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 |
94487c4e | 318 | Latin codes (especially for dealing with Microsoft codes). */ |
3f003981 | 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 | ||
ed29121d EZ |
324 | /* Flag to make buffer-file-coding-system inherit from process-coding. */ |
325 | int inherit_process_coding_system; | |
326 | ||
c4825358 | 327 | /* Coding system to be used to encode text for terminal display. */ |
4ed46869 KH |
328 | struct coding_system terminal_coding; |
329 | ||
c4825358 KH |
330 | /* Coding system to be used to encode text for terminal display when |
331 | terminal coding system is nil. */ | |
332 | struct coding_system safe_terminal_coding; | |
333 | ||
334 | /* Coding system of what is sent from terminal keyboard. */ | |
4ed46869 KH |
335 | struct coding_system keyboard_coding; |
336 | ||
02ba4723 KH |
337 | Lisp_Object Vfile_coding_system_alist; |
338 | Lisp_Object Vprocess_coding_system_alist; | |
339 | Lisp_Object Vnetwork_coding_system_alist; | |
4ed46869 KH |
340 | |
341 | #endif /* emacs */ | |
342 | ||
d46c5b12 | 343 | Lisp_Object Qcoding_category, Qcoding_category_index; |
4ed46869 KH |
344 | |
345 | /* List of symbols `coding-category-xxx' ordered by priority. */ | |
346 | Lisp_Object Vcoding_category_list; | |
347 | ||
d46c5b12 KH |
348 | /* Table of coding categories (Lisp symbols). */ |
349 | Lisp_Object Vcoding_category_table; | |
4ed46869 KH |
350 | |
351 | /* Table of names of symbol for each coding-category. */ | |
352 | char *coding_category_name[CODING_CATEGORY_IDX_MAX] = { | |
0ef69138 | 353 | "coding-category-emacs-mule", |
4ed46869 KH |
354 | "coding-category-sjis", |
355 | "coding-category-iso-7", | |
d46c5b12 | 356 | "coding-category-iso-7-tight", |
4ed46869 KH |
357 | "coding-category-iso-8-1", |
358 | "coding-category-iso-8-2", | |
7717c392 KH |
359 | "coding-category-iso-7-else", |
360 | "coding-category-iso-8-else", | |
4ed46869 | 361 | "coding-category-big5", |
27901516 | 362 | "coding-category-raw-text", |
4ed46869 KH |
363 | "coding-category-binary" |
364 | }; | |
365 | ||
66cfb530 | 366 | /* Table of pointers to coding systems corresponding to each coding |
d46c5b12 KH |
367 | categories. */ |
368 | struct coding_system *coding_system_table[CODING_CATEGORY_IDX_MAX]; | |
369 | ||
66cfb530 KH |
370 | /* Table of coding category masks. Nth element is a mask for a coding |
371 | cateogry of which priority is Nth. */ | |
372 | static | |
373 | int coding_priorities[CODING_CATEGORY_IDX_MAX]; | |
374 | ||
f967223b KH |
375 | /* Flag to tell if we look up translation table on character code |
376 | conversion. */ | |
84fbb8a0 | 377 | Lisp_Object Venable_character_translation; |
f967223b KH |
378 | /* Standard translation table to look up on decoding (reading). */ |
379 | Lisp_Object Vstandard_translation_table_for_decode; | |
380 | /* Standard translation table to look up on encoding (writing). */ | |
381 | Lisp_Object Vstandard_translation_table_for_encode; | |
84fbb8a0 | 382 | |
f967223b KH |
383 | Lisp_Object Qtranslation_table; |
384 | Lisp_Object Qtranslation_table_id; | |
385 | Lisp_Object Qtranslation_table_for_decode; | |
386 | Lisp_Object Qtranslation_table_for_encode; | |
4ed46869 KH |
387 | |
388 | /* Alist of charsets vs revision number. */ | |
389 | Lisp_Object Vcharset_revision_alist; | |
390 | ||
02ba4723 KH |
391 | /* Default coding systems used for process I/O. */ |
392 | Lisp_Object Vdefault_process_coding_system; | |
393 | ||
4ed46869 | 394 | \f |
0ef69138 | 395 | /*** 2. Emacs internal format (emacs-mule) handlers ***/ |
4ed46869 KH |
396 | |
397 | /* Emacs' internal format for encoding multiple character sets is a | |
f4dee582 RS |
398 | kind of multi-byte encoding, i.e. characters are encoded by |
399 | variable-length sequences of one-byte codes. ASCII characters | |
400 | and control characters (e.g. `tab', `newline') are represented by | |
401 | one-byte sequences which are their ASCII codes, in the range 0x00 | |
402 | through 0x7F. The other characters are represented by a sequence | |
403 | of `base leading-code', optional `extended leading-code', and one | |
404 | or two `position-code's. The length of the sequence is determined | |
405 | by the base leading-code. Leading-code takes the range 0x80 | |
406 | through 0x9F, whereas extended leading-code and position-code take | |
407 | the range 0xA0 through 0xFF. See `charset.h' for more details | |
408 | about leading-code and position-code. | |
409 | ||
410 | There's one exception to this rule. Special leading-code | |
4ed46869 KH |
411 | `leading-code-composition' denotes that the following several |
412 | characters should be composed into one character. Leading-codes of | |
413 | components (except for ASCII) are added 0x20. An ASCII character | |
414 | component is represented by a 2-byte sequence of `0xA0' and | |
f4dee582 RS |
415 | `ASCII-code + 0x80'. See also the comments in `charset.h' for the |
416 | details of composite character. Hence, we can summarize the code | |
4ed46869 KH |
417 | range as follows: |
418 | ||
419 | --- CODE RANGE of Emacs' internal format --- | |
420 | (character set) (range) | |
421 | ASCII 0x00 .. 0x7F | |
422 | ELSE (1st byte) 0x80 .. 0x9F | |
423 | (rest bytes) 0xA0 .. 0xFF | |
424 | --------------------------------------------- | |
425 | ||
426 | */ | |
427 | ||
428 | enum emacs_code_class_type emacs_code_class[256]; | |
429 | ||
430 | /* Go to the next statement only if *SRC is accessible and the code is | |
431 | greater than 0xA0. */ | |
432 | #define CHECK_CODE_RANGE_A0_FF \ | |
433 | do { \ | |
434 | if (src >= src_end) \ | |
435 | goto label_end_of_switch; \ | |
436 | else if (*src++ < 0xA0) \ | |
437 | return 0; \ | |
438 | } while (0) | |
439 | ||
440 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". | |
441 | Check if a text is encoded in Emacs' internal format. If it is, | |
d46c5b12 | 442 | return CODING_CATEGORY_MASK_EMACS_MULE, else return 0. */ |
4ed46869 KH |
443 | |
444 | int | |
0ef69138 | 445 | detect_coding_emacs_mule (src, src_end) |
4ed46869 KH |
446 | unsigned char *src, *src_end; |
447 | { | |
448 | unsigned char c; | |
449 | int composing = 0; | |
450 | ||
451 | while (src < src_end) | |
452 | { | |
453 | c = *src++; | |
454 | ||
455 | if (composing) | |
456 | { | |
457 | if (c < 0xA0) | |
458 | composing = 0; | |
459 | else | |
460 | c -= 0x20; | |
461 | } | |
462 | ||
463 | switch (emacs_code_class[c]) | |
464 | { | |
465 | case EMACS_ascii_code: | |
466 | case EMACS_linefeed_code: | |
467 | break; | |
468 | ||
469 | case EMACS_control_code: | |
470 | if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) | |
471 | return 0; | |
472 | break; | |
473 | ||
474 | case EMACS_invalid_code: | |
475 | return 0; | |
476 | ||
477 | case EMACS_leading_code_composition: /* c == 0x80 */ | |
478 | if (composing) | |
479 | CHECK_CODE_RANGE_A0_FF; | |
480 | else | |
481 | composing = 1; | |
482 | break; | |
483 | ||
484 | case EMACS_leading_code_4: | |
485 | CHECK_CODE_RANGE_A0_FF; | |
486 | /* fall down to check it two more times ... */ | |
487 | ||
488 | case EMACS_leading_code_3: | |
489 | CHECK_CODE_RANGE_A0_FF; | |
490 | /* fall down to check it one more time ... */ | |
491 | ||
492 | case EMACS_leading_code_2: | |
493 | CHECK_CODE_RANGE_A0_FF; | |
494 | break; | |
495 | ||
496 | default: | |
497 | label_end_of_switch: | |
498 | break; | |
499 | } | |
500 | } | |
0ef69138 | 501 | return CODING_CATEGORY_MASK_EMACS_MULE; |
4ed46869 KH |
502 | } |
503 | ||
504 | \f | |
505 | /*** 3. ISO2022 handlers ***/ | |
506 | ||
507 | /* The following note describes the coding system ISO2022 briefly. | |
f4dee582 RS |
508 | Since the intention of this note is to help in understanding of |
509 | the programs in this file, some parts are NOT ACCURATE or OVERLY | |
4ed46869 KH |
510 | SIMPLIFIED. For the thorough understanding, please refer to the |
511 | original document of ISO2022. | |
512 | ||
513 | ISO2022 provides many mechanisms to encode several character sets | |
f4dee582 | 514 | in 7-bit and 8-bit environment. If one chooses 7-bite environment, |
4ed46869 | 515 | all text is encoded by codes of less than 128. This may make the |
f4dee582 RS |
516 | encoded text a little bit longer, but the text gets more stability |
517 | to pass through several gateways (some of them strip off the MSB). | |
4ed46869 | 518 | |
f4dee582 | 519 | There are two kinds of character set: control character set and |
4ed46869 KH |
520 | graphic character set. The former contains control characters such |
521 | as `newline' and `escape' to provide control functions (control | |
f4dee582 | 522 | functions are provided also by escape sequences). The latter |
4ed46869 KH |
523 | contains graphic characters such as ' A' and '-'. Emacs recognizes |
524 | two control character sets and many graphic character sets. | |
525 | ||
526 | Graphic character sets are classified into one of the following | |
527 | four classes, DIMENSION1_CHARS94, DIMENSION1_CHARS96, | |
528 | DIMENSION2_CHARS94, DIMENSION2_CHARS96 according to the number of | |
529 | bytes (DIMENSION) and the number of characters in one dimension | |
530 | (CHARS) of the set. In addition, each character set is assigned an | |
531 | identification tag (called "final character" and denoted as <F> | |
532 | here after) which is unique in each class. <F> of each character | |
533 | set is decided by ECMA(*) when it is registered in ISO. Code range | |
534 | of <F> is 0x30..0x7F (0x30..0x3F are for private use only). | |
535 | ||
536 | Note (*): ECMA = European Computer Manufacturers Association | |
537 | ||
538 | Here are examples of graphic character set [NAME(<F>)]: | |
539 | o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ... | |
540 | o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ... | |
541 | o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ... | |
542 | o DIMENSION2_CHARS96 -- none for the moment | |
543 | ||
544 | A code area (1byte=8bits) is divided into 4 areas, C0, GL, C1, and GR. | |
545 | C0 [0x00..0x1F] -- control character plane 0 | |
546 | GL [0x20..0x7F] -- graphic character plane 0 | |
547 | C1 [0x80..0x9F] -- control character plane 1 | |
548 | GR [0xA0..0xFF] -- graphic character plane 1 | |
549 | ||
550 | A control character set is directly designated and invoked to C0 or | |
551 | C1 by an escape sequence. The most common case is that ISO646's | |
552 | control character set is designated/invoked to C0 and ISO6429's | |
553 | control character set is designated/invoked to C1, and usually | |
554 | these designations/invocations are omitted in a coded text. With | |
555 | 7-bit environment, only C0 can be used, and a control character for | |
556 | C1 is encoded by an appropriate escape sequence to fit in the | |
557 | environment. All control characters for C1 are defined the | |
558 | corresponding escape sequences. | |
559 | ||
560 | A graphic character set is at first designated to one of four | |
561 | graphic registers (G0 through G3), then these graphic registers are | |
562 | invoked to GL or GR. These designations and invocations can be | |
563 | done independently. The most common case is that G0 is invoked to | |
564 | GL, G1 is invoked to GR, and ASCII is designated to G0, and usually | |
565 | these invocations and designations are omitted in a coded text. | |
566 | With 7-bit environment, only GL can be used. | |
567 | ||
568 | When a graphic character set of CHARS94 is invoked to GL, code 0x20 | |
569 | and 0x7F of GL area work as control characters SPACE and DEL | |
570 | respectively, and code 0xA0 and 0xFF of GR area should not be used. | |
571 | ||
572 | There are two ways of invocation: locking-shift and single-shift. | |
573 | With locking-shift, the invocation lasts until the next different | |
574 | invocation, whereas with single-shift, the invocation works only | |
575 | for the following character and doesn't affect locking-shift. | |
576 | Invocations are done by the following control characters or escape | |
577 | sequences. | |
578 | ||
579 | ---------------------------------------------------------------------- | |
580 | function control char escape sequence description | |
581 | ---------------------------------------------------------------------- | |
582 | SI (shift-in) 0x0F none invoke G0 to GL | |
10bff6f1 | 583 | SO (shift-out) 0x0E none invoke G1 to GL |
4ed46869 KH |
584 | LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL |
585 | LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL | |
586 | SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 into GL | |
587 | SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 into GL | |
588 | ---------------------------------------------------------------------- | |
589 | The first four are for locking-shift. Control characters for these | |
590 | functions are defined by macros ISO_CODE_XXX in `coding.h'. | |
591 | ||
592 | Designations are done by the following escape sequences. | |
593 | ---------------------------------------------------------------------- | |
594 | escape sequence description | |
595 | ---------------------------------------------------------------------- | |
596 | ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0 | |
597 | ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1 | |
598 | ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2 | |
599 | ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3 | |
600 | ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*) | |
601 | ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1 | |
602 | ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2 | |
603 | ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3 | |
604 | ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**) | |
605 | ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1 | |
606 | ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2 | |
607 | ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3 | |
608 | ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*) | |
609 | ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1 | |
610 | ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2 | |
611 | ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3 | |
612 | ---------------------------------------------------------------------- | |
613 | ||
614 | In this list, "DIMENSION1_CHARS94<F>" means a graphic character set | |
615 | of dimension 1, chars 94, and final character <F>, and etc. | |
616 | ||
617 | Note (*): Although these designations are not allowed in ISO2022, | |
618 | Emacs accepts them on decoding, and produces them on encoding | |
619 | CHARS96 character set in a coding system which is characterized as | |
620 | 7-bit environment, non-locking-shift, and non-single-shift. | |
621 | ||
622 | Note (**): If <F> is '@', 'A', or 'B', the intermediate character | |
623 | '(' can be omitted. We call this as "short-form" here after. | |
624 | ||
625 | Now you may notice that there are a lot of ways for encoding the | |
f4dee582 | 626 | same multilingual text in ISO2022. Actually, there exists many |
4ed46869 KH |
627 | coding systems such as Compound Text (used in X's inter client |
628 | communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR | |
629 | (used in Korean Internet), EUC (Extended UNIX Code, used in Asian | |
630 | localized platforms), and all of these are variants of ISO2022. | |
631 | ||
632 | In addition to the above, Emacs handles two more kinds of escape | |
633 | sequences: ISO6429's direction specification and Emacs' private | |
634 | sequence for specifying character composition. | |
635 | ||
636 | ISO6429's direction specification takes the following format: | |
637 | o CSI ']' -- end of the current direction | |
638 | o CSI '0' ']' -- end of the current direction | |
639 | o CSI '1' ']' -- start of left-to-right text | |
640 | o CSI '2' ']' -- start of right-to-left text | |
641 | The control character CSI (0x9B: control sequence introducer) is | |
642 | abbreviated to the escape sequence ESC '[' in 7-bit environment. | |
643 | ||
644 | Character composition specification takes the following format: | |
645 | o ESC '0' -- start character composition | |
646 | o ESC '1' -- end character composition | |
647 | Since these are not standard escape sequences of any ISO, the use | |
648 | of them for these meaning is restricted to Emacs only. */ | |
649 | ||
650 | enum iso_code_class_type iso_code_class[256]; | |
651 | ||
704c5781 KH |
652 | #define CHARSET_OK(idx, charset) \ |
653 | (coding_system_table[idx]->safe_charsets[charset] \ | |
654 | || (CODING_SPEC_ISO_REQUESTED_DESIGNATION \ | |
655 | (coding_system_table[idx], charset) \ | |
656 | != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)) | |
d46c5b12 KH |
657 | |
658 | #define SHIFT_OUT_OK(idx) \ | |
659 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding_system_table[idx], 1) >= 0) | |
660 | ||
4ed46869 KH |
661 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
662 | Check if a text is encoded in ISO2022. If it is, returns an | |
663 | integer in which appropriate flag bits any of: | |
664 | CODING_CATEGORY_MASK_ISO_7 | |
d46c5b12 | 665 | CODING_CATEGORY_MASK_ISO_7_TIGHT |
4ed46869 KH |
666 | CODING_CATEGORY_MASK_ISO_8_1 |
667 | CODING_CATEGORY_MASK_ISO_8_2 | |
7717c392 KH |
668 | CODING_CATEGORY_MASK_ISO_7_ELSE |
669 | CODING_CATEGORY_MASK_ISO_8_ELSE | |
4ed46869 KH |
670 | are set. If a code which should never appear in ISO2022 is found, |
671 | returns 0. */ | |
672 | ||
673 | int | |
674 | detect_coding_iso2022 (src, src_end) | |
675 | unsigned char *src, *src_end; | |
676 | { | |
d46c5b12 KH |
677 | int mask = CODING_CATEGORY_MASK_ISO; |
678 | int mask_found = 0; | |
679 | int reg[4], shift_out = 0; | |
680 | int c, c1, i, charset; | |
3f003981 | 681 | |
d46c5b12 | 682 | reg[0] = CHARSET_ASCII, reg[1] = reg[2] = reg[3] = -1; |
3f003981 | 683 | while (mask && src < src_end) |
4ed46869 KH |
684 | { |
685 | c = *src++; | |
686 | switch (c) | |
687 | { | |
688 | case ISO_CODE_ESC: | |
e0e989f6 | 689 | if (src >= src_end) |
4ed46869 KH |
690 | break; |
691 | c = *src++; | |
d46c5b12 | 692 | if (c >= '(' && c <= '/') |
4ed46869 | 693 | { |
bf9cdd4e KH |
694 | /* Designation sequence for a charset of dimension 1. */ |
695 | if (src >= src_end) | |
696 | break; | |
d46c5b12 KH |
697 | c1 = *src++; |
698 | if (c1 < ' ' || c1 >= 0x80 | |
699 | || (charset = iso_charset_table[0][c >= ','][c1]) < 0) | |
700 | /* Invalid designation sequence. Just ignore. */ | |
701 | break; | |
702 | reg[(c - '(') % 4] = charset; | |
bf9cdd4e KH |
703 | } |
704 | else if (c == '$') | |
705 | { | |
706 | /* Designation sequence for a charset of dimension 2. */ | |
707 | if (src >= src_end) | |
708 | break; | |
709 | c = *src++; | |
710 | if (c >= '@' && c <= 'B') | |
711 | /* Designation for JISX0208.1978, GB2312, or JISX0208. */ | |
d46c5b12 | 712 | reg[0] = charset = iso_charset_table[1][0][c]; |
bf9cdd4e | 713 | else if (c >= '(' && c <= '/') |
bcf26d6a | 714 | { |
bf9cdd4e KH |
715 | if (src >= src_end) |
716 | break; | |
d46c5b12 KH |
717 | c1 = *src++; |
718 | if (c1 < ' ' || c1 >= 0x80 | |
719 | || (charset = iso_charset_table[1][c >= ','][c1]) < 0) | |
720 | /* Invalid designation sequence. Just ignore. */ | |
721 | break; | |
722 | reg[(c - '(') % 4] = charset; | |
bcf26d6a | 723 | } |
bf9cdd4e | 724 | else |
d46c5b12 KH |
725 | /* Invalid designation sequence. Just ignore. */ |
726 | break; | |
727 | } | |
728 | else if (c == 'N' || c == 'n') | |
729 | { | |
730 | if (shift_out == 0 | |
731 | && (reg[1] >= 0 | |
732 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE) | |
733 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))) | |
734 | { | |
735 | /* Locking shift out. */ | |
736 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
737 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
738 | shift_out = 1; | |
739 | } | |
740 | break; | |
741 | } | |
742 | else if (c == 'O' || c == 'o') | |
743 | { | |
744 | if (shift_out == 1) | |
745 | { | |
746 | /* Locking shift in. */ | |
747 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
748 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
749 | shift_out = 0; | |
750 | } | |
751 | break; | |
4ed46869 | 752 | } |
bf9cdd4e | 753 | else if (c == '0' || c == '1' || c == '2') |
d46c5b12 KH |
754 | /* Start/end composition. Just ignore. */ |
755 | break; | |
bf9cdd4e | 756 | else |
d46c5b12 KH |
757 | /* Invalid escape sequence. Just ignore. */ |
758 | break; | |
759 | ||
760 | /* We found a valid designation sequence for CHARSET. */ | |
761 | mask &= ~CODING_CATEGORY_MASK_ISO_8BIT; | |
762 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7, charset)) | |
763 | mask_found |= CODING_CATEGORY_MASK_ISO_7; | |
764 | else | |
765 | mask &= ~CODING_CATEGORY_MASK_ISO_7; | |
766 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT, charset)) | |
767 | mask_found |= CODING_CATEGORY_MASK_ISO_7_TIGHT; | |
768 | else | |
769 | mask &= ~CODING_CATEGORY_MASK_ISO_7_TIGHT; | |
770 | if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE, charset)) | |
771 | mask &= ~CODING_CATEGORY_MASK_ISO_7_ELSE; | |
772 | if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE, charset)) | |
773 | mask &= ~CODING_CATEGORY_MASK_ISO_8_ELSE; | |
4ed46869 KH |
774 | break; |
775 | ||
4ed46869 | 776 | case ISO_CODE_SO: |
d46c5b12 KH |
777 | if (shift_out == 0 |
778 | && (reg[1] >= 0 | |
779 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE) | |
780 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))) | |
781 | { | |
782 | /* Locking shift out. */ | |
783 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
784 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
785 | } | |
e0e989f6 KH |
786 | break; |
787 | ||
d46c5b12 KH |
788 | case ISO_CODE_SI: |
789 | if (shift_out == 1) | |
790 | { | |
791 | /* Locking shift in. */ | |
792 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; | |
793 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; | |
794 | } | |
795 | break; | |
796 | ||
4ed46869 KH |
797 | case ISO_CODE_CSI: |
798 | case ISO_CODE_SS2: | |
799 | case ISO_CODE_SS3: | |
3f003981 KH |
800 | { |
801 | int newmask = CODING_CATEGORY_MASK_ISO_8_ELSE; | |
802 | ||
70c22245 KH |
803 | if (c != ISO_CODE_CSI) |
804 | { | |
d46c5b12 KH |
805 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
806 | & CODING_FLAG_ISO_SINGLE_SHIFT) | |
70c22245 | 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_SINGLE_SHIFT) | |
70c22245 KH |
810 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
811 | } | |
3f003981 KH |
812 | if (VECTORP (Vlatin_extra_code_table) |
813 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) | |
814 | { | |
d46c5b12 KH |
815 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
816 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 | 817 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
d46c5b12 KH |
818 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
819 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 KH |
820 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
821 | } | |
822 | mask &= newmask; | |
d46c5b12 | 823 | mask_found |= newmask; |
3f003981 KH |
824 | } |
825 | break; | |
4ed46869 KH |
826 | |
827 | default: | |
828 | if (c < 0x80) | |
829 | break; | |
830 | else if (c < 0xA0) | |
c4825358 | 831 | { |
3f003981 KH |
832 | if (VECTORP (Vlatin_extra_code_table) |
833 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) | |
c4825358 | 834 | { |
3f003981 KH |
835 | int newmask = 0; |
836 | ||
d46c5b12 KH |
837 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
838 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 | 839 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
d46c5b12 KH |
840 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
841 | & CODING_FLAG_ISO_LATIN_EXTRA) | |
3f003981 KH |
842 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
843 | mask &= newmask; | |
d46c5b12 | 844 | mask_found |= newmask; |
c4825358 | 845 | } |
3f003981 KH |
846 | else |
847 | return 0; | |
c4825358 | 848 | } |
4ed46869 KH |
849 | else |
850 | { | |
7717c392 | 851 | unsigned char *src_begin = src; |
4ed46869 | 852 | |
d46c5b12 | 853 | mask &= ~(CODING_CATEGORY_MASK_ISO_7BIT |
7717c392 | 854 | | CODING_CATEGORY_MASK_ISO_7_ELSE); |
d46c5b12 | 855 | mask_found |= CODING_CATEGORY_MASK_ISO_8_1; |
e0e989f6 | 856 | while (src < src_end && *src >= 0xA0) |
7717c392 KH |
857 | src++; |
858 | if ((src - src_begin - 1) & 1 && src < src_end) | |
4ed46869 | 859 | mask &= ~CODING_CATEGORY_MASK_ISO_8_2; |
d46c5b12 KH |
860 | else |
861 | mask_found |= CODING_CATEGORY_MASK_ISO_8_2; | |
4ed46869 KH |
862 | } |
863 | break; | |
864 | } | |
865 | } | |
866 | ||
d46c5b12 | 867 | return (mask & mask_found); |
4ed46869 KH |
868 | } |
869 | ||
870 | /* Decode a character of which charset is CHARSET and the 1st position | |
bdd9fb48 | 871 | code is C1. If dimension of CHARSET is 2, the 2nd position code is |
4ed46869 KH |
872 | fetched from SRC and set to C2. If CHARSET is negative, it means |
873 | that we are decoding ill formed text, and what we can do is just to | |
874 | read C1 as is. */ | |
875 | ||
bdd9fb48 KH |
876 | #define DECODE_ISO_CHARACTER(charset, c1) \ |
877 | do { \ | |
878 | int c_alt, charset_alt = (charset); \ | |
879 | if (COMPOSING_HEAD_P (coding->composing)) \ | |
880 | { \ | |
881 | *dst++ = LEADING_CODE_COMPOSITION; \ | |
882 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ | |
883 | /* To tell composition rules are embeded. */ \ | |
884 | *dst++ = 0xFF; \ | |
885 | coding->composing += 2; \ | |
886 | } \ | |
85bbb134 | 887 | if (charset_alt >= 0) \ |
bdd9fb48 | 888 | { \ |
85bbb134 | 889 | if (CHARSET_DIMENSION (charset_alt) == 2) \ |
70c22245 KH |
890 | { \ |
891 | ONE_MORE_BYTE (c2); \ | |
892 | if (iso_code_class[(c2) & 0x7F] != ISO_0x20_or_0x7F \ | |
893 | && iso_code_class[(c2) & 0x7F] != ISO_graphic_plane_0) \ | |
894 | { \ | |
895 | src--; \ | |
85bbb134 | 896 | charset_alt = CHARSET_ASCII; \ |
70c22245 KH |
897 | } \ |
898 | } \ | |
84fbb8a0 KH |
899 | if (!NILP (translation_table) \ |
900 | && ((c_alt = translate_char (translation_table, \ | |
85bbb134 | 901 | -1, charset_alt, c1, c2)) >= 0)) \ |
bdd9fb48 KH |
902 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
903 | } \ | |
904 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ | |
905 | DECODE_CHARACTER_ASCII (c1); \ | |
906 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
907 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
908 | else \ | |
909 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
910 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ | |
911 | /* To tell a composition rule follows. */ \ | |
912 | coding->composing = COMPOSING_WITH_RULE_RULE; \ | |
4ed46869 KH |
913 | } while (0) |
914 | ||
915 | /* Set designation state into CODING. */ | |
d46c5b12 KH |
916 | #define DECODE_DESIGNATION(reg, dimension, chars, final_char) \ |
917 | do { \ | |
918 | int charset = ISO_CHARSET_TABLE (make_number (dimension), \ | |
919 | make_number (chars), \ | |
920 | make_number (final_char)); \ | |
921 | if (charset >= 0 \ | |
704c5781 KH |
922 | && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) == reg \ |
923 | || coding->safe_charsets[charset])) \ | |
d46c5b12 KH |
924 | { \ |
925 | if (coding->spec.iso2022.last_invalid_designation_register == 0 \ | |
926 | && reg == 0 \ | |
927 | && charset == CHARSET_ASCII) \ | |
928 | { \ | |
929 | /* We should insert this designation sequence as is so \ | |
930 | that it is surely written back to a file. */ \ | |
931 | coding->spec.iso2022.last_invalid_designation_register = -1; \ | |
932 | goto label_invalid_code; \ | |
933 | } \ | |
934 | coding->spec.iso2022.last_invalid_designation_register = -1; \ | |
935 | if ((coding->mode & CODING_MODE_DIRECTION) \ | |
936 | && CHARSET_REVERSE_CHARSET (charset) >= 0) \ | |
937 | charset = CHARSET_REVERSE_CHARSET (charset); \ | |
938 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ | |
939 | } \ | |
940 | else \ | |
941 | { \ | |
942 | coding->spec.iso2022.last_invalid_designation_register = reg; \ | |
943 | goto label_invalid_code; \ | |
944 | } \ | |
4ed46869 KH |
945 | } while (0) |
946 | ||
d46c5b12 KH |
947 | /* Check if the current composing sequence contains only valid codes. |
948 | If the composing sequence doesn't end before SRC_END, return -1. | |
949 | Else, if it contains only valid codes, return 0. | |
950 | Else return the length of the composing sequence. */ | |
951 | ||
84fbb8a0 KH |
952 | int |
953 | check_composing_code (coding, src, src_end) | |
d46c5b12 KH |
954 | struct coding_system *coding; |
955 | unsigned char *src, *src_end; | |
956 | { | |
957 | unsigned char *src_start = src; | |
958 | int invalid_code_found = 0; | |
959 | int charset, c, c1, dim; | |
960 | ||
961 | while (src < src_end) | |
962 | { | |
963 | if (*src++ != ISO_CODE_ESC) continue; | |
964 | if (src >= src_end) break; | |
965 | if ((c = *src++) == '1') /* end of compsition */ | |
966 | return (invalid_code_found ? src - src_start : 0); | |
967 | if (src + 2 >= src_end) break; | |
968 | if (!coding->flags & CODING_FLAG_ISO_DESIGNATION) | |
969 | invalid_code_found = 1; | |
970 | else | |
971 | { | |
972 | dim = 0; | |
973 | if (c == '$') | |
974 | { | |
975 | dim = 1; | |
976 | c = (*src >= '@' && *src <= 'B') ? '(' : *src++; | |
977 | } | |
978 | if (c >= '(' && c <= '/') | |
979 | { | |
980 | c1 = *src++; | |
981 | if ((c1 < ' ' || c1 >= 0x80) | |
982 | || (charset = iso_charset_table[dim][c >= ','][c1]) < 0 | |
704c5781 | 983 | || ! coding->safe_charsets[charset] |
d46c5b12 KH |
984 | || (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
985 | == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)) | |
986 | invalid_code_found = 1; | |
987 | } | |
988 | else | |
989 | invalid_code_found = 1; | |
990 | } | |
991 | } | |
84fbb8a0 KH |
992 | return (invalid_code_found |
993 | ? src - src_start | |
994 | : (coding->mode & CODING_MODE_LAST_BLOCK ? 0 : -1)); | |
d46c5b12 KH |
995 | } |
996 | ||
4ed46869 KH |
997 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */ |
998 | ||
999 | int | |
d46c5b12 | 1000 | decode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
1001 | struct coding_system *coding; |
1002 | unsigned char *source, *destination; | |
1003 | int src_bytes, dst_bytes; | |
4ed46869 KH |
1004 | { |
1005 | unsigned char *src = source; | |
1006 | unsigned char *src_end = source + src_bytes; | |
1007 | unsigned char *dst = destination; | |
1008 | unsigned char *dst_end = destination + dst_bytes; | |
1009 | /* Since the maximum bytes produced by each loop is 7, we subtract 6 | |
1010 | from DST_END to assure that overflow checking is necessary only | |
1011 | at the head of loop. */ | |
1012 | unsigned char *adjusted_dst_end = dst_end - 6; | |
1013 | int charset; | |
1014 | /* Charsets invoked to graphic plane 0 and 1 respectively. */ | |
1015 | int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1016 | int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); | |
84fbb8a0 | 1017 | Lisp_Object translation_table |
f967223b | 1018 | = coding->translation_table_for_decode; |
d46c5b12 | 1019 | int result = CODING_FINISH_NORMAL; |
bdd9fb48 | 1020 | |
84fbb8a0 | 1021 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
f967223b | 1022 | translation_table = Vstandard_translation_table_for_decode; |
4ed46869 | 1023 | |
d46c5b12 | 1024 | coding->produced_char = 0; |
fb88bf2d | 1025 | coding->fake_multibyte = 0; |
d46c5b12 KH |
1026 | while (src < src_end && (dst_bytes |
1027 | ? (dst < adjusted_dst_end) | |
1028 | : (dst < src - 6))) | |
4ed46869 KH |
1029 | { |
1030 | /* SRC_BASE remembers the start position in source in each loop. | |
1031 | The loop will be exited when there's not enough source text | |
1032 | to analyze long escape sequence or 2-byte code (within macros | |
1033 | ONE_MORE_BYTE or TWO_MORE_BYTES). In that case, SRC is reset | |
1034 | to SRC_BASE before exiting. */ | |
1035 | unsigned char *src_base = src; | |
bdd9fb48 | 1036 | int c1 = *src++, c2; |
4ed46869 KH |
1037 | |
1038 | switch (iso_code_class [c1]) | |
1039 | { | |
1040 | case ISO_0x20_or_0x7F: | |
1041 | if (!coding->composing | |
1042 | && (charset0 < 0 || CHARSET_CHARS (charset0) == 94)) | |
1043 | { | |
1044 | /* This is SPACE or DEL. */ | |
1045 | *dst++ = c1; | |
d46c5b12 | 1046 | coding->produced_char++; |
4ed46869 KH |
1047 | break; |
1048 | } | |
1049 | /* This is a graphic character, we fall down ... */ | |
1050 | ||
1051 | case ISO_graphic_plane_0: | |
1052 | if (coding->composing == COMPOSING_WITH_RULE_RULE) | |
1053 | { | |
1054 | /* This is a composition rule. */ | |
1055 | *dst++ = c1 | 0x80; | |
1056 | coding->composing = COMPOSING_WITH_RULE_TAIL; | |
1057 | } | |
1058 | else | |
1059 | DECODE_ISO_CHARACTER (charset0, c1); | |
1060 | break; | |
1061 | ||
1062 | case ISO_0xA0_or_0xFF: | |
d46c5b12 KH |
1063 | if (charset1 < 0 || CHARSET_CHARS (charset1) == 94 |
1064 | || coding->flags & CODING_FLAG_ISO_SEVEN_BITS) | |
fb88bf2d | 1065 | goto label_invalid_code; |
4ed46869 KH |
1066 | /* This is a graphic character, we fall down ... */ |
1067 | ||
1068 | case ISO_graphic_plane_1: | |
d46c5b12 | 1069 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) |
fb88bf2d | 1070 | goto label_invalid_code; |
d46c5b12 KH |
1071 | else |
1072 | DECODE_ISO_CHARACTER (charset1, c1); | |
4ed46869 KH |
1073 | break; |
1074 | ||
1075 | case ISO_control_code: | |
1076 | /* All ISO2022 control characters in this class have the | |
1077 | same representation in Emacs internal format. */ | |
d46c5b12 KH |
1078 | if (c1 == '\n' |
1079 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
1080 | && (coding->eol_type == CODING_EOL_CR | |
1081 | || coding->eol_type == CODING_EOL_CRLF)) | |
1082 | { | |
1083 | result = CODING_FINISH_INCONSISTENT_EOL; | |
1084 | goto label_end_of_loop_2; | |
1085 | } | |
4ed46869 | 1086 | *dst++ = c1; |
d46c5b12 | 1087 | coding->produced_char++; |
4ed46869 KH |
1088 | break; |
1089 | ||
1090 | case ISO_carriage_return: | |
1091 | if (coding->eol_type == CODING_EOL_CR) | |
d46c5b12 | 1092 | *dst++ = '\n'; |
4ed46869 KH |
1093 | else if (coding->eol_type == CODING_EOL_CRLF) |
1094 | { | |
1095 | ONE_MORE_BYTE (c1); | |
1096 | if (c1 == ISO_CODE_LF) | |
1097 | *dst++ = '\n'; | |
1098 | else | |
1099 | { | |
d46c5b12 KH |
1100 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
1101 | { | |
1102 | result = CODING_FINISH_INCONSISTENT_EOL; | |
1103 | goto label_end_of_loop_2; | |
1104 | } | |
4ed46869 | 1105 | src--; |
d46c5b12 | 1106 | *dst++ = '\r'; |
4ed46869 KH |
1107 | } |
1108 | } | |
1109 | else | |
d46c5b12 KH |
1110 | *dst++ = c1; |
1111 | coding->produced_char++; | |
4ed46869 KH |
1112 | break; |
1113 | ||
1114 | case ISO_shift_out: | |
d46c5b12 KH |
1115 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
1116 | || CODING_SPEC_ISO_DESIGNATION (coding, 1) < 0) | |
1117 | goto label_invalid_code; | |
4ed46869 KH |
1118 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; |
1119 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1120 | break; | |
1121 | ||
1122 | case ISO_shift_in: | |
d46c5b12 KH |
1123 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
1124 | goto label_invalid_code; | |
4ed46869 KH |
1125 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; |
1126 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1127 | break; | |
1128 | ||
1129 | case ISO_single_shift_2_7: | |
1130 | case ISO_single_shift_2: | |
d46c5b12 KH |
1131 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
1132 | goto label_invalid_code; | |
4ed46869 KH |
1133 | /* SS2 is handled as an escape sequence of ESC 'N' */ |
1134 | c1 = 'N'; | |
1135 | goto label_escape_sequence; | |
1136 | ||
1137 | case ISO_single_shift_3: | |
d46c5b12 KH |
1138 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
1139 | goto label_invalid_code; | |
4ed46869 KH |
1140 | /* SS2 is handled as an escape sequence of ESC 'O' */ |
1141 | c1 = 'O'; | |
1142 | goto label_escape_sequence; | |
1143 | ||
1144 | case ISO_control_sequence_introducer: | |
1145 | /* CSI is handled as an escape sequence of ESC '[' ... */ | |
1146 | c1 = '['; | |
1147 | goto label_escape_sequence; | |
1148 | ||
1149 | case ISO_escape: | |
1150 | ONE_MORE_BYTE (c1); | |
1151 | label_escape_sequence: | |
1152 | /* Escape sequences handled by Emacs are invocation, | |
1153 | designation, direction specification, and character | |
1154 | composition specification. */ | |
1155 | switch (c1) | |
1156 | { | |
1157 | case '&': /* revision of following character set */ | |
1158 | ONE_MORE_BYTE (c1); | |
1159 | if (!(c1 >= '@' && c1 <= '~')) | |
d46c5b12 | 1160 | goto label_invalid_code; |
4ed46869 KH |
1161 | ONE_MORE_BYTE (c1); |
1162 | if (c1 != ISO_CODE_ESC) | |
d46c5b12 | 1163 | goto label_invalid_code; |
4ed46869 KH |
1164 | ONE_MORE_BYTE (c1); |
1165 | goto label_escape_sequence; | |
1166 | ||
1167 | case '$': /* designation of 2-byte character set */ | |
d46c5b12 KH |
1168 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
1169 | goto label_invalid_code; | |
4ed46869 KH |
1170 | ONE_MORE_BYTE (c1); |
1171 | if (c1 >= '@' && c1 <= 'B') | |
1172 | { /* designation of JISX0208.1978, GB2312.1980, | |
1173 | or JISX0208.1980 */ | |
1174 | DECODE_DESIGNATION (0, 2, 94, c1); | |
1175 | } | |
1176 | else if (c1 >= 0x28 && c1 <= 0x2B) | |
1177 | { /* designation of DIMENSION2_CHARS94 character set */ | |
1178 | ONE_MORE_BYTE (c2); | |
1179 | DECODE_DESIGNATION (c1 - 0x28, 2, 94, c2); | |
1180 | } | |
1181 | else if (c1 >= 0x2C && c1 <= 0x2F) | |
1182 | { /* designation of DIMENSION2_CHARS96 character set */ | |
1183 | ONE_MORE_BYTE (c2); | |
1184 | DECODE_DESIGNATION (c1 - 0x2C, 2, 96, c2); | |
1185 | } | |
1186 | else | |
d46c5b12 | 1187 | goto label_invalid_code; |
4ed46869 KH |
1188 | break; |
1189 | ||
1190 | case 'n': /* invocation of locking-shift-2 */ | |
d46c5b12 KH |
1191 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
1192 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) | |
1193 | goto label_invalid_code; | |
4ed46869 | 1194 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; |
e0e989f6 | 1195 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
4ed46869 KH |
1196 | break; |
1197 | ||
1198 | case 'o': /* invocation of locking-shift-3 */ | |
d46c5b12 KH |
1199 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
1200 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) | |
1201 | goto label_invalid_code; | |
4ed46869 | 1202 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; |
e0e989f6 | 1203 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
4ed46869 KH |
1204 | break; |
1205 | ||
1206 | case 'N': /* invocation of single-shift-2 */ | |
d46c5b12 KH |
1207 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
1208 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) | |
1209 | goto label_invalid_code; | |
4ed46869 KH |
1210 | ONE_MORE_BYTE (c1); |
1211 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 2); | |
1212 | DECODE_ISO_CHARACTER (charset, c1); | |
1213 | break; | |
1214 | ||
1215 | case 'O': /* invocation of single-shift-3 */ | |
d46c5b12 KH |
1216 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
1217 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) | |
1218 | goto label_invalid_code; | |
4ed46869 KH |
1219 | ONE_MORE_BYTE (c1); |
1220 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 3); | |
1221 | DECODE_ISO_CHARACTER (charset, c1); | |
1222 | break; | |
1223 | ||
d46c5b12 KH |
1224 | case '0': case '2': /* start composing */ |
1225 | /* Before processing composing, we must be sure that all | |
1226 | characters being composed are supported by CODING. | |
1227 | If not, we must give up composing and insert the | |
1228 | bunch of codes for composing as is without decoding. */ | |
1229 | { | |
1230 | int result1; | |
1231 | ||
1232 | result1 = check_composing_code (coding, src, src_end); | |
1233 | if (result1 == 0) | |
84fbb8a0 KH |
1234 | { |
1235 | coding->composing = (c1 == '0' | |
1236 | ? COMPOSING_NO_RULE_HEAD | |
1237 | : COMPOSING_WITH_RULE_HEAD); | |
1238 | coding->produced_char++; | |
1239 | } | |
d46c5b12 KH |
1240 | else if (result1 > 0) |
1241 | { | |
1242 | if (result1 + 2 < (dst_bytes ? dst_end : src_base) - dst) | |
1243 | { | |
1244 | bcopy (src_base, dst, result1 + 2); | |
1245 | src += result1; | |
1246 | dst += result1 + 2; | |
1247 | coding->produced_char += result1 + 2; | |
1248 | } | |
1249 | else | |
1250 | { | |
1251 | result = CODING_FINISH_INSUFFICIENT_DST; | |
1252 | goto label_end_of_loop_2; | |
1253 | } | |
1254 | } | |
1255 | else | |
1256 | goto label_end_of_loop; | |
1257 | } | |
4ed46869 KH |
1258 | break; |
1259 | ||
1260 | case '1': /* end composing */ | |
1261 | coding->composing = COMPOSING_NO; | |
4ed46869 KH |
1262 | break; |
1263 | ||
1264 | case '[': /* specification of direction */ | |
d46c5b12 KH |
1265 | if (coding->flags & CODING_FLAG_ISO_NO_DIRECTION) |
1266 | goto label_invalid_code; | |
4ed46869 | 1267 | /* For the moment, nested direction is not supported. |
d46c5b12 KH |
1268 | So, `coding->mode & CODING_MODE_DIRECTION' zero means |
1269 | left-to-right, and nozero means right-to-left. */ | |
4ed46869 KH |
1270 | ONE_MORE_BYTE (c1); |
1271 | switch (c1) | |
1272 | { | |
1273 | case ']': /* end of the current direction */ | |
d46c5b12 | 1274 | coding->mode &= ~CODING_MODE_DIRECTION; |
4ed46869 KH |
1275 | |
1276 | case '0': /* end of the current direction */ | |
1277 | case '1': /* start of left-to-right direction */ | |
1278 | ONE_MORE_BYTE (c1); | |
1279 | if (c1 == ']') | |
d46c5b12 | 1280 | coding->mode &= ~CODING_MODE_DIRECTION; |
4ed46869 | 1281 | else |
d46c5b12 | 1282 | goto label_invalid_code; |
4ed46869 KH |
1283 | break; |
1284 | ||
1285 | case '2': /* start of right-to-left direction */ | |
1286 | ONE_MORE_BYTE (c1); | |
1287 | if (c1 == ']') | |
d46c5b12 | 1288 | coding->mode |= CODING_MODE_DIRECTION; |
4ed46869 | 1289 | else |
d46c5b12 | 1290 | goto label_invalid_code; |
4ed46869 KH |
1291 | break; |
1292 | ||
1293 | default: | |
d46c5b12 | 1294 | goto label_invalid_code; |
4ed46869 KH |
1295 | } |
1296 | break; | |
1297 | ||
1298 | default: | |
d46c5b12 KH |
1299 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
1300 | goto label_invalid_code; | |
4ed46869 KH |
1301 | if (c1 >= 0x28 && c1 <= 0x2B) |
1302 | { /* designation of DIMENSION1_CHARS94 character set */ | |
1303 | ONE_MORE_BYTE (c2); | |
1304 | DECODE_DESIGNATION (c1 - 0x28, 1, 94, c2); | |
1305 | } | |
1306 | else if (c1 >= 0x2C && c1 <= 0x2F) | |
1307 | { /* designation of DIMENSION1_CHARS96 character set */ | |
1308 | ONE_MORE_BYTE (c2); | |
1309 | DECODE_DESIGNATION (c1 - 0x2C, 1, 96, c2); | |
1310 | } | |
1311 | else | |
1312 | { | |
d46c5b12 | 1313 | goto label_invalid_code; |
4ed46869 KH |
1314 | } |
1315 | } | |
1316 | /* We must update these variables now. */ | |
1317 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
1318 | charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); | |
1319 | break; | |
1320 | ||
d46c5b12 | 1321 | label_invalid_code: |
d46c5b12 KH |
1322 | while (src_base < src) |
1323 | *dst++ = *src_base++; | |
fb88bf2d | 1324 | coding->fake_multibyte = 1; |
4ed46869 KH |
1325 | } |
1326 | continue; | |
1327 | ||
1328 | label_end_of_loop: | |
d46c5b12 KH |
1329 | result = CODING_FINISH_INSUFFICIENT_SRC; |
1330 | label_end_of_loop_2: | |
4ed46869 KH |
1331 | src = src_base; |
1332 | break; | |
1333 | } | |
1334 | ||
fb88bf2d | 1335 | if (src < src_end) |
4ed46869 | 1336 | { |
fb88bf2d KH |
1337 | if (result == CODING_FINISH_NORMAL) |
1338 | result = CODING_FINISH_INSUFFICIENT_DST; | |
1339 | else if (result != CODING_FINISH_INCONSISTENT_EOL | |
1340 | && coding->mode & CODING_MODE_LAST_BLOCK) | |
1341 | { | |
1342 | /* This is the last block of the text to be decoded. We had | |
1343 | better just flush out all remaining codes in the text | |
1344 | although they are not valid characters. */ | |
1345 | src_bytes = src_end - src; | |
1346 | if (dst_bytes && (dst_end - dst < src_bytes)) | |
1347 | src_bytes = dst_end - dst; | |
1348 | bcopy (src, dst, src_bytes); | |
1349 | dst += src_bytes; | |
1350 | src += src_bytes; | |
1351 | coding->fake_multibyte = 1; | |
1352 | } | |
4ed46869 | 1353 | } |
fb88bf2d | 1354 | |
d46c5b12 KH |
1355 | coding->consumed = coding->consumed_char = src - source; |
1356 | coding->produced = dst - destination; | |
1357 | return result; | |
4ed46869 KH |
1358 | } |
1359 | ||
f4dee582 | 1360 | /* ISO2022 encoding stuff. */ |
4ed46869 KH |
1361 | |
1362 | /* | |
f4dee582 | 1363 | It is not enough to say just "ISO2022" on encoding, we have to |
d46c5b12 | 1364 | specify more details. In Emacs, each coding system of ISO2022 |
4ed46869 KH |
1365 | variant has the following specifications: |
1366 | 1. Initial designation to G0 thru G3. | |
1367 | 2. Allows short-form designation? | |
1368 | 3. ASCII should be designated to G0 before control characters? | |
1369 | 4. ASCII should be designated to G0 at end of line? | |
1370 | 5. 7-bit environment or 8-bit environment? | |
1371 | 6. Use locking-shift? | |
1372 | 7. Use Single-shift? | |
1373 | And the following two are only for Japanese: | |
1374 | 8. Use ASCII in place of JIS0201-1976-Roman? | |
1375 | 9. Use JISX0208-1983 in place of JISX0208-1978? | |
1376 | These specifications are encoded in `coding->flags' as flag bits | |
1377 | defined by macros CODING_FLAG_ISO_XXX. See `coding.h' for more | |
f4dee582 | 1378 | details. |
4ed46869 KH |
1379 | */ |
1380 | ||
1381 | /* Produce codes (escape sequence) for designating CHARSET to graphic | |
1382 | register REG. If <final-char> of CHARSET is '@', 'A', or 'B' and | |
1383 | the coding system CODING allows, produce designation sequence of | |
1384 | short-form. */ | |
1385 | ||
1386 | #define ENCODE_DESIGNATION(charset, reg, coding) \ | |
1387 | do { \ | |
1388 | unsigned char final_char = CHARSET_ISO_FINAL_CHAR (charset); \ | |
1389 | char *intermediate_char_94 = "()*+"; \ | |
1390 | char *intermediate_char_96 = ",-./"; \ | |
70c22245 KH |
1391 | int revision = CODING_SPEC_ISO_REVISION_NUMBER(coding, charset); \ |
1392 | if (revision < 255) \ | |
1393 | { \ | |
4ed46869 KH |
1394 | *dst++ = ISO_CODE_ESC; \ |
1395 | *dst++ = '&'; \ | |
70c22245 | 1396 | *dst++ = '@' + revision; \ |
4ed46869 KH |
1397 | } \ |
1398 | *dst++ = ISO_CODE_ESC; \ | |
1399 | if (CHARSET_DIMENSION (charset) == 1) \ | |
1400 | { \ | |
1401 | if (CHARSET_CHARS (charset) == 94) \ | |
1402 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ | |
1403 | else \ | |
1404 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ | |
1405 | } \ | |
1406 | else \ | |
1407 | { \ | |
1408 | *dst++ = '$'; \ | |
1409 | if (CHARSET_CHARS (charset) == 94) \ | |
1410 | { \ | |
1411 | if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \ | |
1412 | || reg != 0 \ | |
1413 | || final_char < '@' || final_char > 'B') \ | |
1414 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ | |
1415 | } \ | |
1416 | else \ | |
1417 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ | |
1418 | } \ | |
1419 | *dst++ = final_char; \ | |
1420 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ | |
1421 | } while (0) | |
1422 | ||
1423 | /* The following two macros produce codes (control character or escape | |
1424 | sequence) for ISO2022 single-shift functions (single-shift-2 and | |
1425 | single-shift-3). */ | |
1426 | ||
1427 | #define ENCODE_SINGLE_SHIFT_2 \ | |
1428 | do { \ | |
1429 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1430 | *dst++ = ISO_CODE_ESC, *dst++ = 'N'; \ | |
1431 | else \ | |
fb88bf2d KH |
1432 | { \ |
1433 | *dst++ = ISO_CODE_SS2; \ | |
1434 | coding->fake_multibyte = 1; \ | |
1435 | } \ | |
4ed46869 KH |
1436 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
1437 | } while (0) | |
1438 | ||
fb88bf2d KH |
1439 | #define ENCODE_SINGLE_SHIFT_3 \ |
1440 | do { \ | |
4ed46869 | 1441 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
fb88bf2d KH |
1442 | *dst++ = ISO_CODE_ESC, *dst++ = 'O'; \ |
1443 | else \ | |
1444 | { \ | |
1445 | *dst++ = ISO_CODE_SS3; \ | |
1446 | coding->fake_multibyte = 1; \ | |
1447 | } \ | |
4ed46869 KH |
1448 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
1449 | } while (0) | |
1450 | ||
1451 | /* The following four macros produce codes (control character or | |
1452 | escape sequence) for ISO2022 locking-shift functions (shift-in, | |
1453 | shift-out, locking-shift-2, and locking-shift-3). */ | |
1454 | ||
1455 | #define ENCODE_SHIFT_IN \ | |
1456 | do { \ | |
1457 | *dst++ = ISO_CODE_SI; \ | |
1458 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; \ | |
1459 | } while (0) | |
1460 | ||
1461 | #define ENCODE_SHIFT_OUT \ | |
1462 | do { \ | |
1463 | *dst++ = ISO_CODE_SO; \ | |
1464 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; \ | |
1465 | } while (0) | |
1466 | ||
1467 | #define ENCODE_LOCKING_SHIFT_2 \ | |
1468 | do { \ | |
1469 | *dst++ = ISO_CODE_ESC, *dst++ = 'n'; \ | |
1470 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; \ | |
1471 | } while (0) | |
1472 | ||
1473 | #define ENCODE_LOCKING_SHIFT_3 \ | |
1474 | do { \ | |
1475 | *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \ | |
1476 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; \ | |
1477 | } while (0) | |
1478 | ||
f4dee582 RS |
1479 | /* Produce codes for a DIMENSION1 character whose character set is |
1480 | CHARSET and whose position-code is C1. Designation and invocation | |
4ed46869 KH |
1481 | sequences are also produced in advance if necessary. */ |
1482 | ||
1483 | ||
6e85d753 KH |
1484 | #define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \ |
1485 | do { \ | |
1486 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ | |
1487 | { \ | |
1488 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1489 | *dst++ = c1 & 0x7F; \ | |
1490 | else \ | |
1491 | *dst++ = c1 | 0x80; \ | |
1492 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ | |
1493 | break; \ | |
1494 | } \ | |
1495 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ | |
1496 | { \ | |
1497 | *dst++ = c1 & 0x7F; \ | |
1498 | break; \ | |
1499 | } \ | |
1500 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ | |
1501 | { \ | |
1502 | *dst++ = c1 | 0x80; \ | |
1503 | break; \ | |
1504 | } \ | |
1505 | else if (coding->flags & CODING_FLAG_ISO_SAFE \ | |
70c22245 | 1506 | && !coding->safe_charsets[charset]) \ |
6e85d753 KH |
1507 | { \ |
1508 | /* We should not encode this character, instead produce one or \ | |
1509 | two `?'s. */ \ | |
1510 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1511 | if (CHARSET_WIDTH (charset) == 2) \ | |
1512 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1513 | break; \ | |
1514 | } \ | |
1515 | else \ | |
1516 | /* Since CHARSET is not yet invoked to any graphic planes, we \ | |
1517 | must invoke it, or, at first, designate it to some graphic \ | |
1518 | register. Then repeat the loop to actually produce the \ | |
1519 | character. */ \ | |
1520 | dst = encode_invocation_designation (charset, coding, dst); \ | |
4ed46869 KH |
1521 | } while (1) |
1522 | ||
f4dee582 RS |
1523 | /* Produce codes for a DIMENSION2 character whose character set is |
1524 | CHARSET and whose position-codes are C1 and C2. Designation and | |
4ed46869 KH |
1525 | invocation codes are also produced in advance if necessary. */ |
1526 | ||
6e85d753 KH |
1527 | #define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2) \ |
1528 | do { \ | |
1529 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ | |
1530 | { \ | |
1531 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1532 | *dst++ = c1 & 0x7F, *dst++ = c2 & 0x7F; \ | |
1533 | else \ | |
1534 | *dst++ = c1 | 0x80, *dst++ = c2 | 0x80; \ | |
1535 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ | |
1536 | break; \ | |
1537 | } \ | |
1538 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ | |
1539 | { \ | |
1540 | *dst++ = c1 & 0x7F, *dst++= c2 & 0x7F; \ | |
1541 | break; \ | |
1542 | } \ | |
1543 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ | |
1544 | { \ | |
1545 | *dst++ = c1 | 0x80, *dst++= c2 | 0x80; \ | |
1546 | break; \ | |
1547 | } \ | |
1548 | else if (coding->flags & CODING_FLAG_ISO_SAFE \ | |
70c22245 | 1549 | && !coding->safe_charsets[charset]) \ |
6e85d753 KH |
1550 | { \ |
1551 | /* We should not encode this character, instead produce one or \ | |
1552 | two `?'s. */ \ | |
1553 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1554 | if (CHARSET_WIDTH (charset) == 2) \ | |
1555 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ | |
1556 | break; \ | |
1557 | } \ | |
1558 | else \ | |
1559 | /* Since CHARSET is not yet invoked to any graphic planes, we \ | |
1560 | must invoke it, or, at first, designate it to some graphic \ | |
1561 | register. Then repeat the loop to actually produce the \ | |
1562 | character. */ \ | |
1563 | dst = encode_invocation_designation (charset, coding, dst); \ | |
4ed46869 KH |
1564 | } while (1) |
1565 | ||
84fbb8a0 KH |
1566 | #define ENCODE_ISO_CHARACTER(charset, c1, c2) \ |
1567 | do { \ | |
1568 | int c_alt, charset_alt; \ | |
1569 | if (!NILP (translation_table) \ | |
1570 | && ((c_alt = translate_char (translation_table, -1, \ | |
1571 | charset, c1, c2)) \ | |
1572 | >= 0)) \ | |
1573 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
1574 | else \ | |
1575 | charset_alt = charset; \ | |
1576 | if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
1577 | { \ | |
1578 | if (charset == CHARSET_ASCII \ | |
1579 | && coding->flags & CODING_FLAG_ISO_USE_ROMAN) \ | |
1580 | charset_alt = charset_latin_jisx0201; \ | |
1581 | ENCODE_ISO_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
1582 | } \ | |
1583 | else \ | |
1584 | { \ | |
1585 | if (charset == charset_jisx0208 \ | |
1586 | && coding->flags & CODING_FLAG_ISO_USE_OLDJIS) \ | |
1587 | charset_alt = charset_jisx0208_1978; \ | |
1588 | ENCODE_ISO_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
1589 | } \ | |
1590 | if (! COMPOSING_P (coding->composing)) \ | |
1591 | coding->consumed_char++; \ | |
1592 | } while (0) | |
bdd9fb48 | 1593 | |
4ed46869 KH |
1594 | /* Produce designation and invocation codes at a place pointed by DST |
1595 | to use CHARSET. The element `spec.iso2022' of *CODING is updated. | |
1596 | Return new DST. */ | |
1597 | ||
1598 | unsigned char * | |
1599 | encode_invocation_designation (charset, coding, dst) | |
1600 | int charset; | |
1601 | struct coding_system *coding; | |
1602 | unsigned char *dst; | |
1603 | { | |
1604 | int reg; /* graphic register number */ | |
1605 | ||
1606 | /* At first, check designations. */ | |
1607 | for (reg = 0; reg < 4; reg++) | |
1608 | if (charset == CODING_SPEC_ISO_DESIGNATION (coding, reg)) | |
1609 | break; | |
1610 | ||
1611 | if (reg >= 4) | |
1612 | { | |
1613 | /* CHARSET is not yet designated to any graphic registers. */ | |
1614 | /* At first check the requested designation. */ | |
1615 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); | |
1ba9e4ab KH |
1616 | if (reg == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION) |
1617 | /* Since CHARSET requests no special designation, designate it | |
1618 | to graphic register 0. */ | |
4ed46869 KH |
1619 | reg = 0; |
1620 | ||
1621 | ENCODE_DESIGNATION (charset, reg, coding); | |
1622 | } | |
1623 | ||
1624 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != reg | |
1625 | && CODING_SPEC_ISO_INVOCATION (coding, 1) != reg) | |
1626 | { | |
1627 | /* Since the graphic register REG is not invoked to any graphic | |
1628 | planes, invoke it to graphic plane 0. */ | |
1629 | switch (reg) | |
1630 | { | |
1631 | case 0: /* graphic register 0 */ | |
1632 | ENCODE_SHIFT_IN; | |
1633 | break; | |
1634 | ||
1635 | case 1: /* graphic register 1 */ | |
1636 | ENCODE_SHIFT_OUT; | |
1637 | break; | |
1638 | ||
1639 | case 2: /* graphic register 2 */ | |
1640 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) | |
1641 | ENCODE_SINGLE_SHIFT_2; | |
1642 | else | |
1643 | ENCODE_LOCKING_SHIFT_2; | |
1644 | break; | |
1645 | ||
1646 | case 3: /* graphic register 3 */ | |
1647 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) | |
1648 | ENCODE_SINGLE_SHIFT_3; | |
1649 | else | |
1650 | ENCODE_LOCKING_SHIFT_3; | |
1651 | break; | |
1652 | } | |
1653 | } | |
1654 | return dst; | |
1655 | } | |
1656 | ||
1657 | /* The following two macros produce codes for indicating composition. */ | |
1658 | #define ENCODE_COMPOSITION_NO_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '0' | |
1659 | #define ENCODE_COMPOSITION_WITH_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '2' | |
1660 | #define ENCODE_COMPOSITION_END *dst++ = ISO_CODE_ESC, *dst++ = '1' | |
1661 | ||
1662 | /* The following three macros produce codes for indicating direction | |
1663 | of text. */ | |
1664 | #define ENCODE_CONTROL_SEQUENCE_INTRODUCER \ | |
1665 | do { \ | |
1666 | if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS) \ | |
1667 | *dst++ = ISO_CODE_ESC, *dst++ = '['; \ | |
1668 | else \ | |
1669 | *dst++ = ISO_CODE_CSI; \ | |
1670 | } while (0) | |
1671 | ||
1672 | #define ENCODE_DIRECTION_R2L \ | |
1673 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '2', *dst++ = ']' | |
1674 | ||
1675 | #define ENCODE_DIRECTION_L2R \ | |
1676 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '0', *dst++ = ']' | |
1677 | ||
1678 | /* Produce codes for designation and invocation to reset the graphic | |
1679 | planes and registers to initial state. */ | |
e0e989f6 KH |
1680 | #define ENCODE_RESET_PLANE_AND_REGISTER \ |
1681 | do { \ | |
1682 | int reg; \ | |
1683 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != 0) \ | |
1684 | ENCODE_SHIFT_IN; \ | |
1685 | for (reg = 0; reg < 4; reg++) \ | |
1686 | if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg) >= 0 \ | |
1687 | && (CODING_SPEC_ISO_DESIGNATION (coding, reg) \ | |
1688 | != CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg))) \ | |
1689 | ENCODE_DESIGNATION \ | |
1690 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg), reg, coding); \ | |
4ed46869 KH |
1691 | } while (0) |
1692 | ||
bdd9fb48 | 1693 | /* Produce designation sequences of charsets in the line started from |
d46c5b12 | 1694 | SRC to a place pointed by *DSTP, and update DSTP. |
bdd9fb48 KH |
1695 | |
1696 | If the current block ends before any end-of-line, we may fail to | |
d46c5b12 KH |
1697 | find all the necessary designations. */ |
1698 | ||
dfcf069d | 1699 | void |
bdd9fb48 | 1700 | encode_designation_at_bol (coding, table, src, src_end, dstp) |
e0e989f6 | 1701 | struct coding_system *coding; |
bdd9fb48 | 1702 | Lisp_Object table; |
e0e989f6 KH |
1703 | unsigned char *src, *src_end, **dstp; |
1704 | { | |
bdd9fb48 KH |
1705 | int charset, c, found = 0, reg; |
1706 | /* Table of charsets to be designated to each graphic register. */ | |
1707 | int r[4]; | |
1708 | unsigned char *dst = *dstp; | |
1709 | ||
1710 | for (reg = 0; reg < 4; reg++) | |
1711 | r[reg] = -1; | |
1712 | ||
1713 | while (src < src_end && *src != '\n' && found < 4) | |
e0e989f6 | 1714 | { |
bdd9fb48 KH |
1715 | int bytes = BYTES_BY_CHAR_HEAD (*src); |
1716 | ||
1717 | if (NILP (table)) | |
1718 | charset = CHARSET_AT (src); | |
1719 | else | |
e0e989f6 | 1720 | { |
35cb8686 RS |
1721 | int c_alt; |
1722 | unsigned char c1, c2; | |
bdd9fb48 KH |
1723 | |
1724 | SPLIT_STRING(src, bytes, charset, c1, c2); | |
84fbb8a0 | 1725 | if ((c_alt = translate_char (table, -1, charset, c1, c2)) >= 0) |
bdd9fb48 | 1726 | charset = CHAR_CHARSET (c_alt); |
e0e989f6 | 1727 | } |
bdd9fb48 | 1728 | |
e0e989f6 | 1729 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); |
d46c5b12 | 1730 | if (reg != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION && r[reg] < 0) |
bdd9fb48 KH |
1731 | { |
1732 | found++; | |
1733 | r[reg] = charset; | |
1734 | } | |
1735 | ||
1736 | src += bytes; | |
1737 | } | |
1738 | ||
1739 | if (found) | |
1740 | { | |
1741 | for (reg = 0; reg < 4; reg++) | |
1742 | if (r[reg] >= 0 | |
1743 | && CODING_SPEC_ISO_DESIGNATION (coding, reg) != r[reg]) | |
1744 | ENCODE_DESIGNATION (r[reg], reg, coding); | |
1745 | *dstp = dst; | |
e0e989f6 | 1746 | } |
e0e989f6 KH |
1747 | } |
1748 | ||
4ed46869 KH |
1749 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */ |
1750 | ||
1751 | int | |
d46c5b12 | 1752 | encode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
1753 | struct coding_system *coding; |
1754 | unsigned char *source, *destination; | |
1755 | int src_bytes, dst_bytes; | |
4ed46869 KH |
1756 | { |
1757 | unsigned char *src = source; | |
1758 | unsigned char *src_end = source + src_bytes; | |
1759 | unsigned char *dst = destination; | |
1760 | unsigned char *dst_end = destination + dst_bytes; | |
e0e989f6 | 1761 | /* Since the maximum bytes produced by each loop is 20, we subtract 19 |
4ed46869 KH |
1762 | from DST_END to assure overflow checking is necessary only at the |
1763 | head of loop. */ | |
e0e989f6 | 1764 | unsigned char *adjusted_dst_end = dst_end - 19; |
84fbb8a0 | 1765 | Lisp_Object translation_table |
f967223b | 1766 | = coding->translation_table_for_encode; |
d46c5b12 | 1767 | int result = CODING_FINISH_NORMAL; |
bdd9fb48 | 1768 | |
84fbb8a0 | 1769 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
f967223b | 1770 | translation_table = Vstandard_translation_table_for_encode; |
4ed46869 | 1771 | |
d46c5b12 | 1772 | coding->consumed_char = 0; |
fb88bf2d | 1773 | coding->fake_multibyte = 0; |
d46c5b12 KH |
1774 | while (src < src_end && (dst_bytes |
1775 | ? (dst < adjusted_dst_end) | |
1776 | : (dst < src - 19))) | |
4ed46869 KH |
1777 | { |
1778 | /* SRC_BASE remembers the start position in source in each loop. | |
1779 | The loop will be exited when there's not enough source text | |
1780 | to analyze multi-byte codes (within macros ONE_MORE_BYTE, | |
1781 | TWO_MORE_BYTES, and THREE_MORE_BYTES). In that case, SRC is | |
1782 | reset to SRC_BASE before exiting. */ | |
1783 | unsigned char *src_base = src; | |
bdd9fb48 | 1784 | int charset, c1, c2, c3, c4; |
4ed46869 | 1785 | |
e0e989f6 KH |
1786 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL |
1787 | && CODING_SPEC_ISO_BOL (coding)) | |
1788 | { | |
bdd9fb48 | 1789 | /* We have to produce designation sequences if any now. */ |
84fbb8a0 | 1790 | encode_designation_at_bol (coding, translation_table, |
bdd9fb48 | 1791 | src, src_end, &dst); |
e0e989f6 KH |
1792 | CODING_SPEC_ISO_BOL (coding) = 0; |
1793 | } | |
1794 | ||
1795 | c1 = *src++; | |
4ed46869 | 1796 | /* If we are seeing a component of a composite character, we are |
d46c5b12 KH |
1797 | seeing a leading-code encoded irregularly for composition, or |
1798 | a composition rule if composing with rule. We must set C1 to | |
1799 | a normal leading-code or an ASCII code. If we are not seeing | |
1800 | a composite character, we must reset composition, | |
1801 | designation, and invocation states. */ | |
4ed46869 KH |
1802 | if (COMPOSING_P (coding->composing)) |
1803 | { | |
1804 | if (c1 < 0xA0) | |
1805 | { | |
1806 | /* We are not in a composite character any longer. */ | |
1807 | coding->composing = COMPOSING_NO; | |
d46c5b12 | 1808 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1809 | ENCODE_COMPOSITION_END; |
1810 | } | |
1811 | else | |
1812 | { | |
1813 | if (coding->composing == COMPOSING_WITH_RULE_RULE) | |
1814 | { | |
1815 | *dst++ = c1 & 0x7F; | |
1816 | coding->composing = COMPOSING_WITH_RULE_HEAD; | |
1817 | continue; | |
1818 | } | |
1819 | else if (coding->composing == COMPOSING_WITH_RULE_HEAD) | |
1820 | coding->composing = COMPOSING_WITH_RULE_RULE; | |
1821 | if (c1 == 0xA0) | |
1822 | { | |
1823 | /* This is an ASCII component. */ | |
1824 | ONE_MORE_BYTE (c1); | |
1825 | c1 &= 0x7F; | |
1826 | } | |
1827 | else | |
1828 | /* This is a leading-code of non ASCII component. */ | |
1829 | c1 -= 0x20; | |
1830 | } | |
1831 | } | |
1832 | ||
1833 | /* Now encode one character. C1 is a control character, an | |
1834 | ASCII character, or a leading-code of multi-byte character. */ | |
1835 | switch (emacs_code_class[c1]) | |
1836 | { | |
1837 | case EMACS_ascii_code: | |
bdd9fb48 | 1838 | ENCODE_ISO_CHARACTER (CHARSET_ASCII, c1, /* dummy */ c2); |
4ed46869 KH |
1839 | break; |
1840 | ||
1841 | case EMACS_control_code: | |
1842 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) | |
e0e989f6 | 1843 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 | 1844 | *dst++ = c1; |
d46c5b12 | 1845 | coding->consumed_char++; |
4ed46869 KH |
1846 | break; |
1847 | ||
1848 | case EMACS_carriage_return_code: | |
d46c5b12 | 1849 | if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
4ed46869 KH |
1850 | { |
1851 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) | |
e0e989f6 | 1852 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 | 1853 | *dst++ = c1; |
d46c5b12 | 1854 | coding->consumed_char++; |
4ed46869 KH |
1855 | break; |
1856 | } | |
1857 | /* fall down to treat '\r' as '\n' ... */ | |
1858 | ||
1859 | case EMACS_linefeed_code: | |
1860 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL) | |
e0e989f6 KH |
1861 | ENCODE_RESET_PLANE_AND_REGISTER; |
1862 | if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL) | |
1863 | bcopy (coding->spec.iso2022.initial_designation, | |
1864 | coding->spec.iso2022.current_designation, | |
1865 | sizeof coding->spec.iso2022.initial_designation); | |
4ed46869 | 1866 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 1867 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
1868 | *dst++ = ISO_CODE_LF; |
1869 | else if (coding->eol_type == CODING_EOL_CRLF) | |
1870 | *dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF; | |
1871 | else | |
1872 | *dst++ = ISO_CODE_CR; | |
e0e989f6 | 1873 | CODING_SPEC_ISO_BOL (coding) = 1; |
d46c5b12 | 1874 | coding->consumed_char++; |
4ed46869 KH |
1875 | break; |
1876 | ||
1877 | case EMACS_leading_code_2: | |
1878 | ONE_MORE_BYTE (c2); | |
19a8d9e0 KH |
1879 | if (c2 < 0xA0) |
1880 | { | |
1881 | /* invalid sequence */ | |
1882 | *dst++ = c1; | |
38cf95df RS |
1883 | src--; |
1884 | coding->consumed_char++; | |
19a8d9e0 KH |
1885 | } |
1886 | else | |
1887 | ENCODE_ISO_CHARACTER (c1, c2, /* dummy */ c3); | |
4ed46869 KH |
1888 | break; |
1889 | ||
1890 | case EMACS_leading_code_3: | |
1891 | TWO_MORE_BYTES (c2, c3); | |
19a8d9e0 KH |
1892 | if (c2 < 0xA0 || c3 < 0xA0) |
1893 | { | |
1894 | /* invalid sequence */ | |
1895 | *dst++ = c1; | |
38cf95df RS |
1896 | src -= 2; |
1897 | coding->consumed_char++; | |
19a8d9e0 KH |
1898 | } |
1899 | else if (c1 < LEADING_CODE_PRIVATE_11) | |
bdd9fb48 | 1900 | ENCODE_ISO_CHARACTER (c1, c2, c3); |
4ed46869 | 1901 | else |
bdd9fb48 | 1902 | ENCODE_ISO_CHARACTER (c2, c3, /* dummy */ c4); |
4ed46869 KH |
1903 | break; |
1904 | ||
1905 | case EMACS_leading_code_4: | |
1906 | THREE_MORE_BYTES (c2, c3, c4); | |
19a8d9e0 KH |
1907 | if (c2 < 0xA0 || c3 < 0xA0 || c4 < 0xA0) |
1908 | { | |
1909 | /* invalid sequence */ | |
1910 | *dst++ = c1; | |
38cf95df RS |
1911 | src -= 3; |
1912 | coding->consumed_char++; | |
19a8d9e0 KH |
1913 | } |
1914 | else | |
1915 | ENCODE_ISO_CHARACTER (c2, c3, c4); | |
4ed46869 KH |
1916 | break; |
1917 | ||
1918 | case EMACS_leading_code_composition: | |
19a8d9e0 KH |
1919 | ONE_MORE_BYTE (c2); |
1920 | if (c2 < 0xA0) | |
1921 | { | |
1922 | /* invalid sequence */ | |
1923 | *dst++ = c1; | |
38cf95df RS |
1924 | src--; |
1925 | coding->consumed_char++; | |
19a8d9e0 KH |
1926 | } |
1927 | else if (c2 == 0xFF) | |
4ed46869 | 1928 | { |
d46c5b12 | 1929 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1930 | coding->composing = COMPOSING_WITH_RULE_HEAD; |
1931 | ENCODE_COMPOSITION_WITH_RULE_START; | |
d46c5b12 | 1932 | coding->consumed_char++; |
4ed46869 KH |
1933 | } |
1934 | else | |
1935 | { | |
d46c5b12 | 1936 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1937 | /* Rewind one byte because it is a character code of |
1938 | composition elements. */ | |
1939 | src--; | |
1940 | coding->composing = COMPOSING_NO_RULE_HEAD; | |
1941 | ENCODE_COMPOSITION_NO_RULE_START; | |
d46c5b12 | 1942 | coding->consumed_char++; |
4ed46869 KH |
1943 | } |
1944 | break; | |
1945 | ||
1946 | case EMACS_invalid_code: | |
1947 | *dst++ = c1; | |
d46c5b12 | 1948 | coding->consumed_char++; |
4ed46869 KH |
1949 | break; |
1950 | } | |
1951 | continue; | |
1952 | label_end_of_loop: | |
d46c5b12 KH |
1953 | result = CODING_FINISH_INSUFFICIENT_SRC; |
1954 | src = src_base; | |
4ed46869 KH |
1955 | break; |
1956 | } | |
1957 | ||
49cb52b4 KH |
1958 | if (src < src_end && result == CODING_FINISH_NORMAL) |
1959 | result = CODING_FINISH_INSUFFICIENT_DST; | |
1960 | ||
1961 | /* If this is the last block of the text to be encoded, we must | |
1962 | reset graphic planes and registers to the initial state, and | |
1963 | flush out the carryover if any. */ | |
1964 | if (coding->mode & CODING_MODE_LAST_BLOCK) | |
84fbb8a0 KH |
1965 | { |
1966 | ENCODE_RESET_PLANE_AND_REGISTER; | |
1967 | if (COMPOSING_P (coding->composing)) | |
1968 | ENCODE_COMPOSITION_END; | |
1969 | } | |
d46c5b12 KH |
1970 | coding->consumed = src - source; |
1971 | coding->produced = coding->produced_char = dst - destination; | |
1972 | return result; | |
4ed46869 KH |
1973 | } |
1974 | ||
1975 | \f | |
1976 | /*** 4. SJIS and BIG5 handlers ***/ | |
1977 | ||
f4dee582 | 1978 | /* Although SJIS and BIG5 are not ISO's coding system, they are used |
4ed46869 KH |
1979 | quite widely. So, for the moment, Emacs supports them in the bare |
1980 | C code. But, in the future, they may be supported only by CCL. */ | |
1981 | ||
1982 | /* SJIS is a coding system encoding three character sets: ASCII, right | |
1983 | half of JISX0201-Kana, and JISX0208. An ASCII character is encoded | |
1984 | as is. A character of charset katakana-jisx0201 is encoded by | |
1985 | "position-code + 0x80". A character of charset japanese-jisx0208 | |
1986 | is encoded in 2-byte but two position-codes are divided and shifted | |
1987 | so that it fit in the range below. | |
1988 | ||
1989 | --- CODE RANGE of SJIS --- | |
1990 | (character set) (range) | |
1991 | ASCII 0x00 .. 0x7F | |
1992 | KATAKANA-JISX0201 0xA0 .. 0xDF | |
54f78171 | 1993 | JISX0208 (1st byte) 0x80 .. 0x9F and 0xE0 .. 0xEF |
4ed46869 KH |
1994 | (2nd byte) 0x40 .. 0xFF |
1995 | ------------------------------- | |
1996 | ||
1997 | */ | |
1998 | ||
1999 | /* BIG5 is a coding system encoding two character sets: ASCII and | |
2000 | Big5. An ASCII character is encoded as is. Big5 is a two-byte | |
2001 | character set and is encoded in two-byte. | |
2002 | ||
2003 | --- CODE RANGE of BIG5 --- | |
2004 | (character set) (range) | |
2005 | ASCII 0x00 .. 0x7F | |
2006 | Big5 (1st byte) 0xA1 .. 0xFE | |
2007 | (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE | |
2008 | -------------------------- | |
2009 | ||
2010 | Since the number of characters in Big5 is larger than maximum | |
2011 | characters in Emacs' charset (96x96), it can't be handled as one | |
2012 | charset. So, in Emacs, Big5 is divided into two: `charset-big5-1' | |
2013 | and `charset-big5-2'. Both are DIMENSION2 and CHARS94. The former | |
2014 | contains frequently used characters and the latter contains less | |
2015 | frequently used characters. */ | |
2016 | ||
2017 | /* Macros to decode or encode a character of Big5 in BIG5. B1 and B2 | |
2018 | are the 1st and 2nd position-codes of Big5 in BIG5 coding system. | |
2019 | C1 and C2 are the 1st and 2nd position-codes of of Emacs' internal | |
2020 | format. CHARSET is `charset_big5_1' or `charset_big5_2'. */ | |
2021 | ||
2022 | /* Number of Big5 characters which have the same code in 1st byte. */ | |
2023 | #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) | |
2024 | ||
2025 | #define DECODE_BIG5(b1, b2, charset, c1, c2) \ | |
2026 | do { \ | |
2027 | unsigned int temp \ | |
2028 | = (b1 - 0xA1) * BIG5_SAME_ROW + b2 - (b2 < 0x7F ? 0x40 : 0x62); \ | |
2029 | if (b1 < 0xC9) \ | |
2030 | charset = charset_big5_1; \ | |
2031 | else \ | |
2032 | { \ | |
2033 | charset = charset_big5_2; \ | |
2034 | temp -= (0xC9 - 0xA1) * BIG5_SAME_ROW; \ | |
2035 | } \ | |
2036 | c1 = temp / (0xFF - 0xA1) + 0x21; \ | |
2037 | c2 = temp % (0xFF - 0xA1) + 0x21; \ | |
2038 | } while (0) | |
2039 | ||
2040 | #define ENCODE_BIG5(charset, c1, c2, b1, b2) \ | |
2041 | do { \ | |
2042 | unsigned int temp = (c1 - 0x21) * (0xFF - 0xA1) + (c2 - 0x21); \ | |
2043 | if (charset == charset_big5_2) \ | |
2044 | temp += BIG5_SAME_ROW * (0xC9 - 0xA1); \ | |
2045 | b1 = temp / BIG5_SAME_ROW + 0xA1; \ | |
2046 | b2 = temp % BIG5_SAME_ROW; \ | |
2047 | b2 += b2 < 0x3F ? 0x40 : 0x62; \ | |
2048 | } while (0) | |
2049 | ||
a5d301df KH |
2050 | #define DECODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ |
2051 | do { \ | |
2052 | int c_alt, charset_alt = (charset); \ | |
84fbb8a0 KH |
2053 | if (!NILP (translation_table) \ |
2054 | && ((c_alt = translate_char (translation_table, \ | |
2055 | -1, (charset), c1, c2)) >= 0)) \ | |
a5d301df KH |
2056 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
2057 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ | |
2058 | DECODE_CHARACTER_ASCII (c1); \ | |
2059 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
2060 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
2061 | else \ | |
2062 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
2063 | } while (0) | |
2064 | ||
84fbb8a0 KH |
2065 | #define ENCODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ |
2066 | do { \ | |
2067 | int c_alt, charset_alt; \ | |
2068 | if (!NILP (translation_table) \ | |
2069 | && ((c_alt = translate_char (translation_table, -1, \ | |
2070 | charset, c1, c2)) \ | |
2071 | >= 0)) \ | |
2072 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
2073 | else \ | |
2074 | charset_alt = charset; \ | |
2075 | if (charset_alt == charset_ascii) \ | |
2076 | *dst++ = c1; \ | |
2077 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
2078 | { \ | |
2079 | if (sjis_p && charset_alt == charset_katakana_jisx0201) \ | |
2080 | *dst++ = c1; \ | |
2081 | else \ | |
2082 | { \ | |
2083 | *dst++ = charset_alt, *dst++ = c1; \ | |
2084 | coding->fake_multibyte = 1; \ | |
2085 | } \ | |
2086 | } \ | |
2087 | else \ | |
2088 | { \ | |
2089 | c1 &= 0x7F, c2 &= 0x7F; \ | |
2090 | if (sjis_p && charset_alt == charset_jisx0208) \ | |
2091 | { \ | |
2092 | unsigned char s1, s2; \ | |
2093 | \ | |
2094 | ENCODE_SJIS (c1, c2, s1, s2); \ | |
2095 | *dst++ = s1, *dst++ = s2; \ | |
2096 | coding->fake_multibyte = 1; \ | |
2097 | } \ | |
2098 | else if (!sjis_p \ | |
2099 | && (charset_alt == charset_big5_1 \ | |
2100 | || charset_alt == charset_big5_2)) \ | |
2101 | { \ | |
2102 | unsigned char b1, b2; \ | |
2103 | \ | |
2104 | ENCODE_BIG5 (charset_alt, c1, c2, b1, b2); \ | |
2105 | *dst++ = b1, *dst++ = b2; \ | |
2106 | } \ | |
2107 | else \ | |
2108 | { \ | |
2109 | *dst++ = charset_alt, *dst++ = c1, *dst++ = c2; \ | |
2110 | coding->fake_multibyte = 1; \ | |
2111 | } \ | |
2112 | } \ | |
2113 | coding->consumed_char++; \ | |
a5d301df KH |
2114 | } while (0); |
2115 | ||
4ed46869 KH |
2116 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
2117 | Check if a text is encoded in SJIS. If it is, return | |
2118 | CODING_CATEGORY_MASK_SJIS, else return 0. */ | |
2119 | ||
2120 | int | |
2121 | detect_coding_sjis (src, src_end) | |
2122 | unsigned char *src, *src_end; | |
2123 | { | |
2124 | unsigned char c; | |
2125 | ||
2126 | while (src < src_end) | |
2127 | { | |
2128 | c = *src++; | |
4ed46869 KH |
2129 | if ((c >= 0x80 && c < 0xA0) || c >= 0xE0) |
2130 | { | |
2131 | if (src < src_end && *src++ < 0x40) | |
2132 | return 0; | |
2133 | } | |
2134 | } | |
2135 | return CODING_CATEGORY_MASK_SJIS; | |
2136 | } | |
2137 | ||
2138 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". | |
2139 | Check if a text is encoded in BIG5. If it is, return | |
2140 | CODING_CATEGORY_MASK_BIG5, else return 0. */ | |
2141 | ||
2142 | int | |
2143 | detect_coding_big5 (src, src_end) | |
2144 | unsigned char *src, *src_end; | |
2145 | { | |
2146 | unsigned char c; | |
2147 | ||
2148 | while (src < src_end) | |
2149 | { | |
2150 | c = *src++; | |
4ed46869 KH |
2151 | if (c >= 0xA1) |
2152 | { | |
2153 | if (src >= src_end) | |
2154 | break; | |
2155 | c = *src++; | |
2156 | if (c < 0x40 || (c >= 0x7F && c <= 0xA0)) | |
2157 | return 0; | |
2158 | } | |
2159 | } | |
2160 | return CODING_CATEGORY_MASK_BIG5; | |
2161 | } | |
2162 | ||
2163 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". | |
2164 | If SJIS_P is 1, decode SJIS text, else decode BIG5 test. */ | |
2165 | ||
2166 | int | |
2167 | decode_coding_sjis_big5 (coding, source, destination, | |
d46c5b12 | 2168 | src_bytes, dst_bytes, sjis_p) |
4ed46869 KH |
2169 | struct coding_system *coding; |
2170 | unsigned char *source, *destination; | |
2171 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2172 | int sjis_p; |
2173 | { | |
2174 | unsigned char *src = source; | |
2175 | unsigned char *src_end = source + src_bytes; | |
2176 | unsigned char *dst = destination; | |
2177 | unsigned char *dst_end = destination + dst_bytes; | |
2178 | /* Since the maximum bytes produced by each loop is 4, we subtract 3 | |
2179 | from DST_END to assure overflow checking is necessary only at the | |
2180 | head of loop. */ | |
2181 | unsigned char *adjusted_dst_end = dst_end - 3; | |
84fbb8a0 | 2182 | Lisp_Object translation_table |
f967223b | 2183 | = coding->translation_table_for_decode; |
d46c5b12 | 2184 | int result = CODING_FINISH_NORMAL; |
a5d301df | 2185 | |
84fbb8a0 | 2186 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
f967223b | 2187 | translation_table = Vstandard_translation_table_for_decode; |
4ed46869 | 2188 | |
d46c5b12 | 2189 | coding->produced_char = 0; |
fb88bf2d | 2190 | coding->fake_multibyte = 0; |
d46c5b12 KH |
2191 | while (src < src_end && (dst_bytes |
2192 | ? (dst < adjusted_dst_end) | |
2193 | : (dst < src - 3))) | |
4ed46869 KH |
2194 | { |
2195 | /* SRC_BASE remembers the start position in source in each loop. | |
2196 | The loop will be exited when there's not enough source text | |
2197 | to analyze two-byte character (within macro ONE_MORE_BYTE). | |
2198 | In that case, SRC is reset to SRC_BASE before exiting. */ | |
2199 | unsigned char *src_base = src; | |
2200 | unsigned char c1 = *src++, c2, c3, c4; | |
2201 | ||
d46c5b12 | 2202 | if (c1 < 0x20) |
4ed46869 | 2203 | { |
d46c5b12 | 2204 | if (c1 == '\r') |
4ed46869 | 2205 | { |
d46c5b12 KH |
2206 | if (coding->eol_type == CODING_EOL_CRLF) |
2207 | { | |
2208 | ONE_MORE_BYTE (c2); | |
2209 | if (c2 == '\n') | |
2210 | *dst++ = c2; | |
2211 | else if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
2212 | { | |
2213 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2214 | goto label_end_of_loop_2; | |
2215 | } | |
2216 | else | |
2217 | /* To process C2 again, SRC is subtracted by 1. */ | |
2218 | *dst++ = c1, src--; | |
2219 | } | |
2220 | else if (coding->eol_type == CODING_EOL_CR) | |
2221 | *dst++ = '\n'; | |
4ed46869 | 2222 | else |
d46c5b12 KH |
2223 | *dst++ = c1; |
2224 | } | |
2225 | else if (c1 == '\n' | |
2226 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
2227 | && (coding->eol_type == CODING_EOL_CR | |
2228 | || coding->eol_type == CODING_EOL_CRLF)) | |
2229 | { | |
2230 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2231 | goto label_end_of_loop_2; | |
4ed46869 KH |
2232 | } |
2233 | else | |
2234 | *dst++ = c1; | |
d46c5b12 | 2235 | coding->produced_char++; |
4ed46869 | 2236 | } |
a5d301df KH |
2237 | else if (c1 < 0x80) |
2238 | DECODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); | |
54f78171 | 2239 | else |
4ed46869 | 2240 | { |
4ed46869 KH |
2241 | if (sjis_p) |
2242 | { | |
54f78171 | 2243 | if (c1 < 0xA0 || (c1 >= 0xE0 && c1 < 0xF0)) |
fb88bf2d | 2244 | { |
54f78171 KH |
2245 | /* SJIS -> JISX0208 */ |
2246 | ONE_MORE_BYTE (c2); | |
2247 | if (c2 >= 0x40) | |
2248 | { | |
2249 | DECODE_SJIS (c1, c2, c3, c4); | |
2250 | DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208, c3, c4); | |
2251 | } | |
2252 | else | |
2253 | goto label_invalid_code_2; | |
fb88bf2d | 2254 | } |
54f78171 KH |
2255 | else if (c1 < 0xE0) |
2256 | /* SJIS -> JISX0201-Kana */ | |
2257 | DECODE_SJIS_BIG5_CHARACTER (charset_katakana_jisx0201, c1, | |
2258 | /* dummy */ c2); | |
fb88bf2d | 2259 | else |
54f78171 | 2260 | goto label_invalid_code_1; |
4ed46869 | 2261 | } |
fb88bf2d | 2262 | else |
fb88bf2d | 2263 | { |
54f78171 KH |
2264 | /* BIG5 -> Big5 */ |
2265 | if (c1 >= 0xA1 && c1 <= 0xFE) | |
fb88bf2d | 2266 | { |
54f78171 KH |
2267 | ONE_MORE_BYTE (c2); |
2268 | if ((c2 >= 0x40 && c2 <= 0x7E) || (c2 >= 0xA1 && c2 <= 0xFE)) | |
2269 | { | |
2270 | int charset; | |
4ed46869 | 2271 | |
54f78171 KH |
2272 | DECODE_BIG5 (c1, c2, charset, c3, c4); |
2273 | DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4); | |
2274 | } | |
2275 | else | |
2276 | goto label_invalid_code_2; | |
fb88bf2d KH |
2277 | } |
2278 | else | |
54f78171 | 2279 | goto label_invalid_code_1; |
4ed46869 KH |
2280 | } |
2281 | } | |
2282 | continue; | |
2283 | ||
fb88bf2d KH |
2284 | label_invalid_code_1: |
2285 | *dst++ = c1; | |
2286 | coding->produced_char++; | |
2287 | coding->fake_multibyte = 1; | |
2288 | continue; | |
2289 | ||
2290 | label_invalid_code_2: | |
2291 | *dst++ = c1; *dst++= c2; | |
2292 | coding->produced_char += 2; | |
2293 | coding->fake_multibyte = 1; | |
2294 | continue; | |
2295 | ||
4ed46869 | 2296 | label_end_of_loop: |
d46c5b12 KH |
2297 | result = CODING_FINISH_INSUFFICIENT_SRC; |
2298 | label_end_of_loop_2: | |
4ed46869 KH |
2299 | src = src_base; |
2300 | break; | |
2301 | } | |
2302 | ||
fb88bf2d KH |
2303 | if (src < src_end) |
2304 | { | |
2305 | if (result == CODING_FINISH_NORMAL) | |
2306 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2307 | else if (result != CODING_FINISH_INCONSISTENT_EOL | |
2308 | && coding->mode & CODING_MODE_LAST_BLOCK) | |
2309 | { | |
2310 | src_bytes = src_end - src; | |
2311 | if (dst_bytes && (dst_end - dst < src_bytes)) | |
2312 | src_bytes = dst_end - dst; | |
2313 | bcopy (dst, src, src_bytes); | |
2314 | src += src_bytes; | |
2315 | dst += src_bytes; | |
2316 | coding->fake_multibyte = 1; | |
2317 | } | |
2318 | } | |
d46c5b12 KH |
2319 | |
2320 | coding->consumed = coding->consumed_char = src - source; | |
2321 | coding->produced = dst - destination; | |
2322 | return result; | |
4ed46869 KH |
2323 | } |
2324 | ||
2325 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". | |
2326 | This function can encode `charset_ascii', `charset_katakana_jisx0201', | |
2327 | `charset_jisx0208', `charset_big5_1', and `charset_big5-2'. We are | |
2328 | sure that all these charsets are registered as official charset | |
2329 | (i.e. do not have extended leading-codes). Characters of other | |
2330 | charsets are produced without any encoding. If SJIS_P is 1, encode | |
2331 | SJIS text, else encode BIG5 text. */ | |
2332 | ||
2333 | int | |
2334 | encode_coding_sjis_big5 (coding, source, destination, | |
d46c5b12 | 2335 | src_bytes, dst_bytes, sjis_p) |
4ed46869 KH |
2336 | struct coding_system *coding; |
2337 | unsigned char *source, *destination; | |
2338 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2339 | int sjis_p; |
2340 | { | |
2341 | unsigned char *src = source; | |
2342 | unsigned char *src_end = source + src_bytes; | |
2343 | unsigned char *dst = destination; | |
2344 | unsigned char *dst_end = destination + dst_bytes; | |
2345 | /* Since the maximum bytes produced by each loop is 2, we subtract 1 | |
2346 | from DST_END to assure overflow checking is necessary only at the | |
2347 | head of loop. */ | |
2348 | unsigned char *adjusted_dst_end = dst_end - 1; | |
84fbb8a0 | 2349 | Lisp_Object translation_table |
f967223b | 2350 | = coding->translation_table_for_encode; |
d46c5b12 | 2351 | int result = CODING_FINISH_NORMAL; |
a5d301df | 2352 | |
84fbb8a0 | 2353 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
f967223b | 2354 | translation_table = Vstandard_translation_table_for_encode; |
4ed46869 | 2355 | |
d46c5b12 | 2356 | coding->consumed_char = 0; |
fb88bf2d | 2357 | coding->fake_multibyte = 0; |
d46c5b12 KH |
2358 | while (src < src_end && (dst_bytes |
2359 | ? (dst < adjusted_dst_end) | |
2360 | : (dst < src - 1))) | |
4ed46869 KH |
2361 | { |
2362 | /* SRC_BASE remembers the start position in source in each loop. | |
2363 | The loop will be exited when there's not enough source text | |
2364 | to analyze multi-byte codes (within macros ONE_MORE_BYTE and | |
2365 | TWO_MORE_BYTES). In that case, SRC is reset to SRC_BASE | |
2366 | before exiting. */ | |
2367 | unsigned char *src_base = src; | |
2368 | unsigned char c1 = *src++, c2, c3, c4; | |
2369 | ||
2370 | if (coding->composing) | |
2371 | { | |
2372 | if (c1 == 0xA0) | |
2373 | { | |
2374 | ONE_MORE_BYTE (c1); | |
2375 | c1 &= 0x7F; | |
2376 | } | |
2377 | else if (c1 >= 0xA0) | |
2378 | c1 -= 0x20; | |
2379 | else | |
2380 | coding->composing = 0; | |
2381 | } | |
2382 | ||
2383 | switch (emacs_code_class[c1]) | |
2384 | { | |
2385 | case EMACS_ascii_code: | |
a5d301df KH |
2386 | ENCODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); |
2387 | break; | |
2388 | ||
4ed46869 KH |
2389 | case EMACS_control_code: |
2390 | *dst++ = c1; | |
d46c5b12 | 2391 | coding->consumed_char++; |
4ed46869 KH |
2392 | break; |
2393 | ||
2394 | case EMACS_carriage_return_code: | |
d46c5b12 | 2395 | if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
4ed46869 KH |
2396 | { |
2397 | *dst++ = c1; | |
d46c5b12 | 2398 | coding->consumed_char++; |
4ed46869 KH |
2399 | break; |
2400 | } | |
2401 | /* fall down to treat '\r' as '\n' ... */ | |
2402 | ||
2403 | case EMACS_linefeed_code: | |
2404 | if (coding->eol_type == CODING_EOL_LF | |
0ef69138 | 2405 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
2406 | *dst++ = '\n'; |
2407 | else if (coding->eol_type == CODING_EOL_CRLF) | |
2408 | *dst++ = '\r', *dst++ = '\n'; | |
2409 | else | |
2410 | *dst++ = '\r'; | |
d46c5b12 | 2411 | coding->consumed_char++; |
4ed46869 KH |
2412 | break; |
2413 | ||
2414 | case EMACS_leading_code_2: | |
2415 | ONE_MORE_BYTE (c2); | |
a5d301df | 2416 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, /* dummy */ c3); |
4ed46869 KH |
2417 | break; |
2418 | ||
2419 | case EMACS_leading_code_3: | |
2420 | TWO_MORE_BYTES (c2, c3); | |
a5d301df | 2421 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, c3); |
4ed46869 KH |
2422 | break; |
2423 | ||
2424 | case EMACS_leading_code_4: | |
2425 | THREE_MORE_BYTES (c2, c3, c4); | |
a5d301df | 2426 | ENCODE_SJIS_BIG5_CHARACTER (c2, c3, c4); |
4ed46869 KH |
2427 | break; |
2428 | ||
2429 | case EMACS_leading_code_composition: | |
2430 | coding->composing = 1; | |
2431 | break; | |
2432 | ||
2433 | default: /* i.e. case EMACS_invalid_code: */ | |
2434 | *dst++ = c1; | |
d46c5b12 | 2435 | coding->consumed_char++; |
4ed46869 KH |
2436 | } |
2437 | continue; | |
2438 | ||
2439 | label_end_of_loop: | |
d46c5b12 KH |
2440 | result = CODING_FINISH_INSUFFICIENT_SRC; |
2441 | src = src_base; | |
4ed46869 KH |
2442 | break; |
2443 | } | |
2444 | ||
d46c5b12 KH |
2445 | if (result == CODING_FINISH_NORMAL |
2446 | && src < src_end) | |
2447 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2448 | coding->consumed = src - source; | |
2449 | coding->produced = coding->produced_char = dst - destination; | |
2450 | return result; | |
4ed46869 KH |
2451 | } |
2452 | ||
2453 | \f | |
2454 | /*** 5. End-of-line handlers ***/ | |
2455 | ||
2456 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". | |
2457 | This function is called only when `coding->eol_type' is | |
2458 | CODING_EOL_CRLF or CODING_EOL_CR. */ | |
2459 | ||
dfcf069d | 2460 | int |
d46c5b12 | 2461 | decode_eol (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
2462 | struct coding_system *coding; |
2463 | unsigned char *source, *destination; | |
2464 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2465 | { |
2466 | unsigned char *src = source; | |
2467 | unsigned char *src_end = source + src_bytes; | |
2468 | unsigned char *dst = destination; | |
2469 | unsigned char *dst_end = destination + dst_bytes; | |
fb88bf2d | 2470 | unsigned char c; |
d46c5b12 KH |
2471 | int result = CODING_FINISH_NORMAL; |
2472 | ||
fb88bf2d KH |
2473 | coding->fake_multibyte = 0; |
2474 | ||
d46c5b12 KH |
2475 | if (src_bytes <= 0) |
2476 | return result; | |
4ed46869 KH |
2477 | |
2478 | switch (coding->eol_type) | |
2479 | { | |
2480 | case CODING_EOL_CRLF: | |
2481 | { | |
2482 | /* Since the maximum bytes produced by each loop is 2, we | |
2483 | subtract 1 from DST_END to assure overflow checking is | |
2484 | necessary only at the head of loop. */ | |
2485 | unsigned char *adjusted_dst_end = dst_end - 1; | |
2486 | ||
d46c5b12 KH |
2487 | while (src < src_end && (dst_bytes |
2488 | ? (dst < adjusted_dst_end) | |
2489 | : (dst < src - 1))) | |
4ed46869 KH |
2490 | { |
2491 | unsigned char *src_base = src; | |
fb88bf2d KH |
2492 | |
2493 | c = *src++; | |
4ed46869 KH |
2494 | if (c == '\r') |
2495 | { | |
2496 | ONE_MORE_BYTE (c); | |
2497 | if (c != '\n') | |
d46c5b12 KH |
2498 | { |
2499 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) | |
2500 | { | |
2501 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2502 | goto label_end_of_loop_2; | |
2503 | } | |
2504 | *dst++ = '\r'; | |
fb88bf2d KH |
2505 | if (BASE_LEADING_CODE_P (c)) |
2506 | coding->fake_multibyte = 1; | |
d46c5b12 | 2507 | } |
bfd99048 | 2508 | *dst++ = c; |
4ed46869 | 2509 | } |
d46c5b12 KH |
2510 | else if (c == '\n' |
2511 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)) | |
2512 | { | |
2513 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2514 | goto label_end_of_loop_2; | |
2515 | } | |
4ed46869 | 2516 | else |
fb88bf2d KH |
2517 | { |
2518 | *dst++ = c; | |
2519 | if (BASE_LEADING_CODE_P (c)) | |
2520 | coding->fake_multibyte = 1; | |
2521 | } | |
4ed46869 KH |
2522 | continue; |
2523 | ||
2524 | label_end_of_loop: | |
d46c5b12 KH |
2525 | result = CODING_FINISH_INSUFFICIENT_SRC; |
2526 | label_end_of_loop_2: | |
4ed46869 KH |
2527 | src = src_base; |
2528 | break; | |
2529 | } | |
d46c5b12 KH |
2530 | if (result == CODING_FINISH_NORMAL |
2531 | && src < src_end) | |
2532 | result = CODING_FINISH_INSUFFICIENT_DST; | |
4ed46869 | 2533 | } |
d46c5b12 | 2534 | break; |
4ed46869 KH |
2535 | |
2536 | case CODING_EOL_CR: | |
d46c5b12 KH |
2537 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
2538 | { | |
fb88bf2d KH |
2539 | while (src < src_end) |
2540 | { | |
2541 | if ((c = *src++) == '\n') | |
2542 | break; | |
2543 | if (BASE_LEADING_CODE_P (c)) | |
2544 | coding->fake_multibyte = 1; | |
2545 | } | |
d46c5b12 KH |
2546 | if (*--src == '\n') |
2547 | { | |
2548 | src_bytes = src - source; | |
2549 | result = CODING_FINISH_INCONSISTENT_EOL; | |
2550 | } | |
2551 | } | |
2552 | if (dst_bytes && src_bytes > dst_bytes) | |
2553 | { | |
2554 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2555 | src_bytes = dst_bytes; | |
2556 | } | |
2557 | if (dst_bytes) | |
2558 | bcopy (source, destination, src_bytes); | |
2559 | else | |
2560 | safe_bcopy (source, destination, src_bytes); | |
2561 | src = source + src_bytes; | |
2562 | while (src_bytes--) if (*dst++ == '\r') dst[-1] = '\n'; | |
4ed46869 KH |
2563 | break; |
2564 | ||
2565 | default: /* i.e. case: CODING_EOL_LF */ | |
d46c5b12 KH |
2566 | if (dst_bytes && src_bytes > dst_bytes) |
2567 | { | |
2568 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2569 | src_bytes = dst_bytes; | |
2570 | } | |
2571 | if (dst_bytes) | |
2572 | bcopy (source, destination, src_bytes); | |
2573 | else | |
2574 | safe_bcopy (source, destination, src_bytes); | |
2575 | src += src_bytes; | |
993824c9 | 2576 | dst += src_bytes; |
fb88bf2d | 2577 | coding->fake_multibyte = 1; |
4ed46869 KH |
2578 | break; |
2579 | } | |
2580 | ||
d46c5b12 KH |
2581 | coding->consumed = coding->consumed_char = src - source; |
2582 | coding->produced = coding->produced_char = dst - destination; | |
2583 | return result; | |
4ed46869 KH |
2584 | } |
2585 | ||
2586 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". Encode | |
2587 | format of end-of-line according to `coding->eol_type'. If | |
d46c5b12 KH |
2588 | `coding->mode & CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code |
2589 | '\r' in source text also means end-of-line. */ | |
4ed46869 | 2590 | |
dfcf069d | 2591 | int |
d46c5b12 | 2592 | encode_eol (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
2593 | struct coding_system *coding; |
2594 | unsigned char *source, *destination; | |
2595 | int src_bytes, dst_bytes; | |
4ed46869 KH |
2596 | { |
2597 | unsigned char *src = source; | |
2598 | unsigned char *dst = destination; | |
d46c5b12 | 2599 | int result = CODING_FINISH_NORMAL; |
4ed46869 | 2600 | |
fb88bf2d KH |
2601 | coding->fake_multibyte = 0; |
2602 | ||
d46c5b12 KH |
2603 | if (coding->eol_type == CODING_EOL_CRLF) |
2604 | { | |
2605 | unsigned char c; | |
2606 | unsigned char *src_end = source + src_bytes; | |
2607 | unsigned char *dst_end = destination + dst_bytes; | |
2608 | /* Since the maximum bytes produced by each loop is 2, we | |
2609 | subtract 1 from DST_END to assure overflow checking is | |
2610 | necessary only at the head of loop. */ | |
2611 | unsigned char *adjusted_dst_end = dst_end - 1; | |
2612 | ||
2613 | while (src < src_end && (dst_bytes | |
2614 | ? (dst < adjusted_dst_end) | |
2615 | : (dst < src - 1))) | |
2616 | { | |
2617 | c = *src++; | |
2618 | if (c == '\n' | |
2619 | || (c == '\r' && (coding->mode & CODING_MODE_SELECTIVE_DISPLAY))) | |
2620 | *dst++ = '\r', *dst++ = '\n'; | |
2621 | else | |
fb88bf2d KH |
2622 | { |
2623 | *dst++ = c; | |
2624 | if (BASE_LEADING_CODE_P (c)) | |
2625 | coding->fake_multibyte = 1; | |
2626 | } | |
d46c5b12 KH |
2627 | } |
2628 | if (src < src_end) | |
2629 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2630 | } | |
2631 | else | |
4ed46869 | 2632 | { |
fb88bf2d KH |
2633 | unsigned char c; |
2634 | ||
d46c5b12 | 2635 | if (dst_bytes && src_bytes > dst_bytes) |
4ed46869 | 2636 | { |
d46c5b12 KH |
2637 | src_bytes = dst_bytes; |
2638 | result = CODING_FINISH_INSUFFICIENT_DST; | |
2639 | } | |
2640 | if (dst_bytes) | |
2641 | bcopy (source, destination, src_bytes); | |
2642 | else | |
993824c9 RS |
2643 | safe_bcopy (source, destination, src_bytes); |
2644 | dst_bytes = src_bytes; | |
2645 | if (coding->eol_type == CODING_EOL_CR) | |
d46c5b12 KH |
2646 | { |
2647 | while (src_bytes--) | |
fb88bf2d KH |
2648 | { |
2649 | if ((c = *dst++) == '\n') | |
2650 | dst[-1] = '\r'; | |
2651 | else if (BASE_LEADING_CODE_P (c)) | |
993824c9 | 2652 | coding->fake_multibyte = 1; |
fb88bf2d | 2653 | } |
d46c5b12 | 2654 | } |
fb88bf2d | 2655 | else |
d46c5b12 | 2656 | { |
fb88bf2d KH |
2657 | if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY) |
2658 | { | |
2659 | while (src_bytes--) | |
2660 | if (*dst++ == '\r') dst[-1] = '\n'; | |
2661 | } | |
2662 | coding->fake_multibyte = 1; | |
4ed46869 | 2663 | } |
fb88bf2d KH |
2664 | src = source + dst_bytes; |
2665 | dst = destination + dst_bytes; | |
4ed46869 KH |
2666 | } |
2667 | ||
d46c5b12 KH |
2668 | coding->consumed = coding->consumed_char = src - source; |
2669 | coding->produced = coding->produced_char = dst - destination; | |
2670 | return result; | |
4ed46869 KH |
2671 | } |
2672 | ||
2673 | \f | |
2674 | /*** 6. C library functions ***/ | |
2675 | ||
2676 | /* In Emacs Lisp, coding system is represented by a Lisp symbol which | |
2677 | has a property `coding-system'. The value of this property is a | |
2678 | vector of length 5 (called as coding-vector). Among elements of | |
2679 | this vector, the first (element[0]) and the fifth (element[4]) | |
2680 | carry important information for decoding/encoding. Before | |
2681 | decoding/encoding, this information should be set in fields of a | |
2682 | structure of type `coding_system'. | |
2683 | ||
2684 | A value of property `coding-system' can be a symbol of another | |
2685 | subsidiary coding-system. In that case, Emacs gets coding-vector | |
2686 | from that symbol. | |
2687 | ||
2688 | `element[0]' contains information to be set in `coding->type'. The | |
2689 | value and its meaning is as follows: | |
2690 | ||
0ef69138 KH |
2691 | 0 -- coding_type_emacs_mule |
2692 | 1 -- coding_type_sjis | |
2693 | 2 -- coding_type_iso2022 | |
2694 | 3 -- coding_type_big5 | |
2695 | 4 -- coding_type_ccl encoder/decoder written in CCL | |
2696 | nil -- coding_type_no_conversion | |
2697 | t -- coding_type_undecided (automatic conversion on decoding, | |
2698 | no-conversion on encoding) | |
4ed46869 KH |
2699 | |
2700 | `element[4]' contains information to be set in `coding->flags' and | |
2701 | `coding->spec'. The meaning varies by `coding->type'. | |
2702 | ||
2703 | If `coding->type' is `coding_type_iso2022', element[4] is a vector | |
2704 | of length 32 (of which the first 13 sub-elements are used now). | |
2705 | Meanings of these sub-elements are: | |
2706 | ||
2707 | sub-element[N] where N is 0 through 3: to be set in `coding->spec.iso2022' | |
2708 | If the value is an integer of valid charset, the charset is | |
2709 | assumed to be designated to graphic register N initially. | |
2710 | ||
2711 | If the value is minus, it is a minus value of charset which | |
2712 | reserves graphic register N, which means that the charset is | |
2713 | not designated initially but should be designated to graphic | |
2714 | register N just before encoding a character in that charset. | |
2715 | ||
2716 | If the value is nil, graphic register N is never used on | |
2717 | encoding. | |
2718 | ||
2719 | sub-element[N] where N is 4 through 11: to be set in `coding->flags' | |
2720 | Each value takes t or nil. See the section ISO2022 of | |
2721 | `coding.h' for more information. | |
2722 | ||
2723 | If `coding->type' is `coding_type_big5', element[4] is t to denote | |
2724 | BIG5-ETen or nil to denote BIG5-HKU. | |
2725 | ||
2726 | If `coding->type' takes the other value, element[4] is ignored. | |
2727 | ||
2728 | Emacs Lisp's coding system also carries information about format of | |
2729 | end-of-line in a value of property `eol-type'. If the value is | |
2730 | integer, 0 means CODING_EOL_LF, 1 means CODING_EOL_CRLF, and 2 | |
2731 | means CODING_EOL_CR. If it is not integer, it should be a vector | |
2732 | of subsidiary coding systems of which property `eol-type' has one | |
2733 | of above values. | |
2734 | ||
2735 | */ | |
2736 | ||
2737 | /* Extract information for decoding/encoding from CODING_SYSTEM_SYMBOL | |
2738 | and set it in CODING. If CODING_SYSTEM_SYMBOL is invalid, CODING | |
2739 | is setup so that no conversion is necessary and return -1, else | |
2740 | return 0. */ | |
2741 | ||
2742 | int | |
e0e989f6 KH |
2743 | setup_coding_system (coding_system, coding) |
2744 | Lisp_Object coding_system; | |
4ed46869 KH |
2745 | struct coding_system *coding; |
2746 | { | |
d46c5b12 | 2747 | Lisp_Object coding_spec, coding_type, eol_type, plist; |
4608c386 | 2748 | Lisp_Object val; |
70c22245 | 2749 | int i; |
4ed46869 | 2750 | |
d46c5b12 | 2751 | /* Initialize some fields required for all kinds of coding systems. */ |
774324d6 | 2752 | coding->symbol = coding_system; |
d46c5b12 KH |
2753 | coding->common_flags = 0; |
2754 | coding->mode = 0; | |
2755 | coding->heading_ascii = -1; | |
2756 | coding->post_read_conversion = coding->pre_write_conversion = Qnil; | |
4608c386 KH |
2757 | coding_spec = Fget (coding_system, Qcoding_system); |
2758 | if (!VECTORP (coding_spec) | |
2759 | || XVECTOR (coding_spec)->size != 5 | |
2760 | || !CONSP (XVECTOR (coding_spec)->contents[3])) | |
4ed46869 | 2761 | goto label_invalid_coding_system; |
4608c386 | 2762 | |
d46c5b12 KH |
2763 | eol_type = inhibit_eol_conversion ? Qnil : Fget (coding_system, Qeol_type); |
2764 | if (VECTORP (eol_type)) | |
2765 | { | |
2766 | coding->eol_type = CODING_EOL_UNDECIDED; | |
2767 | coding->common_flags = CODING_REQUIRE_DETECTION_MASK; | |
2768 | } | |
2769 | else if (XFASTINT (eol_type) == 1) | |
2770 | { | |
2771 | coding->eol_type = CODING_EOL_CRLF; | |
2772 | coding->common_flags | |
2773 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
2774 | } | |
2775 | else if (XFASTINT (eol_type) == 2) | |
2776 | { | |
2777 | coding->eol_type = CODING_EOL_CR; | |
2778 | coding->common_flags | |
2779 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
2780 | } | |
2781 | else | |
2782 | coding->eol_type = CODING_EOL_LF; | |
2783 | ||
2784 | coding_type = XVECTOR (coding_spec)->contents[0]; | |
2785 | /* Try short cut. */ | |
2786 | if (SYMBOLP (coding_type)) | |
2787 | { | |
2788 | if (EQ (coding_type, Qt)) | |
2789 | { | |
2790 | coding->type = coding_type_undecided; | |
2791 | coding->common_flags |= CODING_REQUIRE_DETECTION_MASK; | |
2792 | } | |
2793 | else | |
2794 | coding->type = coding_type_no_conversion; | |
2795 | return 0; | |
2796 | } | |
2797 | ||
2798 | /* Initialize remaining fields. */ | |
2799 | coding->composing = 0; | |
f967223b KH |
2800 | coding->translation_table_for_decode = Qnil; |
2801 | coding->translation_table_for_encode = Qnil; | |
d46c5b12 KH |
2802 | |
2803 | /* Get values of coding system properties: | |
2804 | `post-read-conversion', `pre-write-conversion', | |
f967223b | 2805 | `translation-table-for-decode', `translation-table-for-encode'. */ |
4608c386 KH |
2806 | plist = XVECTOR (coding_spec)->contents[3]; |
2807 | coding->post_read_conversion = Fplist_get (plist, Qpost_read_conversion); | |
2808 | coding->pre_write_conversion = Fplist_get (plist, Qpre_write_conversion); | |
f967223b | 2809 | val = Fplist_get (plist, Qtranslation_table_for_decode); |
4608c386 | 2810 | if (SYMBOLP (val)) |
f967223b KH |
2811 | val = Fget (val, Qtranslation_table_for_decode); |
2812 | coding->translation_table_for_decode = CHAR_TABLE_P (val) ? val : Qnil; | |
2813 | val = Fplist_get (plist, Qtranslation_table_for_encode); | |
4608c386 | 2814 | if (SYMBOLP (val)) |
f967223b KH |
2815 | val = Fget (val, Qtranslation_table_for_encode); |
2816 | coding->translation_table_for_encode = CHAR_TABLE_P (val) ? val : Qnil; | |
d46c5b12 KH |
2817 | val = Fplist_get (plist, Qcoding_category); |
2818 | if (!NILP (val)) | |
2819 | { | |
2820 | val = Fget (val, Qcoding_category_index); | |
2821 | if (INTEGERP (val)) | |
2822 | coding->category_idx = XINT (val); | |
2823 | else | |
2824 | goto label_invalid_coding_system; | |
2825 | } | |
2826 | else | |
2827 | goto label_invalid_coding_system; | |
4608c386 | 2828 | |
70c22245 KH |
2829 | val = Fplist_get (plist, Qsafe_charsets); |
2830 | if (EQ (val, Qt)) | |
2831 | { | |
2832 | for (i = 0; i <= MAX_CHARSET; i++) | |
2833 | coding->safe_charsets[i] = 1; | |
2834 | } | |
2835 | else | |
2836 | { | |
2837 | bzero (coding->safe_charsets, MAX_CHARSET + 1); | |
2838 | while (CONSP (val)) | |
2839 | { | |
2840 | if ((i = get_charset_id (XCONS (val)->car)) >= 0) | |
2841 | coding->safe_charsets[i] = 1; | |
2842 | val = XCONS (val)->cdr; | |
2843 | } | |
2844 | } | |
2845 | ||
d46c5b12 | 2846 | switch (XFASTINT (coding_type)) |
4ed46869 KH |
2847 | { |
2848 | case 0: | |
0ef69138 | 2849 | coding->type = coding_type_emacs_mule; |
c952af22 KH |
2850 | if (!NILP (coding->post_read_conversion)) |
2851 | coding->common_flags |= CODING_REQUIRE_DECODING_MASK; | |
2852 | if (!NILP (coding->pre_write_conversion)) | |
2853 | coding->common_flags |= CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 KH |
2854 | break; |
2855 | ||
2856 | case 1: | |
2857 | coding->type = coding_type_sjis; | |
c952af22 KH |
2858 | coding->common_flags |
2859 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 KH |
2860 | break; |
2861 | ||
2862 | case 2: | |
2863 | coding->type = coding_type_iso2022; | |
c952af22 KH |
2864 | coding->common_flags |
2865 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 | 2866 | { |
70c22245 | 2867 | Lisp_Object val, temp; |
4ed46869 | 2868 | Lisp_Object *flags; |
d46c5b12 | 2869 | int i, charset, reg_bits = 0; |
4ed46869 | 2870 | |
4608c386 | 2871 | val = XVECTOR (coding_spec)->contents[4]; |
f44d27ce | 2872 | |
4ed46869 KH |
2873 | if (!VECTORP (val) || XVECTOR (val)->size != 32) |
2874 | goto label_invalid_coding_system; | |
2875 | ||
2876 | flags = XVECTOR (val)->contents; | |
2877 | coding->flags | |
2878 | = ((NILP (flags[4]) ? 0 : CODING_FLAG_ISO_SHORT_FORM) | |
2879 | | (NILP (flags[5]) ? 0 : CODING_FLAG_ISO_RESET_AT_EOL) | |
2880 | | (NILP (flags[6]) ? 0 : CODING_FLAG_ISO_RESET_AT_CNTL) | |
2881 | | (NILP (flags[7]) ? 0 : CODING_FLAG_ISO_SEVEN_BITS) | |
2882 | | (NILP (flags[8]) ? 0 : CODING_FLAG_ISO_LOCKING_SHIFT) | |
2883 | | (NILP (flags[9]) ? 0 : CODING_FLAG_ISO_SINGLE_SHIFT) | |
2884 | | (NILP (flags[10]) ? 0 : CODING_FLAG_ISO_USE_ROMAN) | |
2885 | | (NILP (flags[11]) ? 0 : CODING_FLAG_ISO_USE_OLDJIS) | |
e0e989f6 KH |
2886 | | (NILP (flags[12]) ? 0 : CODING_FLAG_ISO_NO_DIRECTION) |
2887 | | (NILP (flags[13]) ? 0 : CODING_FLAG_ISO_INIT_AT_BOL) | |
c4825358 KH |
2888 | | (NILP (flags[14]) ? 0 : CODING_FLAG_ISO_DESIGNATE_AT_BOL) |
2889 | | (NILP (flags[15]) ? 0 : CODING_FLAG_ISO_SAFE) | |
3f003981 | 2890 | | (NILP (flags[16]) ? 0 : CODING_FLAG_ISO_LATIN_EXTRA) |
c4825358 | 2891 | ); |
4ed46869 KH |
2892 | |
2893 | /* Invoke graphic register 0 to plane 0. */ | |
2894 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; | |
2895 | /* Invoke graphic register 1 to plane 1 if we can use full 8-bit. */ | |
2896 | CODING_SPEC_ISO_INVOCATION (coding, 1) | |
2897 | = (coding->flags & CODING_FLAG_ISO_SEVEN_BITS ? -1 : 1); | |
2898 | /* Not single shifting at first. */ | |
6e85d753 | 2899 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; |
e0e989f6 | 2900 | /* Beginning of buffer should also be regarded as bol. */ |
6e85d753 | 2901 | CODING_SPEC_ISO_BOL (coding) = 1; |
4ed46869 | 2902 | |
70c22245 KH |
2903 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
2904 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = 255; | |
2905 | val = Vcharset_revision_alist; | |
2906 | while (CONSP (val)) | |
2907 | { | |
2908 | charset = get_charset_id (Fcar_safe (XCONS (val)->car)); | |
2909 | if (charset >= 0 | |
2910 | && (temp = Fcdr_safe (XCONS (val)->car), INTEGERP (temp)) | |
2911 | && (i = XINT (temp), (i >= 0 && (i + '@') < 128))) | |
2912 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = i; | |
2913 | val = XCONS (val)->cdr; | |
2914 | } | |
2915 | ||
4ed46869 KH |
2916 | /* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations. |
2917 | FLAGS[REG] can be one of below: | |
2918 | integer CHARSET: CHARSET occupies register I, | |
2919 | t: designate nothing to REG initially, but can be used | |
2920 | by any charsets, | |
2921 | list of integer, nil, or t: designate the first | |
2922 | element (if integer) to REG initially, the remaining | |
2923 | elements (if integer) is designated to REG on request, | |
d46c5b12 | 2924 | if an element is t, REG can be used by any charsets, |
4ed46869 | 2925 | nil: REG is never used. */ |
467e7675 | 2926 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
1ba9e4ab KH |
2927 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
2928 | = CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION; | |
4ed46869 KH |
2929 | for (i = 0; i < 4; i++) |
2930 | { | |
2931 | if (INTEGERP (flags[i]) | |
e0e989f6 KH |
2932 | && (charset = XINT (flags[i]), CHARSET_VALID_P (charset)) |
2933 | || (charset = get_charset_id (flags[i])) >= 0) | |
4ed46869 KH |
2934 | { |
2935 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; | |
2936 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) = i; | |
2937 | } | |
2938 | else if (EQ (flags[i], Qt)) | |
2939 | { | |
2940 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
d46c5b12 KH |
2941 | reg_bits |= 1 << i; |
2942 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; | |
4ed46869 KH |
2943 | } |
2944 | else if (CONSP (flags[i])) | |
2945 | { | |
2946 | Lisp_Object tail = flags[i]; | |
2947 | ||
d46c5b12 | 2948 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; |
4ed46869 KH |
2949 | if (INTEGERP (XCONS (tail)->car) |
2950 | && (charset = XINT (XCONS (tail)->car), | |
e0e989f6 KH |
2951 | CHARSET_VALID_P (charset)) |
2952 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) | |
4ed46869 KH |
2953 | { |
2954 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; | |
2955 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) =i; | |
2956 | } | |
2957 | else | |
2958 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
2959 | tail = XCONS (tail)->cdr; | |
2960 | while (CONSP (tail)) | |
2961 | { | |
2962 | if (INTEGERP (XCONS (tail)->car) | |
2963 | && (charset = XINT (XCONS (tail)->car), | |
e0e989f6 KH |
2964 | CHARSET_VALID_P (charset)) |
2965 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) | |
70c22245 KH |
2966 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
2967 | = i; | |
4ed46869 | 2968 | else if (EQ (XCONS (tail)->car, Qt)) |
d46c5b12 | 2969 | reg_bits |= 1 << i; |
4ed46869 KH |
2970 | tail = XCONS (tail)->cdr; |
2971 | } | |
2972 | } | |
2973 | else | |
2974 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
2975 | ||
2976 | CODING_SPEC_ISO_DESIGNATION (coding, i) | |
2977 | = CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i); | |
2978 | } | |
2979 | ||
d46c5b12 | 2980 | if (reg_bits && ! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
4ed46869 KH |
2981 | { |
2982 | /* REG 1 can be used only by locking shift in 7-bit env. */ | |
2983 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) | |
d46c5b12 | 2984 | reg_bits &= ~2; |
4ed46869 KH |
2985 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
2986 | /* Without any shifting, only REG 0 and 1 can be used. */ | |
d46c5b12 | 2987 | reg_bits &= 3; |
4ed46869 KH |
2988 | } |
2989 | ||
d46c5b12 KH |
2990 | if (reg_bits) |
2991 | for (charset = 0; charset <= MAX_CHARSET; charset++) | |
6e85d753 | 2992 | { |
d46c5b12 KH |
2993 | if (CHARSET_VALID_P (charset)) |
2994 | { | |
2995 | /* There exist some default graphic registers to be | |
2996 | used CHARSET. */ | |
2997 | ||
2998 | /* We had better avoid designating a charset of | |
2999 | CHARS96 to REG 0 as far as possible. */ | |
3000 | if (CHARSET_CHARS (charset) == 96) | |
3001 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) | |
3002 | = (reg_bits & 2 | |
3003 | ? 1 : (reg_bits & 4 ? 2 : (reg_bits & 8 ? 3 : 0))); | |
3004 | else | |
3005 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) | |
3006 | = (reg_bits & 1 | |
3007 | ? 0 : (reg_bits & 2 ? 1 : (reg_bits & 4 ? 2 : 3))); | |
3008 | } | |
6e85d753 | 3009 | } |
4ed46869 | 3010 | } |
c952af22 | 3011 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
d46c5b12 | 3012 | coding->spec.iso2022.last_invalid_designation_register = -1; |
4ed46869 KH |
3013 | break; |
3014 | ||
3015 | case 3: | |
3016 | coding->type = coding_type_big5; | |
c952af22 KH |
3017 | coding->common_flags |
3018 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 | 3019 | coding->flags |
4608c386 | 3020 | = (NILP (XVECTOR (coding_spec)->contents[4]) |
4ed46869 KH |
3021 | ? CODING_FLAG_BIG5_HKU |
3022 | : CODING_FLAG_BIG5_ETEN); | |
3023 | break; | |
3024 | ||
3025 | case 4: | |
3026 | coding->type = coding_type_ccl; | |
c952af22 KH |
3027 | coding->common_flags |
3028 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; | |
4ed46869 | 3029 | { |
4608c386 | 3030 | Lisp_Object val = XVECTOR (coding_spec)->contents[4]; |
d21ca14d KH |
3031 | Lisp_Object decoder, encoder; |
3032 | ||
4ed46869 | 3033 | if (CONSP (val) |
d21ca14d KH |
3034 | && SYMBOLP (XCONS (val)->car) |
3035 | && !NILP (decoder = Fget (XCONS (val)->car, Qccl_program_idx)) | |
f82423d7 | 3036 | && !NILP (decoder = Fcdr (Faref (Vccl_program_table, decoder))) |
d21ca14d KH |
3037 | && SYMBOLP (XCONS (val)->cdr) |
3038 | && !NILP (encoder = Fget (XCONS (val)->cdr, Qccl_program_idx)) | |
f82423d7 | 3039 | && !NILP (encoder = Fcdr (Faref (Vccl_program_table, encoder)))) |
4ed46869 | 3040 | { |
d21ca14d KH |
3041 | setup_ccl_program (&(coding->spec.ccl.decoder), decoder); |
3042 | setup_ccl_program (&(coding->spec.ccl.encoder), encoder); | |
4ed46869 KH |
3043 | } |
3044 | else | |
3045 | goto label_invalid_coding_system; | |
3046 | } | |
c952af22 | 3047 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
4ed46869 KH |
3048 | break; |
3049 | ||
27901516 KH |
3050 | case 5: |
3051 | coding->type = coding_type_raw_text; | |
3052 | break; | |
3053 | ||
4ed46869 | 3054 | default: |
d46c5b12 | 3055 | goto label_invalid_coding_system; |
4ed46869 KH |
3056 | } |
3057 | return 0; | |
3058 | ||
3059 | label_invalid_coding_system: | |
3060 | coding->type = coding_type_no_conversion; | |
d46c5b12 | 3061 | coding->category_idx = CODING_CATEGORY_IDX_BINARY; |
c952af22 | 3062 | coding->common_flags = 0; |
dec137e5 | 3063 | coding->eol_type = CODING_EOL_LF; |
d46c5b12 | 3064 | coding->pre_write_conversion = coding->post_read_conversion = Qnil; |
4ed46869 KH |
3065 | return -1; |
3066 | } | |
3067 | ||
54f78171 KH |
3068 | /* Setup raw-text or one of its subsidiaries in the structure |
3069 | coding_system CODING according to the already setup value eol_type | |
3070 | in CODING. CODING should be setup for some coding system in | |
3071 | advance. */ | |
3072 | ||
3073 | void | |
3074 | setup_raw_text_coding_system (coding) | |
3075 | struct coding_system *coding; | |
3076 | { | |
3077 | if (coding->type != coding_type_raw_text) | |
3078 | { | |
3079 | coding->symbol = Qraw_text; | |
3080 | coding->type = coding_type_raw_text; | |
3081 | if (coding->eol_type != CODING_EOL_UNDECIDED) | |
3082 | { | |
3083 | Lisp_Object subsidiaries = Fget (Qraw_text, Qeol_type); | |
3084 | ||
3085 | if (VECTORP (subsidiaries) | |
3086 | && XVECTOR (subsidiaries)->size == 3) | |
3087 | coding->symbol | |
3088 | = XVECTOR (subsidiaries)->contents[coding->eol_type]; | |
3089 | } | |
3090 | } | |
3091 | return; | |
3092 | } | |
3093 | ||
4ed46869 KH |
3094 | /* Emacs has a mechanism to automatically detect a coding system if it |
3095 | is one of Emacs' internal format, ISO2022, SJIS, and BIG5. But, | |
3096 | it's impossible to distinguish some coding systems accurately | |
3097 | because they use the same range of codes. So, at first, coding | |
3098 | systems are categorized into 7, those are: | |
3099 | ||
0ef69138 | 3100 | o coding-category-emacs-mule |
4ed46869 KH |
3101 | |
3102 | The category for a coding system which has the same code range | |
3103 | as Emacs' internal format. Assigned the coding-system (Lisp | |
0ef69138 | 3104 | symbol) `emacs-mule' by default. |
4ed46869 KH |
3105 | |
3106 | o coding-category-sjis | |
3107 | ||
3108 | The category for a coding system which has the same code range | |
3109 | as SJIS. Assigned the coding-system (Lisp | |
7717c392 | 3110 | symbol) `japanese-shift-jis' by default. |
4ed46869 KH |
3111 | |
3112 | o coding-category-iso-7 | |
3113 | ||
3114 | The category for a coding system which has the same code range | |
7717c392 | 3115 | as ISO2022 of 7-bit environment. This doesn't use any locking |
d46c5b12 KH |
3116 | shift and single shift functions. This can encode/decode all |
3117 | charsets. Assigned the coding-system (Lisp symbol) | |
3118 | `iso-2022-7bit' by default. | |
3119 | ||
3120 | o coding-category-iso-7-tight | |
3121 | ||
3122 | Same as coding-category-iso-7 except that this can | |
3123 | encode/decode only the specified charsets. | |
4ed46869 KH |
3124 | |
3125 | o coding-category-iso-8-1 | |
3126 | ||
3127 | The category for a coding system which has the same code range | |
3128 | as ISO2022 of 8-bit environment and graphic plane 1 used only | |
7717c392 KH |
3129 | for DIMENSION1 charset. This doesn't use any locking shift |
3130 | and single shift functions. Assigned the coding-system (Lisp | |
3131 | symbol) `iso-latin-1' by default. | |
4ed46869 KH |
3132 | |
3133 | o coding-category-iso-8-2 | |
3134 | ||
3135 | The category for a coding system which has the same code range | |
3136 | as ISO2022 of 8-bit environment and graphic plane 1 used only | |
7717c392 KH |
3137 | for DIMENSION2 charset. This doesn't use any locking shift |
3138 | and single shift functions. Assigned the coding-system (Lisp | |
3139 | symbol) `japanese-iso-8bit' by default. | |
4ed46869 | 3140 | |
7717c392 | 3141 | o coding-category-iso-7-else |
4ed46869 KH |
3142 | |
3143 | The category for a coding system which has the same code range | |
7717c392 KH |
3144 | as ISO2022 of 7-bit environemnt but uses locking shift or |
3145 | single shift functions. Assigned the coding-system (Lisp | |
3146 | symbol) `iso-2022-7bit-lock' by default. | |
3147 | ||
3148 | o coding-category-iso-8-else | |
3149 | ||
3150 | The category for a coding system which has the same code range | |
3151 | as ISO2022 of 8-bit environemnt but uses locking shift or | |
3152 | single shift functions. Assigned the coding-system (Lisp | |
3153 | symbol) `iso-2022-8bit-ss2' by default. | |
4ed46869 KH |
3154 | |
3155 | o coding-category-big5 | |
3156 | ||
3157 | The category for a coding system which has the same code range | |
3158 | as BIG5. Assigned the coding-system (Lisp symbol) | |
e0e989f6 | 3159 | `cn-big5' by default. |
4ed46869 KH |
3160 | |
3161 | o coding-category-binary | |
3162 | ||
3163 | The category for a coding system not categorized in any of the | |
3164 | above. Assigned the coding-system (Lisp symbol) | |
e0e989f6 | 3165 | `no-conversion' by default. |
4ed46869 KH |
3166 | |
3167 | Each of them is a Lisp symbol and the value is an actual | |
3168 | `coding-system's (this is also a Lisp symbol) assigned by a user. | |
3169 | What Emacs does actually is to detect a category of coding system. | |
3170 | Then, it uses a `coding-system' assigned to it. If Emacs can't | |
3171 | decide only one possible category, it selects a category of the | |
3172 | highest priority. Priorities of categories are also specified by a | |
3173 | user in a Lisp variable `coding-category-list'. | |
3174 | ||
3175 | */ | |
3176 | ||
66cfb530 KH |
3177 | static |
3178 | int ascii_skip_code[256]; | |
3179 | ||
d46c5b12 | 3180 | /* Detect how a text of length SRC_BYTES pointed by SOURCE is encoded. |
4ed46869 KH |
3181 | If it detects possible coding systems, return an integer in which |
3182 | appropriate flag bits are set. Flag bits are defined by macros | |
d46c5b12 | 3183 | CODING_CATEGORY_MASK_XXX in `coding.h'. |
4ed46869 | 3184 | |
d46c5b12 KH |
3185 | How many ASCII characters are at the head is returned as *SKIP. */ |
3186 | ||
3187 | static int | |
3188 | detect_coding_mask (source, src_bytes, priorities, skip) | |
3189 | unsigned char *source; | |
3190 | int src_bytes, *priorities, *skip; | |
4ed46869 KH |
3191 | { |
3192 | register unsigned char c; | |
d46c5b12 | 3193 | unsigned char *src = source, *src_end = source + src_bytes; |
66cfb530 | 3194 | unsigned int mask; |
d46c5b12 | 3195 | int i; |
4ed46869 KH |
3196 | |
3197 | /* At first, skip all ASCII characters and control characters except | |
3198 | for three ISO2022 specific control characters. */ | |
66cfb530 KH |
3199 | ascii_skip_code[ISO_CODE_SO] = 0; |
3200 | ascii_skip_code[ISO_CODE_SI] = 0; | |
3201 | ascii_skip_code[ISO_CODE_ESC] = 0; | |
3202 | ||
bcf26d6a | 3203 | label_loop_detect_coding: |
66cfb530 | 3204 | while (src < src_end && ascii_skip_code[*src]) src++; |
d46c5b12 | 3205 | *skip = src - source; |
4ed46869 KH |
3206 | |
3207 | if (src >= src_end) | |
3208 | /* We found nothing other than ASCII. There's nothing to do. */ | |
d46c5b12 | 3209 | return 0; |
4ed46869 | 3210 | |
8a8147d6 | 3211 | c = *src; |
4ed46869 KH |
3212 | /* The text seems to be encoded in some multilingual coding system. |
3213 | Now, try to find in which coding system the text is encoded. */ | |
3214 | if (c < 0x80) | |
bcf26d6a KH |
3215 | { |
3216 | /* i.e. (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) */ | |
3217 | /* C is an ISO2022 specific control code of C0. */ | |
3218 | mask = detect_coding_iso2022 (src, src_end); | |
1b2af4b0 | 3219 | if (mask == 0) |
d46c5b12 KH |
3220 | { |
3221 | /* No valid ISO2022 code follows C. Try again. */ | |
3222 | src++; | |
66cfb530 KH |
3223 | if (c == ISO_CODE_ESC) |
3224 | ascii_skip_code[ISO_CODE_ESC] = 1; | |
3225 | else | |
3226 | ascii_skip_code[ISO_CODE_SO] = ascii_skip_code[ISO_CODE_SI] = 1; | |
d46c5b12 KH |
3227 | goto label_loop_detect_coding; |
3228 | } | |
3229 | if (priorities) | |
3230 | goto label_return_highest_only; | |
bcf26d6a | 3231 | } |
d46c5b12 | 3232 | else |
c4825358 | 3233 | { |
d46c5b12 | 3234 | int try; |
4ed46869 | 3235 | |
d46c5b12 KH |
3236 | if (c < 0xA0) |
3237 | { | |
3238 | /* C is the first byte of SJIS character code, | |
3239 | or a leading-code of Emacs' internal format (emacs-mule). */ | |
3240 | try = CODING_CATEGORY_MASK_SJIS | CODING_CATEGORY_MASK_EMACS_MULE; | |
3241 | ||
3242 | /* Or, if C is a special latin extra code, | |
3243 | or is an ISO2022 specific control code of C1 (SS2 or SS3), | |
3244 | or is an ISO2022 control-sequence-introducer (CSI), | |
3245 | we should also consider the possibility of ISO2022 codings. */ | |
3246 | if ((VECTORP (Vlatin_extra_code_table) | |
3247 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) | |
3248 | || (c == ISO_CODE_SS2 || c == ISO_CODE_SS3) | |
3249 | || (c == ISO_CODE_CSI | |
3250 | && (src < src_end | |
3251 | && (*src == ']' | |
3252 | || ((*src == '0' || *src == '1' || *src == '2') | |
3253 | && src + 1 < src_end | |
3254 | && src[1] == ']'))))) | |
3255 | try |= (CODING_CATEGORY_MASK_ISO_8_ELSE | |
3256 | | CODING_CATEGORY_MASK_ISO_8BIT); | |
3257 | } | |
c4825358 | 3258 | else |
d46c5b12 KH |
3259 | /* C is a character of ISO2022 in graphic plane right, |
3260 | or a SJIS's 1-byte character code (i.e. JISX0201), | |
3261 | or the first byte of BIG5's 2-byte code. */ | |
3262 | try = (CODING_CATEGORY_MASK_ISO_8_ELSE | |
3263 | | CODING_CATEGORY_MASK_ISO_8BIT | |
3264 | | CODING_CATEGORY_MASK_SJIS | |
3265 | | CODING_CATEGORY_MASK_BIG5); | |
3266 | ||
3267 | mask = 0; | |
3268 | if (priorities) | |
3269 | { | |
3270 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) | |
3271 | { | |
5ab13dd0 | 3272 | if (priorities[i] & try & CODING_CATEGORY_MASK_ISO) |
d46c5b12 | 3273 | mask = detect_coding_iso2022 (src, src_end); |
5ab13dd0 | 3274 | else if (priorities[i] & try & CODING_CATEGORY_MASK_SJIS) |
d46c5b12 | 3275 | mask = detect_coding_sjis (src, src_end); |
5ab13dd0 | 3276 | else if (priorities[i] & try & CODING_CATEGORY_MASK_BIG5) |
d46c5b12 | 3277 | mask = detect_coding_big5 (src, src_end); |
5ab13dd0 | 3278 | else if (priorities[i] & try & CODING_CATEGORY_MASK_EMACS_MULE) |
d46c5b12 | 3279 | mask = detect_coding_emacs_mule (src, src_end); |
5ab13dd0 RS |
3280 | else if (priorities[i] & CODING_CATEGORY_MASK_RAW_TEXT) |
3281 | mask = CODING_CATEGORY_MASK_RAW_TEXT; | |
3282 | else if (priorities[i] & CODING_CATEGORY_MASK_BINARY) | |
3283 | mask = CODING_CATEGORY_MASK_BINARY; | |
d46c5b12 KH |
3284 | if (mask) |
3285 | goto label_return_highest_only; | |
3286 | } | |
3287 | return CODING_CATEGORY_MASK_RAW_TEXT; | |
3288 | } | |
3289 | if (try & CODING_CATEGORY_MASK_ISO) | |
3290 | mask |= detect_coding_iso2022 (src, src_end); | |
3291 | if (try & CODING_CATEGORY_MASK_SJIS) | |
3292 | mask |= detect_coding_sjis (src, src_end); | |
3293 | if (try & CODING_CATEGORY_MASK_BIG5) | |
3294 | mask |= detect_coding_big5 (src, src_end); | |
3295 | if (try & CODING_CATEGORY_MASK_EMACS_MULE) | |
3296 | mask |= detect_coding_emacs_mule (src, src_end); | |
c4825358 | 3297 | } |
5ab13dd0 | 3298 | return (mask | CODING_CATEGORY_MASK_RAW_TEXT | CODING_CATEGORY_MASK_BINARY); |
d46c5b12 KH |
3299 | |
3300 | label_return_highest_only: | |
3301 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) | |
3302 | { | |
3303 | if (mask & priorities[i]) | |
3304 | return priorities[i]; | |
3305 | } | |
3306 | return CODING_CATEGORY_MASK_RAW_TEXT; | |
4ed46869 KH |
3307 | } |
3308 | ||
3309 | /* Detect how a text of length SRC_BYTES pointed by SRC is encoded. | |
3310 | The information of the detected coding system is set in CODING. */ | |
3311 | ||
3312 | void | |
3313 | detect_coding (coding, src, src_bytes) | |
3314 | struct coding_system *coding; | |
3315 | unsigned char *src; | |
3316 | int src_bytes; | |
3317 | { | |
d46c5b12 KH |
3318 | unsigned int idx; |
3319 | int skip, mask, i; | |
27901516 | 3320 | Lisp_Object val = Vcoding_category_list; |
4ed46869 | 3321 | |
66cfb530 | 3322 | mask = detect_coding_mask (src, src_bytes, coding_priorities, &skip); |
d46c5b12 | 3323 | coding->heading_ascii = skip; |
4ed46869 | 3324 | |
d46c5b12 KH |
3325 | if (!mask) return; |
3326 | ||
3327 | /* We found a single coding system of the highest priority in MASK. */ | |
3328 | idx = 0; | |
3329 | while (mask && ! (mask & 1)) mask >>= 1, idx++; | |
3330 | if (! mask) | |
3331 | idx = CODING_CATEGORY_IDX_RAW_TEXT; | |
4ed46869 | 3332 | |
d46c5b12 KH |
3333 | val = XSYMBOL (XVECTOR (Vcoding_category_table)->contents[idx])->value; |
3334 | ||
3335 | if (coding->eol_type != CODING_EOL_UNDECIDED) | |
27901516 | 3336 | { |
d46c5b12 KH |
3337 | Lisp_Object tmp = Fget (val, Qeol_type); |
3338 | ||
3339 | if (VECTORP (tmp)) | |
3340 | val = XVECTOR (tmp)->contents[coding->eol_type]; | |
4ed46869 | 3341 | } |
d46c5b12 KH |
3342 | setup_coding_system (val, coding); |
3343 | /* Set this again because setup_coding_system reset this member. */ | |
3344 | coding->heading_ascii = skip; | |
4ed46869 KH |
3345 | } |
3346 | ||
d46c5b12 KH |
3347 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by |
3348 | SOURCE is encoded. Return one of CODING_EOL_LF, CODING_EOL_CRLF, | |
3349 | CODING_EOL_CR, and CODING_EOL_UNDECIDED. | |
3350 | ||
3351 | How many non-eol characters are at the head is returned as *SKIP. */ | |
4ed46869 | 3352 | |
bc4bc72a RS |
3353 | #define MAX_EOL_CHECK_COUNT 3 |
3354 | ||
d46c5b12 KH |
3355 | static int |
3356 | detect_eol_type (source, src_bytes, skip) | |
3357 | unsigned char *source; | |
3358 | int src_bytes, *skip; | |
4ed46869 | 3359 | { |
d46c5b12 | 3360 | unsigned char *src = source, *src_end = src + src_bytes; |
4ed46869 | 3361 | unsigned char c; |
bc4bc72a RS |
3362 | int total = 0; /* How many end-of-lines are found so far. */ |
3363 | int eol_type = CODING_EOL_UNDECIDED; | |
3364 | int this_eol_type; | |
4ed46869 | 3365 | |
d46c5b12 KH |
3366 | *skip = 0; |
3367 | ||
bc4bc72a | 3368 | while (src < src_end && total < MAX_EOL_CHECK_COUNT) |
4ed46869 KH |
3369 | { |
3370 | c = *src++; | |
bc4bc72a | 3371 | if (c == '\n' || c == '\r') |
4ed46869 | 3372 | { |
d46c5b12 KH |
3373 | if (*skip == 0) |
3374 | *skip = src - 1 - source; | |
bc4bc72a RS |
3375 | total++; |
3376 | if (c == '\n') | |
3377 | this_eol_type = CODING_EOL_LF; | |
3378 | else if (src >= src_end || *src != '\n') | |
3379 | this_eol_type = CODING_EOL_CR; | |
4ed46869 | 3380 | else |
bc4bc72a RS |
3381 | this_eol_type = CODING_EOL_CRLF, src++; |
3382 | ||
3383 | if (eol_type == CODING_EOL_UNDECIDED) | |
3384 | /* This is the first end-of-line. */ | |
3385 | eol_type = this_eol_type; | |
3386 | else if (eol_type != this_eol_type) | |
d46c5b12 KH |
3387 | { |
3388 | /* The found type is different from what found before. */ | |
3389 | eol_type = CODING_EOL_INCONSISTENT; | |
3390 | break; | |
3391 | } | |
4ed46869 KH |
3392 | } |
3393 | } | |
bc4bc72a | 3394 | |
d46c5b12 KH |
3395 | if (*skip == 0) |
3396 | *skip = src_end - source; | |
85a02ca4 | 3397 | return eol_type; |
4ed46869 KH |
3398 | } |
3399 | ||
3400 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC | |
3401 | is encoded. If it detects an appropriate format of end-of-line, it | |
3402 | sets the information in *CODING. */ | |
3403 | ||
3404 | void | |
3405 | detect_eol (coding, src, src_bytes) | |
3406 | struct coding_system *coding; | |
3407 | unsigned char *src; | |
3408 | int src_bytes; | |
3409 | { | |
4608c386 | 3410 | Lisp_Object val; |
d46c5b12 KH |
3411 | int skip; |
3412 | int eol_type = detect_eol_type (src, src_bytes, &skip); | |
3413 | ||
3414 | if (coding->heading_ascii > skip) | |
3415 | coding->heading_ascii = skip; | |
3416 | else | |
3417 | skip = coding->heading_ascii; | |
4ed46869 | 3418 | |
0ef69138 | 3419 | if (eol_type == CODING_EOL_UNDECIDED) |
4ed46869 | 3420 | return; |
27901516 KH |
3421 | if (eol_type == CODING_EOL_INCONSISTENT) |
3422 | { | |
3423 | #if 0 | |
3424 | /* This code is suppressed until we find a better way to | |
992f23f2 | 3425 | distinguish raw text file and binary file. */ |
27901516 KH |
3426 | |
3427 | /* If we have already detected that the coding is raw-text, the | |
3428 | coding should actually be no-conversion. */ | |
3429 | if (coding->type == coding_type_raw_text) | |
3430 | { | |
3431 | setup_coding_system (Qno_conversion, coding); | |
3432 | return; | |
3433 | } | |
3434 | /* Else, let's decode only text code anyway. */ | |
3435 | #endif /* 0 */ | |
1b2af4b0 | 3436 | eol_type = CODING_EOL_LF; |
27901516 KH |
3437 | } |
3438 | ||
4608c386 | 3439 | val = Fget (coding->symbol, Qeol_type); |
4ed46869 | 3440 | if (VECTORP (val) && XVECTOR (val)->size == 3) |
d46c5b12 KH |
3441 | { |
3442 | setup_coding_system (XVECTOR (val)->contents[eol_type], coding); | |
3443 | coding->heading_ascii = skip; | |
3444 | } | |
3445 | } | |
3446 | ||
3447 | #define CONVERSION_BUFFER_EXTRA_ROOM 256 | |
3448 | ||
3449 | #define DECODING_BUFFER_MAG(coding) \ | |
3450 | (coding->type == coding_type_iso2022 \ | |
3451 | ? 3 \ | |
3452 | : ((coding->type == coding_type_sjis || coding->type == coding_type_big5) \ | |
3453 | ? 2 \ | |
3454 | : (coding->type == coding_type_raw_text \ | |
3455 | ? 1 \ | |
3456 | : (coding->type == coding_type_ccl \ | |
3457 | ? coding->spec.ccl.decoder.buf_magnification \ | |
3458 | : 2)))) | |
3459 | ||
3460 | /* Return maximum size (bytes) of a buffer enough for decoding | |
3461 | SRC_BYTES of text encoded in CODING. */ | |
3462 | ||
3463 | int | |
3464 | decoding_buffer_size (coding, src_bytes) | |
3465 | struct coding_system *coding; | |
3466 | int src_bytes; | |
3467 | { | |
3468 | return (src_bytes * DECODING_BUFFER_MAG (coding) | |
3469 | + CONVERSION_BUFFER_EXTRA_ROOM); | |
3470 | } | |
3471 | ||
3472 | /* Return maximum size (bytes) of a buffer enough for encoding | |
3473 | SRC_BYTES of text to CODING. */ | |
3474 | ||
3475 | int | |
3476 | encoding_buffer_size (coding, src_bytes) | |
3477 | struct coding_system *coding; | |
3478 | int src_bytes; | |
3479 | { | |
3480 | int magnification; | |
3481 | ||
3482 | if (coding->type == coding_type_ccl) | |
3483 | magnification = coding->spec.ccl.encoder.buf_magnification; | |
3484 | else | |
3485 | magnification = 3; | |
3486 | ||
3487 | return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM); | |
3488 | } | |
3489 | ||
3490 | #ifndef MINIMUM_CONVERSION_BUFFER_SIZE | |
3491 | #define MINIMUM_CONVERSION_BUFFER_SIZE 1024 | |
3492 | #endif | |
3493 | ||
3494 | char *conversion_buffer; | |
3495 | int conversion_buffer_size; | |
3496 | ||
3497 | /* Return a pointer to a SIZE bytes of buffer to be used for encoding | |
3498 | or decoding. Sufficient memory is allocated automatically. If we | |
3499 | run out of memory, return NULL. */ | |
3500 | ||
3501 | char * | |
3502 | get_conversion_buffer (size) | |
3503 | int size; | |
3504 | { | |
3505 | if (size > conversion_buffer_size) | |
3506 | { | |
3507 | char *buf; | |
3508 | int real_size = conversion_buffer_size * 2; | |
3509 | ||
3510 | while (real_size < size) real_size *= 2; | |
3511 | buf = (char *) xmalloc (real_size); | |
3512 | xfree (conversion_buffer); | |
3513 | conversion_buffer = buf; | |
3514 | conversion_buffer_size = real_size; | |
3515 | } | |
3516 | return conversion_buffer; | |
3517 | } | |
3518 | ||
3519 | int | |
3520 | ccl_coding_driver (coding, source, destination, src_bytes, dst_bytes, encodep) | |
3521 | struct coding_system *coding; | |
3522 | unsigned char *source, *destination; | |
3523 | int src_bytes, dst_bytes, encodep; | |
3524 | { | |
3525 | struct ccl_program *ccl | |
3526 | = encodep ? &coding->spec.ccl.encoder : &coding->spec.ccl.decoder; | |
3527 | int result; | |
3528 | ||
3529 | coding->produced = ccl_driver (ccl, source, destination, | |
3530 | src_bytes, dst_bytes, &(coding->consumed)); | |
3531 | if (encodep) | |
3532 | { | |
3533 | coding->produced_char = coding->produced; | |
3534 | coding->consumed_char | |
3535 | = multibyte_chars_in_text (source, coding->consumed); | |
3536 | } | |
3537 | else | |
3538 | { | |
3539 | coding->produced_char | |
3540 | = multibyte_chars_in_text (destination, coding->produced); | |
3541 | coding->consumed_char = coding->consumed; | |
3542 | } | |
3543 | switch (ccl->status) | |
3544 | { | |
3545 | case CCL_STAT_SUSPEND_BY_SRC: | |
3546 | result = CODING_FINISH_INSUFFICIENT_SRC; | |
3547 | break; | |
3548 | case CCL_STAT_SUSPEND_BY_DST: | |
3549 | result = CODING_FINISH_INSUFFICIENT_DST; | |
3550 | break; | |
3551 | default: | |
3552 | result = CODING_FINISH_NORMAL; | |
3553 | break; | |
3554 | } | |
3555 | return result; | |
4ed46869 KH |
3556 | } |
3557 | ||
3558 | /* See "GENERAL NOTES about `decode_coding_XXX ()' functions". Before | |
3559 | decoding, it may detect coding system and format of end-of-line if | |
3560 | those are not yet decided. */ | |
3561 | ||
3562 | int | |
d46c5b12 | 3563 | decode_coding (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
3564 | struct coding_system *coding; |
3565 | unsigned char *source, *destination; | |
3566 | int src_bytes, dst_bytes; | |
4ed46869 | 3567 | { |
d46c5b12 | 3568 | int result; |
4ed46869 KH |
3569 | |
3570 | if (src_bytes <= 0) | |
3571 | { | |
d46c5b12 KH |
3572 | coding->produced = coding->produced_char = 0; |
3573 | coding->consumed = coding->consumed_char = 0; | |
fb88bf2d | 3574 | coding->fake_multibyte = 0; |
d46c5b12 | 3575 | return CODING_FINISH_NORMAL; |
4ed46869 KH |
3576 | } |
3577 | ||
0ef69138 | 3578 | if (coding->type == coding_type_undecided) |
4ed46869 KH |
3579 | detect_coding (coding, source, src_bytes); |
3580 | ||
0ef69138 | 3581 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
3582 | detect_eol (coding, source, src_bytes); |
3583 | ||
4ed46869 KH |
3584 | switch (coding->type) |
3585 | { | |
0ef69138 KH |
3586 | case coding_type_emacs_mule: |
3587 | case coding_type_undecided: | |
27901516 | 3588 | case coding_type_raw_text: |
4ed46869 | 3589 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 3590 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 | 3591 | goto label_no_conversion; |
d46c5b12 | 3592 | result = decode_eol (coding, source, destination, src_bytes, dst_bytes); |
4ed46869 KH |
3593 | break; |
3594 | ||
3595 | case coding_type_sjis: | |
d46c5b12 KH |
3596 | result = decode_coding_sjis_big5 (coding, source, destination, |
3597 | src_bytes, dst_bytes, 1); | |
4ed46869 KH |
3598 | break; |
3599 | ||
3600 | case coding_type_iso2022: | |
d46c5b12 KH |
3601 | result = decode_coding_iso2022 (coding, source, destination, |
3602 | src_bytes, dst_bytes); | |
4ed46869 KH |
3603 | break; |
3604 | ||
3605 | case coding_type_big5: | |
d46c5b12 KH |
3606 | result = decode_coding_sjis_big5 (coding, source, destination, |
3607 | src_bytes, dst_bytes, 0); | |
4ed46869 KH |
3608 | break; |
3609 | ||
3610 | case coding_type_ccl: | |
d46c5b12 KH |
3611 | result = ccl_coding_driver (coding, source, destination, |
3612 | src_bytes, dst_bytes, 0); | |
3613 | break; | |
3614 | ||
3615 | default: /* i.e. case coding_type_no_conversion: */ | |
3616 | label_no_conversion: | |
3617 | if (dst_bytes && src_bytes > dst_bytes) | |
3618 | { | |
3619 | coding->produced = dst_bytes; | |
3620 | result = CODING_FINISH_INSUFFICIENT_DST; | |
3621 | } | |
3622 | else | |
3623 | { | |
3624 | coding->produced = src_bytes; | |
3625 | result = CODING_FINISH_NORMAL; | |
3626 | } | |
3627 | if (dst_bytes) | |
3628 | bcopy (source, destination, coding->produced); | |
3629 | else | |
3630 | safe_bcopy (source, destination, coding->produced); | |
fb88bf2d | 3631 | coding->fake_multibyte = 1; |
d46c5b12 KH |
3632 | coding->consumed |
3633 | = coding->consumed_char = coding->produced_char = coding->produced; | |
4ed46869 KH |
3634 | break; |
3635 | } | |
3636 | ||
d46c5b12 | 3637 | return result; |
4ed46869 KH |
3638 | } |
3639 | ||
3640 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". */ | |
3641 | ||
3642 | int | |
d46c5b12 | 3643 | encode_coding (coding, source, destination, src_bytes, dst_bytes) |
4ed46869 KH |
3644 | struct coding_system *coding; |
3645 | unsigned char *source, *destination; | |
3646 | int src_bytes, dst_bytes; | |
4ed46869 | 3647 | { |
d46c5b12 | 3648 | int result; |
4ed46869 | 3649 | |
d46c5b12 | 3650 | if (src_bytes <= 0) |
4ed46869 | 3651 | { |
d46c5b12 KH |
3652 | coding->produced = coding->produced_char = 0; |
3653 | coding->consumed = coding->consumed_char = 0; | |
fb88bf2d | 3654 | coding->fake_multibyte = 0; |
d46c5b12 KH |
3655 | return CODING_FINISH_NORMAL; |
3656 | } | |
4ed46869 | 3657 | |
d46c5b12 KH |
3658 | switch (coding->type) |
3659 | { | |
0ef69138 KH |
3660 | case coding_type_emacs_mule: |
3661 | case coding_type_undecided: | |
27901516 | 3662 | case coding_type_raw_text: |
4ed46869 | 3663 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 3664 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 | 3665 | goto label_no_conversion; |
d46c5b12 | 3666 | result = encode_eol (coding, source, destination, src_bytes, dst_bytes); |
4ed46869 KH |
3667 | break; |
3668 | ||
3669 | case coding_type_sjis: | |
d46c5b12 KH |
3670 | result = encode_coding_sjis_big5 (coding, source, destination, |
3671 | src_bytes, dst_bytes, 1); | |
4ed46869 KH |
3672 | break; |
3673 | ||
3674 | case coding_type_iso2022: | |
d46c5b12 KH |
3675 | result = encode_coding_iso2022 (coding, source, destination, |
3676 | src_bytes, dst_bytes); | |
4ed46869 KH |
3677 | break; |
3678 | ||
3679 | case coding_type_big5: | |
d46c5b12 KH |
3680 | result = encode_coding_sjis_big5 (coding, source, destination, |
3681 | src_bytes, dst_bytes, 0); | |
4ed46869 KH |
3682 | break; |
3683 | ||
3684 | case coding_type_ccl: | |
d46c5b12 KH |
3685 | result = ccl_coding_driver (coding, source, destination, |
3686 | src_bytes, dst_bytes, 1); | |
3687 | break; | |
3688 | ||
3689 | default: /* i.e. case coding_type_no_conversion: */ | |
3690 | label_no_conversion: | |
3691 | if (dst_bytes && src_bytes > dst_bytes) | |
3692 | { | |
3693 | coding->produced = dst_bytes; | |
3694 | result = CODING_FINISH_INSUFFICIENT_DST; | |
3695 | } | |
3696 | else | |
3697 | { | |
3698 | coding->produced = src_bytes; | |
3699 | result = CODING_FINISH_NORMAL; | |
3700 | } | |
3701 | if (dst_bytes) | |
3702 | bcopy (source, destination, coding->produced); | |
3703 | else | |
3704 | safe_bcopy (source, destination, coding->produced); | |
3705 | if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY) | |
3706 | { | |
3707 | unsigned char *p = destination, *pend = p + coding->produced; | |
3708 | while (p < pend) | |
3709 | if (*p++ == '\015') p[-1] = '\n'; | |
3710 | } | |
fb88bf2d | 3711 | coding->fake_multibyte = 1; |
d46c5b12 KH |
3712 | coding->consumed |
3713 | = coding->consumed_char = coding->produced_char = coding->produced; | |
4ed46869 KH |
3714 | break; |
3715 | } | |
3716 | ||
d46c5b12 | 3717 | return result; |
4ed46869 KH |
3718 | } |
3719 | ||
fb88bf2d KH |
3720 | /* Scan text in the region between *BEG and *END (byte positions), |
3721 | skip characters which we don't have to decode by coding system | |
3722 | CODING at the head and tail, then set *BEG and *END to the region | |
3723 | of the text we actually have to convert. The caller should move | |
3724 | the gap out of the region in advance. | |
4ed46869 | 3725 | |
d46c5b12 KH |
3726 | If STR is not NULL, *BEG and *END are indices into STR. */ |
3727 | ||
3728 | static void | |
3729 | shrink_decoding_region (beg, end, coding, str) | |
3730 | int *beg, *end; | |
3731 | struct coding_system *coding; | |
3732 | unsigned char *str; | |
3733 | { | |
fb88bf2d | 3734 | unsigned char *begp_orig, *begp, *endp_orig, *endp, c; |
d46c5b12 KH |
3735 | int eol_conversion; |
3736 | ||
3737 | if (coding->type == coding_type_ccl | |
3738 | || coding->type == coding_type_undecided | |
3739 | || !NILP (coding->post_read_conversion)) | |
3740 | { | |
3741 | /* We can't skip any data. */ | |
3742 | return; | |
3743 | } | |
3744 | else if (coding->type == coding_type_no_conversion) | |
3745 | { | |
fb88bf2d KH |
3746 | /* We need no conversion, but don't have to skip any data here. |
3747 | Decoding routine handles them effectively anyway. */ | |
d46c5b12 KH |
3748 | return; |
3749 | } | |
3750 | ||
aa60dea6 KH |
3751 | eol_conversion = (coding->eol_type != CODING_EOL_LF); |
3752 | ||
3753 | if ((! eol_conversion) && (coding->heading_ascii >= 0)) | |
d46c5b12 KH |
3754 | /* Detection routine has already found how much we can skip at the |
3755 | head. */ | |
3756 | *beg += coding->heading_ascii; | |
3757 | ||
3758 | if (str) | |
3759 | { | |
3760 | begp_orig = begp = str + *beg; | |
3761 | endp_orig = endp = str + *end; | |
3762 | } | |
3763 | else | |
3764 | { | |
fb88bf2d | 3765 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
d46c5b12 KH |
3766 | endp_orig = endp = begp + *end - *beg; |
3767 | } | |
3768 | ||
d46c5b12 KH |
3769 | switch (coding->type) |
3770 | { | |
3771 | case coding_type_emacs_mule: | |
3772 | case coding_type_raw_text: | |
3773 | if (eol_conversion) | |
3774 | { | |
3775 | if (coding->heading_ascii < 0) | |
fb88bf2d | 3776 | while (begp < endp && *begp != '\r' && *begp < 0x80) begp++; |
ee59c65f | 3777 | while (begp < endp && endp[-1] != '\r' && endp[-1] < 0x80) |
fb88bf2d | 3778 | endp--; |
ee59c65f RS |
3779 | /* Do not consider LF as ascii if preceded by CR, since that |
3780 | confuses eol decoding. */ | |
3781 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
3782 | endp++; | |
d46c5b12 KH |
3783 | } |
3784 | else | |
3785 | begp = endp; | |
3786 | break; | |
3787 | ||
3788 | case coding_type_sjis: | |
3789 | case coding_type_big5: | |
3790 | /* We can skip all ASCII characters at the head. */ | |
3791 | if (coding->heading_ascii < 0) | |
3792 | { | |
3793 | if (eol_conversion) | |
de9d083c | 3794 | while (begp < endp && *begp < 0x80 && *begp != '\r') begp++; |
d46c5b12 KH |
3795 | else |
3796 | while (begp < endp && *begp < 0x80) begp++; | |
3797 | } | |
3798 | /* We can skip all ASCII characters at the tail except for the | |
3799 | second byte of SJIS or BIG5 code. */ | |
3800 | if (eol_conversion) | |
de9d083c | 3801 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\r') endp--; |
d46c5b12 KH |
3802 | else |
3803 | while (begp < endp && endp[-1] < 0x80) endp--; | |
ee59c65f RS |
3804 | /* Do not consider LF as ascii if preceded by CR, since that |
3805 | confuses eol decoding. */ | |
3806 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
3807 | endp++; | |
d46c5b12 KH |
3808 | if (begp < endp && endp < endp_orig && endp[-1] >= 0x80) |
3809 | endp++; | |
3810 | break; | |
3811 | ||
3812 | default: /* i.e. case coding_type_iso2022: */ | |
3813 | if (coding->heading_ascii < 0) | |
3814 | { | |
d46c5b12 KH |
3815 | /* We can skip all ASCII characters at the head except for a |
3816 | few control codes. */ | |
3817 | while (begp < endp && (c = *begp) < 0x80 | |
3818 | && c != ISO_CODE_CR && c != ISO_CODE_SO | |
3819 | && c != ISO_CODE_SI && c != ISO_CODE_ESC | |
3820 | && (!eol_conversion || c != ISO_CODE_LF)) | |
3821 | begp++; | |
3822 | } | |
3823 | switch (coding->category_idx) | |
3824 | { | |
3825 | case CODING_CATEGORY_IDX_ISO_8_1: | |
3826 | case CODING_CATEGORY_IDX_ISO_8_2: | |
3827 | /* We can skip all ASCII characters at the tail. */ | |
3828 | if (eol_conversion) | |
de9d083c | 3829 | while (begp < endp && (c = endp[-1]) < 0x80 && c != '\r') endp--; |
d46c5b12 KH |
3830 | else |
3831 | while (begp < endp && endp[-1] < 0x80) endp--; | |
ee59c65f RS |
3832 | /* Do not consider LF as ascii if preceded by CR, since that |
3833 | confuses eol decoding. */ | |
3834 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
3835 | endp++; | |
d46c5b12 KH |
3836 | break; |
3837 | ||
3838 | case CODING_CATEGORY_IDX_ISO_7: | |
3839 | case CODING_CATEGORY_IDX_ISO_7_TIGHT: | |
3840 | /* We can skip all charactes at the tail except for ESC and | |
3841 | the following 2-byte at the tail. */ | |
3842 | if (eol_conversion) | |
fb88bf2d | 3843 | while (begp < endp |
de9d083c | 3844 | && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC && c != '\r') |
d46c5b12 KH |
3845 | endp--; |
3846 | else | |
fb88bf2d KH |
3847 | while (begp < endp |
3848 | && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC) | |
d46c5b12 | 3849 | endp--; |
ee59c65f RS |
3850 | /* Do not consider LF as ascii if preceded by CR, since that |
3851 | confuses eol decoding. */ | |
3852 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
3853 | endp++; | |
d46c5b12 KH |
3854 | if (begp < endp && endp[-1] == ISO_CODE_ESC) |
3855 | { | |
3856 | if (endp + 1 < endp_orig && end[0] == '(' && end[1] == 'B') | |
3857 | /* This is an ASCII designation sequence. We can | |
3858 | surely skip the tail. */ | |
3859 | endp += 2; | |
3860 | else | |
3861 | /* Hmmm, we can't skip the tail. */ | |
3862 | endp = endp_orig; | |
3863 | } | |
3864 | } | |
3865 | } | |
3866 | *beg += begp - begp_orig; | |
3867 | *end += endp - endp_orig; | |
3868 | return; | |
3869 | } | |
3870 | ||
3871 | /* Like shrink_decoding_region but for encoding. */ | |
3872 | ||
3873 | static void | |
3874 | shrink_encoding_region (beg, end, coding, str) | |
3875 | int *beg, *end; | |
3876 | struct coding_system *coding; | |
3877 | unsigned char *str; | |
3878 | { | |
3879 | unsigned char *begp_orig, *begp, *endp_orig, *endp; | |
3880 | int eol_conversion; | |
3881 | ||
3882 | if (coding->type == coding_type_ccl) | |
3883 | /* We can't skip any data. */ | |
3884 | return; | |
3885 | else if (coding->type == coding_type_no_conversion) | |
3886 | { | |
3887 | /* We need no conversion. */ | |
3888 | *beg = *end; | |
3889 | return; | |
3890 | } | |
3891 | ||
3892 | if (str) | |
3893 | { | |
3894 | begp_orig = begp = str + *beg; | |
3895 | endp_orig = endp = str + *end; | |
3896 | } | |
3897 | else | |
3898 | { | |
fb88bf2d | 3899 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
d46c5b12 KH |
3900 | endp_orig = endp = begp + *end - *beg; |
3901 | } | |
3902 | ||
3903 | eol_conversion = (coding->eol_type == CODING_EOL_CR | |
3904 | || coding->eol_type == CODING_EOL_CRLF); | |
3905 | ||
3906 | /* Here, we don't have to check coding->pre_write_conversion because | |
3907 | the caller is expected to have handled it already. */ | |
3908 | switch (coding->type) | |
3909 | { | |
3910 | case coding_type_undecided: | |
3911 | case coding_type_emacs_mule: | |
3912 | case coding_type_raw_text: | |
3913 | if (eol_conversion) | |
3914 | { | |
3915 | while (begp < endp && *begp != '\n') begp++; | |
3916 | while (begp < endp && endp[-1] != '\n') endp--; | |
3917 | } | |
3918 | else | |
3919 | begp = endp; | |
3920 | break; | |
3921 | ||
3922 | case coding_type_iso2022: | |
3923 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL) | |
3924 | { | |
3925 | unsigned char *bol = begp; | |
3926 | while (begp < endp && *begp < 0x80) | |
3927 | { | |
3928 | begp++; | |
3929 | if (begp[-1] == '\n') | |
3930 | bol = begp; | |
3931 | } | |
3932 | begp = bol; | |
3933 | goto label_skip_tail; | |
3934 | } | |
3935 | /* fall down ... */ | |
3936 | ||
3937 | default: | |
3938 | /* We can skip all ASCII characters at the head and tail. */ | |
3939 | if (eol_conversion) | |
3940 | while (begp < endp && *begp < 0x80 && *begp != '\n') begp++; | |
3941 | else | |
3942 | while (begp < endp && *begp < 0x80) begp++; | |
3943 | label_skip_tail: | |
3944 | if (eol_conversion) | |
3945 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\n') endp--; | |
3946 | else | |
3947 | while (begp < endp && *(endp - 1) < 0x80) endp--; | |
3948 | break; | |
3949 | } | |
3950 | ||
3951 | *beg += begp - begp_orig; | |
3952 | *end += endp - endp_orig; | |
3953 | return; | |
3954 | } | |
3955 | ||
3956 | /* Decode (if ENCODEP is zero) or encode (if ENCODEP is nonzero) the | |
fb88bf2d KH |
3957 | text from FROM to TO (byte positions are FROM_BYTE and TO_BYTE) by |
3958 | coding system CODING, and return the status code of code conversion | |
3959 | (currently, this value has no meaning). | |
3960 | ||
3961 | How many characters (and bytes) are converted to how many | |
3962 | characters (and bytes) are recorded in members of the structure | |
3963 | CODING. | |
d46c5b12 | 3964 | |
6e44253b | 3965 | If REPLACE is nonzero, we do various things as if the original text |
d46c5b12 | 3966 | is deleted and a new text is inserted. See the comments in |
6e44253b | 3967 | replace_range (insdel.c) to know what we are doing. */ |
4ed46869 KH |
3968 | |
3969 | int | |
6e44253b KH |
3970 | code_convert_region (from, from_byte, to, to_byte, coding, encodep, replace) |
3971 | int from, from_byte, to, to_byte, encodep, replace; | |
4ed46869 | 3972 | struct coding_system *coding; |
4ed46869 | 3973 | { |
fb88bf2d KH |
3974 | int len = to - from, len_byte = to_byte - from_byte; |
3975 | int require, inserted, inserted_byte; | |
12410ef1 | 3976 | int head_skip, tail_skip, total_skip; |
d46c5b12 | 3977 | Lisp_Object saved_coding_symbol = Qnil; |
fb88bf2d KH |
3978 | int multibyte = !NILP (current_buffer->enable_multibyte_characters); |
3979 | int first = 1; | |
3980 | int fake_multibyte = 0; | |
3981 | unsigned char *src, *dst; | |
12410ef1 | 3982 | Lisp_Object deletion = Qnil; |
d46c5b12 | 3983 | |
83fa074f KH |
3984 | if (from < PT && PT < to) |
3985 | SET_PT_BOTH (from, from_byte); | |
3986 | ||
6e44253b | 3987 | if (replace) |
d46c5b12 | 3988 | { |
fb88bf2d KH |
3989 | int saved_from = from; |
3990 | ||
d46c5b12 | 3991 | prepare_to_modify_buffer (from, to, &from); |
fb88bf2d KH |
3992 | if (saved_from != from) |
3993 | { | |
3994 | to = from + len; | |
3995 | if (multibyte) | |
3996 | from_byte = CHAR_TO_BYTE (from), to_byte = CHAR_TO_BYTE (to); | |
3997 | else | |
3998 | from_byte = from, to_byte = to; | |
3999 | len_byte = to_byte - from_byte; | |
4000 | } | |
d46c5b12 | 4001 | } |
d46c5b12 KH |
4002 | |
4003 | if (! encodep && CODING_REQUIRE_DETECTION (coding)) | |
4004 | { | |
12410ef1 | 4005 | /* We must detect encoding of text and eol format. */ |
d46c5b12 KH |
4006 | |
4007 | if (from < GPT && to > GPT) | |
4008 | move_gap_both (from, from_byte); | |
4009 | if (coding->type == coding_type_undecided) | |
4010 | { | |
fb88bf2d | 4011 | detect_coding (coding, BYTE_POS_ADDR (from_byte), len_byte); |
d46c5b12 | 4012 | if (coding->type == coding_type_undecided) |
12410ef1 KH |
4013 | /* It seems that the text contains only ASCII, but we |
4014 | should not left it undecided because the deeper | |
4015 | decoding routine (decode_coding) tries to detect the | |
4016 | encodings again in vain. */ | |
d46c5b12 KH |
4017 | coding->type = coding_type_emacs_mule; |
4018 | } | |
4019 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
4020 | { | |
4021 | saved_coding_symbol = coding->symbol; | |
4022 | detect_eol (coding, BYTE_POS_ADDR (from_byte), len_byte); | |
4023 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
4024 | coding->eol_type = CODING_EOL_LF; | |
4025 | /* We had better recover the original eol format if we | |
4026 | encounter an inconsitent eol format while decoding. */ | |
4027 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; | |
4028 | } | |
4029 | } | |
4030 | ||
fb88bf2d KH |
4031 | coding->consumed_char = len, coding->consumed = len_byte; |
4032 | ||
d46c5b12 KH |
4033 | if (encodep |
4034 | ? ! CODING_REQUIRE_ENCODING (coding) | |
4035 | : ! CODING_REQUIRE_DECODING (coding)) | |
fb88bf2d KH |
4036 | { |
4037 | coding->produced = len_byte; | |
12410ef1 KH |
4038 | if (multibyte |
4039 | && ! replace | |
4040 | /* See the comment of the member heading_ascii in coding.h. */ | |
4041 | && coding->heading_ascii < len_byte) | |
fb88bf2d | 4042 | { |
6e44253b KH |
4043 | /* We still may have to combine byte at the head and the |
4044 | tail of the text in the region. */ | |
12410ef1 | 4045 | if (from < GPT && GPT < to) |
6e44253b | 4046 | move_gap_both (to, to_byte); |
12410ef1 KH |
4047 | len = multibyte_chars_in_text (BYTE_POS_ADDR (from_byte), len_byte); |
4048 | adjust_after_insert (from, from_byte, to, to_byte, len); | |
4049 | coding->produced_char = len; | |
fb88bf2d KH |
4050 | } |
4051 | else | |
68e3a8f1 AS |
4052 | { |
4053 | if (!replace) | |
4054 | adjust_after_insert (from, from_byte, to, to_byte, len_byte); | |
4055 | coding->produced_char = len_byte; | |
4056 | } | |
fb88bf2d KH |
4057 | return 0; |
4058 | } | |
d46c5b12 KH |
4059 | |
4060 | /* Now we convert the text. */ | |
4061 | ||
4062 | /* For encoding, we must process pre-write-conversion in advance. */ | |
4063 | if (encodep | |
d46c5b12 KH |
4064 | && ! NILP (coding->pre_write_conversion) |
4065 | && SYMBOLP (coding->pre_write_conversion) | |
4066 | && ! NILP (Ffboundp (coding->pre_write_conversion))) | |
4067 | { | |
2b4f9037 KH |
4068 | /* The function in pre-write-conversion may put a new text in a |
4069 | new buffer. */ | |
d46c5b12 KH |
4070 | struct buffer *prev = current_buffer, *new; |
4071 | ||
b39f748c AS |
4072 | call2 (coding->pre_write_conversion, |
4073 | make_number (from), make_number (to)); | |
d46c5b12 KH |
4074 | if (current_buffer != prev) |
4075 | { | |
4076 | len = ZV - BEGV; | |
4077 | new = current_buffer; | |
4078 | set_buffer_internal_1 (prev); | |
ddbc19ff | 4079 | del_range_2 (from, from_byte, to, to_byte); |
d46c5b12 KH |
4080 | insert_from_buffer (new, BEG, len, 0); |
4081 | to = from + len; | |
fb88bf2d | 4082 | to_byte = multibyte ? CHAR_TO_BYTE (to) : to; |
d46c5b12 KH |
4083 | len_byte = to_byte - from_byte; |
4084 | } | |
4085 | } | |
4086 | ||
12410ef1 KH |
4087 | if (replace) |
4088 | deletion = make_buffer_string_both (from, from_byte, to, to_byte, 1); | |
4089 | ||
d46c5b12 | 4090 | /* Try to skip the heading and tailing ASCIIs. */ |
12410ef1 KH |
4091 | { |
4092 | int from_byte_orig = from_byte, to_byte_orig = to_byte; | |
4093 | ||
4094 | if (from < GPT && GPT < to) | |
4095 | move_gap_both (from, from_byte); | |
4096 | if (encodep) | |
4097 | shrink_encoding_region (&from_byte, &to_byte, coding, NULL); | |
4098 | else | |
4099 | shrink_decoding_region (&from_byte, &to_byte, coding, NULL); | |
4100 | if (from_byte == to_byte) | |
4101 | { | |
4102 | coding->produced = len_byte; | |
4103 | coding->produced_char = multibyte ? len : len_byte; | |
4104 | if (!replace) | |
4105 | /* We must record and adjust for this new text now. */ | |
4106 | adjust_after_insert (from, from_byte_orig, to, to_byte_orig, len); | |
4107 | return 0; | |
4108 | } | |
fb88bf2d | 4109 | |
12410ef1 KH |
4110 | head_skip = from_byte - from_byte_orig; |
4111 | tail_skip = to_byte_orig - to_byte; | |
4112 | total_skip = head_skip + tail_skip; | |
4113 | from += head_skip; | |
4114 | to -= tail_skip; | |
4115 | len -= total_skip; len_byte -= total_skip; | |
4116 | } | |
d46c5b12 | 4117 | |
fb88bf2d KH |
4118 | /* For converion, we must put the gap before the text in addition to |
4119 | making the gap larger for efficient decoding. The required gap | |
4120 | size starts from 2000 which is the magic number used in make_gap. | |
4121 | But, after one batch of conversion, it will be incremented if we | |
4122 | find that it is not enough . */ | |
d46c5b12 KH |
4123 | require = 2000; |
4124 | ||
4125 | if (GAP_SIZE < require) | |
4126 | make_gap (require - GAP_SIZE); | |
4127 | move_gap_both (from, from_byte); | |
4128 | ||
d46c5b12 KH |
4129 | if (GPT - BEG < beg_unchanged) |
4130 | beg_unchanged = GPT - BEG; | |
4131 | if (Z - GPT < end_unchanged) | |
4132 | end_unchanged = Z - GPT; | |
4133 | ||
4134 | inserted = inserted_byte = 0; | |
fb88bf2d KH |
4135 | src = GAP_END_ADDR, dst = GPT_ADDR; |
4136 | ||
4137 | GAP_SIZE += len_byte; | |
4138 | ZV -= len; | |
4139 | Z -= len; | |
4140 | ZV_BYTE -= len_byte; | |
4141 | Z_BYTE -= len_byte; | |
4142 | ||
d46c5b12 KH |
4143 | for (;;) |
4144 | { | |
fb88bf2d | 4145 | int result; |
d46c5b12 KH |
4146 | |
4147 | /* The buffer memory is changed from: | |
fb88bf2d KH |
4148 | +--------+converted-text+---------+-------original-text------+---+ |
4149 | |<-from->|<--inserted-->|---------|<-----------len---------->|---| | |
4150 | |<------------------- GAP_SIZE -------------------->| */ | |
d46c5b12 | 4151 | if (encodep) |
fb88bf2d | 4152 | result = encode_coding (coding, src, dst, len_byte, 0); |
d46c5b12 | 4153 | else |
fb88bf2d | 4154 | result = decode_coding (coding, src, dst, len_byte, 0); |
d46c5b12 KH |
4155 | /* to: |
4156 | +--------+-------converted-text--------+--+---original-text--+---+ | |
fb88bf2d KH |
4157 | |<-from->|<--inserted-->|<--produced-->|--|<-(len-consumed)->|---| |
4158 | |<------------------- GAP_SIZE -------------------->| */ | |
4159 | if (coding->fake_multibyte) | |
4160 | fake_multibyte = 1; | |
d46c5b12 | 4161 | |
fb88bf2d KH |
4162 | if (!encodep && !multibyte) |
4163 | coding->produced_char = coding->produced; | |
d46c5b12 KH |
4164 | inserted += coding->produced_char; |
4165 | inserted_byte += coding->produced; | |
d46c5b12 | 4166 | len_byte -= coding->consumed; |
fb88bf2d KH |
4167 | src += coding->consumed; |
4168 | dst += inserted_byte; | |
d46c5b12 KH |
4169 | |
4170 | if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL) | |
4171 | { | |
fb88bf2d | 4172 | unsigned char *pend = dst, *p = pend - inserted_byte; |
d46c5b12 KH |
4173 | |
4174 | /* Encode LFs back to the original eol format (CR or CRLF). */ | |
4175 | if (coding->eol_type == CODING_EOL_CR) | |
4176 | { | |
4177 | while (p < pend) if (*p++ == '\n') p[-1] = '\r'; | |
4178 | } | |
4179 | else | |
4180 | { | |
d46c5b12 KH |
4181 | int count = 0; |
4182 | ||
fb88bf2d KH |
4183 | while (p < pend) if (*p++ == '\n') count++; |
4184 | if (src - dst < count) | |
d46c5b12 | 4185 | { |
fb88bf2d KH |
4186 | /* We don't have sufficient room for putting LFs |
4187 | back to CRLF. We must record converted and | |
4188 | not-yet-converted text back to the buffer | |
4189 | content, enlarge the gap, then record them out of | |
4190 | the buffer contents again. */ | |
4191 | int add = len_byte + inserted_byte; | |
4192 | ||
4193 | GAP_SIZE -= add; | |
4194 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; | |
4195 | GPT += inserted_byte; GPT_BYTE += inserted_byte; | |
4196 | make_gap (count - GAP_SIZE); | |
4197 | GAP_SIZE += add; | |
4198 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; | |
4199 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; | |
4200 | /* Don't forget to update SRC, DST, and PEND. */ | |
4201 | src = GAP_END_ADDR - len_byte; | |
4202 | dst = GPT_ADDR + inserted_byte; | |
4203 | pend = dst; | |
d46c5b12 | 4204 | } |
d46c5b12 KH |
4205 | inserted += count; |
4206 | inserted_byte += count; | |
fb88bf2d KH |
4207 | coding->produced += count; |
4208 | p = dst = pend + count; | |
4209 | while (count) | |
4210 | { | |
4211 | *--p = *--pend; | |
4212 | if (*p == '\n') count--, *--p = '\r'; | |
4213 | } | |
d46c5b12 KH |
4214 | } |
4215 | ||
4216 | /* Suppress eol-format conversion in the further conversion. */ | |
4217 | coding->eol_type = CODING_EOL_LF; | |
4218 | ||
4219 | /* Restore the original symbol. */ | |
4220 | coding->symbol = saved_coding_symbol; | |
fb88bf2d KH |
4221 | |
4222 | continue; | |
d46c5b12 KH |
4223 | } |
4224 | if (len_byte <= 0) | |
4225 | break; | |
4226 | if (result == CODING_FINISH_INSUFFICIENT_SRC) | |
4227 | { | |
4228 | /* The source text ends in invalid codes. Let's just | |
4229 | make them valid buffer contents, and finish conversion. */ | |
fb88bf2d | 4230 | inserted += len_byte; |
d46c5b12 | 4231 | inserted_byte += len_byte; |
fb88bf2d | 4232 | while (len_byte--) |
ee59c65f | 4233 | *dst++ = *src++; |
fb88bf2d | 4234 | fake_multibyte = 1; |
d46c5b12 KH |
4235 | break; |
4236 | } | |
fb88bf2d KH |
4237 | if (first) |
4238 | { | |
4239 | /* We have just done the first batch of conversion which was | |
4240 | stoped because of insufficient gap. Let's reconsider the | |
4241 | required gap size (i.e. SRT - DST) now. | |
4242 | ||
4243 | We have converted ORIG bytes (== coding->consumed) into | |
4244 | NEW bytes (coding->produced). To convert the remaining | |
4245 | LEN bytes, we may need REQUIRE bytes of gap, where: | |
4246 | REQUIRE + LEN_BYTE = LEN_BYTE * (NEW / ORIG) | |
4247 | REQUIRE = LEN_BYTE * (NEW - ORIG) / ORIG | |
4248 | Here, we are sure that NEW >= ORIG. */ | |
6e44253b KH |
4249 | float ratio = coding->produced - coding->consumed; |
4250 | ratio /= coding->consumed; | |
4251 | require = len_byte * ratio; | |
fb88bf2d KH |
4252 | first = 0; |
4253 | } | |
4254 | if ((src - dst) < (require + 2000)) | |
4255 | { | |
4256 | /* See the comment above the previous call of make_gap. */ | |
4257 | int add = len_byte + inserted_byte; | |
4258 | ||
4259 | GAP_SIZE -= add; | |
4260 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; | |
4261 | GPT += inserted_byte; GPT_BYTE += inserted_byte; | |
4262 | make_gap (require + 2000); | |
4263 | GAP_SIZE += add; | |
4264 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; | |
4265 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; | |
4266 | /* Don't forget to update SRC, DST. */ | |
4267 | src = GAP_END_ADDR - len_byte; | |
4268 | dst = GPT_ADDR + inserted_byte; | |
4269 | } | |
d46c5b12 | 4270 | } |
fb88bf2d KH |
4271 | if (src - dst > 0) *dst = 0; /* Put an anchor. */ |
4272 | ||
2b4f9037 | 4273 | if (multibyte |
12410ef1 KH |
4274 | && (fake_multibyte |
4275 | || !encodep && (to - from) != (to_byte - from_byte))) | |
2b4f9037 | 4276 | inserted = multibyte_chars_in_text (GPT_ADDR, inserted_byte); |
7553d0e1 | 4277 | |
12410ef1 KH |
4278 | /* If we have shrinked the conversion area, adjust it now. */ |
4279 | if (total_skip > 0) | |
4280 | { | |
4281 | if (tail_skip > 0) | |
4282 | safe_bcopy (GAP_END_ADDR, GPT_ADDR + inserted_byte, tail_skip); | |
4283 | inserted += total_skip; inserted_byte += total_skip; | |
4284 | GAP_SIZE += total_skip; | |
4285 | GPT -= head_skip; GPT_BYTE -= head_skip; | |
4286 | ZV -= total_skip; ZV_BYTE -= total_skip; | |
4287 | Z -= total_skip; Z_BYTE -= total_skip; | |
4288 | from -= head_skip; from_byte -= head_skip; | |
4289 | to += tail_skip; to_byte += tail_skip; | |
4290 | } | |
4291 | ||
4292 | adjust_after_replace (from, from_byte, deletion, inserted, inserted_byte); | |
4ed46869 | 4293 | |
2b4f9037 | 4294 | if (! encodep && ! NILP (coding->post_read_conversion)) |
d46c5b12 | 4295 | { |
2b4f9037 KH |
4296 | Lisp_Object val; |
4297 | int orig_inserted = inserted, pos = PT; | |
4ed46869 | 4298 | |
2b4f9037 KH |
4299 | if (from != pos) |
4300 | temp_set_point_both (current_buffer, from, from_byte); | |
4301 | val = call1 (coding->post_read_conversion, make_number (inserted)); | |
4302 | if (! NILP (val)) | |
d46c5b12 | 4303 | { |
2b4f9037 KH |
4304 | CHECK_NUMBER (val, 0); |
4305 | inserted = XFASTINT (val); | |
d46c5b12 | 4306 | } |
2b4f9037 KH |
4307 | if (pos >= from + orig_inserted) |
4308 | temp_set_point (current_buffer, pos + (inserted - orig_inserted)); | |
d46c5b12 | 4309 | } |
4ed46869 | 4310 | |
2b4f9037 KH |
4311 | signal_after_change (from, to - from, inserted); |
4312 | ||
fb88bf2d | 4313 | { |
12410ef1 KH |
4314 | coding->consumed = to_byte - from_byte; |
4315 | coding->consumed_char = to - from; | |
4316 | coding->produced = inserted_byte; | |
4317 | coding->produced_char = inserted; | |
fb88bf2d | 4318 | } |
7553d0e1 | 4319 | |
fb88bf2d | 4320 | return 0; |
d46c5b12 KH |
4321 | } |
4322 | ||
4323 | Lisp_Object | |
4324 | code_convert_string (str, coding, encodep, nocopy) | |
4325 | Lisp_Object str; | |
4ed46869 | 4326 | struct coding_system *coding; |
d46c5b12 | 4327 | int encodep, nocopy; |
4ed46869 | 4328 | { |
d46c5b12 KH |
4329 | int len; |
4330 | char *buf; | |
fc932ac6 RS |
4331 | int from = 0, to = XSTRING (str)->size; |
4332 | int to_byte = STRING_BYTES (XSTRING (str)); | |
d46c5b12 KH |
4333 | struct gcpro gcpro1; |
4334 | Lisp_Object saved_coding_symbol = Qnil; | |
4335 | int result; | |
4ed46869 | 4336 | |
d46c5b12 KH |
4337 | if (encodep && !NILP (coding->pre_write_conversion) |
4338 | || !encodep && !NILP (coding->post_read_conversion)) | |
4339 | { | |
4340 | /* Since we have to call Lisp functions which assume target text | |
4341 | is in a buffer, after setting a temporary buffer, call | |
4342 | code_convert_region. */ | |
4343 | int count = specpdl_ptr - specpdl; | |
4344 | struct buffer *prev = current_buffer; | |
4345 | ||
4346 | record_unwind_protect (Fset_buffer, Fcurrent_buffer ()); | |
4347 | temp_output_buffer_setup (" *code-converting-work*"); | |
4348 | set_buffer_internal (XBUFFER (Vstandard_output)); | |
4349 | if (encodep) | |
4350 | insert_from_string (str, 0, 0, to, to_byte, 0); | |
4351 | else | |
4352 | { | |
4353 | /* We must insert the contents of STR as is without | |
4354 | unibyte<->multibyte conversion. */ | |
4355 | current_buffer->enable_multibyte_characters = Qnil; | |
4356 | insert_from_string (str, 0, 0, to_byte, to_byte, 0); | |
4357 | current_buffer->enable_multibyte_characters = Qt; | |
4358 | } | |
fb88bf2d | 4359 | code_convert_region (BEGV, BEGV_BYTE, ZV, ZV_BYTE, coding, encodep, 1); |
d46c5b12 KH |
4360 | if (encodep) |
4361 | /* We must return the buffer contents as unibyte string. */ | |
4362 | current_buffer->enable_multibyte_characters = Qnil; | |
4363 | str = make_buffer_string (BEGV, ZV, 0); | |
4364 | set_buffer_internal (prev); | |
4365 | return unbind_to (count, str); | |
4366 | } | |
4ed46869 | 4367 | |
d46c5b12 KH |
4368 | if (! encodep && CODING_REQUIRE_DETECTION (coding)) |
4369 | { | |
4370 | /* See the comments in code_convert_region. */ | |
4371 | if (coding->type == coding_type_undecided) | |
4372 | { | |
4373 | detect_coding (coding, XSTRING (str)->data, to_byte); | |
4374 | if (coding->type == coding_type_undecided) | |
4375 | coding->type = coding_type_emacs_mule; | |
4376 | } | |
4377 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
4378 | { | |
4379 | saved_coding_symbol = coding->symbol; | |
4380 | detect_eol (coding, XSTRING (str)->data, to_byte); | |
4381 | if (coding->eol_type == CODING_EOL_UNDECIDED) | |
4382 | coding->eol_type = CODING_EOL_LF; | |
4383 | /* We had better recover the original eol format if we | |
4384 | encounter an inconsitent eol format while decoding. */ | |
4385 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; | |
4386 | } | |
4387 | } | |
4ed46869 | 4388 | |
d46c5b12 KH |
4389 | if (encodep |
4390 | ? ! CODING_REQUIRE_ENCODING (coding) | |
4391 | : ! CODING_REQUIRE_DECODING (coding)) | |
4392 | from = to_byte; | |
4393 | else | |
4394 | { | |
4395 | /* Try to skip the heading and tailing ASCIIs. */ | |
4396 | if (encodep) | |
4397 | shrink_encoding_region (&from, &to_byte, coding, XSTRING (str)->data); | |
4398 | else | |
4399 | shrink_decoding_region (&from, &to_byte, coding, XSTRING (str)->data); | |
4400 | } | |
4401 | if (from == to_byte) | |
4402 | return (nocopy ? str : Fcopy_sequence (str)); | |
4ed46869 | 4403 | |
d46c5b12 KH |
4404 | if (encodep) |
4405 | len = encoding_buffer_size (coding, to_byte - from); | |
4406 | else | |
4407 | len = decoding_buffer_size (coding, to_byte - from); | |
fc932ac6 | 4408 | len += from + STRING_BYTES (XSTRING (str)) - to_byte; |
d46c5b12 KH |
4409 | GCPRO1 (str); |
4410 | buf = get_conversion_buffer (len); | |
4411 | UNGCPRO; | |
4ed46869 | 4412 | |
d46c5b12 KH |
4413 | if (from > 0) |
4414 | bcopy (XSTRING (str)->data, buf, from); | |
4415 | result = (encodep | |
4416 | ? encode_coding (coding, XSTRING (str)->data + from, | |
4417 | buf + from, to_byte - from, len) | |
4418 | : decode_coding (coding, XSTRING (str)->data + from, | |
f30cc612 | 4419 | buf + from, to_byte - from, len)); |
d46c5b12 | 4420 | if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL) |
4ed46869 | 4421 | { |
d46c5b12 KH |
4422 | /* We simple try to decode the whole string again but without |
4423 | eol-conversion this time. */ | |
4424 | coding->eol_type = CODING_EOL_LF; | |
4425 | coding->symbol = saved_coding_symbol; | |
4426 | return code_convert_string (str, coding, encodep, nocopy); | |
4ed46869 | 4427 | } |
d46c5b12 KH |
4428 | |
4429 | bcopy (XSTRING (str)->data + to_byte, buf + from + coding->produced, | |
fc932ac6 | 4430 | STRING_BYTES (XSTRING (str)) - to_byte); |
d46c5b12 | 4431 | |
fc932ac6 | 4432 | len = from + STRING_BYTES (XSTRING (str)) - to_byte; |
d46c5b12 KH |
4433 | if (encodep) |
4434 | str = make_unibyte_string (buf, len + coding->produced); | |
4435 | else | |
bbdf84bd RS |
4436 | str = make_string_from_bytes (buf, len + coding->produced_char, |
4437 | len + coding->produced); | |
d46c5b12 | 4438 | return str; |
4ed46869 KH |
4439 | } |
4440 | ||
4441 | \f | |
4442 | #ifdef emacs | |
4443 | /*** 7. Emacs Lisp library functions ***/ | |
4444 | ||
4ed46869 KH |
4445 | DEFUN ("coding-system-p", Fcoding_system_p, Scoding_system_p, 1, 1, 0, |
4446 | "Return t if OBJECT is nil or a coding-system.\n\ | |
3a73fa5d RS |
4447 | See the documentation of `make-coding-system' for information\n\ |
4448 | about coding-system objects.") | |
4ed46869 KH |
4449 | (obj) |
4450 | Lisp_Object obj; | |
4451 | { | |
4608c386 KH |
4452 | if (NILP (obj)) |
4453 | return Qt; | |
4454 | if (!SYMBOLP (obj)) | |
4455 | return Qnil; | |
4456 | /* Get coding-spec vector for OBJ. */ | |
4457 | obj = Fget (obj, Qcoding_system); | |
4458 | return ((VECTORP (obj) && XVECTOR (obj)->size == 5) | |
4459 | ? Qt : Qnil); | |
4ed46869 KH |
4460 | } |
4461 | ||
9d991de8 RS |
4462 | DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system, |
4463 | Sread_non_nil_coding_system, 1, 1, 0, | |
e0e989f6 | 4464 | "Read a coding system from the minibuffer, prompting with string PROMPT.") |
4ed46869 KH |
4465 | (prompt) |
4466 | Lisp_Object prompt; | |
4467 | { | |
e0e989f6 | 4468 | Lisp_Object val; |
9d991de8 RS |
4469 | do |
4470 | { | |
4608c386 KH |
4471 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
4472 | Qt, Qnil, Qcoding_system_history, Qnil, Qnil); | |
9d991de8 RS |
4473 | } |
4474 | while (XSTRING (val)->size == 0); | |
e0e989f6 | 4475 | return (Fintern (val, Qnil)); |
4ed46869 KH |
4476 | } |
4477 | ||
9b787f3e RS |
4478 | DEFUN ("read-coding-system", Fread_coding_system, Sread_coding_system, 1, 2, 0, |
4479 | "Read a coding system from the minibuffer, prompting with string PROMPT.\n\ | |
4480 | If the user enters null input, return second argument DEFAULT-CODING-SYSTEM.") | |
4481 | (prompt, default_coding_system) | |
4482 | Lisp_Object prompt, default_coding_system; | |
4ed46869 | 4483 | { |
f44d27ce | 4484 | Lisp_Object val; |
9b787f3e RS |
4485 | if (SYMBOLP (default_coding_system)) |
4486 | XSETSTRING (default_coding_system, XSYMBOL (default_coding_system)->name); | |
4608c386 | 4487 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
9b787f3e RS |
4488 | Qt, Qnil, Qcoding_system_history, |
4489 | default_coding_system, Qnil); | |
e0e989f6 | 4490 | return (XSTRING (val)->size == 0 ? Qnil : Fintern (val, Qnil)); |
4ed46869 KH |
4491 | } |
4492 | ||
4493 | DEFUN ("check-coding-system", Fcheck_coding_system, Scheck_coding_system, | |
4494 | 1, 1, 0, | |
4495 | "Check validity of CODING-SYSTEM.\n\ | |
3a73fa5d RS |
4496 | If valid, return CODING-SYSTEM, else signal a `coding-system-error' error.\n\ |
4497 | It is valid if it is a symbol with a non-nil `coding-system' property.\n\ | |
4ed46869 KH |
4498 | The value of property should be a vector of length 5.") |
4499 | (coding_system) | |
4500 | Lisp_Object coding_system; | |
4501 | { | |
4502 | CHECK_SYMBOL (coding_system, 0); | |
4503 | if (!NILP (Fcoding_system_p (coding_system))) | |
4504 | return coding_system; | |
4505 | while (1) | |
02ba4723 | 4506 | Fsignal (Qcoding_system_error, Fcons (coding_system, Qnil)); |
4ed46869 | 4507 | } |
3a73fa5d | 4508 | \f |
d46c5b12 KH |
4509 | Lisp_Object |
4510 | detect_coding_system (src, src_bytes, highest) | |
4511 | unsigned char *src; | |
4512 | int src_bytes, highest; | |
4ed46869 KH |
4513 | { |
4514 | int coding_mask, eol_type; | |
d46c5b12 KH |
4515 | Lisp_Object val, tmp; |
4516 | int dummy; | |
4ed46869 | 4517 | |
d46c5b12 KH |
4518 | coding_mask = detect_coding_mask (src, src_bytes, NULL, &dummy); |
4519 | eol_type = detect_eol_type (src, src_bytes, &dummy); | |
4520 | if (eol_type == CODING_EOL_INCONSISTENT) | |
4521 | eol_type == CODING_EOL_UNDECIDED; | |
4ed46869 | 4522 | |
d46c5b12 | 4523 | if (!coding_mask) |
4ed46869 | 4524 | { |
27901516 | 4525 | val = Qundecided; |
d46c5b12 | 4526 | if (eol_type != CODING_EOL_UNDECIDED) |
4ed46869 | 4527 | { |
f44d27ce RS |
4528 | Lisp_Object val2; |
4529 | val2 = Fget (Qundecided, Qeol_type); | |
4ed46869 KH |
4530 | if (VECTORP (val2)) |
4531 | val = XVECTOR (val2)->contents[eol_type]; | |
4532 | } | |
80e803b4 | 4533 | return (highest ? val : Fcons (val, Qnil)); |
4ed46869 | 4534 | } |
4ed46869 | 4535 | |
d46c5b12 KH |
4536 | /* At first, gather possible coding systems in VAL. */ |
4537 | val = Qnil; | |
4538 | for (tmp = Vcoding_category_list; !NILP (tmp); tmp = XCONS (tmp)->cdr) | |
4ed46869 | 4539 | { |
d46c5b12 KH |
4540 | int idx |
4541 | = XFASTINT (Fget (XCONS (tmp)->car, Qcoding_category_index)); | |
4542 | if (coding_mask & (1 << idx)) | |
4ed46869 | 4543 | { |
d46c5b12 KH |
4544 | val = Fcons (Fsymbol_value (XCONS (tmp)->car), val); |
4545 | if (highest) | |
4546 | break; | |
4ed46869 KH |
4547 | } |
4548 | } | |
d46c5b12 KH |
4549 | if (!highest) |
4550 | val = Fnreverse (val); | |
4ed46869 | 4551 | |
65059037 | 4552 | /* Then, replace the elements with subsidiary coding systems. */ |
d46c5b12 | 4553 | for (tmp = val; !NILP (tmp); tmp = XCONS (tmp)->cdr) |
4ed46869 | 4554 | { |
65059037 RS |
4555 | if (eol_type != CODING_EOL_UNDECIDED |
4556 | && eol_type != CODING_EOL_INCONSISTENT) | |
4ed46869 | 4557 | { |
d46c5b12 KH |
4558 | Lisp_Object eol; |
4559 | eol = Fget (XCONS (tmp)->car, Qeol_type); | |
4560 | if (VECTORP (eol)) | |
4561 | XCONS (tmp)->car = XVECTOR (eol)->contents[eol_type]; | |
4ed46869 KH |
4562 | } |
4563 | } | |
d46c5b12 KH |
4564 | return (highest ? XCONS (val)->car : val); |
4565 | } | |
4ed46869 | 4566 | |
d46c5b12 KH |
4567 | DEFUN ("detect-coding-region", Fdetect_coding_region, Sdetect_coding_region, |
4568 | 2, 3, 0, | |
4569 | "Detect coding system of the text in the region between START and END.\n\ | |
4570 | Return a list of possible coding systems ordered by priority.\n\ | |
4571 | \n\ | |
80e803b4 KH |
4572 | If only ASCII characters are found, it returns a list of single element\n\ |
4573 | `undecided' or its subsidiary coding system according to a detected\n\ | |
4574 | end-of-line format.\n\ | |
d46c5b12 KH |
4575 | \n\ |
4576 | If optional argument HIGHEST is non-nil, return the coding system of\n\ | |
4577 | highest priority.") | |
4578 | (start, end, highest) | |
4579 | Lisp_Object start, end, highest; | |
4580 | { | |
4581 | int from, to; | |
4582 | int from_byte, to_byte; | |
6289dd10 | 4583 | |
d46c5b12 KH |
4584 | CHECK_NUMBER_COERCE_MARKER (start, 0); |
4585 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
4ed46869 | 4586 | |
d46c5b12 KH |
4587 | validate_region (&start, &end); |
4588 | from = XINT (start), to = XINT (end); | |
4589 | from_byte = CHAR_TO_BYTE (from); | |
4590 | to_byte = CHAR_TO_BYTE (to); | |
6289dd10 | 4591 | |
d46c5b12 KH |
4592 | if (from < GPT && to >= GPT) |
4593 | move_gap_both (to, to_byte); | |
4ed46869 | 4594 | |
d46c5b12 KH |
4595 | return detect_coding_system (BYTE_POS_ADDR (from_byte), |
4596 | to_byte - from_byte, | |
4597 | !NILP (highest)); | |
4598 | } | |
6289dd10 | 4599 | |
d46c5b12 KH |
4600 | DEFUN ("detect-coding-string", Fdetect_coding_string, Sdetect_coding_string, |
4601 | 1, 2, 0, | |
4602 | "Detect coding system of the text in STRING.\n\ | |
4603 | Return a list of possible coding systems ordered by priority.\n\ | |
4604 | \n\ | |
80e803b4 KH |
4605 | If only ASCII characters are found, it returns a list of single element\n\ |
4606 | `undecided' or its subsidiary coding system according to a detected\n\ | |
4607 | end-of-line format.\n\ | |
d46c5b12 KH |
4608 | \n\ |
4609 | If optional argument HIGHEST is non-nil, return the coding system of\n\ | |
4610 | highest priority.") | |
4611 | (string, highest) | |
4612 | Lisp_Object string, highest; | |
4613 | { | |
4614 | CHECK_STRING (string, 0); | |
4ed46869 | 4615 | |
d46c5b12 | 4616 | return detect_coding_system (XSTRING (string)->data, |
fc932ac6 | 4617 | STRING_BYTES (XSTRING (string)), |
d46c5b12 | 4618 | !NILP (highest)); |
4ed46869 KH |
4619 | } |
4620 | ||
4031e2bf KH |
4621 | Lisp_Object |
4622 | code_convert_region1 (start, end, coding_system, encodep) | |
d46c5b12 | 4623 | Lisp_Object start, end, coding_system; |
4031e2bf | 4624 | int encodep; |
3a73fa5d RS |
4625 | { |
4626 | struct coding_system coding; | |
4031e2bf | 4627 | int from, to, len; |
3a73fa5d | 4628 | |
d46c5b12 KH |
4629 | CHECK_NUMBER_COERCE_MARKER (start, 0); |
4630 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
3a73fa5d RS |
4631 | CHECK_SYMBOL (coding_system, 2); |
4632 | ||
d46c5b12 KH |
4633 | validate_region (&start, &end); |
4634 | from = XFASTINT (start); | |
4635 | to = XFASTINT (end); | |
4636 | ||
3a73fa5d | 4637 | if (NILP (coding_system)) |
d46c5b12 KH |
4638 | return make_number (to - from); |
4639 | ||
3a73fa5d | 4640 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
d46c5b12 | 4641 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); |
3a73fa5d | 4642 | |
d46c5b12 | 4643 | coding.mode |= CODING_MODE_LAST_BLOCK; |
fb88bf2d KH |
4644 | code_convert_region (from, CHAR_TO_BYTE (from), to, CHAR_TO_BYTE (to), |
4645 | &coding, encodep, 1); | |
f072a3e8 | 4646 | Vlast_coding_system_used = coding.symbol; |
fb88bf2d | 4647 | return make_number (coding.produced_char); |
4031e2bf KH |
4648 | } |
4649 | ||
4650 | DEFUN ("decode-coding-region", Fdecode_coding_region, Sdecode_coding_region, | |
4651 | 3, 3, "r\nzCoding system: ", | |
4652 | "Decode the current region by specified coding system.\n\ | |
4653 | When called from a program, takes three arguments:\n\ | |
4654 | START, END, and CODING-SYSTEM. START and END are buffer positions.\n\ | |
f072a3e8 RS |
4655 | This function sets `last-coding-system-used' to the precise coding system\n\ |
4656 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ | |
4657 | not fully specified.)\n\ | |
4658 | It returns the length of the decoded text.") | |
4031e2bf KH |
4659 | (start, end, coding_system) |
4660 | Lisp_Object start, end, coding_system; | |
4661 | { | |
4662 | return code_convert_region1 (start, end, coding_system, 0); | |
3a73fa5d RS |
4663 | } |
4664 | ||
4665 | DEFUN ("encode-coding-region", Fencode_coding_region, Sencode_coding_region, | |
4666 | 3, 3, "r\nzCoding system: ", | |
d46c5b12 | 4667 | "Encode the current region by specified coding system.\n\ |
3a73fa5d | 4668 | When called from a program, takes three arguments:\n\ |
d46c5b12 | 4669 | START, END, and CODING-SYSTEM. START and END are buffer positions.\n\ |
f072a3e8 RS |
4670 | This function sets `last-coding-system-used' to the precise coding system\n\ |
4671 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ | |
4672 | not fully specified.)\n\ | |
4673 | It returns the length of the encoded text.") | |
d46c5b12 KH |
4674 | (start, end, coding_system) |
4675 | Lisp_Object start, end, coding_system; | |
3a73fa5d | 4676 | { |
4031e2bf KH |
4677 | return code_convert_region1 (start, end, coding_system, 1); |
4678 | } | |
3a73fa5d | 4679 | |
4031e2bf KH |
4680 | Lisp_Object |
4681 | code_convert_string1 (string, coding_system, nocopy, encodep) | |
4682 | Lisp_Object string, coding_system, nocopy; | |
4683 | int encodep; | |
4684 | { | |
4685 | struct coding_system coding; | |
3a73fa5d | 4686 | |
4031e2bf KH |
4687 | CHECK_STRING (string, 0); |
4688 | CHECK_SYMBOL (coding_system, 1); | |
4ed46869 | 4689 | |
d46c5b12 | 4690 | if (NILP (coding_system)) |
4031e2bf | 4691 | return (NILP (nocopy) ? Fcopy_sequence (string) : string); |
4ed46869 | 4692 | |
d46c5b12 KH |
4693 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
4694 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); | |
5f1cd180 | 4695 | |
d46c5b12 | 4696 | coding.mode |= CODING_MODE_LAST_BLOCK; |
f072a3e8 | 4697 | Vlast_coding_system_used = coding.symbol; |
4031e2bf | 4698 | return code_convert_string (string, &coding, encodep, !NILP (nocopy)); |
4ed46869 KH |
4699 | } |
4700 | ||
4ed46869 | 4701 | DEFUN ("decode-coding-string", Fdecode_coding_string, Sdecode_coding_string, |
e0e989f6 KH |
4702 | 2, 3, 0, |
4703 | "Decode STRING which is encoded in CODING-SYSTEM, and return the result.\n\ | |
fe487a71 | 4704 | Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\ |
f072a3e8 RS |
4705 | if the decoding operation is trivial.\n\ |
4706 | This function sets `last-coding-system-used' to the precise coding system\n\ | |
4707 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ | |
4708 | not fully specified.)") | |
e0e989f6 KH |
4709 | (string, coding_system, nocopy) |
4710 | Lisp_Object string, coding_system, nocopy; | |
4ed46869 | 4711 | { |
f072a3e8 | 4712 | return code_convert_string1 (string, coding_system, nocopy, 0); |
4ed46869 KH |
4713 | } |
4714 | ||
4715 | DEFUN ("encode-coding-string", Fencode_coding_string, Sencode_coding_string, | |
e0e989f6 KH |
4716 | 2, 3, 0, |
4717 | "Encode STRING to CODING-SYSTEM, and return the result.\n\ | |
fe487a71 | 4718 | Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\ |
f072a3e8 RS |
4719 | if the encoding operation is trivial.\n\ |
4720 | This function sets `last-coding-system-used' to the precise coding system\n\ | |
4721 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ | |
4722 | not fully specified.)") | |
e0e989f6 KH |
4723 | (string, coding_system, nocopy) |
4724 | Lisp_Object string, coding_system, nocopy; | |
4ed46869 | 4725 | { |
f072a3e8 | 4726 | return code_convert_string1 (string, coding_system, nocopy, 1); |
4ed46869 | 4727 | } |
4031e2bf | 4728 | |
ecec61c1 KH |
4729 | /* Encode or decode STRING according to CODING_SYSTEM. |
4730 | Do not set Vlast_coding_system_used. */ | |
4731 | ||
4732 | Lisp_Object | |
4733 | code_convert_string_norecord (string, coding_system, encodep) | |
4734 | Lisp_Object string, coding_system; | |
4735 | int encodep; | |
4736 | { | |
4737 | struct coding_system coding; | |
4738 | ||
4739 | CHECK_STRING (string, 0); | |
4740 | CHECK_SYMBOL (coding_system, 1); | |
4741 | ||
4742 | if (NILP (coding_system)) | |
4743 | return string; | |
4744 | ||
4745 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) | |
4746 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); | |
4747 | ||
4748 | coding.mode |= CODING_MODE_LAST_BLOCK; | |
4749 | return code_convert_string (string, &coding, encodep, Qt); | |
4750 | } | |
3a73fa5d | 4751 | \f |
4ed46869 | 4752 | DEFUN ("decode-sjis-char", Fdecode_sjis_char, Sdecode_sjis_char, 1, 1, 0, |
e0e989f6 | 4753 | "Decode a JISX0208 character of shift-jis encoding.\n\ |
4ed46869 KH |
4754 | CODE is the character code in SJIS.\n\ |
4755 | Return the corresponding character.") | |
4756 | (code) | |
4757 | Lisp_Object code; | |
4758 | { | |
4759 | unsigned char c1, c2, s1, s2; | |
4760 | Lisp_Object val; | |
4761 | ||
4762 | CHECK_NUMBER (code, 0); | |
4763 | s1 = (XFASTINT (code)) >> 8, s2 = (XFASTINT (code)) & 0xFF; | |
4764 | DECODE_SJIS (s1, s2, c1, c2); | |
4765 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset_jisx0208, c1, c2)); | |
4766 | return val; | |
4767 | } | |
4768 | ||
4769 | DEFUN ("encode-sjis-char", Fencode_sjis_char, Sencode_sjis_char, 1, 1, 0, | |
d46c5b12 | 4770 | "Encode a JISX0208 character CHAR to SJIS coding system.\n\ |
4ed46869 KH |
4771 | Return the corresponding character code in SJIS.") |
4772 | (ch) | |
4773 | Lisp_Object ch; | |
4774 | { | |
bcf26d6a | 4775 | int charset, c1, c2, s1, s2; |
4ed46869 KH |
4776 | Lisp_Object val; |
4777 | ||
4778 | CHECK_NUMBER (ch, 0); | |
4779 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
4780 | if (charset == charset_jisx0208) | |
4781 | { | |
4782 | ENCODE_SJIS (c1, c2, s1, s2); | |
bcf26d6a | 4783 | XSETFASTINT (val, (s1 << 8) | s2); |
4ed46869 KH |
4784 | } |
4785 | else | |
4786 | XSETFASTINT (val, 0); | |
4787 | return val; | |
4788 | } | |
4789 | ||
4790 | DEFUN ("decode-big5-char", Fdecode_big5_char, Sdecode_big5_char, 1, 1, 0, | |
d46c5b12 | 4791 | "Decode a Big5 character CODE of BIG5 coding system.\n\ |
4ed46869 KH |
4792 | CODE is the character code in BIG5.\n\ |
4793 | Return the corresponding character.") | |
4794 | (code) | |
4795 | Lisp_Object code; | |
4796 | { | |
4797 | int charset; | |
4798 | unsigned char b1, b2, c1, c2; | |
4799 | Lisp_Object val; | |
4800 | ||
4801 | CHECK_NUMBER (code, 0); | |
4802 | b1 = (XFASTINT (code)) >> 8, b2 = (XFASTINT (code)) & 0xFF; | |
4803 | DECODE_BIG5 (b1, b2, charset, c1, c2); | |
4804 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset, c1, c2)); | |
4805 | return val; | |
4806 | } | |
4807 | ||
4808 | DEFUN ("encode-big5-char", Fencode_big5_char, Sencode_big5_char, 1, 1, 0, | |
d46c5b12 | 4809 | "Encode the Big5 character CHAR to BIG5 coding system.\n\ |
4ed46869 KH |
4810 | Return the corresponding character code in Big5.") |
4811 | (ch) | |
4812 | Lisp_Object ch; | |
4813 | { | |
bcf26d6a | 4814 | int charset, c1, c2, b1, b2; |
4ed46869 KH |
4815 | Lisp_Object val; |
4816 | ||
4817 | CHECK_NUMBER (ch, 0); | |
4818 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
4819 | if (charset == charset_big5_1 || charset == charset_big5_2) | |
4820 | { | |
4821 | ENCODE_BIG5 (charset, c1, c2, b1, b2); | |
bcf26d6a | 4822 | XSETFASTINT (val, (b1 << 8) | b2); |
4ed46869 KH |
4823 | } |
4824 | else | |
4825 | XSETFASTINT (val, 0); | |
4826 | return val; | |
4827 | } | |
3a73fa5d | 4828 | \f |
1ba9e4ab KH |
4829 | DEFUN ("set-terminal-coding-system-internal", |
4830 | Fset_terminal_coding_system_internal, | |
4831 | Sset_terminal_coding_system_internal, 1, 1, 0, "") | |
4ed46869 KH |
4832 | (coding_system) |
4833 | Lisp_Object coding_system; | |
4834 | { | |
4835 | CHECK_SYMBOL (coding_system, 0); | |
4836 | setup_coding_system (Fcheck_coding_system (coding_system), &terminal_coding); | |
70c22245 | 4837 | /* We had better not send unsafe characters to terminal. */ |
6e85d753 KH |
4838 | terminal_coding.flags |= CODING_FLAG_ISO_SAFE; |
4839 | ||
4ed46869 KH |
4840 | return Qnil; |
4841 | } | |
4842 | ||
c4825358 KH |
4843 | DEFUN ("set-safe-terminal-coding-system-internal", |
4844 | Fset_safe_terminal_coding_system_internal, | |
4845 | Sset_safe_terminal_coding_system_internal, 1, 1, 0, "") | |
4846 | (coding_system) | |
4847 | Lisp_Object coding_system; | |
4848 | { | |
4849 | CHECK_SYMBOL (coding_system, 0); | |
4850 | setup_coding_system (Fcheck_coding_system (coding_system), | |
4851 | &safe_terminal_coding); | |
4852 | return Qnil; | |
4853 | } | |
4854 | ||
4ed46869 KH |
4855 | DEFUN ("terminal-coding-system", |
4856 | Fterminal_coding_system, Sterminal_coding_system, 0, 0, 0, | |
3a73fa5d | 4857 | "Return coding system specified for terminal output.") |
4ed46869 KH |
4858 | () |
4859 | { | |
4860 | return terminal_coding.symbol; | |
4861 | } | |
4862 | ||
1ba9e4ab KH |
4863 | DEFUN ("set-keyboard-coding-system-internal", |
4864 | Fset_keyboard_coding_system_internal, | |
4865 | Sset_keyboard_coding_system_internal, 1, 1, 0, "") | |
4ed46869 KH |
4866 | (coding_system) |
4867 | Lisp_Object coding_system; | |
4868 | { | |
4869 | CHECK_SYMBOL (coding_system, 0); | |
4870 | setup_coding_system (Fcheck_coding_system (coding_system), &keyboard_coding); | |
4871 | return Qnil; | |
4872 | } | |
4873 | ||
4874 | DEFUN ("keyboard-coding-system", | |
4875 | Fkeyboard_coding_system, Skeyboard_coding_system, 0, 0, 0, | |
3a73fa5d | 4876 | "Return coding system specified for decoding keyboard input.") |
4ed46869 KH |
4877 | () |
4878 | { | |
4879 | return keyboard_coding.symbol; | |
4880 | } | |
4881 | ||
4882 | \f | |
a5d301df KH |
4883 | DEFUN ("find-operation-coding-system", Ffind_operation_coding_system, |
4884 | Sfind_operation_coding_system, 1, MANY, 0, | |
4885 | "Choose a coding system for an operation based on the target name.\n\ | |
9ce27fde KH |
4886 | The value names a pair of coding systems: (DECODING-SYSTEM ENCODING-SYSTEM).\n\ |
4887 | DECODING-SYSTEM is the coding system to use for decoding\n\ | |
4888 | \(in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system\n\ | |
4889 | for encoding (in case OPERATION does encoding).\n\ | |
ccdb79f5 RS |
4890 | \n\ |
4891 | The first argument OPERATION specifies an I/O primitive:\n\ | |
4892 | For file I/O, `insert-file-contents' or `write-region'.\n\ | |
4893 | For process I/O, `call-process', `call-process-region', or `start-process'.\n\ | |
4894 | For network I/O, `open-network-stream'.\n\ | |
4895 | \n\ | |
4896 | The remaining arguments should be the same arguments that were passed\n\ | |
4897 | to the primitive. Depending on which primitive, one of those arguments\n\ | |
4898 | is selected as the TARGET. For example, if OPERATION does file I/O,\n\ | |
4899 | whichever argument specifies the file name is TARGET.\n\ | |
4900 | \n\ | |
4901 | TARGET has a meaning which depends on OPERATION:\n\ | |
4ed46869 KH |
4902 | For file I/O, TARGET is a file name.\n\ |
4903 | For process I/O, TARGET is a process name.\n\ | |
4904 | For network I/O, TARGET is a service name or a port number\n\ | |
4905 | \n\ | |
02ba4723 KH |
4906 | This function looks up what specified for TARGET in,\n\ |
4907 | `file-coding-system-alist', `process-coding-system-alist',\n\ | |
4908 | or `network-coding-system-alist' depending on OPERATION.\n\ | |
4909 | They may specify a coding system, a cons of coding systems,\n\ | |
4910 | or a function symbol to call.\n\ | |
4911 | In the last case, we call the function with one argument,\n\ | |
9ce27fde | 4912 | which is a list of all the arguments given to this function.") |
4ed46869 KH |
4913 | (nargs, args) |
4914 | int nargs; | |
4915 | Lisp_Object *args; | |
4916 | { | |
4917 | Lisp_Object operation, target_idx, target, val; | |
4918 | register Lisp_Object chain; | |
4919 | ||
4920 | if (nargs < 2) | |
4921 | error ("Too few arguments"); | |
4922 | operation = args[0]; | |
4923 | if (!SYMBOLP (operation) | |
4924 | || !INTEGERP (target_idx = Fget (operation, Qtarget_idx))) | |
4925 | error ("Invalid first arguement"); | |
4926 | if (nargs < 1 + XINT (target_idx)) | |
4927 | error ("Too few arguments for operation: %s", | |
4928 | XSYMBOL (operation)->name->data); | |
4929 | target = args[XINT (target_idx) + 1]; | |
4930 | if (!(STRINGP (target) | |
4931 | || (EQ (operation, Qopen_network_stream) && INTEGERP (target)))) | |
4932 | error ("Invalid %dth argument", XINT (target_idx) + 1); | |
4933 | ||
2e34157c RS |
4934 | chain = ((EQ (operation, Qinsert_file_contents) |
4935 | || EQ (operation, Qwrite_region)) | |
02ba4723 | 4936 | ? Vfile_coding_system_alist |
2e34157c | 4937 | : (EQ (operation, Qopen_network_stream) |
02ba4723 KH |
4938 | ? Vnetwork_coding_system_alist |
4939 | : Vprocess_coding_system_alist)); | |
4ed46869 KH |
4940 | if (NILP (chain)) |
4941 | return Qnil; | |
4942 | ||
02ba4723 | 4943 | for (; CONSP (chain); chain = XCONS (chain)->cdr) |
4ed46869 | 4944 | { |
f44d27ce RS |
4945 | Lisp_Object elt; |
4946 | elt = XCONS (chain)->car; | |
4ed46869 KH |
4947 | |
4948 | if (CONSP (elt) | |
4949 | && ((STRINGP (target) | |
4950 | && STRINGP (XCONS (elt)->car) | |
4951 | && fast_string_match (XCONS (elt)->car, target) >= 0) | |
4952 | || (INTEGERP (target) && EQ (target, XCONS (elt)->car)))) | |
02ba4723 KH |
4953 | { |
4954 | val = XCONS (elt)->cdr; | |
b19fd4c5 KH |
4955 | /* Here, if VAL is both a valid coding system and a valid |
4956 | function symbol, we return VAL as a coding system. */ | |
02ba4723 KH |
4957 | if (CONSP (val)) |
4958 | return val; | |
4959 | if (! SYMBOLP (val)) | |
4960 | return Qnil; | |
4961 | if (! NILP (Fcoding_system_p (val))) | |
4962 | return Fcons (val, val); | |
b19fd4c5 KH |
4963 | if (! NILP (Ffboundp (val))) |
4964 | { | |
4965 | val = call1 (val, Flist (nargs, args)); | |
4966 | if (CONSP (val)) | |
4967 | return val; | |
4968 | if (SYMBOLP (val) && ! NILP (Fcoding_system_p (val))) | |
4969 | return Fcons (val, val); | |
4970 | } | |
02ba4723 KH |
4971 | return Qnil; |
4972 | } | |
4ed46869 KH |
4973 | } |
4974 | return Qnil; | |
4975 | } | |
4976 | ||
d46c5b12 KH |
4977 | DEFUN ("update-iso-coding-systems", Fupdate_iso_coding_systems, |
4978 | Supdate_iso_coding_systems, 0, 0, 0, | |
4979 | "Update internal database for ISO2022 based coding systems.\n\ | |
4980 | When values of the following coding categories are changed, you must\n\ | |
4981 | call this function:\n\ | |
4982 | coding-category-iso-7, coding-category-iso-7-tight,\n\ | |
4983 | coding-category-iso-8-1, coding-category-iso-8-2,\n\ | |
4984 | coding-category-iso-7-else, coding-category-iso-8-else") | |
4985 | () | |
4986 | { | |
4987 | int i; | |
4988 | ||
4989 | for (i = CODING_CATEGORY_IDX_ISO_7; i <= CODING_CATEGORY_IDX_ISO_8_ELSE; | |
4990 | i++) | |
4991 | { | |
4992 | if (! coding_system_table[i]) | |
4993 | coding_system_table[i] | |
4994 | = (struct coding_system *) xmalloc (sizeof (struct coding_system)); | |
4995 | setup_coding_system | |
4996 | (XSYMBOL (XVECTOR (Vcoding_category_table)->contents[i])->value, | |
4997 | coding_system_table[i]); | |
4998 | } | |
4999 | return Qnil; | |
5000 | } | |
5001 | ||
66cfb530 KH |
5002 | DEFUN ("set-coding-priority-internal", Fset_coding_priority_internal, |
5003 | Sset_coding_priority_internal, 0, 0, 0, | |
5004 | "Update internal database for the current value of `coding-category-list'.\n\ | |
5005 | This function is internal use only.") | |
5006 | () | |
5007 | { | |
5008 | int i = 0, idx; | |
5009 | Lisp_Object val = Vcoding_category_list; | |
5010 | ||
5011 | while (CONSP (val) && i < CODING_CATEGORY_IDX_MAX) | |
5012 | { | |
5013 | if (! SYMBOLP (XCONS (val)->car)) | |
5014 | break; | |
5015 | idx = XFASTINT (Fget (XCONS (val)->car, Qcoding_category_index)); | |
5016 | if (idx >= CODING_CATEGORY_IDX_MAX) | |
5017 | break; | |
5018 | coding_priorities[i++] = (1 << idx); | |
5019 | val = XCONS (val)->cdr; | |
5020 | } | |
5021 | /* If coding-category-list is valid and contains all coding | |
5022 | categories, `i' should be CODING_CATEGORY_IDX_MAX now. If not, | |
5023 | the following code saves Emacs from craching. */ | |
5024 | while (i < CODING_CATEGORY_IDX_MAX) | |
5025 | coding_priorities[i++] = CODING_CATEGORY_MASK_RAW_TEXT; | |
5026 | ||
5027 | return Qnil; | |
5028 | } | |
5029 | ||
4ed46869 KH |
5030 | #endif /* emacs */ |
5031 | ||
5032 | \f | |
5033 | /*** 8. Post-amble ***/ | |
5034 | ||
dfcf069d | 5035 | void |
4ed46869 KH |
5036 | init_coding_once () |
5037 | { | |
5038 | int i; | |
5039 | ||
0ef69138 | 5040 | /* Emacs' internal format specific initialize routine. */ |
4ed46869 KH |
5041 | for (i = 0; i <= 0x20; i++) |
5042 | emacs_code_class[i] = EMACS_control_code; | |
5043 | emacs_code_class[0x0A] = EMACS_linefeed_code; | |
5044 | emacs_code_class[0x0D] = EMACS_carriage_return_code; | |
5045 | for (i = 0x21 ; i < 0x7F; i++) | |
5046 | emacs_code_class[i] = EMACS_ascii_code; | |
5047 | emacs_code_class[0x7F] = EMACS_control_code; | |
5048 | emacs_code_class[0x80] = EMACS_leading_code_composition; | |
5049 | for (i = 0x81; i < 0xFF; i++) | |
5050 | emacs_code_class[i] = EMACS_invalid_code; | |
5051 | emacs_code_class[LEADING_CODE_PRIVATE_11] = EMACS_leading_code_3; | |
5052 | emacs_code_class[LEADING_CODE_PRIVATE_12] = EMACS_leading_code_3; | |
5053 | emacs_code_class[LEADING_CODE_PRIVATE_21] = EMACS_leading_code_4; | |
5054 | emacs_code_class[LEADING_CODE_PRIVATE_22] = EMACS_leading_code_4; | |
5055 | ||
5056 | /* ISO2022 specific initialize routine. */ | |
5057 | for (i = 0; i < 0x20; i++) | |
5058 | iso_code_class[i] = ISO_control_code; | |
5059 | for (i = 0x21; i < 0x7F; i++) | |
5060 | iso_code_class[i] = ISO_graphic_plane_0; | |
5061 | for (i = 0x80; i < 0xA0; i++) | |
5062 | iso_code_class[i] = ISO_control_code; | |
5063 | for (i = 0xA1; i < 0xFF; i++) | |
5064 | iso_code_class[i] = ISO_graphic_plane_1; | |
5065 | iso_code_class[0x20] = iso_code_class[0x7F] = ISO_0x20_or_0x7F; | |
5066 | iso_code_class[0xA0] = iso_code_class[0xFF] = ISO_0xA0_or_0xFF; | |
5067 | iso_code_class[ISO_CODE_CR] = ISO_carriage_return; | |
5068 | iso_code_class[ISO_CODE_SO] = ISO_shift_out; | |
5069 | iso_code_class[ISO_CODE_SI] = ISO_shift_in; | |
5070 | iso_code_class[ISO_CODE_SS2_7] = ISO_single_shift_2_7; | |
5071 | iso_code_class[ISO_CODE_ESC] = ISO_escape; | |
5072 | iso_code_class[ISO_CODE_SS2] = ISO_single_shift_2; | |
5073 | iso_code_class[ISO_CODE_SS3] = ISO_single_shift_3; | |
5074 | iso_code_class[ISO_CODE_CSI] = ISO_control_sequence_introducer; | |
5075 | ||
e0e989f6 KH |
5076 | conversion_buffer_size = MINIMUM_CONVERSION_BUFFER_SIZE; |
5077 | conversion_buffer = (char *) xmalloc (MINIMUM_CONVERSION_BUFFER_SIZE); | |
5078 | ||
5079 | setup_coding_system (Qnil, &keyboard_coding); | |
5080 | setup_coding_system (Qnil, &terminal_coding); | |
c4825358 | 5081 | setup_coding_system (Qnil, &safe_terminal_coding); |
9ce27fde | 5082 | |
d46c5b12 KH |
5083 | bzero (coding_system_table, sizeof coding_system_table); |
5084 | ||
66cfb530 KH |
5085 | bzero (ascii_skip_code, sizeof ascii_skip_code); |
5086 | for (i = 0; i < 128; i++) | |
5087 | ascii_skip_code[i] = 1; | |
5088 | ||
9ce27fde KH |
5089 | #if defined (MSDOS) || defined (WINDOWSNT) |
5090 | system_eol_type = CODING_EOL_CRLF; | |
5091 | #else | |
5092 | system_eol_type = CODING_EOL_LF; | |
5093 | #endif | |
e0e989f6 KH |
5094 | } |
5095 | ||
5096 | #ifdef emacs | |
5097 | ||
dfcf069d | 5098 | void |
e0e989f6 KH |
5099 | syms_of_coding () |
5100 | { | |
5101 | Qtarget_idx = intern ("target-idx"); | |
5102 | staticpro (&Qtarget_idx); | |
5103 | ||
bb0115a2 RS |
5104 | Qcoding_system_history = intern ("coding-system-history"); |
5105 | staticpro (&Qcoding_system_history); | |
5106 | Fset (Qcoding_system_history, Qnil); | |
5107 | ||
9ce27fde | 5108 | /* Target FILENAME is the first argument. */ |
e0e989f6 | 5109 | Fput (Qinsert_file_contents, Qtarget_idx, make_number (0)); |
9ce27fde | 5110 | /* Target FILENAME is the third argument. */ |
e0e989f6 KH |
5111 | Fput (Qwrite_region, Qtarget_idx, make_number (2)); |
5112 | ||
5113 | Qcall_process = intern ("call-process"); | |
5114 | staticpro (&Qcall_process); | |
9ce27fde | 5115 | /* Target PROGRAM is the first argument. */ |
e0e989f6 KH |
5116 | Fput (Qcall_process, Qtarget_idx, make_number (0)); |
5117 | ||
5118 | Qcall_process_region = intern ("call-process-region"); | |
5119 | staticpro (&Qcall_process_region); | |
9ce27fde | 5120 | /* Target PROGRAM is the third argument. */ |
e0e989f6 KH |
5121 | Fput (Qcall_process_region, Qtarget_idx, make_number (2)); |
5122 | ||
5123 | Qstart_process = intern ("start-process"); | |
5124 | staticpro (&Qstart_process); | |
9ce27fde | 5125 | /* Target PROGRAM is the third argument. */ |
e0e989f6 KH |
5126 | Fput (Qstart_process, Qtarget_idx, make_number (2)); |
5127 | ||
5128 | Qopen_network_stream = intern ("open-network-stream"); | |
5129 | staticpro (&Qopen_network_stream); | |
9ce27fde | 5130 | /* Target SERVICE is the fourth argument. */ |
e0e989f6 KH |
5131 | Fput (Qopen_network_stream, Qtarget_idx, make_number (3)); |
5132 | ||
4ed46869 KH |
5133 | Qcoding_system = intern ("coding-system"); |
5134 | staticpro (&Qcoding_system); | |
5135 | ||
5136 | Qeol_type = intern ("eol-type"); | |
5137 | staticpro (&Qeol_type); | |
5138 | ||
5139 | Qbuffer_file_coding_system = intern ("buffer-file-coding-system"); | |
5140 | staticpro (&Qbuffer_file_coding_system); | |
5141 | ||
5142 | Qpost_read_conversion = intern ("post-read-conversion"); | |
5143 | staticpro (&Qpost_read_conversion); | |
5144 | ||
5145 | Qpre_write_conversion = intern ("pre-write-conversion"); | |
5146 | staticpro (&Qpre_write_conversion); | |
5147 | ||
27901516 KH |
5148 | Qno_conversion = intern ("no-conversion"); |
5149 | staticpro (&Qno_conversion); | |
5150 | ||
5151 | Qundecided = intern ("undecided"); | |
5152 | staticpro (&Qundecided); | |
5153 | ||
4ed46869 KH |
5154 | Qcoding_system_p = intern ("coding-system-p"); |
5155 | staticpro (&Qcoding_system_p); | |
5156 | ||
5157 | Qcoding_system_error = intern ("coding-system-error"); | |
5158 | staticpro (&Qcoding_system_error); | |
5159 | ||
5160 | Fput (Qcoding_system_error, Qerror_conditions, | |
5161 | Fcons (Qcoding_system_error, Fcons (Qerror, Qnil))); | |
5162 | Fput (Qcoding_system_error, Qerror_message, | |
9ce27fde | 5163 | build_string ("Invalid coding system")); |
4ed46869 | 5164 | |
d46c5b12 KH |
5165 | Qcoding_category = intern ("coding-category"); |
5166 | staticpro (&Qcoding_category); | |
4ed46869 KH |
5167 | Qcoding_category_index = intern ("coding-category-index"); |
5168 | staticpro (&Qcoding_category_index); | |
5169 | ||
d46c5b12 KH |
5170 | Vcoding_category_table |
5171 | = Fmake_vector (make_number (CODING_CATEGORY_IDX_MAX), Qnil); | |
5172 | staticpro (&Vcoding_category_table); | |
4ed46869 KH |
5173 | { |
5174 | int i; | |
5175 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) | |
5176 | { | |
d46c5b12 KH |
5177 | XVECTOR (Vcoding_category_table)->contents[i] |
5178 | = intern (coding_category_name[i]); | |
5179 | Fput (XVECTOR (Vcoding_category_table)->contents[i], | |
5180 | Qcoding_category_index, make_number (i)); | |
4ed46869 KH |
5181 | } |
5182 | } | |
5183 | ||
f967223b KH |
5184 | Qtranslation_table = intern ("translation-table"); |
5185 | staticpro (&Qtranslation_table); | |
5186 | Fput (Qtranslation_table, Qchar_table_extra_slots, make_number (0)); | |
bdd9fb48 | 5187 | |
f967223b KH |
5188 | Qtranslation_table_id = intern ("translation-table-id"); |
5189 | staticpro (&Qtranslation_table_id); | |
84fbb8a0 | 5190 | |
f967223b KH |
5191 | Qtranslation_table_for_decode = intern ("translation-table-for-decode"); |
5192 | staticpro (&Qtranslation_table_for_decode); | |
a5d301df | 5193 | |
f967223b KH |
5194 | Qtranslation_table_for_encode = intern ("translation-table-for-encode"); |
5195 | staticpro (&Qtranslation_table_for_encode); | |
a5d301df | 5196 | |
70c22245 KH |
5197 | Qsafe_charsets = intern ("safe-charsets"); |
5198 | staticpro (&Qsafe_charsets); | |
5199 | ||
9ce27fde KH |
5200 | Qemacs_mule = intern ("emacs-mule"); |
5201 | staticpro (&Qemacs_mule); | |
5202 | ||
d46c5b12 KH |
5203 | Qraw_text = intern ("raw-text"); |
5204 | staticpro (&Qraw_text); | |
5205 | ||
4ed46869 KH |
5206 | defsubr (&Scoding_system_p); |
5207 | defsubr (&Sread_coding_system); | |
5208 | defsubr (&Sread_non_nil_coding_system); | |
5209 | defsubr (&Scheck_coding_system); | |
5210 | defsubr (&Sdetect_coding_region); | |
d46c5b12 | 5211 | defsubr (&Sdetect_coding_string); |
4ed46869 KH |
5212 | defsubr (&Sdecode_coding_region); |
5213 | defsubr (&Sencode_coding_region); | |
5214 | defsubr (&Sdecode_coding_string); | |
5215 | defsubr (&Sencode_coding_string); | |
5216 | defsubr (&Sdecode_sjis_char); | |
5217 | defsubr (&Sencode_sjis_char); | |
5218 | defsubr (&Sdecode_big5_char); | |
5219 | defsubr (&Sencode_big5_char); | |
1ba9e4ab | 5220 | defsubr (&Sset_terminal_coding_system_internal); |
c4825358 | 5221 | defsubr (&Sset_safe_terminal_coding_system_internal); |
4ed46869 | 5222 | defsubr (&Sterminal_coding_system); |
1ba9e4ab | 5223 | defsubr (&Sset_keyboard_coding_system_internal); |
4ed46869 | 5224 | defsubr (&Skeyboard_coding_system); |
a5d301df | 5225 | defsubr (&Sfind_operation_coding_system); |
d46c5b12 | 5226 | defsubr (&Supdate_iso_coding_systems); |
66cfb530 | 5227 | defsubr (&Sset_coding_priority_internal); |
4ed46869 | 5228 | |
4608c386 KH |
5229 | DEFVAR_LISP ("coding-system-list", &Vcoding_system_list, |
5230 | "List of coding systems.\n\ | |
5231 | \n\ | |
5232 | Do not alter the value of this variable manually. This variable should be\n\ | |
5233 | updated by the functions `make-coding-system' and\n\ | |
5234 | `define-coding-system-alias'."); | |
5235 | Vcoding_system_list = Qnil; | |
5236 | ||
5237 | DEFVAR_LISP ("coding-system-alist", &Vcoding_system_alist, | |
5238 | "Alist of coding system names.\n\ | |
5239 | Each element is one element list of coding system name.\n\ | |
5240 | This variable is given to `completing-read' as TABLE argument.\n\ | |
5241 | \n\ | |
5242 | Do not alter the value of this variable manually. This variable should be\n\ | |
5243 | updated by the functions `make-coding-system' and\n\ | |
5244 | `define-coding-system-alias'."); | |
5245 | Vcoding_system_alist = Qnil; | |
5246 | ||
4ed46869 KH |
5247 | DEFVAR_LISP ("coding-category-list", &Vcoding_category_list, |
5248 | "List of coding-categories (symbols) ordered by priority."); | |
5249 | { | |
5250 | int i; | |
5251 | ||
5252 | Vcoding_category_list = Qnil; | |
5253 | for (i = CODING_CATEGORY_IDX_MAX - 1; i >= 0; i--) | |
5254 | Vcoding_category_list | |
d46c5b12 KH |
5255 | = Fcons (XVECTOR (Vcoding_category_table)->contents[i], |
5256 | Vcoding_category_list); | |
4ed46869 KH |
5257 | } |
5258 | ||
5259 | DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read, | |
10bff6f1 | 5260 | "Specify the coding system for read operations.\n\ |
2ebb362d | 5261 | It is useful to bind this variable with `let', but do not set it globally.\n\ |
4ed46869 | 5262 | If the value is a coding system, it is used for decoding on read operation.\n\ |
a67a9c66 | 5263 | If not, an appropriate element is used from one of the coding system alists:\n\ |
10bff6f1 | 5264 | There are three such tables, `file-coding-system-alist',\n\ |
a67a9c66 | 5265 | `process-coding-system-alist', and `network-coding-system-alist'."); |
4ed46869 KH |
5266 | Vcoding_system_for_read = Qnil; |
5267 | ||
5268 | DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write, | |
10bff6f1 | 5269 | "Specify the coding system for write operations.\n\ |
2ebb362d | 5270 | It is useful to bind this variable with `let', but do not set it globally.\n\ |
4ed46869 | 5271 | If the value is a coding system, it is used for encoding on write operation.\n\ |
a67a9c66 | 5272 | If not, an appropriate element is used from one of the coding system alists:\n\ |
10bff6f1 | 5273 | There are three such tables, `file-coding-system-alist',\n\ |
a67a9c66 | 5274 | `process-coding-system-alist', and `network-coding-system-alist'."); |
4ed46869 KH |
5275 | Vcoding_system_for_write = Qnil; |
5276 | ||
5277 | DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used, | |
a67a9c66 | 5278 | "Coding system used in the latest file or process I/O."); |
4ed46869 KH |
5279 | Vlast_coding_system_used = Qnil; |
5280 | ||
9ce27fde KH |
5281 | DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion, |
5282 | "*Non-nil inhibit code conversion of end-of-line format in any cases."); | |
5283 | inhibit_eol_conversion = 0; | |
5284 | ||
ed29121d EZ |
5285 | DEFVAR_BOOL ("inherit-process-coding-system", &inherit_process_coding_system, |
5286 | "Non-nil means process buffer inherits coding system of process output.\n\ | |
5287 | Bind it to t if the process output is to be treated as if it were a file\n\ | |
5288 | read from some filesystem."); | |
5289 | inherit_process_coding_system = 0; | |
5290 | ||
02ba4723 KH |
5291 | DEFVAR_LISP ("file-coding-system-alist", &Vfile_coding_system_alist, |
5292 | "Alist to decide a coding system to use for a file I/O operation.\n\ | |
5293 | The format is ((PATTERN . VAL) ...),\n\ | |
5294 | where PATTERN is a regular expression matching a file name,\n\ | |
5295 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
5296 | If VAL is a coding system, it is used for both decoding and encoding\n\ | |
5297 | the file contents.\n\ | |
5298 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
5299 | and the cdr part is used for encoding.\n\ | |
5300 | If VAL is a function symbol, the function must return a coding system\n\ | |
5301 | or a cons of coding systems which are used as above.\n\ | |
e0e989f6 | 5302 | \n\ |
9ce27fde | 5303 | See also the function `find-operation-coding-system'."); |
02ba4723 KH |
5304 | Vfile_coding_system_alist = Qnil; |
5305 | ||
5306 | DEFVAR_LISP ("process-coding-system-alist", &Vprocess_coding_system_alist, | |
5307 | "Alist to decide a coding system to use for a process I/O operation.\n\ | |
5308 | The format is ((PATTERN . VAL) ...),\n\ | |
5309 | where PATTERN is a regular expression matching a program name,\n\ | |
5310 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
5311 | If VAL is a coding system, it is used for both decoding what received\n\ | |
5312 | from the program and encoding what sent to the program.\n\ | |
5313 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
5314 | and the cdr part is used for encoding.\n\ | |
5315 | If VAL is a function symbol, the function must return a coding system\n\ | |
5316 | or a cons of coding systems which are used as above.\n\ | |
4ed46869 | 5317 | \n\ |
9ce27fde | 5318 | See also the function `find-operation-coding-system'."); |
02ba4723 KH |
5319 | Vprocess_coding_system_alist = Qnil; |
5320 | ||
5321 | DEFVAR_LISP ("network-coding-system-alist", &Vnetwork_coding_system_alist, | |
5322 | "Alist to decide a coding system to use for a network I/O operation.\n\ | |
5323 | The format is ((PATTERN . VAL) ...),\n\ | |
5324 | where PATTERN is a regular expression matching a network service name\n\ | |
5325 | or is a port number to connect to,\n\ | |
5326 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
5327 | If VAL is a coding system, it is used for both decoding what received\n\ | |
5328 | from the network stream and encoding what sent to the network stream.\n\ | |
5329 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
5330 | and the cdr part is used for encoding.\n\ | |
5331 | If VAL is a function symbol, the function must return a coding system\n\ | |
5332 | or a cons of coding systems which are used as above.\n\ | |
4ed46869 | 5333 | \n\ |
9ce27fde | 5334 | See also the function `find-operation-coding-system'."); |
02ba4723 | 5335 | Vnetwork_coding_system_alist = Qnil; |
4ed46869 KH |
5336 | |
5337 | DEFVAR_INT ("eol-mnemonic-unix", &eol_mnemonic_unix, | |
5338 | "Mnemonic character indicating UNIX-like end-of-line format (i.e. LF) ."); | |
458822a0 | 5339 | eol_mnemonic_unix = ':'; |
4ed46869 KH |
5340 | |
5341 | DEFVAR_INT ("eol-mnemonic-dos", &eol_mnemonic_dos, | |
5342 | "Mnemonic character indicating DOS-like end-of-line format (i.e. CRLF)."); | |
458822a0 | 5343 | eol_mnemonic_dos = '\\'; |
4ed46869 KH |
5344 | |
5345 | DEFVAR_INT ("eol-mnemonic-mac", &eol_mnemonic_mac, | |
5346 | "Mnemonic character indicating MAC-like end-of-line format (i.e. CR)."); | |
458822a0 | 5347 | eol_mnemonic_mac = '/'; |
4ed46869 KH |
5348 | |
5349 | DEFVAR_INT ("eol-mnemonic-undecided", &eol_mnemonic_undecided, | |
5350 | "Mnemonic character indicating end-of-line format is not yet decided."); | |
458822a0 | 5351 | eol_mnemonic_undecided = ':'; |
4ed46869 | 5352 | |
84fbb8a0 | 5353 | DEFVAR_LISP ("enable-character-translation", &Venable_character_translation, |
f967223b | 5354 | "*Non-nil enables character translation while encoding and decoding."); |
84fbb8a0 | 5355 | Venable_character_translation = Qt; |
bdd9fb48 | 5356 | |
f967223b KH |
5357 | DEFVAR_LISP ("standard-translation-table-for-decode", |
5358 | &Vstandard_translation_table_for_decode, | |
84fbb8a0 | 5359 | "Table for translating characters while decoding."); |
f967223b | 5360 | Vstandard_translation_table_for_decode = Qnil; |
bdd9fb48 | 5361 | |
f967223b KH |
5362 | DEFVAR_LISP ("standard-translation-table-for-encode", |
5363 | &Vstandard_translation_table_for_encode, | |
84fbb8a0 | 5364 | "Table for translationg characters while encoding."); |
f967223b | 5365 | Vstandard_translation_table_for_encode = Qnil; |
4ed46869 KH |
5366 | |
5367 | DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist, | |
5368 | "Alist of charsets vs revision numbers.\n\ | |
5369 | While encoding, if a charset (car part of an element) is found,\n\ | |
5370 | designate it with the escape sequence identifing revision (cdr part of the element)."); | |
5371 | Vcharset_revision_alist = Qnil; | |
02ba4723 KH |
5372 | |
5373 | DEFVAR_LISP ("default-process-coding-system", | |
5374 | &Vdefault_process_coding_system, | |
5375 | "Cons of coding systems used for process I/O by default.\n\ | |
5376 | The car part is used for decoding a process output,\n\ | |
5377 | the cdr part is used for encoding a text to be sent to a process."); | |
5378 | Vdefault_process_coding_system = Qnil; | |
c4825358 | 5379 | |
3f003981 KH |
5380 | DEFVAR_LISP ("latin-extra-code-table", &Vlatin_extra_code_table, |
5381 | "Table of extra Latin codes in the range 128..159 (inclusive).\n\ | |
c4825358 KH |
5382 | This is a vector of length 256.\n\ |
5383 | If Nth element is non-nil, the existence of code N in a file\n\ | |
bb0115a2 | 5384 | \(or output of subprocess) doesn't prevent it to be detected as\n\ |
3f003981 KH |
5385 | a coding system of ISO 2022 variant which has a flag\n\ |
5386 | `accept-latin-extra-code' t (e.g. iso-latin-1) on reading a file\n\ | |
c4825358 KH |
5387 | or reading output of a subprocess.\n\ |
5388 | Only 128th through 159th elements has a meaning."); | |
3f003981 | 5389 | Vlatin_extra_code_table = Fmake_vector (make_number (256), Qnil); |
d46c5b12 KH |
5390 | |
5391 | DEFVAR_LISP ("select-safe-coding-system-function", | |
5392 | &Vselect_safe_coding_system_function, | |
5393 | "Function to call to select safe coding system for encoding a text.\n\ | |
5394 | \n\ | |
5395 | If set, this function is called to force a user to select a proper\n\ | |
5396 | coding system which can encode the text in the case that a default\n\ | |
5397 | coding system used in each operation can't encode the text.\n\ | |
5398 | \n\ | |
5399 | The default value is `select-safe-codign-system' (which see)."); | |
5400 | Vselect_safe_coding_system_function = Qnil; | |
5401 | ||
4ed46869 KH |
5402 | } |
5403 | ||
5404 | #endif /* emacs */ |