1007 | unification_table = Vstandard_character_unification_table_for_decode; |
4ed46869 | 1008 | |
d46c5b12 | 1009 | coding->produced_char = 0; |
fb88bf2d | 1010 | coding->fake_multibyte = 0; |
d46c5b12 KH |
1011 | while (src < src_end && (dst_bytes |
1012 | ? (dst < adjusted_dst_end) | |
1013 | : (dst < src - 6))) | |
4ed46869 KH |
1014 | { |
1015 | /* SRC_BASE remembers the start position in source in each loop. | |
1016 | The loop will be exited when there's not enough source text | |
1017 | to analyze long escape sequence or 2-byte code (within macros | |
1018 | ONE_MORE_BYTE or TWO_MORE_BYTES). In that case, SRC is reset | |
1019 | to SRC_BASE before exiting. */ | |
1020 | unsigned char *src_base = src; | |
bdd9fb48 | 1021 | int c1 = *src++, c2; |
4ed46869 KH |
1022 | |
1023 | switch (iso_code_class [c1]) | |
1024 | { | |
1025 | case ISO_0x20_or_0x7F: | |
1026 | if (!coding->composing | |
1027 | && (charset0 < 0 || CHARSET_CHARS (charset0) == 94)) | |
1028 | { | |
1029 | /* This is SPACE or DEL. */ | |
1030 | *dst++ = c1; | |
d46c5b12 | 1031 | coding->produced_char++; |
4ed46869 KH |
1032 | break; |
1033 | } | |
1034 | /* This is a graphic character, we fall down ... */ | |
1035 | ||
1036 | case ISO_graphic_plane_0: | |
1037 | if (coding->composing == COMPOSING_WITH_RULE_RULE) | |
1038 | { | |
1039 | /* This is a composition rule. */ | |
1040 | *dst++ = c1 | 0x80; | |
1041 | coding->composing = COMPOSING_WITH_RULE_TAIL; | |
1042 | } | |
1043 | else | |
1044 | DECODE_ISO_CHARACTER (charset0, c1); | |
1045 | break; | |
1046 | ||
1047 | case ISO_0xA0_or_0xFF: | |
d46c5b12 KH |
1048 | if (charset1 < 0 || CHARSET_CHARS (charset1) == 94 |
1049 | || coding->flags & CODING_FLAG_ISO_SEVEN_BITS) | |
fb88bf2d | 1050 | goto label_invalid_code; |
4ed46869 KH |
1051 | /* This is a graphic character, we fall down ... */ |
1052 | ||
1053 | case ISO_graphic_plane_1: | |
d46c5b12 | 1054 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) |
fb88bf2d | 1055 | goto label_invalid_code; |
d46c5b12 KH |
1056 | else |
1057 | DECODE_ISO_CHARACTER (charset1, c1); | |
4ed46869 KH |
1058 | break; |
1059 | ||
1060 | case ISO_control_code: | |
1061 | /* All ISO2022 control characters in this class have the | |
1062 | same representation in Emacs internal format. */ | |
d46c5b12 KH |
1063 | if (c1 == '\n' |
1064 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
1065 | && (coding->eol_type == CODING_EOL_CR | |
1066 | || coding->eol_type == CODING_EOL_CRLF)) | |
1067 | { | |
1068 | result = CODING_FINISH_INCONSISTENT_EOL; | |
1069 | goto label_end_of_loop_2; | |
1070 | } | |
4ed46869 | 1071 | *dst++ = c1; |
d46c5b12 | 1072 | coding->produced_char++; |
4ed46869 KH |
1073 | break; |
1074 | ||
1075 | case ISO_carriage_return: | |
1076 | if (coding->eol_type == CODING_EOL_CR) | |
d46c5b12 | 1077 | *dst++ = '\n'; |
4ed46869 KH |
1078 | else if (coding->eol_type == CODING_EOL_CRLF) |
1079 | { | |
1080 | ONE_MORE_BYTE (c1); | |
1081 | if (c1 == ISO_CODE_LF) | |
1082 | *dst++ = '\n'; | |
1083 | else | |
1084 | { | |
d46c5b12 KH |
1085 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
1086 | { | |
1087 | result = CODING_FINISH_INCONSISTENT_EOL; | |
1088 | goto label_end_of_loop_2; | |
1089 | } | |
4ed46869 | 1090 | src--; |
d46c5b12 | 1091 | *dst++ = '\r'; |
4ed46869 KH |
1092 | } |
1093 | } | |
1094 | else | |
d46c5b12 KH |
1095 | *dst++ = c1; |
1096 | coding->produced_char++; | |
4ed46869 KH |
1097 | break; |
1098 | ||
1099 | case ISO_shift_out: | |
d46c5b12 KH |
1100 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
1101 | || CODING_SPEC_ISO_DESIGNATION (coding, 1) < 0) | |
1102 | goto label_invalid_code; | |
4ed46869 KH |
1103 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; |
1104 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1105 | break; | |
1106 | ||
1107 | case ISO_shift_in: | |
d46c5b12 KH |
1108 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
1109 | goto label_invalid_code; | |
4ed46869 KH |
1110 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; |
1111 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1112 | break; | |
1113 | ||
1114 | case ISO_single_shift_2_7: | |
1115 | case ISO_single_shift_2: | |
d46c5b12 KH |
1116 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
1117 | goto label_invalid_code; | |
4ed46869 KH |
1118 | /* SS2 is handled as an escape sequence of ESC 'N' */ |
1119 | c1 = 'N'; | |
1120 | goto label_escape_sequence; | |
1121 | ||
1122 | case ISO_single_shift_3: | |
d46c5b12 KH |
1123 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
1124 | goto label_invalid_code; | |
4ed46869 KH |
1125 | /* SS2 is handled as an escape sequence of ESC 'O' */ |
1126 | c1 = 'O'; | |
1127 | goto label_escape_sequence; | |
1128 | ||
1129 | case ISO_control_sequence_introducer: | |
1130 | /* CSI is handled as an escape sequence of ESC '[' ... */ | |
1131 | c1 = '['; | |
1132 | goto label_escape_sequence; | |
1133 | ||
1134 | case ISO_escape: | |
1135 | ONE_MORE_BYTE (c1); | |
1136 | label_escape_sequence: | |
1137 | /* Escape sequences handled by Emacs are invocation, | |
1138 | designation, direction specification, and character | |
1139 | composition specification. */ | |
1140 | switch (c1) | |
1141 | { | |
1142 | case '&': /* revision of following character set */ | |
1143 | ONE_MORE_BYTE (c1); | |
1144 | if (!(c1 >= '@' && c1 <= '~')) | |
d46c5b12 | 1145 | goto label_invalid_code; |
4ed46869 KH |
1146 | ONE_MORE_BYTE (c1); |
1147 | if (c1 != ISO_CODE_ESC) | |
d46c5b12 | 1148 | goto label_invalid_code; |
4ed46869 KH |
1149 | ONE_MORE_BYTE (c1); |
1150 | goto label_escape_sequence; | |
1151 | ||
1152 | case '$': /* designation of 2-byte character set */ | |
d46c5b12 KH |
1153 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
1154 | goto label_invalid_code; | |
4ed46869 KH |
1155 | ONE_MORE_BYTE (c1); |
1156 | if (c1 >= '@' && c1 <= 'B') | |
1157 | { /* designation of JISX0208.1978, GB2312.1980, | |
1158 | or JISX0208.1980 */ | |
1159 | DECODE_DESIGNATION (0, 2, 94, c1); | |
1160 | } | |
1161 | else if (c1 >= 0x28 && c1 <= 0x2B) | |
1162 | { /* designation of DIMENSION2_CHARS94 character set */ | |
1163 | ONE_MORE_BYTE (c2); | |
1164 | DECODE_DESIGNATION (c1 - 0x28, 2, 94, c2); | |
1165 | } | |
1166 | else if (c1 >= 0x2C && c1 <= 0x2F) | |
1167 | { /* designation of DIMENSION2_CHARS96 character set */ | |
1168 | ONE_MORE_BYTE (c2); | |
1169 | DECODE_DESIGNATION (c1 - 0x2C, 2, 96, c2); | |
1170 | } | |
1171 | else | |
d46c5b12 | 1172 | goto label_invalid_code; |
4ed46869 KH |
1173 | break; |
1174 | ||
1175 | case 'n': /* invocation of locking-shift-2 */ | |
d46c5b12 KH |
1176 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
1177 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) | |
1178 | goto label_invalid_code; | |
4ed46869 | 1179 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; |
e0e989f6 | 1180 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
4ed46869 KH |
1181 | break; |
1182 | ||
1183 | case 'o': /* invocation of locking-shift-3 */ | |
d46c5b12 KH |
1184 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
1185 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) | |
1186 | goto label_invalid_code; | |
4ed46869 | 1187 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; |
e0e989f6 | 1188 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
4ed46869 KH |
1189 | break; |
1190 | ||
1191 | case 'N': /* invocation of single-shift-2 */ | |
d46c5b12 KH |
1192 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
1193 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) | |
1194 | goto label_invalid_code; | |
4ed46869 KH |
1195 | ONE_MORE_BYTE (c1); |
1196 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 2); | |
1197 | DECODE_ISO_CHARACTER (charset, c1); | |
1198 | break; | |
1199 | ||
1200 | case 'O': /* invocation of single-shift-3 */ | |
d46c5b12 KH |
1201 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
1202 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) | |
1203 | goto label_invalid_code; | |
4ed46869 KH |
1204 | ONE_MORE_BYTE (c1); |
1205 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 3); | |
1206 | DECODE_ISO_CHARACTER (charset, c1); | |
1207 | break; | |
1208 | ||
d46c5b12 KH |
1209 | case '0': case '2': /* start composing */ |
1210 | /* Before processing composing, we must be sure that all | |
1211 | characters being composed are supported by CODING. | |
1212 | If not, we must give up composing and insert the | |
1213 | bunch of codes for composing as is without decoding. */ | |
1214 | { | |
1215 | int result1; | |
1216 | ||
1217 | result1 = check_composing_code (coding, src, src_end); | |
1218 | if (result1 == 0) | |
1219 | coding->composing = (c1 == '0' | |
1220 | ? COMPOSING_NO_RULE_HEAD | |
1221 | : COMPOSING_WITH_RULE_HEAD); | |
1222 | else if (result1 > 0) | |
1223 | { | |
1224 | if (result1 + 2 < (dst_bytes ? dst_end : src_base) - dst) | |
1225 | { | |
1226 | bcopy (src_base, dst, result1 + 2); | |
1227 | src += result1; | |
1228 | dst += result1 + 2; | |
1229 | coding->produced_char += result1 + 2; | |
1230 | } | |
1231 | else | |
1232 | { | |
1233 | result = CODING_FINISH_INSUFFICIENT_DST; | |
1234 | goto label_end_of_loop_2; | |
1235 | } | |
1236 | } | |
1237 | else | |
1238 | goto label_end_of_loop; | |
1239 | } | |
4ed46869 KH |
1240 | break; |
1241 | ||
1242 | case '1': /* end composing */ | |
1243 | coding->composing = COMPOSING_NO; | |
d46c5b12 | 1244 | coding->produced_char++; |
4ed46869 KH |
1245 | break; |
1246 | ||
1247 | case '[': /* specification of direction */ | |
d46c5b12 KH |
1248 | if (coding->flags & CODING_FLAG_ISO_NO_DIRECTION) |
1249 | goto label_invalid_code; | |
4ed46869 | 1250 | /* For the moment, nested direction is not supported. |
d46c5b12 KH |
1251 | So, `coding->mode & CODING_MODE_DIRECTION' zero means |
1252 | left-to-right, and nozero means right-to-left. */ | |
4ed46869 KH |
1253 | ONE_MORE_BYTE (c1); |
1254 | switch (c1) | |
1255 | { | |
1256 | case ']': /* end of the current direction */ | |
d46c5b12 | 1257 | coding->mode &= ~CODING_MODE_DIRECTION; |
4ed46869 KH |
1258 | |
1259 | case '0': /* end of the current direction */ | |
1260 | case '1': /* start of left-to-right direction */ | |
1261 | ONE_MORE_BYTE (c1); | |
1262 | if (c1 == ']') | |
d46c5b12 | 1263 | coding->mode &= ~CODING_MODE_DIRECTION; |
4ed46869 | 1264 | else |
d46c5b12 | 1265 | goto label_invalid_code; |
4ed46869 KH |
1266 | break; |
1267 | ||
1268 | case '2': /* start of right-to-left direction */ | |
1269 | ONE_MORE_BYTE (c1); | |
1270 | if (c1 == ']') | |
d46c5b12 | 1271 | coding->mode |= CODING_MODE_DIRECTION; |
4ed46869 | 1272 | else |
d46c5b12 | 1273 | goto label_invalid_code; |
4ed46869 KH |
1274 | break; |
1275 | ||
1276 | default: | |
d46c5b12 | 1277 | goto label_invalid_code; |
4ed46869 KH |
1278 | } |
1279 | break; | |
1280 | ||
1281 | default: | |
d46c5b12 KH |
1282 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
1283 | goto label_invalid_code; | |
4ed46869 KH |
1284 | if (c1 >= 0x28 && c1 <= 0x2B) |
1285 | { /* designation of DIMENSION1_CHARS94 character set */ | |
1286 | ONE_MORE_BYTE (c2); | |
1287 | DECODE_DESIGNATION (c1 - 0x28, 1, 94, c2); | |
1288 | } | |
1289 | else if (c1 >= 0x2C && c1 <= 0x2F) | |
1290 | { /* designation of DIMENSION1_CHARS96 character set */ | |
1291 | ONE_MORE_BYTE (c2); | |
1292 | DECODE_DESIGNATION (c1 - 0x2C, 1, 96, c2); | |
1293 | } | |
1294 | else | |
1295 | { | |
d46c5b12 | 1296 | goto label_invalid_code; |
4ed46869 KH |
1297 | } |
1298 | } | |
1299 | /* We must update these variables now. */ | |
1300 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1301 | charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); | |
1302 | break; | |
1303 | ||
d46c5b12 | 1304 | label_invalid_code: |
d46c5b12 KH |
1305 | while (src_base < src) |
1306 | *dst++ = *src_base++; | |
fb88bf2d | 1307 | coding->fake_multibyte = 1; |
4ed46869 KH |
1308 | } |
1309 | continue; | |
1310 | ||
1311 | label_end_of_loop: | |
d46c5b12 KH |
1312 | result = CODING_FINISH_INSUFFICIENT_SRC; |
1313 | label_end_of_loop_2: | |
4ed46869 KH |
1314 | src = src_base; |
1315 | break; | |
1316 | } | |
1317 | ||
fb88bf2d | 1318 | if (src < src_end) |
4ed46869 | 1319 | { |
fb88bf2d KH |
1320 | if (result == CODING_FINISH_NORMAL) |
1321 | result = CODING_FINISH_INSUFFICIENT_DST; | |
1322 | else if (result != CODING_FINISH_INCONSISTENT_EOL | |
1323 | && coding->mode & CODING_MODE_LAST_BLOCK) | |
1324 | { | |
1325 | /* This is the last block of the text to be decoded. We had | |
1326 | better just flush out all remaining codes in the text | |
1327 | although they are not valid characters. */ | |
1328 | src_bytes = src_end - src; | |
1329 | if (dst_bytes && (dst_end - dst < src_bytes)) | |
1330 | src_bytes = dst_end - dst; | |
1331 | bcopy (src, dst, src_bytes); | |
1332 | dst += src_bytes; | |
1333 | src += src_bytes; | |
1334 | coding->fake_multibyte = 1; | |
1335 | } | |
4ed46869 | 1336 | } |
fb88bf2d | 1337 | |
d46c5b12 KH |
1338 | coding->consumed = coding->consumed_char = src - source; |
1339 | coding->produced = dst - destination; | |
1340 | return result; | |
4ed46869 KH |
1341 | } |
1342 | ||
f4dee582 | 1343 | /* ISO2022 encoding stuff. */ |
4ed46869 KH |
1344 | |
1345 | /* | |
f4dee582 | 1346 | It is not enough to say just "ISO2022" on encoding, we have to |
d46c5b12 | 1347 | specify more details. In Emacs, each coding system of ISO2022 |
4ed46869 KH |
1348 | variant has the following specifications: |
1349 | 1. Initial designation to G0 thru G3. | |
1350 | 2. Allows short-form designation? | |
1351 | 3. ASCII should be designated to G0 before control characters? | |
1352 | 4. ASCII should be designated to G0 at end of line? | |
1353 | 5. 7-bit environment or 8-bit environment? | |
1354 | 6. Use locking-shift? | |
1355 | 7. Use Single-shift? | |
1356 | And the following two are only for Japanese: | |
1357 | 8. Use ASCII in place of JIS0201-1976-Roman? | |
1358 | 9. Use JISX0208-1983 in place of JISX0208-1978? | |
1359 | These specifications are encoded in `coding->flags' as flag bits | |
1360 | defined by macros CODING_FLAG_ISO_XXX. See `coding.h' for more | |
f4dee582 | 1361 | details. |
4ed46869 KH |
1362 | */ |
1363 | ||
1364 | /* Produce codes (escape sequence) for designating CHARSET to graphic | |
1365 | register REG. If <final-char> of CHARSET is '@', 'A', or 'B' and | |
1366 | the coding system CODING allows, produce designation sequence of | |
1367 | short-form. */ | |
1368 | ||
1369 | #define ENCODE_DESIGNATION(charset, reg, coding) \ | |
1370 | do { \ | |
1371 | unsigned char final_char = CHARSET_ISO_FINAL_CHAR (charset); \ | |
1372 | char *intermediate_char_94 = "()*+"; \ | |
1373 | char *intermediate_char_96 = ",-./"; \ | |
70c22245 KH |
1374 | int revision = CODING_SPEC_ISO_REVISION_NUMBER(coding, charset); \ |
1375 | if (revision < 255) \ | |
1376 | { \ | |
4ed46869 KH |
1377 | *dst++ = ISO_CODE_ESC; \ |
1378 | *dst++ = '&'; \ | |
70c22245 | 1379 | *dst++ = '@' + revision; \ |
4ed46869 KH |
1380 | } \ |
1381 | *dst++ = ISO_CODE_ESC; \ | |
1382 | if (CHARSET_DIMENSION (charset) == 1) \ | |
1383 | { \ | |
1384 | if (CHARSET_CHARS (charset) == 94) \ | |
1385 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ | |
1386 | else \ | |
1387 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ | |
1388 | } \ | |
1389 | else \ | |
1390 | { \ | |
1391 | *dst++ = '$'; \ | |
1392 | if (CHARSET_CHARS (charset) == 94) \ | |
1393 | { \ | |
1394 | if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \ | |
1395 | || reg != 0 \ | |
1396 | || final_char < '@' || final_char > 'B') \ | |
1397 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ | |
1398 | } \ | |
1399 | else \ | |
1400 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ | |
1401 | } \ | |
1402 | *dst++ = final_char; \ | |
1403 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ | |
1404 | } while (0) | |
1405 | ||
1406 | /* The following two macros produce codes (control character or escape | |
1407 | sequence) for ISO2022 single-shift functions (single-shift-2 and | |
1408 | single-shift-3). */ | |
1409 | ||
1410 | #define ENCODE_SINGLE_SHIFT_2 \ | |
1411 | do { \ | |
1412 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1413 | *dst++ = ISO_CODE_ESC, *dst++ = 'N'; \ | |
1414 | else \ | |
fb88bf2d KH |
1415 | { \ |
1416 | *dst++ = ISO_CODE_SS2; \ | |
1417 | coding->fake_multibyte = 1; \ | |
1418 | } \ | |
4ed46869 KH |
1419 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
1420 | } while (0) | |
1421 | ||
fb88bf2d KH |
1422 | #define ENCODE_SINGLE_SHIFT_3 \ |
1423 | do { \ | |
4ed46869 | 1424 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
fb88bf2d KH |
1425 | *dst++ = ISO_CODE_ESC, *dst++ = 'O'; \ |
1426 | else \ | |
1427 | { \ | |
1428 | *dst++ = ISO_CODE_SS3; \ | |
1429 | coding->fake_multibyte = 1; \ | |
1430 | } \ | |
4ed46869 KH |
1431 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
1432 | } while (0) | |
1433 | ||
1434 | /* The following four macros produce codes (control character or | |
1435 | escape sequence) for ISO2022 locking-shift functions (shift-in, | |
1436 | shift-out, locking-shift-2, and locking-shift-3). */ | |
1437 | ||
1438 | #define ENCODE_SHIFT_IN \ | |
1439 | do { \ | |
1440 | *dst++ = ISO_CODE_SI; \ | |
1441 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; \ | |
1442 | } while (0) | |
1443 | ||
1444 | #define ENCODE_SHIFT_OUT \ | |
1445 | do { \ | |
1446 | *dst++ = ISO_CODE_SO; \ | |
1447 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; \ | |
1448 | } while (0) | |
1449 | ||
1450 | #define ENCODE_LOCKING_SHIFT_2 \ | |
1451 | do { \ | |
1452 | *dst++ = ISO_CODE_ESC, *dst++ = 'n'; \ | |
1453 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; \ | |
1454 | } while (0) | |
1455 | ||
1456 | #define ENCODE_LOCKING_SHIFT_3 \ | |
1457 | do { \ | |
1458 | *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \ | |
1459 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; \ | |
1460 | } while (0) | |
1461 | ||
f4dee582 RS |
1462 | /* Produce codes for a DIMENSION1 character whose character set is |
1463 | CHARSET and whose position-code is C1. Designation and invocation | |
4ed46869 KH |
1464 | sequences are also produced in advance if necessary. */ |
1465 | ||
1466 | ||
6e85d753 KH |
1467 | #define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \ |
1468 | do { \ | |
1469 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ | |
1470 | { \ | |
1471 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1472 | *dst++ = c1 & 0x7F; \ | |
1473 | else \ | |
1474 | *dst++ = c1 | 0x80; \ | |
1475 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ | |
1476 | break; \ | |
1477 | } \ | |
1478 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ | |
1479 | { \ | |
1480 | *dst++ = c1 & 0x7F; \ | |
1481 | break; \ | |
1482 | } \ | |
1483 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ | |
1484 | { \ | |
1485 | *dst++ = c1 | 0x80; \ | |
1486 | break; \ | |
1487 | } \ | |
1488 | else if (coding->flags & CODING_FLAG_ISO_SAFE \ | |
70c22245 | 1489 | && !coding->safe_charsets[charset]) \ |
6e85d753 KH |
1490 | { \ |
1491 | /* We should not encode this character, instead produce one or \ | |
1492 | two `?'s. */ \ | |
1493 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1494 | if (CHARSET_WIDTH (charset) == 2) \ | |
1495 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1496 | break; \ | |
1497 | } \ | |
1498 | else \ | |
1499 | /* Since CHARSET is not yet invoked to any graphic planes, we \ | |
1500 | must invoke it, or, at first, designate it to some graphic \ | |
1501 | register. Then repeat the loop to actually produce the \ | |
1502 | character. */ \ | |
1503 | dst = encode_invocation_designation (charset, coding, dst); \ | |
4ed46869 KH |
1504 | } while (1) |
1505 | ||
f4dee582 RS |
1506 | /* Produce codes for a DIMENSION2 character whose character set is |
1507 | CHARSET and whose position-codes are C1 and C2. Designation and | |
4ed46869 KH |
1508 | invocation codes are also produced in advance if necessary. */ |
1509 | ||
6e85d753 KH |
1510 | #define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2) \ |
1511 | do { \ | |
1512 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ | |
1513 | { \ | |
1514 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1515 | *dst++ = c1 & 0x7F, *dst++ = c2 & 0x7F; \ | |
1516 | else \ | |
1517 | *dst++ = c1 | 0x80, *dst++ = c2 | 0x80; \ | |
1518 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ | |
1519 | break; \ | |
1520 | } \ | |
1521 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ | |
1522 | { \ | |
1523 | *dst++ = c1 & 0x7F, *dst++= c2 & 0x7F; \ | |
1524 | break; \ | |
1525 | } \ | |
1526 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ | |
1527 | { \ | |
1528 | *dst++ = c1 | 0x80, *dst++= c2 | 0x80; \ | |
1529 | break; \ | |
1530 | } \ | |
1531 | else if (coding->flags & CODING_FLAG_ISO_SAFE \ | |
70c22245 | 1532 | && !coding->safe_charsets[charset]) \ |
6e85d753 KH |
1533 | { \ |
1534 | /* We should not encode this character, instead produce one or \ | |
1535 | two `?'s. */ \ | |
1536 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1537 | if (CHARSET_WIDTH (charset) == 2) \ | |
1538 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1539 | break; \ | |
1540 | } \ | |
1541 | else \ | |
1542 | /* Since CHARSET is not yet invoked to any graphic planes, we \ | |
1543 | must invoke it, or, at first, designate it to some graphic \ | |
1544 | register. Then repeat the loop to actually produce the \ | |
1545 | character. */ \ | |
1546 | dst = encode_invocation_designation (charset, coding, dst); \ | |
4ed46869 KH |
1547 | } while (1) |
1548 | ||
bdd9fb48 KH |
1549 | #define ENCODE_ISO_CHARACTER(charset, c1, c2) \ |
1550 | do { \ | |
1551 | int c_alt, charset_alt; \ | |
1552 | if (!NILP (unification_table) \ | |
1553 | && ((c_alt = unify_char (unification_table, -1, charset, c1, c2)) \ | |
a5d301df | 1554 | >= 0)) \ |
bdd9fb48 KH |
1555 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
1556 | else \ | |
1557 | charset_alt = charset; \ | |
1558 | if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
4031e2bf KH |
1559 | { \ |
1560 | if (charset == CHARSET_ASCII \ | |
1561 | && coding->flags & CODING_FLAG_ISO_USE_ROMAN) \ | |
1562 | charset_alt = charset_latin_jisx0201; \ | |
1563 | ENCODE_ISO_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
1564 | } \ | |
bdd9fb48 | 1565 | else \ |
4031e2bf KH |
1566 | { \ |
1567 | if (charset == charset_jisx0208 \ | |
1568 | && coding->flags & CODING_FLAG_ISO_USE_OLDJIS) \ | |
1569 | charset_alt = charset_jisx0208_1978; \ | |
1570 | ENCODE_ISO_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
1571 | } \ | |
d46c5b12 KH |
1572 | if (! COMPOSING_P (coding->composing)) \ |
1573 | coding->consumed_char++; \ | |
4031e2bf | 1574 | } while (0) |
bdd9fb48 | 1575 | |
4ed46869 KH |
1576 | /* Produce designation and invocation codes at a place pointed by DST |
1577 | to use CHARSET. The element `spec.iso2022' of *CODING is updated. | |
1578 | Return new DST. */ | |
1579 | ||
1580 | unsigned char * | |
1581 | encode_invocation_designation (charset, coding, dst) | |
1582 | int charset; | |
1583 | struct coding_system *coding; | |
1584 | unsigned char *dst; | |
1585 | { | |
1586 | int reg; /* graphic register number */ | |
1587 | ||
1588 | /* At first, check designations. */ | |
1589 | for (reg = 0; reg < 4; reg++) | |
1590 | if (charset == CODING_SPEC_ISO_DESIGNATION (coding, reg)) | |
1591 | break; | |
1592 | ||
1593 | if (reg >= 4) | |
1594 | { | |
1595 | /* CHARSET is not yet designated to any graphic registers. */ | |
1596 | /* At first check the requested designation. */ | |
1597 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); | |
1ba9e4ab KH |
1598 | if (reg == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION) |
1599 | /* Since CHARSET requests no special designation, designate it | |
1600 | to graphic register 0. */ | |
4ed46869 KH |
1601 | reg = 0; |
1602 | ||
1603 | ENCODE_DESIGNATION (charset, reg, coding); | |
1604 | } | |
1605 | ||
1606 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != reg | |
1607 | && CODING_SPEC_ISO_INVOCATION (coding, 1) != reg) | |
1608 | { | |
1609 | /* Since the graphic register REG is not invoked to any graphic | |
1610 | planes, invoke it to graphic plane 0. */ | |
1611 | switch (reg) | |
1612 | { | |
1613 | case 0: /* graphic register 0 */ | |
1614 | ENCODE_SHIFT_IN; | |
1615 | break; | |
1616 | ||
1617 | case 1: /* graphic register 1 */ | |
1618 | ENCODE_SHIFT_OUT; | |
1619 | break; | |
1620 | ||
1621 | case 2: /* graphic register 2 */ | |
1622 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) | |
1623 | ENCODE_SINGLE_SHIFT_2; | |
1624 | else | |
1625 | ENCODE_LOCKING_SHIFT_2; | |
1626 | break; | |
1627 | ||
1628 | case 3: /* graphic register 3 */ | |
1629 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) | |
1630 | ENCODE_SINGLE_SHIFT_3; | |
1631 | else | |
1632 | ENCODE_LOCKING_SHIFT_3; | |
1633 | break; | |
1634 | } | |
1635 | } | |
1636 | return dst; | |
1637 | } | |
1638 | ||
1639 | /* The following two macros produce codes for indicating composition. */ | |
1640 | #define ENCODE_COMPOSITION_NO_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '0' | |
1641 | #define ENCODE_COMPOSITION_WITH_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '2' | |
1642 | #define ENCODE_COMPOSITION_END *dst++ = ISO_CODE_ESC, *dst++ = '1' | |
1643 | ||
1644 | /* The following three macros produce codes for indicating direction | |
1645 | of text. */ | |
1646 | #define ENCODE_CONTROL_SEQUENCE_INTRODUCER \ | |
1647 | do { \ | |
1648 | if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS) \ | |
1649 | *dst++ = ISO_CODE_ESC, *dst++ = '['; \ | |
1650 | else \ | |
1651 | *dst++ = ISO_CODE_CSI; \ | |
1652 | } while (0) | |
1653 | ||
1654 | #define ENCODE_DIRECTION_R2L \ | |
1655 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '2', *dst++ = ']' | |
1656 | ||
1657 | #define ENCODE_DIRECTION_L2R \ | |
1658 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '0', *dst++ = ']' | |
1659 | ||
1660 | /* Produce codes for designation and invocation to reset the graphic | |
1661 | planes and registers to initial state. */ | |
e0e989f6 KH |
1662 | #define ENCODE_RESET_PLANE_AND_REGISTER \ |
1663 | do { \ | |
1664 | int reg; \ | |
1665 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != 0) \ | |
1666 | ENCODE_SHIFT_IN; \ | |
1667 | for (reg = 0; reg < 4; reg++) \ | |
1668 | if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg) >= 0 \ | |
1669 | && (CODING_SPEC_ISO_DESIGNATION (coding, reg) \ | |
1670 | != CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg))) \ | |
1671 | ENCODE_DESIGNATION \ | |
1672 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg), reg, coding); \ | |
4ed46869 KH |
1673 | } while (0) |
1674 | ||
bdd9fb48 | 1675 | /* Produce designation sequences of charsets in the line started from |
d46c5b12 | 1676 | SRC to a place pointed by *DSTP, and update DSTP. |
bdd9fb48 KH |
1677 | |
1678 | If the current block ends before any end-of-line, we may fail to | |
d46c5b12 KH |
1679 | find all the necessary designations. */ |
1680 | ||
bdd9fb48 | 1681 | encode_designation_at_bol (coding, table, src, src_end, dstp) |
e0e989f6 | 1682 | struct coding_system *coding; |
bdd9fb48 | 1683 | Lisp_Object table; |
e0e989f6 KH |
1684 | unsigned char *src, *src_end, **dstp; |
1685 | { | |
bdd9fb48 KH |
1686 | int charset, c, found = 0, reg; |
1687 | /* Table of charsets to be designated to each graphic register. */ | |
1688 | int r[4]; | |
1689 | unsigned char *dst = *dstp; | |
1690 | ||
1691 | for (reg = 0; reg < 4; reg++) | |
1692 | r[reg] = -1; | |
1693 | ||
1694 | while (src < src_end && *src != '\n' && found < 4) | |
e0e989f6 | 1695 | { |
bdd9fb48 KH |
1696 | int bytes = BYTES_BY_CHAR_HEAD (*src); |
1697 | ||
1698 | if (NILP (table)) | |
1699 | charset = CHARSET_AT (src); | |
1700 | else | |
e0e989f6 | 1701 | { |
35cb8686 RS |
1702 | int c_alt; |
1703 | unsigned char c1, c2; | |
bdd9fb48 KH |
1704 | |
1705 | SPLIT_STRING(src, bytes, charset, c1, c2); | |
1706 | if ((c_alt = unify_char (table, -1, charset, c1, c2)) >= 0) | |
1707 | charset = CHAR_CHARSET (c_alt); | |
e0e989f6 | 1708 | } |
bdd9fb48 | 1709 | |
e0e989f6 | 1710 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); |
d46c5b12 | 1711 | if (reg != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION && r[reg] < 0) |
bdd9fb48 KH |
1712 | { |
1713 | found++; | |
1714 | r[reg] = charset; | |
1715 | } | |
1716 | ||
1717 | src += bytes; | |
1718 | } | |
1719 | ||
1720 | if (found) | |
1721 | { | |
1722 | for (reg = 0; reg < 4; reg++) | |
1723 | if (r[reg] >= 0 | |
1724 | && CODING_SPEC_ISO_DESIGNATION (coding, reg) != r[reg]) | |
1725 | ENCODE_DESIGNATION (r[reg], reg, coding); | |
1726 | *dstp = dst; | |
e0e989f6 | 1727 | } |
e0e989f6 KH |
1728 | } |
1729 | ||
4ed46869 KH |
1730 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */ |
1731 | ||
1732 | int | |
d46c5b12 | 1733 | encode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
1734 | struct coding_system *coding; |
1735 | unsigned char *source, *destination; | |
1736 | int src_bytes, dst_bytes; | |
4ed46869 KH |
1737 | { |
1738 | unsigned char *src = source; | |
1739 | unsigned char *src_end = source + src_bytes; | |
1740 | unsigned char *dst = destination; | |
1741 | unsigned char *dst_end = destination + dst_bytes; | |
e0e989f6 | 1742 | /* Since the maximum bytes produced by each loop is 20, we subtract 19 |
4ed46869 KH |
1743 | from DST_END to assure overflow checking is necessary only at the |
1744 | head of loop. */ | |
e0e989f6 | 1745 | unsigned char *adjusted_dst_end = dst_end - 19; |
a5d301df KH |
1746 | Lisp_Object unification_table |
1747 | = coding->character_unification_table_for_encode; | |
d46c5b12 | 1748 | int result = CODING_FINISH_NORMAL; |
bdd9fb48 KH |
1749 | |
1750 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
a5d301df | 1751 | unification_table = Vstandard_character_unification_table_for_encode; |
4ed46869 | 1752 | |
d46c5b12 | 1753 | coding->consumed_char = 0; |
fb88bf2d | 1754 | coding->fake_multibyte = 0; |
d46c5b12 KH |
1755 | while (src < src_end && (dst_bytes |
1756 | ? (dst < adjusted_dst_end) | |
1757 | : (dst < src - 19))) | |
4ed46869 KH |
1758 | { |
1759 | /* SRC_BASE remembers the start position in source in each loop. | |
1760 | The loop will be exited when there's not enough source text | |
1761 | to analyze multi-byte codes (within macros ONE_MORE_BYTE, | |
1762 | TWO_MORE_BYTES, and THREE_MORE_BYTES). In that case, SRC is | |
1763 | reset to SRC_BASE before exiting. */ | |
1764 | unsigned char *src_base = src; | |
bdd9fb48 | 1765 | int charset, c1, c2, c3, c4; |
4ed46869 | 1766 | |
e0e989f6 KH |
1767 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL |
1768 | && CODING_SPEC_ISO_BOL (coding)) | |
1769 | { | |
bdd9fb48 KH |
1770 | /* We have to produce designation sequences if any now. */ |
1771 | encode_designation_at_bol (coding, unification_table, | |
1772 | src, src_end, &dst); | |
e0e989f6 KH |
1773 | CODING_SPEC_ISO_BOL (coding) = 0; |
1774 | } | |
1775 | ||
1776 | c1 = *src++; | |
4ed46869 | 1777 | /* If we are seeing a component of a composite character, we are |
d46c5b12 KH |
1778 | seeing a leading-code encoded irregularly for composition, or |
1779 | a composition rule if composing with rule. We must set C1 to | |
1780 | a normal leading-code or an ASCII code. If we are not seeing | |
1781 | a composite character, we must reset composition, | |
1782 | designation, and invocation states. */ | |
4ed46869 KH |
1783 | if (COMPOSING_P (coding->composing)) |
1784 | { | |
1785 | if (c1 < 0xA0) | |
1786 | { | |
1787 | /* We are not in a composite character any longer. */ | |
1788 | coding->composing = COMPOSING_NO; | |
d46c5b12 | 1789 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1790 | ENCODE_COMPOSITION_END; |
1791 | } | |
1792 | else | |
1793 | { | |
1794 | if (coding->composing == COMPOSING_WITH_RULE_RULE) | |
1795 | { | |
1796 | *dst++ = c1 & 0x7F; | |
1797 | coding->composing = COMPOSING_WITH_RULE_HEAD; | |
1798 | continue; | |
1799 | } | |
1800 | else if (coding->composing == COMPOSING_WITH_RULE_HEAD) | |
1801 | coding->composing = COMPOSING_WITH_RULE_RULE; | |
1802 | if (c1 == 0xA0) | |
1803 | { | |
1804 | /* This is an ASCII component. */ | |
1805 | ONE_MORE_BYTE (c1); | |
1806 | c1 &= 0x7F; | |
1807 | } | |
1808 | else | |
1809 | /* This is a leading-code of non ASCII component. */ | |
1810 | c1 -= 0x20; | |
1811 | } | |
1812 | } | |
1813 | ||
1814 | /* Now encode one character. C1 is a control character, an | |
1815 | ASCII character, or a leading-code of multi-byte character. */ | |
1816 | switch (emacs_code_class[c1]) | |
1817 | { | |
1818 | case EMACS_ascii_code: | |
bdd9fb48 | 1819 | ENCODE_ISO_CHARACTER (CHARSET_ASCII, c1, /* dummy */ c2); |
4ed46869 KH |
1820 | break; |
1821 | ||
1822 | case EMACS_control_code: | |
1823 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) | |
e0e989f6 | 1824 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 | 1825 | *dst++ = c1; |
d46c5b12 | 1826 | coding->consumed_char++; |
4ed46869 KH |
1827 | break; |
1828 | ||
1829 | case EMACS_carriage_return_code: | |
d46c5b12 | 1830 | if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
4ed46869 KH |
1831 | { |
1832 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) | |
e0e989f6 | 1833 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 | 1834 | *dst++ = c1; |
d46c5b12 | 1835 | coding->consumed_char++; |
4ed46869 KH |
1836 | break; |
1837 | } | |
1838 | /* fall down to treat '\r' as '\n' ... */ | |
1839 | ||
1840 | case EMACS_linefeed_code: | |
1841 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL) | |
e0e989f6 KH |
1842 | ENCODE_RESET_PLANE_AND_REGISTER; |
1843 | if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL) | |
1844 | bcopy (coding->spec.iso2022.initial_designation, | |
1845 | coding->spec.iso2022.current_designation, | |
1846 | sizeof coding->spec.iso2022.initial_designation); | |
4ed46869 | 1847 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 1848 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
1849 | *dst++ = ISO_CODE_LF; |
1850 | else if (coding->eol_type == CODING_EOL_CRLF) | |
1851 | *dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF; | |
1852 | else | |
1853 | *dst++ = ISO_CODE_CR; | |
e0e989f6 | 1854 | CODING_SPEC_ISO_BOL (coding) = 1; |
d46c5b12 | 1855 | coding->consumed_char++; |
4ed46869 KH |
1856 | break; |
1857 | ||
1858 | case EMACS_leading_code_2: | |
1859 | ONE_MORE_BYTE (c2); | |
19a8d9e0 KH |
1860 | if (c2 < 0xA0) |
1861 | { | |
1862 | /* invalid sequence */ | |
1863 | *dst++ = c1; | |
1864 | *dst++ = c2; | |
d46c5b12 | 1865 | coding->consumed_char += 2; |
19a8d9e0 KH |
1866 | } |
1867 | else | |
1868 | ENCODE_ISO_CHARACTER (c1, c2, /* dummy */ c3); | |
4ed46869 KH |
1869 | break; |
1870 | ||
1871 | case EMACS_leading_code_3: | |
1872 | TWO_MORE_BYTES (c2, c3); | |
19a8d9e0 KH |
1873 | if (c2 < 0xA0 || c3 < 0xA0) |
1874 | { | |
1875 | /* invalid sequence */ | |
1876 | *dst++ = c1; | |
1877 | *dst++ = c2; | |
1878 | *dst++ = c3; | |
d46c5b12 | 1879 | coding->consumed_char += 3; |
19a8d9e0 KH |
1880 | } |
1881 | else if (c1 < LEADING_CODE_PRIVATE_11) | |
bdd9fb48 | 1882 | ENCODE_ISO_CHARACTER (c1, c2, c3); |
4ed46869 | 1883 | else |
bdd9fb48 | 1884 | ENCODE_ISO_CHARACTER (c2, c3, /* dummy */ c4); |
4ed46869 KH |
1885 | break; |
1886 | ||
1887 | case EMACS_leading_code_4: | |
1888 | THREE_MORE_BYTES (c2, c3, c4); | |
19a8d9e0 KH |
1889 | if (c2 < 0xA0 || c3 < 0xA0 || c4 < 0xA0) |
1890 | { | |
1891 | /* invalid sequence */ | |
1892 | *dst++ = c1; | |
1893 | *dst++ = c2; | |
1894 | *dst++ = c3; | |
1895 | *dst++ = c4; | |
d46c5b12 | 1896 | coding->consumed_char += 4; |
19a8d9e0 KH |
1897 | } |
1898 | else | |
1899 | ENCODE_ISO_CHARACTER (c2, c3, c4); | |
4ed46869 KH |
1900 | break; |
1901 | ||
1902 | case EMACS_leading_code_composition: | |
19a8d9e0 KH |
1903 | ONE_MORE_BYTE (c2); |
1904 | if (c2 < 0xA0) | |
1905 | { | |
1906 | /* invalid sequence */ | |
1907 | *dst++ = c1; | |
1908 | *dst++ = c2; | |
d46c5b12 | 1909 | coding->consumed_char += 2; |
19a8d9e0 KH |
1910 | } |
1911 | else if (c2 == 0xFF) | |
4ed46869 | 1912 | { |
d46c5b12 | 1913 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1914 | coding->composing = COMPOSING_WITH_RULE_HEAD; |
1915 | ENCODE_COMPOSITION_WITH_RULE_START; | |
d46c5b12 | 1916 | coding->consumed_char++; |
4ed46869 KH |
1917 | } |
1918 | else | |
1919 | { | |
d46c5b12 | 1920 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1921 | /* Rewind one byte because it is a character code of |
1922 | composition elements. */ | |
1923 | src--; | |
1924 | coding->composing = COMPOSING_NO_RULE_HEAD; | |
1925 | ENCODE_COMPOSITION_NO_RULE_START; | |
d46c5b12 | 1926 | coding->consumed_char++; |
4ed46869 KH |
1927 | } |
1928 | break; | |
1929 | ||
1930 | case EMACS_invalid_code: | |
1931 | *dst++ = c1; | |
d46c5b12 | 1932 | coding->consumed_char++; |
4ed46869 KH |
1933 | break; |
1934 | } | |
1935 | continue; | |
1936 | label_end_of_loop: | |
d46c5b12 KH |
1937 | result = CODING_FINISH_INSUFFICIENT_SRC; |
1938 | src = src_base; | |
4ed46869 KH |
1939 | break; |
1940 | } | |
1941 | ||
fb88bf2d KH |
1942 | if (src < src_end) |
1943 | { | |
1944 | if (result == CODING_FINISH_NORMAL) | |
1945 | result = CODING_FINISH_INSUFFICIENT_DST; | |
1946 | else | |
1947 | /* If this is the last block of the text to be encoded, we | |
1948 | must reset graphic planes and registers to the initial | |
1949 | state, and flush out the carryover if any. */ | |
1950 | if (coding->mode & CODING_MODE_LAST_BLOCK) | |
1951 | ENCODE_RESET_PLANE_AND_REGISTER; | |
1952 | } | |
d46c5b12 KH |
1953 | |
1954 | coding->consumed = src - source; | |
1955 | coding->produced = coding->produced_char = dst - destination; | |
1956 | return result; | |
4ed46869 KH |
1957 | } |
1958 | ||
1959 | \f | |
1960 | /*** 4. SJIS and BIG5 handlers ***/ | |
1961 | ||
f4dee582 | 1962 | /* Although SJIS and BIG5 are not ISO's coding system, they are used |
4ed46869 KH |
1963 | quite widely. So, for the moment, Emacs supports them in the bare |
1964 | C code. But, in the future, they may be supported only by CCL. */ | |
1965 | ||
1966 | /* SJIS is a coding system encoding three character sets: ASCII, right | |
1967 | half of JISX0201-Kana, and JISX0208. An ASCII character is encoded | |
1968 | as is. A character of charset katakana-jisx0201 is encoded by | |
1969 | "position-code + 0x80". A character of charset japanese-jisx0208 | |
1970 | is encoded in 2-byte but two position-codes are divided and shifted | |
1971 | so that it fit in the range below. | |
1972 | ||
1973 | --- CODE RANGE of SJIS --- | |
1974 | (character set) (range) | |
1975 | ASCII 0x00 .. 0x7F | |
1976 | KATAKANA-JISX0201 0xA0 .. 0xDF | |
1977 | JISX0208 (1st byte) 0x80 .. 0x9F and 0xE0 .. 0xFF | |
1978 | (2nd byte) 0x40 .. 0xFF | |
1979 | ------------------------------- | |
1980 | ||
1981 | */ | |
1982 | ||
1983 | /* BIG5 is a coding system encoding two character sets: ASCII and | |
1984 | Big5. An ASCII character is encoded as is. Big5 is a two-byte | |
1985 | character set and is encoded in two-byte. | |
1986 | ||
1987 | --- CODE RANGE of BIG5 --- | |
1988 | (character set) (range) | |
1989 | ASCII 0x00 .. 0x7F | |
1990 | Big5 (1st byte) 0xA1 .. 0xFE | |
1991 | (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE | |
1992 | -------------------------- | |
1993 | ||
1994 | Since the number of characters in Big5 is larger than maximum | |
1995 | characters in Emacs' charset (96x96), it can't be handled as one | |
1996 | charset. So, in Emacs, Big5 is divided into two: `charset-big5-1' | |
1997 | and `charset-big5-2'. Both are DIMENSION2 and CHARS94. The former | |
1998 | contains frequently used characters and the latter contains less | |
1999 | frequently used characters. */ | |
2000 | ||
2001 | /* Macros to decode or encode a character of Big5 in BIG5. B1 and B2 | |
2002 | are the 1st and 2nd position-codes of Big5 in BIG5 coding system. | |
2003 | C1 and C2 are the 1st and 2nd position-codes of of Emacs' internal | |
2004 | format. CHARSET is `charset_big5_1' or `charset_big5_2'. */ | |
2005 | ||
2006 | /* Number of Big5 characters which have the same code in 1st byte. */ | |
2007 | #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) | |
2008 | ||
2009 | #define DECODE_BIG5(b1, b2, charset, c1, c2) \ | |
2010 | do { \ | |
2011 | unsigned int temp \ | |
2012 | = (b1 - 0xA1) * BIG5_SAME_ROW + b2 - (b2 < 0x7F ? 0x40 : 0x62); \ | |
2013 | if (b1 < 0xC9) \ | |
2014 | charset = charset_big5_1; \ | |
2015 | else \ | |
2016 | { \ | |
2017 | charset = charset_big5_2; \ | |
2018 | temp -= (0xC9 - 0xA1) * BIG5_SAME_ROW; \ | |
2019 | } \ | |
2020 | c1 = temp / (0xFF - 0xA1) + 0x21; \ | |
2021 | c2 = temp % (0xFF - 0xA1) + 0x21; \ | |
2022 | } while (0) | |
2023 | ||
2024 | #define ENCODE_BIG5(charset, c1, c2, b1, b2) \ | |
2025 | do { \ | |
2026 | unsigned int temp = (c1 - 0x21) * (0xFF - 0xA1) + (c2 - 0x21); \ | |
2027 | if (charset == charset_big5_2) \ | |
2028 | temp += BIG5_SAME_ROW * (0xC9 - 0xA1); \ | |
2029 | b1 = temp / BIG5_SAME_ROW + 0xA1; \ | |
2030 | b2 = temp % BIG5_SAME_ROW; \ | |
2031 | b2 += b2 < 0x3F ? 0x40 : 0x62; \ | |
2032 | } while (0) | |
2033 | ||
a5d301df KH |
2034 | #define DECODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ |
2035 | do { \ | |
2036 | int c_alt, charset_alt = (charset); \ | |
2037 | if (!NILP (unification_table) \ | |
2038 | && ((c_alt = unify_char (unification_table, \ | |
2039 | -1, (charset), c1, c2)) >= 0)) \ | |
2040 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
2041 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ | |
2042 | DECODE_CHARACTER_ASCII (c1); \ | |
2043 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
2044 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
2045 | else \ | |
2046 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
2047 | } while (0) | |
2048 | ||
2049 | #define ENCODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ | |
2050 | do { \ | |
2051 | int c_alt, charset_alt; \ | |
2052 | if (!NILP (unification_table) \ | |
2053 | && ((c_alt = unify_char (unification_table, -1, charset, c1, c2)) \ | |
2054 | >= 0)) \ | |
2055 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
2056 | else \ | |
2057 | charset_alt = charset; \ | |
2058 | if (charset_alt == charset_ascii) \ | |
2059 | *dst++ = c1; \ | |
2060 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
2061 | { \ | |
2062 | if (sjis_p && charset_alt == charset_katakana_jisx0201) \ | |
2063 | *dst++ = c1; \ | |
2064 | else \ | |
fb88bf2d KH |
2065 | { \ |
2066 | *dst++ = charset_alt, *dst++ = c1; \ | |
2067 | coding->fake_multibyte = 1; \ | |
2068 | } \ | |
a5d301df KH |
2069 | } \ |
2070 | else \ | |
2071 | { \ | |
2072 | c1 &= 0x7F, c2 &= 0x7F; \ | |
2073 | if (sjis_p && charset_alt == charset_jisx0208) \ | |
2074 | { \ | |
2075 | unsigned char s1, s2; \ | |
fb88bf2d | 2076 | \ |
a5d301df KH |
2077 | ENCODE_SJIS (c1, c2, s1, s2); \ |
2078 | *dst++ = s1, *dst++ = s2; \ | |
fb88bf2d | 2079 | coding->fake_multibyte = 1; \ |
a5d301df KH |
2080 | } \ |
2081 | else if (!sjis_p \ | |
2082 | && (charset_alt == charset_big5_1 \ | |
2083 | || charset_alt == charset_big5_2)) \ | |
2084 | { \ | |
2085 | unsigned char b1, b2; \ | |
fb88bf2d | 2086 | \ |
9ce27fde | 2087 | ENCODE_BIG5 (charset_alt, c1, c2, b1, b2); \ |
a5d301df KH |
2088 | *dst++ = b1, *dst++ = b2; \ |
2089 | } \ | |
2090 | else \ | |
fb88bf2d KH |
2091 | { \ |
2092 | *dst++ = charset_alt, *dst++ = c1, *dst++ = c2; \ | |
2093 | coding->fake_multibyte = 1; \ | |
2094 | } \ | |
a5d301df | 2095 | } \ |
d46c5b12 | 2096 | coding->consumed_char++; \ |
a5d301df KH |
2097 | } while (0); |
2098 | ||
4ed46869 KH |
2099 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
2100 | Check if a text is encoded in SJIS. If it is, return | |
2101 | CODING_CATEGORY_MASK_SJIS, else return 0. */ | |
2102 | ||
2103 | int | |
2104 | detect_coding_sjis (src, src_end) | |
2105 | unsigned char *src, *src_end; | |
2106 | { | |
2107 | unsigned char c; | |
2108 | ||
2109 | while (src < src_end) | |
2110 | { | |
2111 | c = *src++; | |
4ed46869 KH |
2112 | if ((c >= 0x80 && c < 0xA0) || c >= 0xE0) |
2113 | { | |
2114 | if (src < src_end && *src++ < 0x40) | |
2115 | return 0; | |
2116 | } | |
2117 | } | |
2118 | return CODING_CATEGORY_MASK_SJIS; | |
2119 | } | |
2120 | ||
2121 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". | |
2122 | Check if a text is encoded in BIG5. If it is, return | |
2123 | CODING_CATEGORY_MASK_BIG5, else return 0. */ | |
2124 | ||
2125 | int | |
2126 | detect_coding_big5 (src, src_end) | |
2127 | unsigned char *src, *src_end; | |
2128 | { | |
2129 | unsigned char c; | |
2130 | ||
2131 | while (src < src_end) | |
2132 | { | |
2133 | c = *src++; | |
4ed46869 KH |
2134 | if (c >= 0xA1) |
2135 | { | |
2136 | if (src >= src_end) | |
2137 | break; | |
2138 | c = *src++; | |
2139 | if (c < 0x40 || (c >= 0x7F && c <= 0xA0)) | |
2140 | return 0; | |
2141 | } | |
2142 | } | |
2143 | return CODING_CATEGORY_MASK_BIG5; | |
2144 | } | |
2145 | ||
2146 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". | |
2147 | If SJIS_P is 1, decode SJIS text, else decode BIG5 test. */ | |
2148 | ||
2149 | int | |
2150 | decode_coding_sjis_big5 (coding, source, destination, | |
d46c5b12 | 2151 | src_bytes, dst_bytes, sjis_p) |
4ed46869 KH |
2152 | struct coding_system *coding; |
2153 | unsigned char *source, *destination; | |
2154 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2155 | int sjis_p; |
2156 | { | |
2157 | unsigned char *src = source; | |
2158 | unsigned char *src_end = source + src_bytes; | |
2159 | unsigned char *dst = destination; | |
2160 | unsigned char *dst_end = destination + dst_bytes; | |
2161 | /* Since the maximum bytes produced by each loop is 4, we subtract 3 | |
2162 | from DST_END to assure overflow checking is necessary only at the | |
2163 | head of loop. */ | |
2164 | unsigned char *adjusted_dst_end = dst_end - 3; | |
a5d301df KH |
2165 | Lisp_Object unification_table |
2166 | = coding->character_unification_table_for_decode; | |
d46c5b12 | 2167 | int result = CODING_FINISH_NORMAL; |
a5d301df KH |
2168 | |
2169 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
2170 | unification_table = Vstandard_character_unification_table_for_decode; | |
4ed46869 | 2171 | |
d46c5b12 | 2172 | coding->produced_char = 0; |
fb88bf2d | 2173 | coding->fake_multibyte = 0; |
d46c5b12 KH |
2174 | while (src < src_end && (dst_bytes |
2175 | ? (dst < adjusted_dst_end) | |
2176 | : (dst < src - 3))) | |
4ed46869 KH |
2177 | { |
2178 | /* SRC_BASE remembers the start position in source in each loop. | |
2179 | The loop will be exited when there's not enough source text | |
2180 | to analyze two-byte character (within macro ONE_MORE_BYTE). | |
2181 | In that case, SRC is reset to SRC_BASE before exiting. */ | |
2182 | unsigned char *src_base = src; | |
2183 | unsigned char c1 = *src++, c2, c3, c4; | |
2184 | ||
d46c5b12 | 2185 | if (c1 < 0x20) |
4ed46869 | 2186 | { |
d46c5b12 | 2187 | if (c1 == '\r') |
4ed46869 | 2188 | { |
d46c5b12 KH |
2189 | if (coding->eol_type == CODING_EOL_CRLF) |
2190 | { | |
2191 | ONE_MORE_BYTE (c2); | |
2192 | if (c2 == '\n') | |
2193 | *dst++ = c2; | |
2194 | else if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
2195 | { | |
2196 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2197 | goto label_end_of_loop_2; | |
2198 | } | |
2199 | else | |
2200 | /* To process C2 again, SRC is subtracted by 1. */ | |
2201 | *dst++ = c1, src--; | |
2202 | } | |
2203 | else if (coding->eol_type == CODING_EOL_CR) | |
2204 | *dst++ = '\n'; | |
4ed46869 | 2205 | else |
d46c5b12 KH |
2206 | *dst++ = c1; |
2207 | } | |
2208 | else if (c1 == '\n' | |
2209 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
2210 | && (coding->eol_type == CODING_EOL_CR | |
2211 | || coding->eol_type == CODING_EOL_CRLF)) | |
2212 | { | |
2213 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2214 | goto label_end_of_loop_2; | |
4ed46869 KH |
2215 | } |
2216 | else | |
2217 | *dst++ = c1; | |
d46c5b12 | 2218 | coding->produced_char++; |
4ed46869 | 2219 | } |
a5d301df KH |
2220 | else if (c1 < 0x80) |
2221 | DECODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); | |
fb88bf2d | 2222 | else if (c1 < 0xA0) |
4ed46869 | 2223 | { |
fb88bf2d | 2224 | /* SJIS -> JISX0208 */ |
4ed46869 KH |
2225 | if (sjis_p) |
2226 | { | |
2227 | ONE_MORE_BYTE (c2); | |
fb88bf2d KH |
2228 | if (c2 >= 0x40) |
2229 | { | |
2230 | DECODE_SJIS (c1, c2, c3, c4); | |
2231 | DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208, c3, c4); | |
2232 | } | |
2233 | else | |
2234 | goto label_invalid_code_2; | |
4ed46869 | 2235 | } |
fb88bf2d KH |
2236 | else |
2237 | goto label_invalid_code_1; | |
2238 | } | |
2239 | else if (c1 < 0xE0) | |
2240 | { | |
2241 | /* SJIS -> JISX0201-Kana, BIG5 -> Big5 */ | |
2242 | if (sjis_p) | |
2243 | DECODE_SJIS_BIG5_CHARACTER (charset_katakana_jisx0201, c1, | |
2244 | /* dummy */ c2); | |
2245 | else | |
4ed46869 KH |
2246 | { |
2247 | int charset; | |
2248 | ||
2249 | ONE_MORE_BYTE (c2); | |
fb88bf2d KH |
2250 | if ((c2 >= 0x40 && c2 <= 0x7E) || (c2 >= 0xA1 && c2 <= 0xFE)) |
2251 | { | |
2252 | DECODE_BIG5 (c1, c2, charset, c3, c4); | |
2253 | DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4); | |
2254 | } | |
2255 | else | |
2256 | goto label_invalid_code_2; | |
d46c5b12 | 2257 | } |
4ed46869 | 2258 | } |
fb88bf2d | 2259 | else /* C1 >= 0xE0 */ |
4ed46869 | 2260 | { |
fb88bf2d | 2261 | /* SJIS -> JISX0208, BIG5 -> Big5 */ |
4ed46869 | 2262 | if (sjis_p) |
fb88bf2d KH |
2263 | { |
2264 | ONE_MORE_BYTE (c2); | |
2265 | if (c2 >= 0x40) | |
2266 | { | |
2267 | DECODE_SJIS (c1, c2, c3, c4); | |
2268 | DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208, c3, c4); | |
2269 | } | |
2270 | else | |
2271 | goto label_invalid_code_2; | |
2272 | } | |
4ed46869 KH |
2273 | else |
2274 | { | |
2275 | int charset; | |
2276 | ||
2277 | ONE_MORE_BYTE (c2); | |
fb88bf2d KH |
2278 | if ((c2 >= 0x40 && c2 <= 0x7E) || (c2 >= 0xA1 && c2 <= 0xFE)) |
2279 | { | |
2280 | DECODE_BIG5 (c1, c2, charset, c3, c4); | |
2281 | DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4); | |
2282 | } | |
2283 | else | |
2284 | goto label_invalid_code_2; | |
4ed46869 KH |
2285 | } |
2286 | } | |
2287 | continue; | |
2288 | ||
fb88bf2d KH |
2289 | label_invalid_code_1: |
2290 | *dst++ = c1; | |
2291 | coding->produced_char++; | |
2292 | coding->fake_multibyte = 1; | |
2293 | continue; | |
2294 | ||
2295 | label_invalid_code_2: | |
2296 | *dst++ = c1; *dst++= c2; | |
2297 | coding->produced_char += 2; | |
2298 | coding->fake_multibyte = 1; | |
2299 | continue; | |
2300 | ||
4ed46869 | 2301 | label_end_of_loop: |
d46c5b12 KH |
2302 | result = CODING_FINISH_INSUFFICIENT_SRC; |
2303 | label_end_of_loop_2: | |
4ed46869 KH |
2304 | src = src_base; |
2305 | break; | |
2306 | } | |
2307 | ||
fb88bf2d KH |
2308 | if (src < src_end) |
2309 | { | |
2310 | if (result == CODING_FINISH_NORMAL) | |
2311 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2312 | else if (result != CODING_FINISH_INCONSISTENT_EOL | |
2313 | && coding->mode & CODING_MODE_LAST_BLOCK) | |
2314 | { | |
2315 | src_bytes = src_end - src; | |
2316 | if (dst_bytes && (dst_end - dst < src_bytes)) | |
2317 | src_bytes = dst_end - dst; | |
2318 | bcopy (dst, src, src_bytes); | |
2319 | src += src_bytes; | |
2320 | dst += src_bytes; | |
2321 | coding->fake_multibyte = 1; | |
2322 | } | |
2323 | } | |
d46c5b12 KH |
2324 | |
2325 | coding->consumed = coding->consumed_char = src - source; | |
2326 | coding->produced = dst - destination; | |
2327 | return result; | |
4ed46869 KH |
2328 | } |
2329 | ||
2330 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". | |
2331 | This function can encode `charset_ascii', `charset_katakana_jisx0201', | |
2332 | `charset_jisx0208', `charset_big5_1', and `charset_big5-2'. We are | |
2333 | sure that all these charsets are registered as official charset | |
2334 | (i.e. do not have extended leading-codes). Characters of other | |
2335 | charsets are produced without any encoding. If SJIS_P is 1, encode | |
2336 | SJIS text, else encode BIG5 text. */ | |
2337 | ||
2338 | int | |
2339 | encode_coding_sjis_big5 (coding, source, destination, | |
d46c5b12 | 2340 | src_bytes, dst_bytes, sjis_p) |
4ed46869 KH |
2341 | struct coding_system *coding; |
2342 | unsigned char *source, *destination; | |
2343 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2344 | int sjis_p; |
2345 | { | |
2346 | unsigned char *src = source; | |
2347 | unsigned char *src_end = source + src_bytes; | |
2348 | unsigned char *dst = destination; | |
2349 | unsigned char *dst_end = destination + dst_bytes; | |
2350 | /* Since the maximum bytes produced by each loop is 2, we subtract 1 | |
2351 | from DST_END to assure overflow checking is necessary only at the | |
2352 | head of loop. */ | |
2353 | unsigned char *adjusted_dst_end = dst_end - 1; | |
a5d301df KH |
2354 | Lisp_Object unification_table |
2355 | = coding->character_unification_table_for_encode; | |
d46c5b12 | 2356 | int result = CODING_FINISH_NORMAL; |
a5d301df KH |
2357 | |
2358 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
2359 | unification_table = Vstandard_character_unification_table_for_encode; | |
4ed46869 | 2360 | |
d46c5b12 | 2361 | coding->consumed_char = 0; |
fb88bf2d | 2362 | coding->fake_multibyte = 0; |
d46c5b12 KH |
2363 | while (src < src_end && (dst_bytes |
2364 | ? (dst < adjusted_dst_end) | |
2365 | : (dst < src - 1))) | |
4ed46869 KH |
2366 | { |
2367 | /* SRC_BASE remembers the start position in source in each loop. | |
2368 | The loop will be exited when there's not enough source text | |
2369 | to analyze multi-byte codes (within macros ONE_MORE_BYTE and | |
2370 | TWO_MORE_BYTES). In that case, SRC is reset to SRC_BASE | |
2371 | before exiting. */ | |
2372 | unsigned char *src_base = src; | |
2373 | unsigned char c1 = *src++, c2, c3, c4; | |
2374 | ||
2375 | if (coding->composing) | |
2376 | { | |
2377 | if (c1 == 0xA0) | |
2378 | { | |
2379 | ONE_MORE_BYTE (c1); | |
2380 | c1 &= 0x7F; | |
2381 | } | |
2382 | else if (c1 >= 0xA0) | |
2383 | c1 -= 0x20; | |
2384 | else | |
2385 | coding->composing = 0; | |
2386 | } | |
2387 | ||
2388 | switch (emacs_code_class[c1]) | |
2389 | { | |
2390 | case EMACS_ascii_code: | |
a5d301df KH |
2391 | ENCODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); |
2392 | break; | |
2393 | ||
4ed46869 KH |
2394 | case EMACS_control_code: |
2395 | *dst++ = c1; | |
d46c5b12 | 2396 | coding->consumed_char++; |
4ed46869 KH |
2397 | break; |
2398 | ||
2399 | case EMACS_carriage_return_code: | |
d46c5b12 | 2400 | if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
4ed46869 KH |
2401 | { |
2402 | *dst++ = c1; | |
d46c5b12 | 2403 | coding->consumed_char++; |
4ed46869 KH |
2404 | break; |
2405 | } | |
2406 | /* fall down to treat '\r' as '\n' ... */ | |
2407 | ||
2408 | case EMACS_linefeed_code: | |
2409 | if (coding->eol_type == CODING_EOL_LF | |
0ef69138 | 2410 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
2411 | *dst++ = '\n'; |
2412 | else if (coding->eol_type == CODING_EOL_CRLF) | |
2413 | *dst++ = '\r', *dst++ = '\n'; | |
2414 | else | |
2415 | *dst++ = '\r'; | |
d46c5b12 | 2416 | coding->consumed_char++; |
4ed46869 KH |
2417 | break; |
2418 | ||
2419 | case EMACS_leading_code_2: | |
2420 | ONE_MORE_BYTE (c2); | |
a5d301df | 2421 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, /* dummy */ c3); |
4ed46869 KH |
2422 | break; |
2423 | ||
2424 | case EMACS_leading_code_3: | |
2425 | TWO_MORE_BYTES (c2, c3); | |
a5d301df | 2426 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, c3); |
4ed46869 KH |
2427 | break; |
2428 | ||
2429 | case EMACS_leading_code_4: | |
2430 | THREE_MORE_BYTES (c2, c3, c4); | |
a5d301df | 2431 | ENCODE_SJIS_BIG5_CHARACTER (c2, c3, c4); |
4ed46869 KH |
2432 | break; |
2433 | ||
2434 | case EMACS_leading_code_composition: | |
2435 | coding->composing = 1; | |
2436 | break; | |
2437 | ||
2438 | default: /* i.e. case EMACS_invalid_code: */ | |
2439 | *dst++ = c1; | |
d46c5b12 | 2440 | coding->consumed_char++; |
4ed46869 KH |
2441 | } |
2442 | continue; | |
2443 | ||
2444 | label_end_of_loop: | |
d46c5b12 KH |
2445 | result = CODING_FINISH_INSUFFICIENT_SRC; |
2446 | src = src_base; | |
4ed46869 KH |
2447 | break; |
2448 | } | |
2449 | ||
d46c5b12 KH |
2450 | if (result == CODING_FINISH_NORMAL |
2451 | && src < src_end) | |
2452 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2453 | coding->consumed = src - source; | |
2454 | coding->produced = coding->produced_char = dst - destination; | |
2455 | return result; | |
4ed46869 KH |
2456 | } |
2457 | ||
2458 | \f | |
2459 | /*** 5. End-of-line handlers ***/ | |
2460 | ||
2461 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". | |
2462 | This function is called only when `coding->eol_type' is | |
2463 | CODING_EOL_CRLF or CODING_EOL_CR. */ | |
2464 | ||
d46c5b12 | 2465 | decode_eol (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
2466 | struct coding_system *coding; |
2467 | unsigned char *source, *destination; | |
2468 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2469 | { |
2470 | unsigned char *src = source; | |
2471 | unsigned char *src_end = source + src_bytes; | |
2472 | unsigned char *dst = destination; | |
2473 | unsigned char *dst_end = destination + dst_bytes; | |
fb88bf2d | 2474 | unsigned char c; |
d46c5b12 KH |
2475 | int result = CODING_FINISH_NORMAL; |
2476 | ||
fb88bf2d KH |
2477 | coding->fake_multibyte = 0; |
2478 | ||
d46c5b12 KH |
2479 | if (src_bytes <= 0) |
2480 | return result; | |
4ed46869 KH |
2481 | |
2482 | switch (coding->eol_type) | |
2483 | { | |
2484 | case CODING_EOL_CRLF: | |
2485 | { | |
2486 | /* Since the maximum bytes produced by each loop is 2, we | |
2487 | subtract 1 from DST_END to assure overflow checking is | |
2488 | necessary only at the head of loop. */ | |
2489 | unsigned char *adjusted_dst_end = dst_end - 1; | |
2490 | ||
d46c5b12 KH |
2491 | while (src < src_end && (dst_bytes |
2492 | ? (dst < adjusted_dst_end) | |
2493 | : (dst < src - 1))) | |
4ed46869 KH |
2494 | { |
2495 | unsigned char *src_base = src; | |
fb88bf2d KH |
2496 | |
2497 | c = *src++; | |
4ed46869 KH |
2498 | if (c == '\r') |
2499 | { | |
2500 | ONE_MORE_BYTE (c); | |
2501 | if (c != '\n') | |
d46c5b12 KH |
2502 | { |
2503 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
2504 | { | |
2505 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2506 | goto label_end_of_loop_2; | |
2507 | } | |
2508 | *dst++ = '\r'; | |
fb88bf2d KH |
2509 | if (BASE_LEADING_CODE_P (c)) |
2510 | coding->fake_multibyte = 1; | |
d46c5b12 | 2511 | } |
bfd99048 | 2512 | *dst++ = c; |
4ed46869 | 2513 | } |
d46c5b12 KH |
2514 | else if (c == '\n' |
2515 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)) | |
2516 | { | |
2517 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2518 | goto label_end_of_loop_2; | |
2519 | } | |
4ed46869 | 2520 | else |
fb88bf2d KH |
2521 | { |
2522 | *dst++ = c; | |
2523 | if (BASE_LEADING_CODE_P (c)) | |
2524 | coding->fake_multibyte = 1; | |
2525 | } | |
4ed46869 KH |
2526 | continue; |
2527 | ||
2528 | label_end_of_loop: | |
d46c5b12 KH |
2529 | result = CODING_FINISH_INSUFFICIENT_SRC; |
2530 | label_end_of_loop_2: | |
4ed46869 KH |
2531 | src = src_base; |
2532 | break; | |
2533 | } | |
d46c5b12 KH |
2534 | if (result == CODING_FINISH_NORMAL |
2535 | && src < src_end) | |
2536 | result = CODING_FINISH_INSUFFICIENT_DST; | |
4ed46869 | 2537 | } |
d46c5b12 | 2538 | break; |
4ed46869 KH |
2539 | |
2540 | case CODING_EOL_CR: | |
d46c5b12 KH |
2541 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
2542 | { | |
fb88bf2d KH |
2543 | while (src < src_end) |
2544 | { | |
2545 | if ((c = *src++) == '\n') | |
2546 | break; | |
2547 | if (BASE_LEADING_CODE_P (c)) | |
2548 | coding->fake_multibyte = 1; | |
2549 | } | |
d46c5b12 KH |
2550 | if (*--src == '\n') |
2551 | { | |
2552 | src_bytes = src - source; | |
2553 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2554 | } | |
2555 | } | |
2556 | if (dst_bytes && src_bytes > dst_bytes) | |
2557 | { | |
2558 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2559 | src_bytes = dst_bytes; | |
2560 | } | |
2561 | if (dst_bytes) | |
2562 | bcopy (source, destination, src_bytes); | |
2563 | else | |
2564 | safe_bcopy (source, destination, src_bytes); | |
2565 | src = source + src_bytes; | |
2566 | while (src_bytes--) if (*dst++ == '\r') dst[-1] = '\n'; | |
4ed46869 KH |
2567 | break; |
2568 | ||
2569 | default: /* i.e. case: CODING_EOL_LF */ | |
d46c5b12 KH |
2570 | if (dst_bytes && src_bytes > dst_bytes) |
2571 | { | |
2572 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2573 | src_bytes = dst_bytes; | |
2574 | } | |
2575 | if (dst_bytes) | |
2576 | bcopy (source, destination, src_bytes); | |
2577 | else | |
2578 | safe_bcopy (source, destination, src_bytes); | |
2579 | src += src_bytes; | |
2580 | dst += dst_bytes; | |
fb88bf2d | 2581 | coding->fake_multibyte = 1; |
4ed46869 KH |
2582 | break; |
2583 | } | |
2584 | ||
d46c5b12 KH |
2585 | coding->consumed = coding->consumed_char = src - source; |
2586 | coding->produced = coding->produced_char = dst - destination; | |
2587 | return result; | |
4ed46869 KH |
2588 | } |
2589 | ||
2590 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". Encode | |
2591 | format of end-of-line according to `coding->eol_type'. If | |
d46c5b12 KH |
2592 | `coding->mode & CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code |
2593 | '\r' in source text also means end-of-line. */ | |
4ed46869 | 2594 | |
d46c5b12 | 2595 | encode_eol (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
2596 | struct coding_system *coding; |
2597 | unsigned char *source, *destination; | |
2598 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2599 | { |
2600 | unsigned char *src = source; | |
2601 | unsigned char *dst = destination; | |
d46c5b12 | 2602 | int result = CODING_FINISH_NORMAL; |
4ed46869 | 2603 | |
fb88bf2d KH |
2604 | coding->fake_multibyte = 0; |
2605 | ||
d46c5b12 KH |
2606 | if (coding->eol_type == CODING_EOL_CRLF) |
2607 | { | |
2608 | unsigned char c; | |
2609 | unsigned char *src_end = source + src_bytes; | |
2610 | unsigned char *dst_end = destination + dst_bytes; | |
2611 | /* Since the maximum bytes produced by each loop is 2, we | |
2612 | subtract 1 from DST_END to assure overflow checking is | |
2613 | necessary only at the head of loop. */ | |
2614 | unsigned char *adjusted_dst_end = dst_end - 1; | |
2615 | ||
2616 | while (src < src_end && (dst_bytes | |
2617 | ? (dst < adjusted_dst_end) | |
2618 | : (dst < src - 1))) | |
2619 | { | |
2620 | c = *src++; | |
2621 | if (c == '\n' | |
2622 | || (c == '\r' && (coding->mode & CODING_MODE_SELECTIVE_DISPLAY))) | |
2623 | *dst++ = '\r', *dst++ = '\n'; | |
2624 | else | |
fb88bf2d KH |
2625 | { |
2626 | *dst++ = c; | |
2627 | if (BASE_LEADING_CODE_P (c)) | |
2628 | coding->fake_multibyte = 1; | |
2629 | } | |
d46c5b12 KH |
2630 | } |
2631 | if (src < src_end) | |
2632 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2633 | } | |
2634 | else | |
4ed46869 | 2635 | { |
fb88bf2d KH |
2636 | unsigned char c; |
2637 | ||
d46c5b12 | 2638 | if (dst_bytes && src_bytes > dst_bytes) |
4ed46869 | 2639 | { |
d46c5b12 KH |
2640 | src_bytes = dst_bytes; |
2641 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2642 | } | |
2643 | if (dst_bytes) | |
2644 | bcopy (source, destination, src_bytes); | |
2645 | else | |
fb88bf2d KH |
2646 | { |
2647 | safe_bcopy (source, destination, src_bytes); | |
2648 | dst_bytes = src_bytes; | |
2649 | } | |
d46c5b12 KH |
2650 | if (coding->eol_type == CODING_EOL_CRLF) |
2651 | { | |
2652 | while (src_bytes--) | |
fb88bf2d KH |
2653 | { |
2654 | if ((c = *dst++) == '\n') | |
2655 | dst[-1] = '\r'; | |
2656 | else if (BASE_LEADING_CODE_P (c)) | |
2657 | coding->fake_multibyte = 1; | |
2658 | } | |
d46c5b12 | 2659 | } |
fb88bf2d | 2660 | else |
d46c5b12 | 2661 | { |
fb88bf2d KH |
2662 | if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY) |
2663 | { | |
2664 | while (src_bytes--) | |
2665 | if (*dst++ == '\r') dst[-1] = '\n'; | |
2666 | } | |
2667 | coding->fake_multibyte = 1; | |
4ed46869 | 2668 | } |
fb88bf2d KH |
2669 | src = source + dst_bytes; |
2670 | dst = destination + dst_bytes; | |
4ed46869 KH |
2671 | } |
2672 | ||
d46c5b12 KH |
2673 | coding->consumed = coding->consumed_char = src - source; |
2674 | coding->produced = coding->produced_char = dst - destination; | |
2675 | return result; | |
4ed46869 KH |
2676 | } |
2677 | ||
2678 | \f | |
2679 | /*** 6. C library functions ***/ | |
2680 | ||
2681 | /* In Emacs Lisp, coding system is represented by a Lisp symbol which | |
2682 | has a property `coding-system'. The value of this property is a | |
2683 | vector of length 5 (called as coding-vector). Among elements of | |
2684 | this vector, the first (element[0]) and the fifth (element[4]) | |
2685 | carry important information for decoding/encoding. Before | |
2686 | decoding/encoding, this information should be set in fields of a | |
2687 | structure of type `coding_system'. | |
2688 | ||
2689 | A value of property `coding-system' can be a symbol of another | |
2690 | subsidiary coding-system. In that case, Emacs gets coding-vector | |
2691 | from that symbol. | |
2692 | ||
2693 | `element[0]' contains information to be set in `coding->type'. The | |
2694 | value and its meaning is as follows: | |
2695 | ||
0ef69138 KH |
2696 | 0 -- coding_type_emacs_mule |
2697 | 1 -- coding_type_sjis | |
2698 | 2 -- coding_type_iso2022 | |
2699 | 3 -- coding_type_big5 | |
2700 | 4 -- coding_type_ccl encoder/decoder written in CCL | |
2701 | nil -- coding_type_no_conversion | |
2702 | t -- coding_type_undecided (automatic conversion on decoding, | |
2703 | no-conversion on encoding) | |
4ed46869 KH |
2704 | |
2705 | `element[4]' contains information to be set in `coding->flags' and | |
2706 | `coding->spec'. The meaning varies by `coding->type'. | |
2707 | ||
2708 | If `coding->type' is `coding_type_iso2022', element[4] is a vector | |
2709 | of length 32 (of which the first 13 sub-elements are used now). | |
2710 | Meanings of these sub-elements are: | |
2711 | ||
2712 | sub-element[N] where N is 0 through 3: to be set in `coding->spec.iso2022' | |
2713 | If the value is an integer of valid charset, the charset is | |
2714 | assumed to be designated to graphic register N initially. | |
2715 | ||
2716 | If the value is minus, it is a minus value of charset which | |
2717 | reserves graphic register N, which means that the charset is | |
2718 | not designated initially but should be designated to graphic | |
2719 | register N just before encoding a character in that charset. | |
2720 | ||
2721 | If the value is nil, graphic register N is never used on | |
2722 | encoding. | |
2723 | ||
2724 | sub-element[N] where N is 4 through 11: to be set in `coding->flags' | |
2725 | Each value takes t or nil. See the section ISO2022 of | |
2726 | `coding.h' for more information. | |
2727 | ||
2728 | If `coding->type' is `coding_type_big5', element[4] is t to denote | |
2729 | BIG5-ETen or nil to denote BIG5-HKU. | |
2730 | ||
2731 | If `coding->type' takes the other value, element[4] is ignored. | |
2732 | ||
2733 | Emacs Lisp's coding system also carries information about format of | |
2734 | end-of-line in a value of property `eol-type'. If the value is | |
2735 | integer, 0 means CODING_EOL_LF, 1 means CODING_EOL_CRLF, and 2 | |
2736 | means CODING_EOL_CR. If it is not integer, it should be a vector | |
2737 | of subsidiary coding systems of which property `eol-type' has one | |
2738 | of above values. | |
2739 | ||
2740 | */ | |
2741 | ||
2742 | /* Extract information for decoding/encoding from CODING_SYSTEM_SYMBOL | |
2743 | and set it in CODING. If CODING_SYSTEM_SYMBOL is invalid, CODING | |
2744 | is setup so that no conversion is necessary and return -1, else | |
2745 | return 0. */ | |
2746 | ||
2747 | int | |
e0e989f6 KH |
2748 | setup_coding_system (coding_system, coding) |
2749 | Lisp_Object coding_system; | |
4ed46869 KH |
2750 | struct coding_system *coding; |
2751 | { | |
d46c5b12 | 2752 | Lisp_Object coding_spec, coding_type, eol_type, plist; |
4608c386 | 2753 | Lisp_Object val; |
70c22245 | 2754 | int i; |
4ed46869 | 2755 | |
d46c5b12 | 2756 | /* Initialize some fields required for all kinds of coding systems. */ |
774324d6 | 2757 | coding->symbol = coding_system; |
d46c5b12 KH |
2758 | coding->common_flags = 0; |
2759 | coding->mode = 0; | |
2760 | coding->heading_ascii = -1; | |
2761 | coding->post_read_conversion = coding->pre_write_conversion = Qnil; | |
4608c386 KH |
2762 | coding_spec = Fget (coding_system, Qcoding_system); |
2763 | if (!VECTORP (coding_spec) | |
2764 | || XVECTOR (coding_spec)->size != 5 | |
2765 | || !CONSP (XVECTOR (coding_spec)->contents[3])) | |
4ed46869 | 2766 | goto label_invalid_coding_system; |
4608c386 | 2767 | |
d46c5b12 KH |
2768 | eol_type = inhibit_eol_conversion ? Qnil : Fget (coding_system, Qeol_type); |
2769 | if (VECTORP (eol_type)) | |
2770 | { | |
2771 | coding->eol_type = CODING_EOL_UNDECIDED; | |
2772 | coding->common_flags = CODING_REQUIRE_DETECTION_MASK; | |
2773 | } | |
2774 | else if (XFASTINT (eol_type) == 1) | |
2775 | { | |
2776 | coding->eol_type = CODING_EOL_CRLF; | |
2777 | coding->common_flags | |
2778 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
2779 | } | |
2780 | else if (XFASTINT (eol_type) == 2) | |
2781 | { | |
2782 | coding->eol_type = CODING_EOL_CR; | |
2783 | coding->common_flags | |
2784 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
2785 | } | |
2786 | else | |
2787 | coding->eol_type = CODING_EOL_LF; | |
2788 | ||
2789 | coding_type = XVECTOR (coding_spec)->contents[0]; | |
2790 | /* Try short cut. */ | |
2791 | if (SYMBOLP (coding_type)) | |
2792 | { | |
2793 | if (EQ (coding_type, Qt)) | |
2794 | { | |
2795 | coding->type = coding_type_undecided; | |
2796 | coding->common_flags |= CODING_REQUIRE_DETECTION_MASK; | |
2797 | } | |
2798 | else | |
2799 | coding->type = coding_type_no_conversion; | |
2800 | return 0; | |
2801 | } | |
2802 | ||
2803 | /* Initialize remaining fields. */ | |
2804 | coding->composing = 0; | |
2805 | coding->character_unification_table_for_decode = Qnil; | |
2806 | coding->character_unification_table_for_encode = Qnil; | |
2807 | ||
2808 | /* Get values of coding system properties: | |
2809 | `post-read-conversion', `pre-write-conversion', | |
2810 | `character-unification-table-for-decode', | |
2811 | `character-unification-table-for-encode'. */ | |
4608c386 KH |
2812 | plist = XVECTOR (coding_spec)->contents[3]; |
2813 | coding->post_read_conversion = Fplist_get (plist, Qpost_read_conversion); | |
2814 | coding->pre_write_conversion = Fplist_get (plist, Qpre_write_conversion); | |
2815 | val = Fplist_get (plist, Qcharacter_unification_table_for_decode); | |
2816 | if (SYMBOLP (val)) | |
2817 | val = Fget (val, Qcharacter_unification_table_for_decode); | |
2818 | coding->character_unification_table_for_decode | |
2819 | = CHAR_TABLE_P (val) ? val : Qnil; | |
2820 | val = Fplist_get (plist, Qcharacter_unification_table_for_encode); | |
2821 | if (SYMBOLP (val)) | |
2822 | val = Fget (val, Qcharacter_unification_table_for_encode); | |
2823 | coding->character_unification_table_for_encode | |
2824 | = CHAR_TABLE_P (val) ? val : Qnil; | |
d46c5b12 KH |
2825 | val = Fplist_get (plist, Qcoding_category); |
2826 | if (!NILP (val)) | |
2827 | { | |
2828 | val = Fget (val, Qcoding_category_index); | |
2829 | if (INTEGERP (val)) | |
2830 | coding->category_idx = XINT (val); | |
2831 | else | |
2832 | goto label_invalid_coding_system; | |
2833 | } | |
2834 | else | |
2835 | goto label_invalid_coding_system; | |
4608c386 | 2836 | |
70c22245 KH |
2837 | val = Fplist_get (plist, Qsafe_charsets); |
2838 | if (EQ (val, Qt)) | |
2839 | { | |
2840 | for (i = 0; i <= MAX_CHARSET; i++) | |
2841 | coding->safe_charsets[i] = 1; | |
2842 | } | |
2843 | else | |
2844 | { | |
2845 | bzero (coding->safe_charsets, MAX_CHARSET + 1); | |
2846 | while (CONSP (val)) | |
2847 | { | |
2848 | if ((i = get_charset_id (XCONS (val)->car)) >= 0) | |
2849 | coding->safe_charsets[i] = 1; | |
2850 | val = XCONS (val)->cdr; | |
2851 | } | |
2852 | } | |
2853 | ||
d46c5b12 | 2854 | switch (XFASTINT (coding_type)) |
4ed46869 KH |
2855 | { |
2856 | case 0: | |
0ef69138 | 2857 | coding->type = coding_type_emacs_mule; |
c952af22 KH |
2858 | if (!NILP (coding->post_read_conversion)) |
2859 | coding->common_flags |= CODING_REQUIRE_DECODING_MASK; | |
2860 | if (!NILP (coding->pre_write_conversion)) | |
2861 | coding->common_flags |= CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 KH |
2862 | break; |
2863 | ||
2864 | case 1: | |
2865 | coding->type = coding_type_sjis; | |
c952af22 KH |
2866 | coding->common_flags |
2867 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 KH |
2868 | break; |
2869 | ||
2870 | case 2: | |
2871 | coding->type = coding_type_iso2022; | |
c952af22 KH |
2872 | coding->common_flags |
2873 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 | 2874 | { |
70c22245 | 2875 | Lisp_Object val, temp; |
4ed46869 | 2876 | Lisp_Object *flags; |
d46c5b12 | 2877 | int i, charset, reg_bits = 0; |
4ed46869 | 2878 | |
4608c386 | 2879 | val = XVECTOR (coding_spec)->contents[4]; |
f44d27ce | 2880 | |
4ed46869 KH |
2881 | if (!VECTORP (val) || XVECTOR (val)->size != 32) |
2882 | goto label_invalid_coding_system; | |
2883 | ||
2884 | flags = XVECTOR (val)->contents; | |
2885 | coding->flags | |
2886 | = ((NILP (flags[4]) ? 0 : CODING_FLAG_ISO_SHORT_FORM) | |
2887 | | (NILP (flags[5]) ? 0 : CODING_FLAG_ISO_RESET_AT_EOL) | |
2888 | | (NILP (flags[6]) ? 0 : CODING_FLAG_ISO_RESET_AT_CNTL) | |
2889 | | (NILP (flags[7]) ? 0 : CODING_FLAG_ISO_SEVEN_BITS) | |
2890 | | (NILP (flags[8]) ? 0 : CODING_FLAG_ISO_LOCKING_SHIFT) | |
2891 | | (NILP (flags[9]) ? 0 : CODING_FLAG_ISO_SINGLE_SHIFT) | |
2892 | | (NILP (flags[10]) ? 0 : CODING_FLAG_ISO_USE_ROMAN) | |
2893 | | (NILP (flags[11]) ? 0 : CODING_FLAG_ISO_USE_OLDJIS) | |
e0e989f6 KH |
2894 | | (NILP (flags[12]) ? 0 : CODING_FLAG_ISO_NO_DIRECTION) |
2895 | | (NILP (flags[13]) ? 0 : CODING_FLAG_ISO_INIT_AT_BOL) | |
c4825358 KH |
2896 | | (NILP (flags[14]) ? 0 : CODING_FLAG_ISO_DESIGNATE_AT_BOL) |
2897 | | (NILP (flags[15]) ? 0 : CODING_FLAG_ISO_SAFE) | |
3f003981 | 2898 | | (NILP (flags[16]) ? 0 : CODING_FLAG_ISO_LATIN_EXTRA) |
c4825358 | 2899 | ); |
4ed46869 KH |
2900 | |
2901 | /* Invoke graphic register 0 to plane 0. */ | |
2902 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; | |
2903 | /* Invoke graphic register 1 to plane 1 if we can use full 8-bit. */ | |
2904 | CODING_SPEC_ISO_INVOCATION (coding, 1) | |
2905 | = (coding->flags & CODING_FLAG_ISO_SEVEN_BITS ? -1 : 1); | |
2906 | /* Not single shifting at first. */ | |
6e85d753 | 2907 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; |
e0e989f6 | 2908 | /* Beginning of buffer should also be regarded as bol. */ |
6e85d753 | 2909 | CODING_SPEC_ISO_BOL (coding) = 1; |
4ed46869 | 2910 | |
70c22245 KH |
2911 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
2912 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = 255; | |
2913 | val = Vcharset_revision_alist; | |
2914 | while (CONSP (val)) | |
2915 | { | |
2916 | charset = get_charset_id (Fcar_safe (XCONS (val)->car)); | |
2917 | if (charset >= 0 | |
2918 | && (temp = Fcdr_safe (XCONS (val)->car), INTEGERP (temp)) | |
2919 | && (i = XINT (temp), (i >= 0 && (i + '@') < 128))) | |
2920 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = i; | |
2921 | val = XCONS (val)->cdr; | |
2922 | } | |
2923 | ||
4ed46869 KH |
2924 | /* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations. |
2925 | FLAGS[REG] can be one of below: | |
2926 | integer CHARSET: CHARSET occupies register I, | |
2927 | t: designate nothing to REG initially, but can be used | |
2928 | by any charsets, | |
2929 | list of integer, nil, or t: designate the first | |
2930 | element (if integer) to REG initially, the remaining | |
2931 | elements (if integer) is designated to REG on request, | |
d46c5b12 | 2932 | if an element is t, REG can be used by any charsets, |
4ed46869 | 2933 | nil: REG is never used. */ |
467e7675 | 2934 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
1ba9e4ab KH |
2935 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
2936 | = CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION; | |
4ed46869 KH |
2937 | for (i = 0; i < 4; i++) |
2938 | { | |
2939 | if (INTEGERP (flags[i]) | |
e0e989f6 KH |
2940 | && (charset = XINT (flags[i]), CHARSET_VALID_P (charset)) |
2941 | || (charset = get_charset_id (flags[i])) >= 0) | |
4ed46869 KH |
2942 | { |
2943 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; | |
2944 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) = i; | |
2945 | } | |
2946 | else if (EQ (flags[i], Qt)) | |
2947 | { | |
2948 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
d46c5b12 KH |
2949 | reg_bits |= 1 << i; |
2950 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; | |
4ed46869 KH |
2951 | } |
2952 | else if (CONSP (flags[i])) | |
2953 | { | |
2954 | Lisp_Object tail = flags[i]; | |
2955 | ||
d46c5b12 | 2956 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; |
4ed46869 KH |
2957 | if (INTEGERP (XCONS (tail)->car) |
2958 | && (charset = XINT (XCONS (tail)->car), | |
e0e989f6 KH |
2959 | CHARSET_VALID_P (charset)) |
2960 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) | |
4ed46869 KH |
2961 | { |
2962 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; | |
2963 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) =i; | |
2964 | } | |
2965 | else | |
2966 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
2967 | tail = XCONS (tail)->cdr; | |
2968 | while (CONSP (tail)) | |
2969 | { | |
2970 | if (INTEGERP (XCONS (tail)->car) | |
2971 | && (charset = XINT (XCONS (tail)->car), | |
e0e989f6 KH |
2972 | CHARSET_VALID_P (charset)) |
2973 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) | |
70c22245 KH |
2974 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
2975 | = i; | |
4ed46869 | 2976 | else if (EQ (XCONS (tail)->car, Qt)) |
d46c5b12 | 2977 | reg_bits |= 1 << i; |
4ed46869 KH |
2978 | tail = XCONS (tail)->cdr; |
2979 | } | |
2980 | } | |
2981 | else | |
2982 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
2983 | ||
2984 | CODING_SPEC_ISO_DESIGNATION (coding, i) | |
2985 | = CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i); | |
2986 | } | |
2987 | ||
d46c5b12 | 2988 | if (reg_bits && ! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
4ed46869 KH |
2989 | { |
2990 | /* REG 1 can be used only by locking shift in 7-bit env. */ | |
2991 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) | |
d46c5b12 | 2992 | reg_bits &= ~2; |
4ed46869 KH |
2993 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
2994 | /* Without any shifting, only REG 0 and 1 can be used. */ | |
d46c5b12 | 2995 | reg_bits &= 3; |
4ed46869 KH |
2996 | } |
2997 | ||
d46c5b12 KH |
2998 | if (reg_bits) |
2999 | for (charset = 0; charset <= MAX_CHARSET; charset++) | |
6e85d753 | 3000 | { |
d46c5b12 KH |
3001 | if (CHARSET_VALID_P (charset)) |
3002 | { | |
3003 | /* There exist some default graphic registers to be | |
3004 | used CHARSET. */ | |
3005 | ||
3006 | /* We had better avoid designating a charset of | |
3007 | CHARS96 to REG 0 as far as possible. */ | |
3008 | if (CHARSET_CHARS (charset) == 96) | |
3009 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) | |
3010 | = (reg_bits & 2 | |
3011 | ? 1 : (reg_bits & 4 ? 2 : (reg_bits & 8 ? 3 : 0))); | |
3012 | else | |
3013 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) | |
3014 | = (reg_bits & 1 | |
3015 | ? 0 : (reg_bits & 2 ? 1 : (reg_bits & 4 ? 2 : 3))); | |
3016 | } | |
6e85d753 | 3017 | } |
4ed46869 | 3018 | } |
c952af22 | 3019 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
d46c5b12 | 3020 | coding->spec.iso2022.last_invalid_designation_register = -1; |
4ed46869 KH |
3021 | break; |
3022 | ||
3023 | case 3: | |
3024 | coding->type = coding_type_big5; | |
c952af22 KH |
3025 | coding->common_flags |
3026 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 | 3027 | coding->flags |
4608c386 | 3028 | = (NILP (XVECTOR (coding_spec)->contents[4]) |
4ed46869 KH |
3029 | ? CODING_FLAG_BIG5_HKU |
3030 | : CODING_FLAG_BIG5_ETEN); | |
3031 | break; | |
3032 | ||
3033 | case 4: | |
3034 | coding->type = coding_type_ccl; | |
c952af22 KH |
3035 | coding->common_flags |
3036 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 | 3037 | { |
4608c386 | 3038 | Lisp_Object val = XVECTOR (coding_spec)->contents[4]; |
4ed46869 KH |
3039 | if (CONSP (val) |
3040 | && VECTORP (XCONS (val)->car) | |
3041 | && VECTORP (XCONS (val)->cdr)) | |
3042 | { | |
3043 | setup_ccl_program (&(coding->spec.ccl.decoder), XCONS (val)->car); | |
3044 | setup_ccl_program (&(coding->spec.ccl.encoder), XCONS (val)->cdr); | |
3045 | } | |
3046 | else | |
3047 | goto label_invalid_coding_system; | |
3048 | } | |
c952af22 | 3049 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
4ed46869 KH |
3050 | break; |
3051 | ||
27901516 KH |
3052 | case 5: |
3053 | coding->type = coding_type_raw_text; | |
3054 | break; | |
3055 | ||
4ed46869 | 3056 | default: |
d46c5b12 | 3057 | goto label_invalid_coding_system; |
4ed46869 KH |
3058 | } |
3059 | return 0; | |
3060 | ||
3061 | label_invalid_coding_system: | |
3062 | coding->type = coding_type_no_conversion; | |
d46c5b12 | 3063 | coding->category_idx = CODING_CATEGORY_IDX_BINARY; |
c952af22 | 3064 | coding->common_flags = 0; |
dec137e5 | 3065 | coding->eol_type = CODING_EOL_LF; |
d46c5b12 | 3066 | coding->pre_write_conversion = coding->post_read_conversion = Qnil; |
4ed46869 KH |
3067 | return -1; |
3068 | } | |
3069 | ||
3070 | /* Emacs has a mechanism to automatically detect a coding system if it | |
3071 | is one of Emacs' internal format, ISO2022, SJIS, and BIG5. But, | |
3072 | it's impossible to distinguish some coding systems accurately | |
3073 | because they use the same range of codes. So, at first, coding | |
3074 | systems are categorized into 7, those are: | |
3075 | ||
0ef69138 | 3076 | o coding-category-emacs-mule |
4ed46869 KH |
3077 | |
3078 | The category for a coding system which has the same code range | |
3079 | as Emacs' internal format. Assigned the coding-system (Lisp | |
0ef69138 | 3080 | symbol) `emacs-mule' by default. |
4ed46869 KH |
3081 | |
3082 | o coding-category-sjis | |
3083 | ||
3084 | The category for a coding system which has the same code range | |
3085 | as SJIS. Assigned the coding-system (Lisp | |
7717c392 | 3086 | symbol) `japanese-shift-jis' by default. |
4ed46869 KH |
3087 | |
3088 | o coding-category-iso-7 | |
3089 | ||
3090 | The category for a coding system which has the same code range | |
7717c392 | 3091 | as ISO2022 of 7-bit environment. This doesn't use any locking |
d46c5b12 KH |
3092 | shift and single shift functions. This can encode/decode all |
3093 | charsets. Assigned the coding-system (Lisp symbol) | |
3094 | `iso-2022-7bit' by default. | |
3095 | ||
3096 | o coding-category-iso-7-tight | |
3097 | ||
3098 | Same as coding-category-iso-7 except that this can | |
3099 | encode/decode only the specified charsets. | |
4ed46869 KH |
3100 | |
3101 | o coding-category-iso-8-1 | |
3102 | ||
3103 | The category for a coding system which has the same code range | |
3104 | as ISO2022 of 8-bit environment and graphic plane 1 used only | |
7717c392 KH |
3105 | for DIMENSION1 charset. This doesn't use any locking shift |
3106 | and single shift functions. Assigned the coding-system (Lisp | |
3107 | symbol) `iso-latin-1' by default. | |
4ed46869 KH |
3108 | |
3109 | o coding-category-iso-8-2 | |
3110 | ||
3111 | The category for a coding system which has the same code range | |
3112 | as ISO2022 of 8-bit environment and graphic plane 1 used only | |
7717c392 KH |
3113 | for DIMENSION2 charset. This doesn't use any locking shift |
3114 | and single shift functions. Assigned the coding-system (Lisp | |
3115 | symbol) `japanese-iso-8bit' by default. | |
4ed46869 | 3116 | |
7717c392 | 3117 | o coding-category-iso-7-else |
4ed46869 KH |
3118 | |
3119 | The category for a coding system which has the same code range | |
7717c392 KH |
3120 | as ISO2022 of 7-bit environemnt but uses locking shift or |
3121 | single shift functions. Assigned the coding-system (Lisp | |
3122 | symbol) `iso-2022-7bit-lock' by default. | |
3123 | ||
3124 | o coding-category-iso-8-else | |
3125 | ||
3126 | The category for a coding system which has the same code range | |
3127 | as ISO2022 of 8-bit environemnt but uses locking shift or | |
3128 | single shift functions. Assigned the coding-system (Lisp | |
3129 | symbol) `iso-2022-8bit-ss2' by default. | |
4ed46869 KH |
3130 | |
3131 | o coding-category-big5 | |
3132 | ||
3133 | The category for a coding system which has the same code range | |
3134 | as BIG5. Assigned the coding-system (Lisp symbol) | |
e0e989f6 | 3135 | `cn-big5' by default. |
4ed46869 KH |
3136 | |
3137 | o coding-category-binary | |
3138 | ||
3139 | The category for a coding system not categorized in any of the | |
3140 | above. Assigned the coding-system (Lisp symbol) | |
e0e989f6 | 3141 | `no-conversion' by default. |
4ed46869 KH |
3142 | |
3143 | Each of them is a Lisp symbol and the value is an actual | |
3144 | `coding-system's (this is also a Lisp symbol) assigned by a user. | |
3145 | What Emacs does actually is to detect a category of coding system. | |
3146 | Then, it uses a `coding-system' assigned to it. If Emacs can't | |
3147 | decide only one possible category, it selects a category of the | |
3148 | highest priority. Priorities of categories are also specified by a | |
3149 | user in a Lisp variable `coding-category-list'. | |
3150 | ||
3151 | */ | |
3152 | ||
d46c5b12 | 3153 | /* Detect how a text of length SRC_BYTES pointed by SOURCE is encoded. |
4ed46869 KH |
3154 | If it detects possible coding systems, return an integer in which |
3155 | appropriate flag bits are set. Flag bits are defined by macros | |
d46c5b12 | 3156 | CODING_CATEGORY_MASK_XXX in `coding.h'. |
4ed46869 | 3157 | |
d46c5b12 KH |
3158 | How many ASCII characters are at the head is returned as *SKIP. */ |
3159 | ||
3160 | static int | |
3161 | detect_coding_mask (source, src_bytes, priorities, skip) | |
3162 | unsigned char *source; | |
3163 | int src_bytes, *priorities, *skip; | |
4ed46869 KH |
3164 | { |
3165 | register unsigned char c; | |
d46c5b12 KH |
3166 | unsigned char *src = source, *src_end = source + src_bytes; |
3167 | unsigned int mask = (CODING_CATEGORY_MASK_ISO_7BIT | |
3168 | | CODING_CATEGORY_MASK_ISO_SHIFT); | |
3169 | int i; | |
4ed46869 KH |
3170 | |
3171 | /* At first, skip all ASCII characters and control characters except | |
3172 | for three ISO2022 specific control characters. */ | |
bcf26d6a | 3173 | label_loop_detect_coding: |
4ed46869 KH |
3174 | while (src < src_end) |
3175 | { | |
3176 | c = *src; | |
3177 | if (c >= 0x80 | |
d46c5b12 KH |
3178 | || ((mask & CODING_CATEGORY_MASK_ISO_7BIT) |
3179 | && c == ISO_CODE_ESC) | |
3180 | || ((mask & CODING_CATEGORY_MASK_ISO_SHIFT) | |
3181 | && (c == ISO_CODE_SI || c == ISO_CODE_SO))) | |
4ed46869 KH |
3182 | break; |
3183 | src++; | |
3184 | } | |
d46c5b12 | 3185 | *skip = src - source; |
4ed46869 KH |
3186 | |
3187 | if (src >= src_end) | |
3188 | /* We found nothing other than ASCII. There's nothing to do. */ | |
d46c5b12 | 3189 | return 0; |
4ed46869 KH |
3190 | |
3191 | /* The text seems to be encoded in some multilingual coding system. | |
3192 | Now, try to find in which coding system the text is encoded. */ | |
3193 | if (c < 0x80) | |
bcf26d6a KH |
3194 | { |
3195 | /* i.e. (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) */ | |
3196 | /* C is an ISO2022 specific control code of C0. */ | |
3197 | mask = detect_coding_iso2022 (src, src_end); | |
1b2af4b0 | 3198 | if (mask == 0) |
d46c5b12 KH |
3199 | { |
3200 | /* No valid ISO2022 code follows C. Try again. */ | |
3201 | src++; | |
3202 | mask = (c != ISO_CODE_ESC | |
3203 | ? CODING_CATEGORY_MASK_ISO_7BIT | |
3204 | : CODING_CATEGORY_MASK_ISO_SHIFT); | |
3205 | goto label_loop_detect_coding; | |
3206 | } | |
3207 | if (priorities) | |
3208 | goto label_return_highest_only; | |
bcf26d6a | 3209 | } |
d46c5b12 | 3210 | else |
c4825358 | 3211 | { |
d46c5b12 | 3212 | int try; |
4ed46869 | 3213 | |
d46c5b12 KH |
3214 | if (c < 0xA0) |
3215 | { | |
3216 | /* C is the first byte of SJIS character code, | |
3217 | or a leading-code of Emacs' internal format (emacs-mule). */ | |
3218 | try = CODING_CATEGORY_MASK_SJIS | CODING_CATEGORY_MASK_EMACS_MULE; | |
3219 | ||
3220 | /* Or, if C is a special latin extra code, | |
3221 | or is an ISO2022 specific control code of C1 (SS2 or SS3), | |
3222 | or is an ISO2022 control-sequence-introducer (CSI), | |
3223 | we should also consider the possibility of ISO2022 codings. */ | |
3224 | if ((VECTORP (Vlatin_extra_code_table) | |
3225 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) | |
3226 | || (c == ISO_CODE_SS2 || c == ISO_CODE_SS3) | |
3227 | || (c == ISO_CODE_CSI | |
3228 | && (src < src_end | |
3229 | && (*src == ']' | |
3230 | || ((*src == '0' || *src == '1' || *src == '2') | |
3231 | && src + 1 < src_end | |
3232 | && src[1] == ']'))))) | |
3233 | try |= (CODING_CATEGORY_MASK_ISO_8_ELSE | |
3234 | | CODING_CATEGORY_MASK_ISO_8BIT); | |
3235 | } | |
c4825358 | 3236 | else |
d46c5b12 KH |
3237 | /* C is a character of ISO2022 in graphic plane right, |
3238 | or a SJIS's 1-byte character code (i.e. JISX0201), | |
3239 | or the first byte of BIG5's 2-byte code. */ | |
3240 | try = (CODING_CATEGORY_MASK_ISO_8_ELSE | |
3241 | | CODING_CATEGORY_MASK_ISO_8BIT | |
3242 | | CODING_CATEGORY_MASK_SJIS | |
3243 | | CODING_CATEGORY_MASK_BIG5); | |
3244 | ||
3245 | mask = 0; | |
3246 | if (priorities) | |
3247 | { | |
3248 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) | |
3249 | { | |
3250 | priorities[i] &= try; | |
3251 | if (priorities[i] & CODING_CATEGORY_MASK_ISO) | |
3252 | mask = detect_coding_iso2022 (src, src_end); | |
3253 | else if (priorities[i] & CODING_CATEGORY_MASK_SJIS) | |
3254 | mask = detect_coding_sjis (src, src_end); | |
3255 | else if (priorities[i] & CODING_CATEGORY_MASK_BIG5) | |
3256 | mask = detect_coding_big5 (src, src_end); | |
3257 | else if (priorities[i] & CODING_CATEGORY_MASK_EMACS_MULE) | |
3258 | mask = detect_coding_emacs_mule (src, src_end); | |
3259 | if (mask) | |
3260 | goto label_return_highest_only; | |
3261 | } | |
3262 | return CODING_CATEGORY_MASK_RAW_TEXT; | |
3263 | } | |
3264 | if (try & CODING_CATEGORY_MASK_ISO) | |
3265 | mask |= detect_coding_iso2022 (src, src_end); | |
3266 | if (try & CODING_CATEGORY_MASK_SJIS) | |
3267 | mask |= detect_coding_sjis (src, src_end); | |
3268 | if (try & CODING_CATEGORY_MASK_BIG5) | |
3269 | mask |= detect_coding_big5 (src, src_end); | |
3270 | if (try & CODING_CATEGORY_MASK_EMACS_MULE) | |
3271 | mask |= detect_coding_emacs_mule (src, src_end); | |
c4825358 | 3272 | } |
d46c5b12 KH |
3273 | return (mask | CODING_CATEGORY_MASK_RAW_TEXT); |
3274 | ||
3275 | label_return_highest_only: | |
3276 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) | |
3277 | { | |
3278 | if (mask & priorities[i]) | |
3279 | return priorities[i]; | |
3280 | } | |
3281 | return CODING_CATEGORY_MASK_RAW_TEXT; | |
4ed46869 KH |
3282 | } |
3283 | ||
3284 | /* Detect how a text of length SRC_BYTES pointed by SRC is encoded. | |
3285 | The information of the detected coding system is set in CODING. */ | |
3286 | ||
3287 | void | |
3288 | detect_coding (coding, src, src_bytes) | |
3289 | struct coding_system *coding; | |
3290 | unsigned char *src; | |
3291 | int src_bytes; | |
3292 | { | |
d46c5b12 KH |
3293 | unsigned int idx; |
3294 | int skip, mask, i; | |
3295 | int priorities[CODING_CATEGORY_IDX_MAX]; | |
27901516 | 3296 | Lisp_Object val = Vcoding_category_list; |
4ed46869 | 3297 | |
d46c5b12 KH |
3298 | i = 0; |
3299 | while (CONSP (val) && i < CODING_CATEGORY_IDX_MAX) | |
3300 | { | |
3301 | if (! SYMBOLP (XCONS (val)->car)) | |
3302 | break; | |
3303 | idx = XFASTINT (Fget (XCONS (val)->car, Qcoding_category_index)); | |
3304 | if (idx >= CODING_CATEGORY_IDX_MAX) | |
3305 | break; | |
3306 | priorities[i++] = (1 << idx); | |
3307 | val = XCONS (val)->cdr; | |
3308 | } | |
3309 | /* If coding-category-list is valid and contains all coding | |
3310 | categories, `i' should be CODING_CATEGORY_IDX_MAX now. If not, | |
3311 | the following code saves Emacs from craching. */ | |
3312 | while (i < CODING_CATEGORY_IDX_MAX) | |
3313 | priorities[i++] = CODING_CATEGORY_MASK_RAW_TEXT; | |
4ed46869 | 3314 | |
d46c5b12 KH |
3315 | mask = detect_coding_mask (src, src_bytes, priorities, &skip); |
3316 | coding->heading_ascii = skip; | |
4ed46869 | 3317 | |
d46c5b12 KH |
3318 | if (!mask) return; |
3319 | ||
3320 | /* We found a single coding system of the highest priority in MASK. */ | |
3321 | idx = 0; | |
3322 | while (mask && ! (mask & 1)) mask >>= 1, idx++; | |
3323 | if (! mask) | |
3324 | idx = CODING_CATEGORY_IDX_RAW_TEXT; | |
4ed46869 | 3325 | |
d46c5b12 KH |
3326 | val = XSYMBOL (XVECTOR (Vcoding_category_table)->contents[idx])->value; |
3327 | ||
3328 | if (coding->eol_type != CODING_EOL_UNDECIDED) | |
27901516 | 3329 | { |
d46c5b12 KH |
3330 | Lisp_Object tmp = Fget (val, Qeol_type); |
3331 | ||
3332 | if (VECTORP (tmp)) | |
3333 | val = XVECTOR (tmp)->contents[coding->eol_type]; | |
4ed46869 | 3334 | } |
d46c5b12 KH |
3335 | setup_coding_system (val, coding); |
3336 | /* Set this again because setup_coding_system reset this member. */ | |
3337 | coding->heading_ascii = skip; | |
4ed46869 KH |
3338 | } |
3339 | ||
d46c5b12 KH |
3340 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by |
3341 | SOURCE is encoded. Return one of CODING_EOL_LF, CODING_EOL_CRLF, | |
3342 | CODING_EOL_CR, and CODING_EOL_UNDECIDED. | |
3343 | ||
3344 | How many non-eol characters are at the head is returned as *SKIP. */ | |
4ed46869 | 3345 | |
bc4bc72a RS |
3346 | #define MAX_EOL_CHECK_COUNT 3 |
3347 | ||
d46c5b12 KH |
3348 | static int |
3349 | detect_eol_type (source, src_bytes, skip) | |
3350 | unsigned char *source; | |
3351 | int src_bytes, *skip; | |
4ed46869 | 3352 | { |
d46c5b12 | 3353 | unsigned char *src = source, *src_end = src + src_bytes; |
4ed46869 | 3354 | unsigned char c; |
bc4bc72a RS |
3355 | int total = 0; /* How many end-of-lines are found so far. */ |
3356 | int eol_type = CODING_EOL_UNDECIDED; | |
3357 | int this_eol_type; | |
4ed46869 | 3358 | |
d46c5b12 KH |
3359 | *skip = 0; |
3360 | ||
bc4bc72a | 3361 | while (src < src_end && total < MAX_EOL_CHECK_COUNT) |
4ed46869 KH |
3362 | { |
3363 | c = *src++; | |
bc4bc72a | 3364 | if (c == '\n' || c == '\r') |
4ed46869 | 3365 | { |
d46c5b12 KH |
3366 | if (*skip == 0) |
3367 | *skip = src - 1 - source; | |
bc4bc72a RS |
3368 | total++; |
3369 | if (c == '\n') | |
3370 | this_eol_type = CODING_EOL_LF; | |
3371 | else if (src >= src_end || *src != '\n') | |
3372 | this_eol_type = CODING_EOL_CR; | |
4ed46869 | 3373 | else |
bc4bc72a RS |
3374 | this_eol_type = CODING_EOL_CRLF, src++; |
3375 | ||
3376 | if (eol_type == CODING_EOL_UNDECIDED) | |
3377 | /* This is the first end-of-line. */ | |
3378 | eol_type = this_eol_type; | |
3379 | else if (eol_type != this_eol_type) | |
d46c5b12 KH |
3380 | { |
3381 | /* The found type is different from what found before. */ | |
3382 | eol_type = CODING_EOL_INCONSISTENT; | |
3383 | break; | |
3384 | } | |
4ed46869 KH |
3385 | } |
3386 | } | |
bc4bc72a | 3387 | |
d46c5b12 KH |
3388 | if (*skip == 0) |
3389 | *skip = src_end - source; | |
85a02ca4 | 3390 | return eol_type; |
4ed46869 KH |
3391 | } |
3392 | ||
3393 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC | |
3394 | is encoded. If it detects an appropriate format of end-of-line, it | |
3395 | sets the information in *CODING. */ | |
3396 | ||
3397 | void | |
3398 | detect_eol (coding, src, src_bytes) | |
3399 | struct coding_system *coding; | |
3400 | unsigned char *src; | |
3401 | int src_bytes; | |
3402 | { | |
4608c386 | 3403 | Lisp_Object val; |
d46c5b12 KH |
3404 | int skip; |
3405 | int eol_type = detect_eol_type (src, src_bytes, &skip); | |
3406 | ||
3407 | if (coding->heading_ascii > skip) | |
3408 | coding->heading_ascii = skip; | |
3409 | else | |
3410 | skip = coding->heading_ascii; | |
4ed46869 | 3411 | |
0ef69138 | 3412 | if (eol_type == CODING_EOL_UNDECIDED) |
4ed46869 | 3413 | return; |
27901516 KH |
3414 | if (eol_type == CODING_EOL_INCONSISTENT) |
3415 | { | |
3416 | #if 0 | |
3417 | /* This code is suppressed until we find a better way to | |
992f23f2 | 3418 | distinguish raw text file and binary file. */ |
27901516 KH |
3419 | |
3420 | /* If we have already detected that the coding is raw-text, the | |
3421 | coding should actually be no-conversion. */ | |
3422 | if (coding->type == coding_type_raw_text) | |
3423 | { | |
3424 | setup_coding_system (Qno_conversion, coding); | |
3425 | return; | |
3426 | } | |
3427 | /* Else, let's decode only text code anyway. */ | |
3428 | #endif /* 0 */ | |
1b2af4b0 | 3429 | eol_type = CODING_EOL_LF; |
27901516 KH |
3430 | } |
3431 | ||
4608c386 | 3432 | val = Fget (coding->symbol, Qeol_type); |
4ed46869 | 3433 | if (VECTORP (val) && XVECTOR (val)->size == 3) |
d46c5b12 KH |
3434 | { |
3435 | setup_coding_system (XVECTOR (val)->contents[eol_type], coding); | |
3436 | coding->heading_ascii = skip; | |
3437 | } | |
3438 | } | |
3439 | ||
3440 | #define CONVERSION_BUFFER_EXTRA_ROOM 256 | |
3441 | ||
3442 | #define DECODING_BUFFER_MAG(coding) \ | |
3443 | (coding->type == coding_type_iso2022 \ | |
3444 | ? 3 \ | |
3445 | : ((coding->type == coding_type_sjis || coding->type == coding_type_big5) \ | |
3446 | ? 2 \ | |
3447 | : (coding->type == coding_type_raw_text \ | |
3448 | ? 1 \ | |
3449 | : (coding->type == coding_type_ccl \ | |
3450 | ? coding->spec.ccl.decoder.buf_magnification \ | |
3451 | : 2)))) | |
3452 | ||
3453 | /* Return maximum size (bytes) of a buffer enough for decoding | |
3454 | SRC_BYTES of text encoded in CODING. */ | |
3455 | ||
3456 | int | |
3457 | decoding_buffer_size (coding, src_bytes) | |
3458 | struct coding_system *coding; | |
3459 | int src_bytes; | |
3460 | { | |
3461 | return (src_bytes * DECODING_BUFFER_MAG (coding) | |
3462 | + CONVERSION_BUFFER_EXTRA_ROOM); | |
3463 | } | |
3464 | ||
3465 | /* Return maximum size (bytes) of a buffer enough for encoding | |
3466 | SRC_BYTES of text to CODING. */ | |
3467 | ||
3468 | int | |
3469 | encoding_buffer_size (coding, src_bytes) | |
3470 | struct coding_system *coding; | |
3471 | int src_bytes; | |
3472 | { | |
3473 | int magnification; | |
3474 | ||
3475 | if (coding->type == coding_type_ccl) | |
3476 | magnification = coding->spec.ccl.encoder.buf_magnification; | |
3477 | else | |
3478 | magnification = 3; | |
3479 | ||
3480 | return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM); | |
3481 | } | |
3482 | ||
3483 | #ifndef MINIMUM_CONVERSION_BUFFER_SIZE | |
3484 | #define MINIMUM_CONVERSION_BUFFER_SIZE 1024 | |
3485 | #endif | |
3486 | ||
3487 | char *conversion_buffer; | |
3488 | int conversion_buffer_size; | |
3489 | ||
3490 | /* Return a pointer to a SIZE bytes of buffer to be used for encoding | |
3491 | or decoding. Sufficient memory is allocated automatically. If we | |
3492 | run out of memory, return NULL. */ | |
3493 | ||
3494 | char * | |
3495 | get_conversion_buffer (size) | |
3496 | int size; | |
3497 | { | |
3498 | if (size > conversion_buffer_size) | |
3499 | { | |
3500 | char *buf; | |
3501 | int real_size = conversion_buffer_size * 2; | |
3502 | ||
3503 | while (real_size < size) real_size *= 2; | |
3504 | buf = (char *) xmalloc (real_size); | |
3505 | xfree (conversion_buffer); | |
3506 | conversion_buffer = buf; | |
3507 | conversion_buffer_size = real_size; | |
3508 | } | |
3509 | return conversion_buffer; | |
3510 | } | |
3511 | ||
3512 | int | |
3513 | ccl_coding_driver (coding, source, destination, src_bytes, dst_bytes, encodep) | |
3514 | struct coding_system *coding; | |
3515 | unsigned char *source, *destination; | |
3516 | int src_bytes, dst_bytes, encodep; | |
3517 | { | |
3518 | struct ccl_program *ccl | |
3519 | = encodep ? &coding->spec.ccl.encoder : &coding->spec.ccl.decoder; | |
3520 | int result; | |
3521 | ||
3522 | coding->produced = ccl_driver (ccl, source, destination, | |
3523 | src_bytes, dst_bytes, &(coding->consumed)); | |
3524 | if (encodep) | |
3525 | { | |
3526 | coding->produced_char = coding->produced; | |
3527 | coding->consumed_char | |
3528 | = multibyte_chars_in_text (source, coding->consumed); | |
3529 | } | |
3530 | else | |
3531 | { | |
3532 | coding->produced_char | |
3533 | = multibyte_chars_in_text (destination, coding->produced); | |
3534 | coding->consumed_char = coding->consumed; | |
3535 | } | |
3536 | switch (ccl->status) | |
3537 | { | |
3538 | case CCL_STAT_SUSPEND_BY_SRC: | |
3539 | result = CODING_FINISH_INSUFFICIENT_SRC; | |
3540 | break; | |
3541 | case CCL_STAT_SUSPEND_BY_DST: | |
3542 | result = CODING_FINISH_INSUFFICIENT_DST; | |
3543 | break; | |
3544 | default: | |
3545 | result = CODING_FINISH_NORMAL; | |
3546 | break; | |
3547 | } | |
3548 | return result; | |
4ed46869 KH |
3549 | } |
3550 | ||
3551 | /* See "GENERAL NOTES about `decode_coding_XXX ()' functions". Before | |
3552 | decoding, it may detect coding system and format of end-of-line if | |
3553 | those are not yet decided. */ | |
3554 | ||
3555 | int | |
d46c5b12 | 3556 | decode_coding (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
3557 | struct coding_system *coding; |
3558 | unsigned char *source, *destination; | |
3559 | int src_bytes, dst_bytes; | |
4ed46869 | 3560 | { |
d46c5b12 | 3561 | int result; |
4ed46869 KH |
3562 | |
3563 | if (src_bytes <= 0) | |
3564 | { | |
d46c5b12 KH |
3565 | coding->produced = coding->produced_char = 0; |
3566 | coding->consumed = coding->consumed_char = 0; | |
fb88bf2d | 3567 | coding->fake_multibyte = 0; |
d46c5b12 | 3568 | return CODING_FINISH_NORMAL; |
4ed46869 KH |
3569 | } |
3570 | ||
0ef69138 | 3571 | if (coding->type == coding_type_undecided) |
4ed46869 KH |
3572 | detect_coding (coding, source, src_bytes); |
3573 | ||
0ef69138 | 3574 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
3575 | detect_eol (coding, source, src_bytes); |
3576 | ||
4ed46869 KH |
3577 | switch (coding->type) |
3578 | { | |
0ef69138 KH |
3579 | case coding_type_emacs_mule: |
3580 | case coding_type_undecided: | |
27901516 | 3581 | case coding_type_raw_text: |
4ed46869 | 3582 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 3583 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 | 3584 | goto label_no_conversion; |
d46c5b12 | 3585 | result = decode_eol (coding, source, destination, src_bytes, dst_bytes); |
4ed46869 KH |
3586 | break; |
3587 | ||
3588 | case coding_type_sjis: | |
d46c5b12 KH |
3589 | result = decode_coding_sjis_big5 (coding, source, destination, |
3590 | src_bytes, dst_bytes, 1); | |
4ed46869 KH |
3591 | break; |
3592 | ||
3593 | case coding_type_iso2022: | |
d46c5b12 KH |
3594 | result = decode_coding_iso2022 (coding, source, destination, |
3595 | src_bytes, dst_bytes); | |
4ed46869 KH |
3596 | break; |
3597 | ||
3598 | case coding_type_big5: | |
d46c5b12 KH |
3599 | result = decode_coding_sjis_big5 (coding, source, destination, |
3600 | src_bytes, dst_bytes, 0); | |
4ed46869 KH |
3601 | break; |
3602 | ||
3603 | case coding_type_ccl: | |
d46c5b12 KH |
3604 | result = ccl_coding_driver (coding, source, destination, |
3605 | src_bytes, dst_bytes, 0); | |
3606 | break; | |
3607 | ||
3608 | default: /* i.e. case coding_type_no_conversion: */ | |
3609 | label_no_conversion: | |
3610 | if (dst_bytes && src_bytes > dst_bytes) | |
3611 | { | |
3612 | coding->produced = dst_bytes; | |
3613 | result = CODING_FINISH_INSUFFICIENT_DST; | |
3614 | } | |
3615 | else | |
3616 | { | |
3617 | coding->produced = src_bytes; | |
3618 | result = CODING_FINISH_NORMAL; | |
3619 | } | |
3620 | if (dst_bytes) | |
3621 | bcopy (source, destination, coding->produced); | |
3622 | else | |
3623 | safe_bcopy (source, destination, coding->produced); | |
fb88bf2d | 3624 | coding->fake_multibyte = 1; |
d46c5b12 KH |
3625 | coding->consumed |
3626 | = coding->consumed_char = coding->produced_char = coding->produced; | |
4ed46869 KH |
3627 | break; |
3628 | } | |
3629 | ||
d46c5b12 | 3630 | return result; |
4ed46869 KH |
3631 | } |
3632 | ||
3633 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". */ | |
3634 | ||
3635 | int | |
d46c5b12 | 3636 | encode_coding (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
3637 | struct coding_system *coding; |
3638 | unsigned char *source, *destination; | |
3639 | int src_bytes, dst_bytes; | |
4ed46869 | 3640 | { |
d46c5b12 | 3641 | int result; |
4ed46869 | 3642 | |
d46c5b12 | 3643 | if (src_bytes <= 0) |
4ed46869 | 3644 | { |
d46c5b12 KH |
3645 | coding->produced = coding->produced_char = 0; |
3646 | coding->consumed = coding->consumed_char = 0; | |
fb88bf2d | 3647 | coding->fake_multibyte = 0; |
d46c5b12 KH |
3648 | return CODING_FINISH_NORMAL; |
3649 | } | |
4ed46869 | 3650 | |
d46c5b12 KH |
3651 | switch (coding->type) |
3652 | { | |
0ef69138 KH |
3653 | case coding_type_emacs_mule: |
3654 | case coding_type_undecided: | |
27901516 | 3655 | case coding_type_raw_text: |
4ed46869 | 3656 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 3657 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 | 3658 | goto label_no_conversion; |
d46c5b12 | 3659 | result = encode_eol (coding, source, destination, src_bytes, dst_bytes); |
4ed46869 KH |
3660 | break; |
3661 | ||
3662 | case coding_type_sjis: | |
d46c5b12 KH |
3663 | result = encode_coding_sjis_big5 (coding, source, destination, |
3664 | src_bytes, dst_bytes, 1); | |
4ed46869 KH |
3665 | break; |
3666 | ||
3667 | case coding_type_iso2022: | |
d46c5b12 KH |
3668 | result = encode_coding_iso2022 (coding, source, destination, |
3669 | src_bytes, dst_bytes); | |
4ed46869 KH |
3670 | break; |
3671 | ||
3672 | case coding_type_big5: | |
d46c5b12 KH |
3673 | result = encode_coding_sjis_big5 (coding, source, destination, |
3674 | src_bytes, dst_bytes, 0); | |
4ed46869 KH |
3675 | break; |
3676 | ||
3677 | case coding_type_ccl: | |
d46c5b12 KH |
3678 | result = ccl_coding_driver (coding, source, destination, |
3679 | src_bytes, dst_bytes, 1); | |
3680 | break; | |
3681 | ||
3682 | default: /* i.e. case coding_type_no_conversion: */ | |
3683 | label_no_conversion: | |
3684 | if (dst_bytes && src_bytes > dst_bytes) | |
3685 | { | |
3686 | coding->produced = dst_bytes; | |
3687 | result = CODING_FINISH_INSUFFICIENT_DST; | |
3688 | } | |
3689 | else | |
3690 | { | |
3691 | coding->produced = src_bytes; | |
3692 | result = CODING_FINISH_NORMAL; | |
3693 | } | |
3694 | if (dst_bytes) | |
3695 | bcopy (source, destination, coding->produced); | |
3696 | else | |
3697 | safe_bcopy (source, destination, coding->produced); | |
3698 | if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY) | |
3699 | { | |
3700 | unsigned char *p = destination, *pend = p + coding->produced; | |
3701 | while (p < pend) | |
3702 | if (*p++ == '\015') p[-1] = '\n'; | |
3703 | } | |
fb88bf2d | 3704 | coding->fake_multibyte = 1; |
d46c5b12 KH |
3705 | coding->consumed |
3706 | = coding->consumed_char = coding->produced_char = coding->produced; | |
4ed46869 KH |
3707 | break; |
3708 | } | |
3709 | ||
d46c5b12 | 3710 | return result; |
4ed46869 KH |
3711 | } |
3712 | ||
fb88bf2d KH |
3713 | /* Scan text in the region between *BEG and *END (byte positions), |
3714 | skip characters which we don't have to decode by coding system | |
3715 | CODING at the head and tail, then set *BEG and *END to the region | |
3716 | of the text we actually have to convert. The caller should move | |
3717 | the gap out of the region in advance. | |
4ed46869 | 3718 | |
d46c5b12 KH |
3719 | If STR is not NULL, *BEG and *END are indices into STR. */ |
3720 | ||
3721 | static void | |
3722 | shrink_decoding_region (beg, end, coding, str) | |
3723 | int *beg, *end; | |
3724 | struct coding_system *coding; | |
3725 | unsigned char *str; | |
3726 | { | |
fb88bf2d | 3727 | unsigned char *begp_orig, *begp, *endp_orig, *endp, c; |
d46c5b12 KH |
3728 | int eol_conversion; |
3729 | ||
3730 | if (coding->type == coding_type_ccl | |
3731 | || coding->type == coding_type_undecided | |
3732 | || !NILP (coding->post_read_conversion)) | |
3733 | { | |
3734 | /* We can't skip any data. */ | |
3735 | return; | |
3736 | } | |
3737 | else if (coding->type == coding_type_no_conversion) | |
3738 | { | |
fb88bf2d KH |
3739 | /* We need no conversion, but don't have to skip any data here. |
3740 | Decoding routine handles them effectively anyway. */ | |
d46c5b12 KH |
3741 | return; |
3742 | } | |
3743 | ||
3744 | if (coding->heading_ascii >= 0) | |
3745 | /* Detection routine has already found how much we can skip at the | |
3746 | head. */ | |
3747 | *beg += coding->heading_ascii; | |
3748 | ||
3749 | if (str) | |
3750 | { | |
3751 | begp_orig = begp = str + *beg; | |
3752 | endp_orig = endp = str + *end; | |
3753 | } | |
3754 | else | |
3755 | { | |
fb88bf2d | 3756 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
d46c5b12 KH |
3757 | endp_orig = endp = begp + *end - *beg; |
3758 | } | |
3759 | ||
3760 | eol_conversion = (coding->eol_type != CODING_EOL_LF); | |
3761 | ||
3762 | switch (coding->type) | |
3763 | { | |
3764 | case coding_type_emacs_mule: | |
3765 | case coding_type_raw_text: | |
3766 | if (eol_conversion) | |
3767 | { | |
3768 | if (coding->heading_ascii < 0) | |
fb88bf2d KH |
3769 | while (begp < endp && *begp != '\r' && *begp < 0x80) begp++; |
3770 | while (begp < endp && *(endp - 1) != '\r' && *(endp - 1) < 0x80) | |
3771 | endp--; | |
d46c5b12 KH |
3772 | } |
3773 | else | |
3774 | begp = endp; | |
3775 | break; | |
3776 | ||
3777 | case coding_type_sjis: | |
3778 | case coding_type_big5: | |
3779 | /* We can skip all ASCII characters at the head. */ | |
3780 | if (coding->heading_ascii < 0) | |
3781 | { | |
3782 | if (eol_conversion) | |
3783 | while (begp < endp && *begp < 0x80 && *begp != '\n') begp++; | |
3784 | else | |
3785 | while (begp < endp && *begp < 0x80) begp++; | |
3786 | } | |
3787 | /* We can skip all ASCII characters at the tail except for the | |
3788 | second byte of SJIS or BIG5 code. */ | |
3789 | if (eol_conversion) | |
3790 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\n') endp--; | |
3791 | else | |
3792 | while (begp < endp && endp[-1] < 0x80) endp--; | |
3793 | if (begp < endp && endp < endp_orig && endp[-1] >= 0x80) | |
3794 | endp++; | |
3795 | break; | |
3796 | ||
3797 | default: /* i.e. case coding_type_iso2022: */ | |
3798 | if (coding->heading_ascii < 0) | |
3799 | { | |
d46c5b12 KH |
3800 | /* We can skip all ASCII characters at the head except for a |
3801 | few control codes. */ | |
3802 | while (begp < endp && (c = *begp) < 0x80 | |
3803 | && c != ISO_CODE_CR && c != ISO_CODE_SO | |
3804 | && c != ISO_CODE_SI && c != ISO_CODE_ESC | |
3805 | && (!eol_conversion || c != ISO_CODE_LF)) | |
3806 | begp++; | |
3807 | } | |
3808 | switch (coding->category_idx) | |
3809 | { | |
3810 | case CODING_CATEGORY_IDX_ISO_8_1: | |
3811 | case CODING_CATEGORY_IDX_ISO_8_2: | |
3812 | /* We can skip all ASCII characters at the tail. */ | |
3813 | if (eol_conversion) | |
fb88bf2d | 3814 | while (begp < endp && (c = endp[-1]) < 0x80 && c != '\n') endp--; |
d46c5b12 KH |
3815 | else |
3816 | while (begp < endp && endp[-1] < 0x80) endp--; | |
3817 | break; | |
3818 | ||
3819 | case CODING_CATEGORY_IDX_ISO_7: | |
3820 | case CODING_CATEGORY_IDX_ISO_7_TIGHT: | |
3821 | /* We can skip all charactes at the tail except for ESC and | |
3822 | the following 2-byte at the tail. */ | |
3823 | if (eol_conversion) | |
fb88bf2d KH |
3824 | while (begp < endp |
3825 | && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC && c != '\n') | |
d46c5b12 KH |
3826 | endp--; |
3827 | else | |
fb88bf2d KH |
3828 | while (begp < endp |
3829 | && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC) | |
d46c5b12 KH |
3830 | endp--; |
3831 | if (begp < endp && endp[-1] == ISO_CODE_ESC) | |
3832 | { | |
3833 | if (endp + 1 < endp_orig && end[0] == '(' && end[1] == 'B') | |
3834 | /* This is an ASCII designation sequence. We can | |
3835 | surely skip the tail. */ | |
3836 | endp += 2; | |
3837 | else | |
3838 | /* Hmmm, we can't skip the tail. */ | |
3839 | endp = endp_orig; | |
3840 | } | |
3841 | } | |
3842 | } | |
3843 | *beg += begp - begp_orig; | |
3844 | *end += endp - endp_orig; | |
3845 | return; | |
3846 | } | |
3847 | ||
3848 | /* Like shrink_decoding_region but for encoding. */ | |
3849 | ||
3850 | static void | |
3851 | shrink_encoding_region (beg, end, coding, str) | |
3852 | int *beg, *end; | |
3853 | struct coding_system *coding; | |
3854 | unsigned char *str; | |
3855 | { | |
3856 | unsigned char *begp_orig, *begp, *endp_orig, *endp; | |
3857 | int eol_conversion; | |
3858 | ||
3859 | if (coding->type == coding_type_ccl) | |
3860 | /* We can't skip any data. */ | |
3861 | return; | |
3862 | else if (coding->type == coding_type_no_conversion) | |
3863 | { | |
3864 | /* We need no conversion. */ | |
3865 | *beg = *end; | |
3866 | return; | |
3867 | } | |
3868 | ||
3869 | if (str) | |
3870 | { | |
3871 | begp_orig = begp = str + *beg; | |
3872 | endp_orig = endp = str + *end; | |
3873 | } | |
3874 | else | |
3875 | { | |
fb88bf2d | 3876 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
d46c5b12 KH |
3877 | endp_orig = endp = begp + *end - *beg; |
3878 | } | |
3879 | ||
3880 | eol_conversion = (coding->eol_type == CODING_EOL_CR | |
3881 | || coding->eol_type == CODING_EOL_CRLF); | |
3882 | ||
3883 | /* Here, we don't have to check coding->pre_write_conversion because | |
3884 | the caller is expected to have handled it already. */ | |
3885 | switch (coding->type) | |
3886 | { | |
3887 | case coding_type_undecided: | |
3888 | case coding_type_emacs_mule: | |
3889 | case coding_type_raw_text: | |
3890 | if (eol_conversion) | |
3891 | { | |
3892 | while (begp < endp && *begp != '\n') begp++; | |
3893 | while (begp < endp && endp[-1] != '\n') endp--; | |
3894 | } | |
3895 | else | |
3896 | begp = endp; | |
3897 | break; | |
3898 | ||
3899 | case coding_type_iso2022: | |
3900 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL) | |
3901 | { | |
3902 | unsigned char *bol = begp; | |
3903 | while (begp < endp && *begp < 0x80) | |
3904 | { | |
3905 | begp++; | |
3906 | if (begp[-1] == '\n') | |
3907 | bol = begp; | |
3908 | } | |
3909 | begp = bol; | |
3910 | goto label_skip_tail; | |
3911 | } | |
3912 | /* fall down ... */ | |
3913 | ||
3914 | default: | |
3915 | /* We can skip all ASCII characters at the head and tail. */ | |
3916 | if (eol_conversion) | |
3917 | while (begp < endp && *begp < 0x80 && *begp != '\n') begp++; | |
3918 | else | |
3919 | while (begp < endp && *begp < 0x80) begp++; | |
3920 | label_skip_tail: | |
3921 | if (eol_conversion) | |
3922 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\n') endp--; | |
3923 | else | |
3924 | while (begp < endp && *(endp - 1) < 0x80) endp--; | |
3925 | break; | |
3926 | } | |
3927 | ||
3928 | *beg += begp - begp_orig; | |
3929 | *end += endp - endp_orig; | |
3930 | return; | |
3931 | } | |
3932 | ||
3933 | /* Decode (if ENCODEP is zero) or encode (if ENCODEP is nonzero) the | |
fb88bf2d KH |
3934 | text from FROM to TO (byte positions are FROM_BYTE and TO_BYTE) by |
3935 | coding system CODING, and return the status code of code conversion | |
3936 | (currently, this value has no meaning). | |
3937 | ||
3938 | How many characters (and bytes) are converted to how many | |
3939 | characters (and bytes) are recorded in members of the structure | |
3940 | CODING. | |
d46c5b12 | 3941 | |
6e44253b | 3942 | If REPLACE is nonzero, we do various things as if the original text |
d46c5b12 | 3943 | is deleted and a new text is inserted. See the comments in |
6e44253b | 3944 | replace_range (insdel.c) to know what we are doing. */ |
4ed46869 KH |
3945 | |
3946 | int | |
6e44253b KH |
3947 | code_convert_region (from, from_byte, to, to_byte, coding, encodep, replace) |
3948 | int from, from_byte, to, to_byte, encodep, replace; | |
4ed46869 | 3949 | struct coding_system *coding; |
4ed46869 | 3950 | { |
fb88bf2d KH |
3951 | int len = to - from, len_byte = to_byte - from_byte; |
3952 | int require, inserted, inserted_byte; | |
d46c5b12 KH |
3953 | int from_byte_orig, to_byte_orig; |
3954 | Lisp_Object saved_coding_symbol = Qnil; | |
fb88bf2d KH |
3955 | int multibyte = !NILP (current_buffer->enable_multibyte_characters); |
3956 | int first = 1; | |
3957 | int fake_multibyte = 0; | |
3958 | unsigned char *src, *dst; | |
d46c5b12 | 3959 | |
6e44253b | 3960 | if (replace) |
d46c5b12 | 3961 | { |
fb88bf2d KH |
3962 | int saved_from = from; |
3963 | ||
d46c5b12 | 3964 | prepare_to_modify_buffer (from, to, &from); |
fb88bf2d KH |
3965 | if (saved_from != from) |
3966 | { | |
3967 | to = from + len; | |
3968 | if (multibyte) | |
3969 | from_byte = CHAR_TO_BYTE (from), to_byte = CHAR_TO_BYTE (to); | |
3970 | else | |
3971 | from_byte = from, to_byte = to; | |
3972 | len_byte = to_byte - from_byte; | |
3973 | } | |
d46c5b12 | 3974 | } |
d46c5b12 KH |
3975 | |
3976 | if (! encodep && CODING_REQUIRE_DETECTION (coding)) | |
3977 | { | |
3978 | /* We must detect encoding of text and eol. Even if detection | |
3979 | routines can't decide the encoding, we should not let them | |
3980 | undecided because the deeper decoding routine (decode_coding) | |
3981 | tries to detect the encodings in vain in that case. */ | |
3982 | ||
3983 | if (from < GPT && to > GPT) | |
3984 | move_gap_both (from, from_byte); | |
3985 | if (coding->type == coding_type_undecided) | |
3986 | { | |
fb88bf2d | 3987 | detect_coding (coding, BYTE_POS_ADDR (from_byte), len_byte); |
d46c5b12 KH |
3988 | if (coding->type == coding_type_undecided) |
3989 | coding->type = coding_type_emacs_mule; | |
3990 | } | |
3991 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
3992 | { | |
3993 | saved_coding_symbol = coding->symbol; | |
3994 | detect_eol (coding, BYTE_POS_ADDR (from_byte), len_byte); | |
3995 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
3996 | coding->eol_type = CODING_EOL_LF; | |
3997 | /* We had better recover the original eol format if we | |
3998 | encounter an inconsitent eol format while decoding. */ | |
3999 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; | |
4000 | } | |
4001 | } | |
4002 | ||
fb88bf2d KH |
4003 | coding->consumed_char = len, coding->consumed = len_byte; |
4004 | ||
d46c5b12 KH |
4005 | if (encodep |
4006 | ? ! CODING_REQUIRE_ENCODING (coding) | |
4007 | : ! CODING_REQUIRE_DECODING (coding)) | |
fb88bf2d KH |
4008 | { |
4009 | coding->produced = len_byte; | |
4010 | if (multibyte) | |
4011 | { | |
6e44253b KH |
4012 | /* We still may have to combine byte at the head and the |
4013 | tail of the text in the region. */ | |
4014 | if (GPT != to) | |
4015 | move_gap_both (to, to_byte); | |
fb88bf2d KH |
4016 | coding->produced_char |
4017 | = multibyte_chars_in_text (BYTE_POS_ADDR (from_byte), len_byte); | |
6e44253b KH |
4018 | GAP_SIZE += len_byte; |
4019 | GPT_BYTE -= len_byte; | |
4020 | ZV_BYTE -= len_byte; | |
4021 | Z_BYTE -= len_byte; | |
4022 | GPT -= len_byte; | |
4023 | ZV -= len_byte; | |
4024 | Z -= len_byte; | |
4025 | adjust_after_replace (from, from_byte, to, to_byte, | |
4026 | coding->produced_char, len_byte, replace); | |
fb88bf2d KH |
4027 | } |
4028 | else | |
4029 | coding->produced_char = len_byte; | |
4030 | return 0; | |
4031 | } | |
d46c5b12 KH |
4032 | |
4033 | /* Now we convert the text. */ | |
4034 | ||
4035 | /* For encoding, we must process pre-write-conversion in advance. */ | |
4036 | if (encodep | |
d46c5b12 KH |
4037 | && ! NILP (coding->pre_write_conversion) |
4038 | && SYMBOLP (coding->pre_write_conversion) | |
4039 | && ! NILP (Ffboundp (coding->pre_write_conversion))) | |
4040 | { | |
2b4f9037 KH |
4041 | /* The function in pre-write-conversion may put a new text in a |
4042 | new buffer. */ | |
d46c5b12 KH |
4043 | struct buffer *prev = current_buffer, *new; |
4044 | ||
4045 | call2 (coding->pre_write_conversion, from, to); | |
4046 | if (current_buffer != prev) | |
4047 | { | |
4048 | len = ZV - BEGV; | |
4049 | new = current_buffer; | |
4050 | set_buffer_internal_1 (prev); | |
ddbc19ff | 4051 | del_range_2 (from, from_byte, to, to_byte); |
d46c5b12 KH |
4052 | insert_from_buffer (new, BEG, len, 0); |
4053 | to = from + len; | |
fb88bf2d | 4054 | to_byte = multibyte ? CHAR_TO_BYTE (to) : to; |
d46c5b12 KH |
4055 | len_byte = to_byte - from_byte; |
4056 | } | |
4057 | } | |
4058 | ||
4059 | /* Try to skip the heading and tailing ASCIIs. */ | |
4060 | from_byte_orig = from_byte; to_byte_orig = to_byte; | |
fb88bf2d KH |
4061 | if (from < GPT && GPT < to) |
4062 | move_gap (from); | |
d46c5b12 KH |
4063 | if (encodep) |
4064 | shrink_encoding_region (&from_byte, &to_byte, coding, NULL); | |
4065 | else | |
4066 | shrink_decoding_region (&from_byte, &to_byte, coding, NULL); | |
4067 | if (from_byte == to_byte) | |
fb88bf2d KH |
4068 | { |
4069 | coding->produced = len_byte; | |
4070 | coding->produced_char = multibyte ? len : len_byte; | |
4071 | return 0; | |
4072 | } | |
4073 | ||
d46c5b12 KH |
4074 | /* Here, the excluded region by shrinking contains only ASCIIs. */ |
4075 | from += (from_byte - from_byte_orig); | |
4076 | to += (to_byte - to_byte_orig); | |
4077 | len = to - from; | |
4078 | len_byte = to_byte - from_byte; | |
4079 | ||
fb88bf2d KH |
4080 | /* For converion, we must put the gap before the text in addition to |
4081 | making the gap larger for efficient decoding. The required gap | |
4082 | size starts from 2000 which is the magic number used in make_gap. | |
4083 | But, after one batch of conversion, it will be incremented if we | |
4084 | find that it is not enough . */ | |
d46c5b12 KH |
4085 | require = 2000; |
4086 | ||
4087 | if (GAP_SIZE < require) | |
4088 | make_gap (require - GAP_SIZE); | |
4089 | move_gap_both (from, from_byte); | |
4090 | ||
6e44253b | 4091 | if (replace) |
d46c5b12 KH |
4092 | adjust_before_replace (from, from_byte, to, to_byte); |
4093 | ||
4094 | if (GPT - BEG < beg_unchanged) | |
4095 | beg_unchanged = GPT - BEG; | |
4096 | if (Z - GPT < end_unchanged) | |
4097 | end_unchanged = Z - GPT; | |
4098 | ||
4099 | inserted = inserted_byte = 0; | |
fb88bf2d KH |
4100 | src = GAP_END_ADDR, dst = GPT_ADDR; |
4101 | ||
4102 | GAP_SIZE += len_byte; | |
4103 | ZV -= len; | |
4104 | Z -= len; | |
4105 | ZV_BYTE -= len_byte; | |
4106 | Z_BYTE -= len_byte; | |
4107 | ||
d46c5b12 KH |
4108 | for (;;) |
4109 | { | |
fb88bf2d | 4110 | int result; |
d46c5b12 KH |
4111 | |
4112 | /* The buffer memory is changed from: | |
fb88bf2d KH |
4113 | +--------+converted-text+---------+-------original-text------+---+ |
4114 | |<-from->|<--inserted-->|---------|<-----------len---------->|---| | |
4115 | |<------------------- GAP_SIZE -------------------->| */ | |
d46c5b12 | 4116 | if (encodep) |
fb88bf2d | 4117 | result = encode_coding (coding, src, dst, len_byte, 0); |
d46c5b12 | 4118 | else |
fb88bf2d | 4119 | result = decode_coding (coding, src, dst, len_byte, 0); |
d46c5b12 KH |
4120 | /* to: |
4121 | +--------+-------converted-text--------+--+---original-text--+---+ | |
fb88bf2d KH |
4122 | |<-from->|<--inserted-->|<--produced-->|--|<-(len-consumed)->|---| |
4123 | |<------------------- GAP_SIZE -------------------->| */ | |
4124 | if (coding->fake_multibyte) | |
4125 | fake_multibyte = 1; | |
d46c5b12 | 4126 | |
fb88bf2d KH |
4127 | if (!encodep && !multibyte) |
4128 | coding->produced_char = coding->produced; | |
d46c5b12 KH |
4129 | inserted += coding->produced_char; |
4130 | inserted_byte += coding->produced; | |
d46c5b12 | 4131 | len_byte -= coding->consumed; |
fb88bf2d KH |
4132 | src += coding->consumed; |
4133 | dst += inserted_byte; | |
d46c5b12 KH |
4134 | |
4135 | if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL) | |
4136 | { | |
fb88bf2d | 4137 | unsigned char *pend = dst, *p = pend - inserted_byte; |
d46c5b12 KH |
4138 | |
4139 | /* Encode LFs back to the original eol format (CR or CRLF). */ | |
4140 | if (coding->eol_type == CODING_EOL_CR) | |
4141 | { | |
4142 | while (p < pend) if (*p++ == '\n') p[-1] = '\r'; | |
4143 | } | |
4144 | else | |
4145 | { | |
d46c5b12 KH |
4146 | int count = 0; |
4147 | ||
fb88bf2d KH |
4148 | while (p < pend) if (*p++ == '\n') count++; |
4149 | if (src - dst < count) | |
d46c5b12 | 4150 | { |
fb88bf2d KH |
4151 | /* We don't have sufficient room for putting LFs |
4152 | back to CRLF. We must record converted and | |
4153 | not-yet-converted text back to the buffer | |
4154 | content, enlarge the gap, then record them out of | |
4155 | the buffer contents again. */ | |
4156 | int add = len_byte + inserted_byte; | |
4157 | ||
4158 | GAP_SIZE -= add; | |
4159 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; | |
4160 | GPT += inserted_byte; GPT_BYTE += inserted_byte; | |
4161 | make_gap (count - GAP_SIZE); | |
4162 | GAP_SIZE += add; | |
4163 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; | |
4164 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; | |
4165 | /* Don't forget to update SRC, DST, and PEND. */ | |
4166 | src = GAP_END_ADDR - len_byte; | |
4167 | dst = GPT_ADDR + inserted_byte; | |
4168 | pend = dst; | |
d46c5b12 | 4169 | } |
d46c5b12 KH |
4170 | inserted += count; |
4171 | inserted_byte += count; | |
fb88bf2d KH |
4172 | coding->produced += count; |
4173 | p = dst = pend + count; | |
4174 | while (count) | |
4175 | { | |
4176 | *--p = *--pend; | |
4177 | if (*p == '\n') count--, *--p = '\r'; | |
4178 | } | |
d46c5b12 KH |
4179 | } |
4180 | ||
4181 | /* Suppress eol-format conversion in the further conversion. */ | |
4182 | coding->eol_type = CODING_EOL_LF; | |
4183 | ||
4184 | /* Restore the original symbol. */ | |
4185 | coding->symbol = saved_coding_symbol; | |
fb88bf2d KH |
4186 | |
4187 | continue; | |
d46c5b12 KH |
4188 | } |
4189 | if (len_byte <= 0) | |
4190 | break; | |
4191 | if (result == CODING_FINISH_INSUFFICIENT_SRC) | |
4192 | { | |
4193 | /* The source text ends in invalid codes. Let's just | |
4194 | make them valid buffer contents, and finish conversion. */ | |
fb88bf2d | 4195 | inserted += len_byte; |
d46c5b12 | 4196 | inserted_byte += len_byte; |
fb88bf2d KH |
4197 | while (len_byte--) |
4198 | *src++ = *dst++; | |
4199 | fake_multibyte = 1; | |
d46c5b12 KH |
4200 | break; |
4201 | } | |
fb88bf2d KH |
4202 | if (first) |
4203 | { | |
4204 | /* We have just done the first batch of conversion which was | |
4205 | stoped because of insufficient gap. Let's reconsider the | |
4206 | required gap size (i.e. SRT - DST) now. | |
4207 | ||
4208 | We have converted ORIG bytes (== coding->consumed) into | |
4209 | NEW bytes (coding->produced). To convert the remaining | |
4210 | LEN bytes, we may need REQUIRE bytes of gap, where: | |
4211 | REQUIRE + LEN_BYTE = LEN_BYTE * (NEW / ORIG) | |
4212 | REQUIRE = LEN_BYTE * (NEW - ORIG) / ORIG | |
4213 | Here, we are sure that NEW >= ORIG. */ | |
6e44253b KH |
4214 | float ratio = coding->produced - coding->consumed; |
4215 | ratio /= coding->consumed; | |
4216 | require = len_byte * ratio; | |
fb88bf2d KH |
4217 | first = 0; |
4218 | } | |
4219 | if ((src - dst) < (require + 2000)) | |
4220 | { | |
4221 | /* See the comment above the previous call of make_gap. */ | |
4222 | int add = len_byte + inserted_byte; | |
4223 | ||
4224 | GAP_SIZE -= add; | |
4225 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; | |
4226 | GPT += inserted_byte; GPT_BYTE += inserted_byte; | |
4227 | make_gap (require + 2000); | |
4228 | GAP_SIZE += add; | |
4229 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; | |
4230 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; | |
4231 | /* Don't forget to update SRC, DST. */ | |
4232 | src = GAP_END_ADDR - len_byte; | |
4233 | dst = GPT_ADDR + inserted_byte; | |
4234 | } | |
d46c5b12 | 4235 | } |
fb88bf2d KH |
4236 | if (src - dst > 0) *dst = 0; /* Put an anchor. */ |
4237 | ||
2b4f9037 KH |
4238 | if (multibyte |
4239 | && (fake_multibyte || !encodep && (to - from) != (to_byte - from_byte))) | |
4240 | inserted = multibyte_chars_in_text (GPT_ADDR, inserted_byte); | |
7553d0e1 | 4241 | |
2b4f9037 | 4242 | adjust_after_replace (from, from_byte, to, to_byte, |
6e44253b | 4243 | inserted, inserted_byte, replace); |
2b4f9037 KH |
4244 | if (from_byte_orig == from_byte) |
4245 | from_byte_orig = from_byte = PT_BYTE; | |
4ed46869 | 4246 | |
2b4f9037 | 4247 | if (! encodep && ! NILP (coding->post_read_conversion)) |
d46c5b12 | 4248 | { |
2b4f9037 KH |
4249 | Lisp_Object val; |
4250 | int orig_inserted = inserted, pos = PT; | |
4ed46869 | 4251 | |
2b4f9037 KH |
4252 | if (from != pos) |
4253 | temp_set_point_both (current_buffer, from, from_byte); | |
4254 | val = call1 (coding->post_read_conversion, make_number (inserted)); | |
4255 | if (! NILP (val)) | |
d46c5b12 | 4256 | { |
2b4f9037 KH |
4257 | CHECK_NUMBER (val, 0); |
4258 | inserted = XFASTINT (val); | |
d46c5b12 | 4259 | } |
2b4f9037 KH |
4260 | if (pos >= from + orig_inserted) |
4261 | temp_set_point (current_buffer, pos + (inserted - orig_inserted)); | |
d46c5b12 | 4262 | } |
4ed46869 | 4263 | |
2b4f9037 KH |
4264 | signal_after_change (from, to - from, inserted); |
4265 | ||
fb88bf2d KH |
4266 | { |
4267 | int skip = (to_byte_orig - to_byte) + (from_byte - from_byte_orig); | |
4268 | ||
4269 | coding->consumed = to_byte_orig - from_byte_orig; | |
4270 | coding->consumed_char = skip + (to - from); | |
4271 | coding->produced = skip + inserted_byte; | |
2b4f9037 | 4272 | coding->produced_char = skip + inserted; |
fb88bf2d | 4273 | } |
7553d0e1 | 4274 | |
fb88bf2d | 4275 | return 0; |
d46c5b12 KH |
4276 | } |
4277 | ||
4278 | Lisp_Object | |
4279 | code_convert_string (str, coding, encodep, nocopy) | |
4280 | Lisp_Object str; | |
4ed46869 | 4281 | struct coding_system *coding; |
d46c5b12 | 4282 | int encodep, nocopy; |
4ed46869 | 4283 | { |
d46c5b12 KH |
4284 | int len; |
4285 | char *buf; | |
4286 | int from = 0, to = XSTRING (str)->size, to_byte = XSTRING (str)->size_byte; | |
4287 | struct gcpro gcpro1; | |
4288 | Lisp_Object saved_coding_symbol = Qnil; | |
4289 | int result; | |
4ed46869 | 4290 | |
d46c5b12 KH |
4291 | if (encodep && !NILP (coding->pre_write_conversion) |
4292 | || !encodep && !NILP (coding->post_read_conversion)) | |
4293 | { | |
4294 | /* Since we have to call Lisp functions which assume target text | |
4295 | is in a buffer, after setting a temporary buffer, call | |
4296 | code_convert_region. */ | |
4297 | int count = specpdl_ptr - specpdl; | |
4298 | struct buffer *prev = current_buffer; | |
4299 | ||
4300 | record_unwind_protect (Fset_buffer, Fcurrent_buffer ()); | |
4301 | temp_output_buffer_setup (" *code-converting-work*"); | |
4302 | set_buffer_internal (XBUFFER (Vstandard_output)); | |
4303 | if (encodep) | |
4304 | insert_from_string (str, 0, 0, to, to_byte, 0); | |
4305 | else | |
4306 | { | |
4307 | /* We must insert the contents of STR as is without | |
4308 | unibyte<->multibyte conversion. */ | |
4309 | current_buffer->enable_multibyte_characters = Qnil; | |
4310 | insert_from_string (str, 0, 0, to_byte, to_byte, 0); | |
4311 | current_buffer->enable_multibyte_characters = Qt; | |
4312 | } | |
fb88bf2d | 4313 | code_convert_region (BEGV, BEGV_BYTE, ZV, ZV_BYTE, coding, encodep, 1); |
d46c5b12 KH |
4314 | if (encodep) |
4315 | /* We must return the buffer contents as unibyte string. */ | |
4316 | current_buffer->enable_multibyte_characters = Qnil; | |
4317 | str = make_buffer_string (BEGV, ZV, 0); | |
4318 | set_buffer_internal (prev); | |
4319 | return unbind_to (count, str); | |
4320 | } | |
4ed46869 | 4321 | |
d46c5b12 KH |
4322 | if (! encodep && CODING_REQUIRE_DETECTION (coding)) |
4323 | { | |
4324 | /* See the comments in code_convert_region. */ | |
4325 | if (coding->type == coding_type_undecided) | |
4326 | { | |
4327 | detect_coding (coding, XSTRING (str)->data, to_byte); | |
4328 | if (coding->type == coding_type_undecided) | |
4329 | coding->type = coding_type_emacs_mule; | |
4330 | } | |
4331 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
4332 | { | |
4333 | saved_coding_symbol = coding->symbol; | |
4334 | detect_eol (coding, XSTRING (str)->data, to_byte); | |
4335 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
4336 | coding->eol_type = CODING_EOL_LF; | |
4337 | /* We had better recover the original eol format if we | |
4338 | encounter an inconsitent eol format while decoding. */ | |
4339 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; | |
4340 | } | |
4341 | } | |
4ed46869 | 4342 | |
d46c5b12 KH |
4343 | if (encodep |
4344 | ? ! CODING_REQUIRE_ENCODING (coding) | |
4345 | : ! CODING_REQUIRE_DECODING (coding)) | |
4346 | from = to_byte; | |
4347 | else | |
4348 | { | |
4349 | /* Try to skip the heading and tailing ASCIIs. */ | |
4350 | if (encodep) | |
4351 | shrink_encoding_region (&from, &to_byte, coding, XSTRING (str)->data); | |
4352 | else | |
4353 | shrink_decoding_region (&from, &to_byte, coding, XSTRING (str)->data); | |
4354 | } | |
4355 | if (from == to_byte) | |
4356 | return (nocopy ? str : Fcopy_sequence (str)); | |
4ed46869 | 4357 | |
d46c5b12 KH |
4358 | if (encodep) |
4359 | len = encoding_buffer_size (coding, to_byte - from); | |
4360 | else | |
4361 | len = decoding_buffer_size (coding, to_byte - from); | |
4362 | len += from + XSTRING (str)->size_byte - to_byte; | |
4363 | GCPRO1 (str); | |
4364 | buf = get_conversion_buffer (len); | |
4365 | UNGCPRO; | |
4ed46869 | 4366 | |
d46c5b12 KH |
4367 | if (from > 0) |
4368 | bcopy (XSTRING (str)->data, buf, from); | |
4369 | result = (encodep | |
4370 | ? encode_coding (coding, XSTRING (str)->data + from, | |
4371 | buf + from, to_byte - from, len) | |
4372 | : decode_coding (coding, XSTRING (str)->data + from, | |
4373 | buf + from, to - from, len)); | |
4374 | if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL) | |
4ed46869 | 4375 | { |
d46c5b12 KH |
4376 | /* We simple try to decode the whole string again but without |
4377 | eol-conversion this time. */ | |
4378 | coding->eol_type = CODING_EOL_LF; | |
4379 | coding->symbol = saved_coding_symbol; | |
4380 | return code_convert_string (str, coding, encodep, nocopy); | |
4ed46869 | 4381 | } |
d46c5b12 KH |
4382 | |
4383 | bcopy (XSTRING (str)->data + to_byte, buf + from + coding->produced, | |
4384 | XSTRING (str)->size_byte - to_byte); | |
4385 | ||
4386 | len = from + XSTRING (str)->size_byte - to_byte; | |
4387 | if (encodep) | |
4388 | str = make_unibyte_string (buf, len + coding->produced); | |
4389 | else | |
4390 | str = make_multibyte_string (buf, len + coding->produced_char, | |
4391 | len + coding->produced); | |
4392 | return str; | |
4ed46869 KH |
4393 | } |
4394 | ||
4395 | \f | |
4396 | #ifdef emacs | |
4397 | /*** 7. Emacs Lisp library functions ***/ | |
4398 | ||
4ed46869 KH |
4399 | DEFUN ("coding-system-p", Fcoding_system_p, Scoding_system_p, 1, 1, 0, |
4400 | "Return t if OBJECT is nil or a coding-system.\n\ | |
3a73fa5d RS |
4401 | See the documentation of `make-coding-system' for information\n\ |
4402 | about coding-system objects.") | |
4ed46869 KH |
4403 | (obj) |
4404 | Lisp_Object obj; | |
4405 | { | |
4608c386 KH |
4406 | if (NILP (obj)) |
4407 | return Qt; | |
4408 | if (!SYMBOLP (obj)) | |
4409 | return Qnil; | |
4410 | /* Get coding-spec vector for OBJ. */ | |
4411 | obj = Fget (obj, Qcoding_system); | |
4412 | return ((VECTORP (obj) && XVECTOR (obj)->size == 5) | |
4413 | ? Qt : Qnil); | |
4ed46869 KH |
4414 | } |
4415 | ||
9d991de8 RS |
4416 | DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system, |
4417 | Sread_non_nil_coding_system, 1, 1, 0, | |
e0e989f6 | 4418 | "Read a coding system from the minibuffer, prompting with string PROMPT.") |
4ed46869 KH |
4419 | (prompt) |
4420 | Lisp_Object prompt; | |
4421 | { | |
e0e989f6 | 4422 | Lisp_Object val; |
9d991de8 RS |
4423 | do |
4424 | { | |
4608c386 KH |
4425 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
4426 | Qt, Qnil, Qcoding_system_history, Qnil, Qnil); | |
9d991de8 RS |
4427 | } |
4428 | while (XSTRING (val)->size == 0); | |
e0e989f6 | 4429 | return (Fintern (val, Qnil)); |
4ed46869 KH |
4430 | } |
4431 | ||
9b787f3e RS |
4432 | DEFUN ("read-coding-system", Fread_coding_system, Sread_coding_system, 1, 2, 0, |
4433 | "Read a coding system from the minibuffer, prompting with string PROMPT.\n\ | |
4434 | If the user enters null input, return second argument DEFAULT-CODING-SYSTEM.") | |
4435 | (prompt, default_coding_system) | |
4436 | Lisp_Object prompt, default_coding_system; | |
4ed46869 | 4437 | { |
f44d27ce | 4438 | Lisp_Object val; |
9b787f3e RS |
4439 | if (SYMBOLP (default_coding_system)) |
4440 | XSETSTRING (default_coding_system, XSYMBOL (default_coding_system)->name); | |
4608c386 | 4441 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
9b787f3e RS |
4442 | Qt, Qnil, Qcoding_system_history, |
4443 | default_coding_system, Qnil); | |
e0e989f6 | 4444 | return (XSTRING (val)->size == 0 ? Qnil : Fintern (val, Qnil)); |
4ed46869 KH |
4445 | } |
4446 | ||
4447 | DEFUN ("check-coding-system", Fcheck_coding_system, Scheck_coding_system, | |
4448 | 1, 1, 0, | |
4449 | "Check validity of CODING-SYSTEM.\n\ | |
3a73fa5d RS |
4450 | If valid, return CODING-SYSTEM, else signal a `coding-system-error' error.\n\ |
4451 | It is valid if it is a symbol with a non-nil `coding-system' property.\n\ | |
4ed46869 KH |
4452 | The value of property should be a vector of length 5.") |
4453 | (coding_system) | |
4454 | Lisp_Object coding_system; | |
4455 | { | |
4456 | CHECK_SYMBOL (coding_system, 0); | |
4457 | if (!NILP (Fcoding_system_p (coding_system))) | |
4458 | return coding_system; | |
4459 | while (1) | |
02ba4723 | 4460 | Fsignal (Qcoding_system_error, Fcons (coding_system, Qnil)); |
4ed46869 | 4461 | } |
3a73fa5d | 4462 | \f |
d46c5b12 KH |
4463 | Lisp_Object |
4464 | detect_coding_system (src, src_bytes, highest) | |
4465 | unsigned char *src; | |
4466 | int src_bytes, highest; | |
4ed46869 KH |
4467 | { |
4468 | int coding_mask, eol_type; | |
d46c5b12 KH |
4469 | Lisp_Object val, tmp; |
4470 | int dummy; | |
4ed46869 | 4471 | |
d46c5b12 KH |
4472 | coding_mask = detect_coding_mask (src, src_bytes, NULL, &dummy); |
4473 | eol_type = detect_eol_type (src, src_bytes, &dummy); | |
4474 | if (eol_type == CODING_EOL_INCONSISTENT) | |
4475 | eol_type == CODING_EOL_UNDECIDED; | |
4ed46869 | 4476 | |
d46c5b12 | 4477 | if (!coding_mask) |
4ed46869 | 4478 | { |
27901516 | 4479 | val = Qundecided; |
d46c5b12 | 4480 | if (eol_type != CODING_EOL_UNDECIDED) |
4ed46869 | 4481 | { |
f44d27ce RS |
4482 | Lisp_Object val2; |
4483 | val2 = Fget (Qundecided, Qeol_type); | |
4ed46869 KH |
4484 | if (VECTORP (val2)) |
4485 | val = XVECTOR (val2)->contents[eol_type]; | |
4486 | } | |
d46c5b12 | 4487 | return val; |
4ed46869 | 4488 | } |
4ed46869 | 4489 | |
d46c5b12 KH |
4490 | /* At first, gather possible coding systems in VAL. */ |
4491 | val = Qnil; | |
4492 | for (tmp = Vcoding_category_list; !NILP (tmp); tmp = XCONS (tmp)->cdr) | |
4ed46869 | 4493 | { |
d46c5b12 KH |
4494 | int idx |
4495 | = XFASTINT (Fget (XCONS (tmp)->car, Qcoding_category_index)); | |
4496 | if (coding_mask & (1 << idx)) | |
4ed46869 | 4497 | { |
d46c5b12 KH |
4498 | val = Fcons (Fsymbol_value (XCONS (tmp)->car), val); |
4499 | if (highest) | |
4500 | break; | |
4ed46869 KH |
4501 | } |
4502 | } | |
d46c5b12 KH |
4503 | if (!highest) |
4504 | val = Fnreverse (val); | |
4ed46869 | 4505 | |
d46c5b12 KH |
4506 | /* Then, substitute the elements by subsidiary coding systems. */ |
4507 | for (tmp = val; !NILP (tmp); tmp = XCONS (tmp)->cdr) | |
4ed46869 | 4508 | { |
d46c5b12 | 4509 | if (eol_type != CODING_EOL_UNDECIDED) |
4ed46869 | 4510 | { |
d46c5b12 KH |
4511 | Lisp_Object eol; |
4512 | eol = Fget (XCONS (tmp)->car, Qeol_type); | |
4513 | if (VECTORP (eol)) | |
4514 | XCONS (tmp)->car = XVECTOR (eol)->contents[eol_type]; | |
4ed46869 KH |
4515 | } |
4516 | } | |
d46c5b12 KH |
4517 | return (highest ? XCONS (val)->car : val); |
4518 | } | |
4ed46869 | 4519 | |
d46c5b12 KH |
4520 | DEFUN ("detect-coding-region", Fdetect_coding_region, Sdetect_coding_region, |
4521 | 2, 3, 0, | |
4522 | "Detect coding system of the text in the region between START and END.\n\ | |
4523 | Return a list of possible coding systems ordered by priority.\n\ | |
4524 | \n\ | |
4525 | If only ASCII characters are found, it returns `undecided'\n\ | |
4526 | or its subsidiary coding system according to a detected end-of-line format.\n\ | |
4527 | \n\ | |
4528 | If optional argument HIGHEST is non-nil, return the coding system of\n\ | |
4529 | highest priority.") | |
4530 | (start, end, highest) | |
4531 | Lisp_Object start, end, highest; | |
4532 | { | |
4533 | int from, to; | |
4534 | int from_byte, to_byte; | |
6289dd10 | 4535 | |
d46c5b12 KH |
4536 | CHECK_NUMBER_COERCE_MARKER (start, 0); |
4537 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
4ed46869 | 4538 | |
d46c5b12 KH |
4539 | validate_region (&start, &end); |
4540 | from = XINT (start), to = XINT (end); | |
4541 | from_byte = CHAR_TO_BYTE (from); | |
4542 | to_byte = CHAR_TO_BYTE (to); | |
6289dd10 | 4543 | |
d46c5b12 KH |
4544 | if (from < GPT && to >= GPT) |
4545 | move_gap_both (to, to_byte); | |
4ed46869 | 4546 | |
d46c5b12 KH |
4547 | return detect_coding_system (BYTE_POS_ADDR (from_byte), |
4548 | to_byte - from_byte, | |
4549 | !NILP (highest)); | |
4550 | } | |
6289dd10 | 4551 | |
d46c5b12 KH |
4552 | DEFUN ("detect-coding-string", Fdetect_coding_string, Sdetect_coding_string, |
4553 | 1, 2, 0, | |
4554 | "Detect coding system of the text in STRING.\n\ | |
4555 | Return a list of possible coding systems ordered by priority.\n\ | |
4556 | \n\ | |
4557 | If only ASCII characters are found, it returns `undecided'\n\ | |
4558 | or its subsidiary coding system according to a detected end-of-line format.\n\ | |
4559 | \n\ | |
4560 | If optional argument HIGHEST is non-nil, return the coding system of\n\ | |
4561 | highest priority.") | |
4562 | (string, highest) | |
4563 | Lisp_Object string, highest; | |
4564 | { | |
4565 | CHECK_STRING (string, 0); | |
4ed46869 | 4566 | |
d46c5b12 KH |
4567 | return detect_coding_system (XSTRING (string)->data, |
4568 | XSTRING (string)->size_byte, | |
4569 | !NILP (highest)); | |
4ed46869 KH |
4570 | } |
4571 | ||
4031e2bf KH |
4572 | Lisp_Object |
4573 | code_convert_region1 (start, end, coding_system, encodep) | |
d46c5b12 | 4574 | Lisp_Object start, end, coding_system; |
4031e2bf | 4575 | int encodep; |
3a73fa5d RS |
4576 | { |
4577 | struct coding_system coding; | |
4031e2bf | 4578 | int from, to, len; |
3a73fa5d | 4579 | |
d46c5b12 KH |
4580 | CHECK_NUMBER_COERCE_MARKER (start, 0); |
4581 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
3a73fa5d RS |
4582 | CHECK_SYMBOL (coding_system, 2); |
4583 | ||
d46c5b12 KH |
4584 | validate_region (&start, &end); |
4585 | from = XFASTINT (start); | |
4586 | to = XFASTINT (end); | |
4587 | ||
3a73fa5d | 4588 | if (NILP (coding_system)) |
d46c5b12 KH |
4589 | return make_number (to - from); |
4590 | ||
3a73fa5d | 4591 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
d46c5b12 | 4592 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); |
3a73fa5d | 4593 | |
d46c5b12 | 4594 | coding.mode |= CODING_MODE_LAST_BLOCK; |
fb88bf2d KH |
4595 | code_convert_region (from, CHAR_TO_BYTE (from), to, CHAR_TO_BYTE (to), |
4596 | &coding, encodep, 1); | |
4597 | return make_number (coding.produced_char); | |
4031e2bf KH |
4598 | } |
4599 | ||
4600 | DEFUN ("decode-coding-region", Fdecode_coding_region, Sdecode_coding_region, | |
4601 | 3, 3, "r\nzCoding system: ", | |
4602 | "Decode the current region by specified coding system.\n\ | |
4603 | When called from a program, takes three arguments:\n\ | |
4604 | START, END, and CODING-SYSTEM. START and END are buffer positions.\n\ | |
4605 | Return length of decoded text.") | |
4606 | (start, end, coding_system) | |
4607 | Lisp_Object start, end, coding_system; | |
4608 | { | |
4609 | return code_convert_region1 (start, end, coding_system, 0); | |
3a73fa5d RS |
4610 | } |
4611 | ||
4612 | DEFUN ("encode-coding-region", Fencode_coding_region, Sencode_coding_region, | |
4613 | 3, 3, "r\nzCoding system: ", | |
d46c5b12 | 4614 | "Encode the current region by specified coding system.\n\ |
3a73fa5d | 4615 | When called from a program, takes three arguments:\n\ |
d46c5b12 | 4616 | START, END, and CODING-SYSTEM. START and END are buffer positions.\n\ |
3a73fa5d | 4617 | Return length of encoded text.") |
d46c5b12 KH |
4618 | (start, end, coding_system) |
4619 | Lisp_Object start, end, coding_system; | |
3a73fa5d | 4620 | { |
4031e2bf KH |
4621 | return code_convert_region1 (start, end, coding_system, 1); |
4622 | } | |
3a73fa5d | 4623 | |
4031e2bf KH |
4624 | Lisp_Object |
4625 | code_convert_string1 (string, coding_system, nocopy, encodep) | |
4626 | Lisp_Object string, coding_system, nocopy; | |
4627 | int encodep; | |
4628 | { | |
4629 | struct coding_system coding; | |
3a73fa5d | 4630 | |
4031e2bf KH |
4631 | CHECK_STRING (string, 0); |
4632 | CHECK_SYMBOL (coding_system, 1); | |
4ed46869 | 4633 | |
d46c5b12 | 4634 | if (NILP (coding_system)) |
4031e2bf | 4635 | return (NILP (nocopy) ? Fcopy_sequence (string) : string); |
4ed46869 | 4636 | |
d46c5b12 KH |
4637 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
4638 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); | |
5f1cd180 | 4639 | |
d46c5b12 | 4640 | coding.mode |= CODING_MODE_LAST_BLOCK; |
4031e2bf | 4641 | return code_convert_string (string, &coding, encodep, !NILP (nocopy)); |
4ed46869 KH |
4642 | } |
4643 | ||
4ed46869 | 4644 | DEFUN ("decode-coding-string", Fdecode_coding_string, Sdecode_coding_string, |
e0e989f6 KH |
4645 | 2, 3, 0, |
4646 | "Decode STRING which is encoded in CODING-SYSTEM, and return the result.\n\ | |
fe487a71 RS |
4647 | Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\ |
4648 | if the decoding operation is trivial.") | |
e0e989f6 KH |
4649 | (string, coding_system, nocopy) |
4650 | Lisp_Object string, coding_system, nocopy; | |
4ed46869 | 4651 | { |
4031e2bf | 4652 | return code_convert_string1(string, coding_system, nocopy, 0); |
4ed46869 KH |
4653 | } |
4654 | ||
4655 | DEFUN ("encode-coding-string", Fencode_coding_string, Sencode_coding_string, | |
e0e989f6 KH |
4656 | 2, 3, 0, |
4657 | "Encode STRING to CODING-SYSTEM, and return the result.\n\ | |
fe487a71 RS |
4658 | Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\ |
4659 | if the encoding operation is trivial.") | |
e0e989f6 KH |
4660 | (string, coding_system, nocopy) |
4661 | Lisp_Object string, coding_system, nocopy; | |
4ed46869 | 4662 | { |
4031e2bf | 4663 | return code_convert_string1(string, coding_system, nocopy, 1); |
4ed46869 | 4664 | } |
4031e2bf | 4665 | |
3a73fa5d | 4666 | \f |
4ed46869 | 4667 | DEFUN ("decode-sjis-char", Fdecode_sjis_char, Sdecode_sjis_char, 1, 1, 0, |
e0e989f6 | 4668 | "Decode a JISX0208 character of shift-jis encoding.\n\ |
4ed46869 KH |
4669 | CODE is the character code in SJIS.\n\ |
4670 | Return the corresponding character.") | |
4671 | (code) | |
4672 | Lisp_Object code; | |
4673 | { | |
4674 | unsigned char c1, c2, s1, s2; | |
4675 | Lisp_Object val; | |
4676 | ||
4677 | CHECK_NUMBER (code, 0); | |
4678 | s1 = (XFASTINT (code)) >> 8, s2 = (XFASTINT (code)) & 0xFF; | |
4679 | DECODE_SJIS (s1, s2, c1, c2); | |
4680 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset_jisx0208, c1, c2)); | |
4681 | return val; | |
4682 | } | |
4683 | ||
4684 | DEFUN ("encode-sjis-char", Fencode_sjis_char, Sencode_sjis_char, 1, 1, 0, | |
d46c5b12 | 4685 | "Encode a JISX0208 character CHAR to SJIS coding system.\n\ |
4ed46869 KH |
4686 | Return the corresponding character code in SJIS.") |
4687 | (ch) | |
4688 | Lisp_Object ch; | |
4689 | { | |
bcf26d6a | 4690 | int charset, c1, c2, s1, s2; |
4ed46869 KH |
4691 | Lisp_Object val; |
4692 | ||
4693 | CHECK_NUMBER (ch, 0); | |
4694 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
4695 | if (charset == charset_jisx0208) | |
4696 | { | |
4697 | ENCODE_SJIS (c1, c2, s1, s2); | |
bcf26d6a | 4698 | XSETFASTINT (val, (s1 << 8) | s2); |
4ed46869 KH |
4699 | } |
4700 | else | |
4701 | XSETFASTINT (val, 0); | |
4702 | return val; | |
4703 | } | |
4704 | ||
4705 | DEFUN ("decode-big5-char", Fdecode_big5_char, Sdecode_big5_char, 1, 1, 0, | |
d46c5b12 | 4706 | "Decode a Big5 character CODE of BIG5 coding system.\n\ |
4ed46869 KH |
4707 | CODE is the character code in BIG5.\n\ |
4708 | Return the corresponding character.") | |
4709 | (code) | |
4710 | Lisp_Object code; | |
4711 | { | |
4712 | int charset; | |
4713 | unsigned char b1, b2, c1, c2; | |
4714 | Lisp_Object val; | |
4715 | ||
4716 | CHECK_NUMBER (code, 0); | |
4717 | b1 = (XFASTINT (code)) >> 8, b2 = (XFASTINT (code)) & 0xFF; | |
4718 | DECODE_BIG5 (b1, b2, charset, c1, c2); | |
4719 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset, c1, c2)); | |
4720 | return val; | |
4721 | } | |
4722 | ||
4723 | DEFUN ("encode-big5-char", Fencode_big5_char, Sencode_big5_char, 1, 1, 0, | |
d46c5b12 | 4724 | "Encode the Big5 character CHAR to BIG5 coding system.\n\ |
4ed46869 KH |
4725 | Return the corresponding character code in Big5.") |
4726 | (ch) | |
4727 | Lisp_Object ch; | |
4728 | { | |
bcf26d6a | 4729 | int charset, c1, c2, b1, b2; |
4ed46869 KH |
4730 | Lisp_Object val; |
4731 | ||
4732 | CHECK_NUMBER (ch, 0); | |
4733 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
4734 | if (charset == charset_big5_1 || charset == charset_big5_2) | |
4735 | { | |
4736 | ENCODE_BIG5 (charset, c1, c2, b1, b2); | |
bcf26d6a | 4737 | XSETFASTINT (val, (b1 << 8) | b2); |
4ed46869 KH |
4738 | } |
4739 | else | |
4740 | XSETFASTINT (val, 0); | |
4741 | return val; | |
4742 | } | |
3a73fa5d | 4743 | \f |
1ba9e4ab KH |
4744 | DEFUN ("set-terminal-coding-system-internal", |
4745 | Fset_terminal_coding_system_internal, | |
4746 | Sset_terminal_coding_system_internal, 1, 1, 0, "") | |
4ed46869 KH |
4747 | (coding_system) |
4748 | Lisp_Object coding_system; | |
4749 | { | |
4750 | CHECK_SYMBOL (coding_system, 0); | |
4751 | setup_coding_system (Fcheck_coding_system (coding_system), &terminal_coding); | |
70c22245 | 4752 | /* We had better not send unsafe characters to terminal. */ |
6e85d753 KH |
4753 | terminal_coding.flags |= CODING_FLAG_ISO_SAFE; |
4754 | ||
4ed46869 KH |
4755 | return Qnil; |
4756 | } | |
4757 | ||
c4825358 KH |
4758 | DEFUN ("set-safe-terminal-coding-system-internal", |
4759 | Fset_safe_terminal_coding_system_internal, | |
4760 | Sset_safe_terminal_coding_system_internal, 1, 1, 0, "") | |
4761 | (coding_system) | |
4762 | Lisp_Object coding_system; | |
4763 | { | |
4764 | CHECK_SYMBOL (coding_system, 0); | |
4765 | setup_coding_system (Fcheck_coding_system (coding_system), | |
4766 | &safe_terminal_coding); | |
4767 | return Qnil; | |
4768 | } | |
4769 | ||
4ed46869 KH |
4770 | DEFUN ("terminal-coding-system", |
4771 | Fterminal_coding_system, Sterminal_coding_system, 0, 0, 0, | |
3a73fa5d | 4772 | "Return coding system specified for terminal output.") |
4ed46869 KH |
4773 | () |
4774 | { | |
4775 | return terminal_coding.symbol; | |
4776 | } | |
4777 | ||
1ba9e4ab KH |
4778 | DEFUN ("set-keyboard-coding-system-internal", |
4779 | Fset_keyboard_coding_system_internal, | |
4780 | Sset_keyboard_coding_system_internal, 1, 1, 0, "") | |
4ed46869 KH |
4781 | (coding_system) |
4782 | Lisp_Object coding_system; | |
4783 | { | |
4784 | CHECK_SYMBOL (coding_system, 0); | |
4785 | setup_coding_system (Fcheck_coding_system (coding_system), &keyboard_coding); | |
4786 | return Qnil; | |
4787 | } | |
4788 | ||
4789 | DEFUN ("keyboard-coding-system", | |
4790 | Fkeyboard_coding_system, Skeyboard_coding_system, 0, 0, 0, | |
3a73fa5d | 4791 | "Return coding system specified for decoding keyboard input.") |
4ed46869 KH |
4792 | () |
4793 | { | |
4794 | return keyboard_coding.symbol; | |
4795 | } | |
4796 | ||
4797 | \f | |
a5d301df KH |
4798 | DEFUN ("find-operation-coding-system", Ffind_operation_coding_system, |
4799 | Sfind_operation_coding_system, 1, MANY, 0, | |
4800 | "Choose a coding system for an operation based on the target name.\n\ | |
9ce27fde KH |
4801 | The value names a pair of coding systems: (DECODING-SYSTEM ENCODING-SYSTEM).\n\ |
4802 | DECODING-SYSTEM is the coding system to use for decoding\n\ | |
4803 | \(in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system\n\ | |
4804 | for encoding (in case OPERATION does encoding).\n\ | |
ccdb79f5 RS |
4805 | \n\ |
4806 | The first argument OPERATION specifies an I/O primitive:\n\ | |
4807 | For file I/O, `insert-file-contents' or `write-region'.\n\ | |
4808 | For process I/O, `call-process', `call-process-region', or `start-process'.\n\ | |
4809 | For network I/O, `open-network-stream'.\n\ | |
4810 | \n\ | |
4811 | The remaining arguments should be the same arguments that were passed\n\ | |
4812 | to the primitive. Depending on which primitive, one of those arguments\n\ | |
4813 | is selected as the TARGET. For example, if OPERATION does file I/O,\n\ | |
4814 | whichever argument specifies the file name is TARGET.\n\ | |
4815 | \n\ | |
4816 | TARGET has a meaning which depends on OPERATION:\n\ | |
4ed46869 KH |
4817 | For file I/O, TARGET is a file name.\n\ |
4818 | For process I/O, TARGET is a process name.\n\ | |
4819 | For network I/O, TARGET is a service name or a port number\n\ | |
4820 | \n\ | |
02ba4723 KH |
4821 | This function looks up what specified for TARGET in,\n\ |
4822 | `file-coding-system-alist', `process-coding-system-alist',\n\ | |
4823 | or `network-coding-system-alist' depending on OPERATION.\n\ | |
4824 | They may specify a coding system, a cons of coding systems,\n\ | |
4825 | or a function symbol to call.\n\ | |
4826 | In the last case, we call the function with one argument,\n\ | |
9ce27fde | 4827 | which is a list of all the arguments given to this function.") |
4ed46869 KH |
4828 | (nargs, args) |
4829 | int nargs; | |
4830 | Lisp_Object *args; | |
4831 | { | |
4832 | Lisp_Object operation, target_idx, target, val; | |
4833 | register Lisp_Object chain; | |
4834 | ||
4835 | if (nargs < 2) | |
4836 | error ("Too few arguments"); | |
4837 | operation = args[0]; | |
4838 | if (!SYMBOLP (operation) | |
4839 | || !INTEGERP (target_idx = Fget (operation, Qtarget_idx))) | |
4840 | error ("Invalid first arguement"); | |
4841 | if (nargs < 1 + XINT (target_idx)) | |
4842 | error ("Too few arguments for operation: %s", | |
4843 | XSYMBOL (operation)->name->data); | |
4844 | target = args[XINT (target_idx) + 1]; | |
4845 | if (!(STRINGP (target) | |
4846 | || (EQ (operation, Qopen_network_stream) && INTEGERP (target)))) | |
4847 | error ("Invalid %dth argument", XINT (target_idx) + 1); | |
4848 | ||
2e34157c RS |
4849 | chain = ((EQ (operation, Qinsert_file_contents) |
4850 | || EQ (operation, Qwrite_region)) | |
02ba4723 | 4851 | ? Vfile_coding_system_alist |
2e34157c | 4852 | : (EQ (operation, Qopen_network_stream) |
02ba4723 KH |
4853 | ? Vnetwork_coding_system_alist |
4854 | : Vprocess_coding_system_alist)); | |
4ed46869 KH |
4855 | if (NILP (chain)) |
4856 | return Qnil; | |
4857 | ||
02ba4723 | 4858 | for (; CONSP (chain); chain = XCONS (chain)->cdr) |
4ed46869 | 4859 | { |
f44d27ce RS |
4860 | Lisp_Object elt; |
4861 | elt = XCONS (chain)->car; | |
4ed46869 KH |
4862 | |
4863 | if (CONSP (elt) | |
4864 | && ((STRINGP (target) | |
4865 | && STRINGP (XCONS (elt)->car) | |
4866 | && fast_string_match (XCONS (elt)->car, target) >= 0) | |
4867 | || (INTEGERP (target) && EQ (target, XCONS (elt)->car)))) | |
02ba4723 KH |
4868 | { |
4869 | val = XCONS (elt)->cdr; | |
b19fd4c5 KH |
4870 | /* Here, if VAL is both a valid coding system and a valid |
4871 | function symbol, we return VAL as a coding system. */ | |
02ba4723 KH |
4872 | if (CONSP (val)) |
4873 | return val; | |
4874 | if (! SYMBOLP (val)) | |
4875 | return Qnil; | |
4876 | if (! NILP (Fcoding_system_p (val))) | |
4877 | return Fcons (val, val); | |
b19fd4c5 KH |
4878 | if (! NILP (Ffboundp (val))) |
4879 | { | |
4880 | val = call1 (val, Flist (nargs, args)); | |
4881 | if (CONSP (val)) | |
4882 | return val; | |
4883 | if (SYMBOLP (val) && ! NILP (Fcoding_system_p (val))) | |
4884 | return Fcons (val, val); | |
4885 | } | |
02ba4723 KH |
4886 | return Qnil; |
4887 | } | |
4ed46869 KH |
4888 | } |
4889 | return Qnil; | |
4890 | } | |
4891 | ||
d46c5b12 KH |
4892 | DEFUN ("update-iso-coding-systems", Fupdate_iso_coding_systems, |
4893 | Supdate_iso_coding_systems, 0, 0, 0, | |
4894 | "Update internal database for ISO2022 based coding systems.\n\ | |
4895 | When values of the following coding categories are changed, you must\n\ | |
4896 | call this function:\n\ | |
4897 | coding-category-iso-7, coding-category-iso-7-tight,\n\ | |
4898 | coding-category-iso-8-1, coding-category-iso-8-2,\n\ | |
4899 | coding-category-iso-7-else, coding-category-iso-8-else") | |
4900 | () | |
4901 | { | |
4902 | int i; | |
4903 | ||
4904 | for (i = CODING_CATEGORY_IDX_ISO_7; i <= CODING_CATEGORY_IDX_ISO_8_ELSE; | |
4905 | i++) | |
4906 | { | |
4907 | if (! coding_system_table[i]) | |
4908 | coding_system_table[i] | |
4909 | = (struct coding_system *) xmalloc (sizeof (struct coding_system)); | |
4910 | setup_coding_system | |
4911 | (XSYMBOL (XVECTOR (Vcoding_category_table)->contents[i])->value, | |
4912 | coding_system_table[i]); | |
4913 | } | |
4914 | return Qnil; | |
4915 | } | |
4916 | ||
4ed46869 KH |
4917 | #endif /* emacs */ |
4918 | ||
4919 | \f | |
4920 | /*** 8. Post-amble ***/ | |
4921 | ||
4922 | init_coding_once () | |
4923 | { | |
4924 | int i; | |
4925 | ||
0ef69138 | 4926 | /* Emacs' internal format specific initialize routine. */ |
4ed46869 KH |
4927 | for (i = 0; i <= 0x20; i++) |
4928 | emacs_code_class[i] = EMACS_control_code; | |
4929 | emacs_code_class[0x0A] = EMACS_linefeed_code; | |
4930 | emacs_code_class[0x0D] = EMACS_carriage_return_code; | |
4931 | for (i = 0x21 ; i < 0x7F; i++) | |
4932 | emacs_code_class[i] = EMACS_ascii_code; | |
4933 | emacs_code_class[0x7F] = EMACS_control_code; | |
4934 | emacs_code_class[0x80] = EMACS_leading_code_composition; | |
4935 | for (i = 0x81; i < 0xFF; i++) | |
4936 | emacs_code_class[i] = EMACS_invalid_code; | |
4937 | emacs_code_class[LEADING_CODE_PRIVATE_11] = EMACS_leading_code_3; | |
4938 | emacs_code_class[LEADING_CODE_PRIVATE_12] = EMACS_leading_code_3; | |
4939 | emacs_code_class[LEADING_CODE_PRIVATE_21] = EMACS_leading_code_4; | |
4940 | emacs_code_class[LEADING_CODE_PRIVATE_22] = EMACS_leading_code_4; | |
4941 | ||
4942 | /* ISO2022 specific initialize routine. */ | |
4943 | for (i = 0; i < 0x20; i++) | |
4944 | iso_code_class[i] = ISO_control_code; | |
4945 | for (i = 0x21; i < 0x7F; i++) | |
4946 | iso_code_class[i] = ISO_graphic_plane_0; | |
4947 | for (i = 0x80; i < 0xA0; i++) | |
4948 | iso_code_class[i] = ISO_control_code; | |
4949 | for (i = 0xA1; i < 0xFF; i++) | |
4950 | iso_code_class[i] = ISO_graphic_plane_1; | |
4951 | iso_code_class[0x20] = iso_code_class[0x7F] = ISO_0x20_or_0x7F; | |
4952 | iso_code_class[0xA0] = iso_code_class[0xFF] = ISO_0xA0_or_0xFF; | |
4953 | iso_code_class[ISO_CODE_CR] = ISO_carriage_return; | |
4954 | iso_code_class[ISO_CODE_SO] = ISO_shift_out; | |
4955 | iso_code_class[ISO_CODE_SI] = ISO_shift_in; | |
4956 | iso_code_class[ISO_CODE_SS2_7] = ISO_single_shift_2_7; | |
4957 | iso_code_class[ISO_CODE_ESC] = ISO_escape; | |
4958 | iso_code_class[ISO_CODE_SS2] = ISO_single_shift_2; | |
4959 | iso_code_class[ISO_CODE_SS3] = ISO_single_shift_3; | |
4960 | iso_code_class[ISO_CODE_CSI] = ISO_control_sequence_introducer; | |
4961 | ||
e0e989f6 KH |
4962 | conversion_buffer_size = MINIMUM_CONVERSION_BUFFER_SIZE; |
4963 | conversion_buffer = (char *) xmalloc (MINIMUM_CONVERSION_BUFFER_SIZE); | |
4964 | ||
4965 | setup_coding_system (Qnil, &keyboard_coding); | |
4966 | setup_coding_system (Qnil, &terminal_coding); | |
c4825358 | 4967 | setup_coding_system (Qnil, &safe_terminal_coding); |
9ce27fde | 4968 | |
d46c5b12 KH |
4969 | bzero (coding_system_table, sizeof coding_system_table); |
4970 | ||
9ce27fde KH |
4971 | #if defined (MSDOS) || defined (WINDOWSNT) |
4972 | system_eol_type = CODING_EOL_CRLF; | |
4973 | #else | |
4974 | system_eol_type = CODING_EOL_LF; | |
4975 | #endif | |
e0e989f6 KH |
4976 | } |
4977 | ||
4978 | #ifdef emacs | |
4979 | ||
4980 | syms_of_coding () | |
4981 | { | |
4982 | Qtarget_idx = intern ("target-idx"); | |
4983 | staticpro (&Qtarget_idx); | |
4984 | ||
bb0115a2 RS |
4985 | Qcoding_system_history = intern ("coding-system-history"); |
4986 | staticpro (&Qcoding_system_history); | |
4987 | Fset (Qcoding_system_history, Qnil); | |
4988 | ||
9ce27fde | 4989 | /* Target FILENAME is the first argument. */ |
e0e989f6 | 4990 | Fput (Qinsert_file_contents, Qtarget_idx, make_number (0)); |
9ce27fde | 4991 | /* Target FILENAME is the third argument. */ |
e0e989f6 KH |
4992 | Fput (Qwrite_region, Qtarget_idx, make_number (2)); |
4993 | ||
4994 | Qcall_process = intern ("call-process"); | |
4995 | staticpro (&Qcall_process); | |
9ce27fde | 4996 | /* Target PROGRAM is the first argument. */ |
e0e989f6 KH |
4997 | Fput (Qcall_process, Qtarget_idx, make_number (0)); |
4998 | ||
4999 | Qcall_process_region = intern ("call-process-region"); | |
5000 | staticpro (&Qcall_process_region); | |
9ce27fde | 5001 | /* Target PROGRAM is the third argument. */ |
e0e989f6 KH |
5002 | Fput (Qcall_process_region, Qtarget_idx, make_number (2)); |
5003 | ||
5004 | Qstart_process = intern ("start-process"); | |
5005 | staticpro (&Qstart_process); | |
9ce27fde | 5006 | /* Target PROGRAM is the third argument. */ |
e0e989f6 KH |
5007 | Fput (Qstart_process, Qtarget_idx, make_number (2)); |
5008 | ||
5009 | Qopen_network_stream = intern ("open-network-stream"); | |
5010 | staticpro (&Qopen_network_stream); | |
9ce27fde | 5011 | /* Target SERVICE is the fourth argument. */ |
e0e989f6 KH |
5012 | Fput (Qopen_network_stream, Qtarget_idx, make_number (3)); |
5013 | ||
4ed46869 KH |
5014 | Qcoding_system = intern ("coding-system"); |
5015 | staticpro (&Qcoding_system); | |
5016 | ||
5017 | Qeol_type = intern ("eol-type"); | |
5018 | staticpro (&Qeol_type); | |
5019 | ||
5020 | Qbuffer_file_coding_system = intern ("buffer-file-coding-system"); | |
5021 | staticpro (&Qbuffer_file_coding_system); | |
5022 | ||
5023 | Qpost_read_conversion = intern ("post-read-conversion"); | |
5024 | staticpro (&Qpost_read_conversion); | |
5025 | ||
5026 | Qpre_write_conversion = intern ("pre-write-conversion"); | |
5027 | staticpro (&Qpre_write_conversion); | |
5028 | ||
27901516 KH |
5029 | Qno_conversion = intern ("no-conversion"); |
5030 | staticpro (&Qno_conversion); | |
5031 | ||
5032 | Qundecided = intern ("undecided"); | |
5033 | staticpro (&Qundecided); | |
5034 | ||
4ed46869 KH |
5035 | Qcoding_system_p = intern ("coding-system-p"); |
5036 | staticpro (&Qcoding_system_p); | |
5037 | ||
5038 | Qcoding_system_error = intern ("coding-system-error"); | |
5039 | staticpro (&Qcoding_system_error); | |
5040 | ||
5041 | Fput (Qcoding_system_error, Qerror_conditions, | |
5042 | Fcons (Qcoding_system_error, Fcons (Qerror, Qnil))); | |
5043 | Fput (Qcoding_system_error, Qerror_message, | |
9ce27fde | 5044 | build_string ("Invalid coding system")); |
4ed46869 | 5045 | |
d46c5b12 KH |
5046 | Qcoding_category = intern ("coding-category"); |
5047 | staticpro (&Qcoding_category); | |
4ed46869 KH |
5048 | Qcoding_category_index = intern ("coding-category-index"); |
5049 | staticpro (&Qcoding_category_index); | |
5050 | ||
d46c5b12 KH |
5051 | Vcoding_category_table |
5052 | = Fmake_vector (make_number (CODING_CATEGORY_IDX_MAX), Qnil); | |
5053 | staticpro (&Vcoding_category_table); | |
4ed46869 KH |
5054 | { |
5055 | int i; | |
5056 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) | |
5057 | { | |
d46c5b12 KH |
5058 | XVECTOR (Vcoding_category_table)->contents[i] |
5059 | = intern (coding_category_name[i]); | |
5060 | Fput (XVECTOR (Vcoding_category_table)->contents[i], | |
5061 | Qcoding_category_index, make_number (i)); | |
4ed46869 KH |
5062 | } |
5063 | } | |
5064 | ||
bdd9fb48 KH |
5065 | Qcharacter_unification_table = intern ("character-unification-table"); |
5066 | staticpro (&Qcharacter_unification_table); | |
5067 | Fput (Qcharacter_unification_table, Qchar_table_extra_slots, | |
5068 | make_number (0)); | |
5069 | ||
a5d301df KH |
5070 | Qcharacter_unification_table_for_decode |
5071 | = intern ("character-unification-table-for-decode"); | |
5072 | staticpro (&Qcharacter_unification_table_for_decode); | |
5073 | ||
5074 | Qcharacter_unification_table_for_encode | |
5075 | = intern ("character-unification-table-for-encode"); | |
5076 | staticpro (&Qcharacter_unification_table_for_encode); | |
5077 | ||
70c22245 KH |
5078 | Qsafe_charsets = intern ("safe-charsets"); |
5079 | staticpro (&Qsafe_charsets); | |
5080 | ||
9ce27fde KH |
5081 | Qemacs_mule = intern ("emacs-mule"); |
5082 | staticpro (&Qemacs_mule); | |
5083 | ||
d46c5b12 KH |
5084 | Qraw_text = intern ("raw-text"); |
5085 | staticpro (&Qraw_text); | |
5086 | ||
4ed46869 KH |
5087 | defsubr (&Scoding_system_p); |
5088 | defsubr (&Sread_coding_system); | |
5089 | defsubr (&Sread_non_nil_coding_system); | |
5090 | defsubr (&Scheck_coding_system); | |
5091 | defsubr (&Sdetect_coding_region); | |
d46c5b12 | 5092 | defsubr (&Sdetect_coding_string); |
4ed46869 KH |
5093 | defsubr (&Sdecode_coding_region); |
5094 | defsubr (&Sencode_coding_region); | |
5095 | defsubr (&Sdecode_coding_string); | |
5096 | defsubr (&Sencode_coding_string); | |
5097 | defsubr (&Sdecode_sjis_char); | |
5098 | defsubr (&Sencode_sjis_char); | |
5099 | defsubr (&Sdecode_big5_char); | |
5100 | defsubr (&Sencode_big5_char); | |
1ba9e4ab | 5101 | defsubr (&Sset_terminal_coding_system_internal); |
c4825358 | 5102 | defsubr (&Sset_safe_terminal_coding_system_internal); |
4ed46869 | 5103 | defsubr (&Sterminal_coding_system); |
1ba9e4ab | 5104 | defsubr (&Sset_keyboard_coding_system_internal); |
4ed46869 | 5105 | defsubr (&Skeyboard_coding_system); |
a5d301df | 5106 | defsubr (&Sfind_operation_coding_system); |
d46c5b12 | 5107 | defsubr (&Supdate_iso_coding_systems); |
4ed46869 | 5108 | |
4608c386 KH |
5109 | DEFVAR_LISP ("coding-system-list", &Vcoding_system_list, |
5110 | "List of coding systems.\n\ | |
5111 | \n\ | |
5112 | Do not alter the value of this variable manually. This variable should be\n\ | |
5113 | updated by the functions `make-coding-system' and\n\ | |
5114 | `define-coding-system-alias'."); | |
5115 | Vcoding_system_list = Qnil; | |
5116 | ||
5117 | DEFVAR_LISP ("coding-system-alist", &Vcoding_system_alist, | |
5118 | "Alist of coding system names.\n\ | |
5119 | Each element is one element list of coding system name.\n\ | |
5120 | This variable is given to `completing-read' as TABLE argument.\n\ | |
5121 | \n\ | |
5122 | Do not alter the value of this variable manually. This variable should be\n\ | |
5123 | updated by the functions `make-coding-system' and\n\ | |
5124 | `define-coding-system-alias'."); | |
5125 | Vcoding_system_alist = Qnil; | |
5126 | ||
4ed46869 KH |
5127 | DEFVAR_LISP ("coding-category-list", &Vcoding_category_list, |
5128 | "List of coding-categories (symbols) ordered by priority."); | |
5129 | { | |
5130 | int i; | |
5131 | ||
5132 | Vcoding_category_list = Qnil; | |
5133 | for (i = CODING_CATEGORY_IDX_MAX - 1; i >= 0; i--) | |
5134 | Vcoding_category_list | |
d46c5b12 KH |
5135 | = Fcons (XVECTOR (Vcoding_category_table)->contents[i], |
5136 | Vcoding_category_list); | |
4ed46869 KH |
5137 | } |
5138 | ||
5139 | DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read, | |
10bff6f1 | 5140 | "Specify the coding system for read operations.\n\ |
2ebb362d | 5141 | It is useful to bind this variable with `let', but do not set it globally.\n\ |
4ed46869 | 5142 | If the value is a coding system, it is used for decoding on read operation.\n\ |
a67a9c66 | 5143 | If not, an appropriate element is used from one of the coding system alists:\n\ |
10bff6f1 | 5144 | There are three such tables, `file-coding-system-alist',\n\ |
a67a9c66 | 5145 | `process-coding-system-alist', and `network-coding-system-alist'."); |
4ed46869 KH |
5146 | Vcoding_system_for_read = Qnil; |
5147 | ||
5148 | DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write, | |
10bff6f1 | 5149 | "Specify the coding system for write operations.\n\ |
2ebb362d | 5150 | It is useful to bind this variable with `let', but do not set it globally.\n\ |
4ed46869 | 5151 | If the value is a coding system, it is used for encoding on write operation.\n\ |
a67a9c66 | 5152 | If not, an appropriate element is used from one of the coding system alists:\n\ |
10bff6f1 | 5153 | There are three such tables, `file-coding-system-alist',\n\ |
a67a9c66 | 5154 | `process-coding-system-alist', and `network-coding-system-alist'."); |
4ed46869 KH |
5155 | Vcoding_system_for_write = Qnil; |
5156 | ||
5157 | DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used, | |
a67a9c66 | 5158 | "Coding system used in the latest file or process I/O."); |
4ed46869 KH |
5159 | Vlast_coding_system_used = Qnil; |
5160 | ||
9ce27fde KH |
5161 | DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion, |
5162 | "*Non-nil inhibit code conversion of end-of-line format in any cases."); | |
5163 | inhibit_eol_conversion = 0; | |
5164 | ||
02ba4723 KH |
5165 | DEFVAR_LISP ("file-coding-system-alist", &Vfile_coding_system_alist, |
5166 | "Alist to decide a coding system to use for a file I/O operation.\n\ | |
5167 | The format is ((PATTERN . VAL) ...),\n\ | |
5168 | where PATTERN is a regular expression matching a file name,\n\ | |
5169 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
5170 | If VAL is a coding system, it is used for both decoding and encoding\n\ | |
5171 | the file contents.\n\ | |
5172 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
5173 | and the cdr part is used for encoding.\n\ | |
5174 | If VAL is a function symbol, the function must return a coding system\n\ | |
5175 | or a cons of coding systems which are used as above.\n\ | |
e0e989f6 | 5176 | \n\ |
9ce27fde | 5177 | See also the function `find-operation-coding-system'."); |
02ba4723 KH |
5178 | Vfile_coding_system_alist = Qnil; |
5179 | ||
5180 | DEFVAR_LISP ("process-coding-system-alist", &Vprocess_coding_system_alist, | |
5181 | "Alist to decide a coding system to use for a process I/O operation.\n\ | |
5182 | The format is ((PATTERN . VAL) ...),\n\ | |
5183 | where PATTERN is a regular expression matching a program name,\n\ | |
5184 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
5185 | If VAL is a coding system, it is used for both decoding what received\n\ | |
5186 | from the program and encoding what sent to the program.\n\ | |
5187 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
5188 | and the cdr part is used for encoding.\n\ | |
5189 | If VAL is a function symbol, the function must return a coding system\n\ | |
5190 | or a cons of coding systems which are used as above.\n\ | |
4ed46869 | 5191 | \n\ |
9ce27fde | 5192 | See also the function `find-operation-coding-system'."); |
02ba4723 KH |
5193 | Vprocess_coding_system_alist = Qnil; |
5194 | ||
5195 | DEFVAR_LISP ("network-coding-system-alist", &Vnetwork_coding_system_alist, | |
5196 | "Alist to decide a coding system to use for a network I/O operation.\n\ | |
5197 | The format is ((PATTERN . VAL) ...),\n\ | |
5198 | where PATTERN is a regular expression matching a network service name\n\ | |
5199 | or is a port number to connect to,\n\ | |
5200 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
5201 | If VAL is a coding system, it is used for both decoding what received\n\ | |
5202 | from the network stream and encoding what sent to the network stream.\n\ | |
5203 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
5204 | and the cdr part is used for encoding.\n\ | |
5205 | If VAL is a function symbol, the function must return a coding system\n\ | |
5206 | or a cons of coding systems which are used as above.\n\ | |
4ed46869 | 5207 | \n\ |
9ce27fde | 5208 | See also the function `find-operation-coding-system'."); |
02ba4723 | 5209 | Vnetwork_coding_system_alist = Qnil; |
4ed46869 KH |
5210 | |
5211 | DEFVAR_INT ("eol-mnemonic-unix", &eol_mnemonic_unix, | |
5212 | "Mnemonic character indicating UNIX-like end-of-line format (i.e. LF) ."); | |
458822a0 | 5213 | eol_mnemonic_unix = ':'; |
4ed46869 KH |
5214 | |
5215 | DEFVAR_INT ("eol-mnemonic-dos", &eol_mnemonic_dos, | |
5216 | "Mnemonic character indicating DOS-like end-of-line format (i.e. CRLF)."); | |
458822a0 | 5217 | eol_mnemonic_dos = '\\'; |
4ed46869 KH |
5218 | |
5219 | DEFVAR_INT ("eol-mnemonic-mac", &eol_mnemonic_mac, | |
5220 | "Mnemonic character indicating MAC-like end-of-line format (i.e. CR)."); | |
458822a0 | 5221 | eol_mnemonic_mac = '/'; |
4ed46869 KH |
5222 | |
5223 | DEFVAR_INT ("eol-mnemonic-undecided", &eol_mnemonic_undecided, | |
5224 | "Mnemonic character indicating end-of-line format is not yet decided."); | |
458822a0 | 5225 | eol_mnemonic_undecided = ':'; |
4ed46869 | 5226 | |
bdd9fb48 KH |
5227 | DEFVAR_LISP ("enable-character-unification", &Venable_character_unification, |
5228 | "Non-nil means ISO 2022 encoder/decoder do character unification."); | |
5229 | Venable_character_unification = Qt; | |
5230 | ||
a5d301df KH |
5231 | DEFVAR_LISP ("standard-character-unification-table-for-decode", |
5232 | &Vstandard_character_unification_table_for_decode, | |
bdd9fb48 | 5233 | "Table for unifying characters when reading."); |
a5d301df | 5234 | Vstandard_character_unification_table_for_decode = Qnil; |
bdd9fb48 | 5235 | |
a5d301df KH |
5236 | DEFVAR_LISP ("standard-character-unification-table-for-encode", |
5237 | &Vstandard_character_unification_table_for_encode, | |
bdd9fb48 | 5238 | "Table for unifying characters when writing."); |
a5d301df | 5239 | Vstandard_character_unification_table_for_encode = Qnil; |
4ed46869 KH |
5240 | |
5241 | DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist, | |
5242 | "Alist of charsets vs revision numbers.\n\ | |
5243 | While encoding, if a charset (car part of an element) is found,\n\ | |
5244 | designate it with the escape sequence identifing revision (cdr part of the element)."); | |
5245 | Vcharset_revision_alist = Qnil; | |
02ba4723 KH |
5246 | |
5247 | DEFVAR_LISP ("default-process-coding-system", | |
5248 | &Vdefault_process_coding_system, | |
5249 | "Cons of coding systems used for process I/O by default.\n\ | |
5250 | The car part is used for decoding a process output,\n\ | |
5251 | the cdr part is used for encoding a text to be sent to a process."); | |
5252 | Vdefault_process_coding_system = Qnil; | |
c4825358 | 5253 | |
3f003981 KH |
5254 | DEFVAR_LISP ("latin-extra-code-table", &Vlatin_extra_code_table, |
5255 | "Table of extra Latin codes in the range 128..159 (inclusive).\n\ | |
c4825358 KH |
5256 | This is a vector of length 256.\n\ |
5257 | If Nth element is non-nil, the existence of code N in a file\n\ | |
bb0115a2 | 5258 | \(or output of subprocess) doesn't prevent it to be detected as\n\ |
3f003981 KH |
5259 | a coding system of ISO 2022 variant which has a flag\n\ |
5260 | `accept-latin-extra-code' t (e.g. iso-latin-1) on reading a file\n\ | |
c4825358 KH |
5261 | or reading output of a subprocess.\n\ |
5262 | Only 128th through 159th elements has a meaning."); | |
3f003981 | 5263 | Vlatin_extra_code_table = Fmake_vector (make_number (256), Qnil); |
d46c5b12 KH |
5264 | |
5265 | DEFVAR_LISP ("select-safe-coding-system-function", | |
5266 | &Vselect_safe_coding_system_function, | |
5267 | "Function to call to select safe coding system for encoding a text.\n\ | |
5268 | \n\ | |
5269 | If set, this function is called to force a user to select a proper\n\ | |
5270 | coding system which can encode the text in the case that a default\n\ | |
5271 | coding system used in each operation can't encode the text.\n\ | |
5272 | \n\ | |
5273 | The default value is `select-safe-codign-system' (which see)."); | |
5274 | Vselect_safe_coding_system_function = Qnil; | |
5275 | ||
4ed46869 KH |
5276 | } |
5277 | ||
5278 | #endif /* emacs */ |