Commit | Line | Data |
---|---|---|
4ed46869 | 1 | /* Coding system handler (conversion, detection, and etc). |
203cb916 RS |
2 | Copyright (C) 1995, 1997 Electrotechnical Laboratory, JAPAN. |
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 | |
f4dee582 | 70 | 4. Other |
4ed46869 | 71 | |
f4dee582 | 72 | If a user wants to read/write a text encoded in a coding system not |
4ed46869 KH |
73 | listed above, he can supply a decoder and an encoder for it in CCL |
74 | (Code Conversion Language) programs. Emacs executes the CCL program | |
75 | while reading/writing. | |
76 | ||
f4dee582 | 77 | Emacs represents a coding-system by a Lisp symbol that has a property |
4ed46869 KH |
78 | `coding-system'. But, before actually using the coding-system, the |
79 | information about it is set in a structure of type `struct | |
f4dee582 | 80 | coding_system' for rapid processing. See section 6 for more details. |
4ed46869 KH |
81 | |
82 | */ | |
83 | ||
84 | /*** GENERAL NOTES on END-OF-LINE FORMAT *** | |
85 | ||
86 | How end-of-line of a text is encoded depends on a system. For | |
87 | instance, Unix's format is just one byte of `line-feed' code, | |
f4dee582 | 88 | whereas DOS's format is two-byte sequence of `carriage-return' and |
4ed46869 KH |
89 | `line-feed' codes. MacOS's format is one byte of `carriage-return'. |
90 | ||
f4dee582 RS |
91 | Since text characters encoding and end-of-line encoding are |
92 | independent, any coding system described above can take | |
4ed46869 | 93 | any format of end-of-line. So, Emacs has information of format of |
f4dee582 | 94 | end-of-line in each coding-system. See section 6 for more details. |
4ed46869 KH |
95 | |
96 | */ | |
97 | ||
98 | /*** GENERAL NOTES on `detect_coding_XXX ()' functions *** | |
99 | ||
100 | These functions check if a text between SRC and SRC_END is encoded | |
101 | in the coding system category XXX. Each returns an integer value in | |
102 | which appropriate flag bits for the category XXX is set. The flag | |
103 | bits are defined in macros CODING_CATEGORY_MASK_XXX. Below is the | |
104 | template of these functions. */ | |
105 | #if 0 | |
106 | int | |
0ef69138 | 107 | detect_coding_emacs_mule (src, src_end) |
4ed46869 KH |
108 | unsigned char *src, *src_end; |
109 | { | |
110 | ... | |
111 | } | |
112 | #endif | |
113 | ||
114 | /*** GENERAL NOTES on `decode_coding_XXX ()' functions *** | |
115 | ||
116 | These functions decode SRC_BYTES length text at SOURCE encoded in | |
0ef69138 | 117 | CODING to Emacs' internal format (emacs-mule). The resulting text |
f4dee582 RS |
118 | goes to a place pointed to by DESTINATION, the length of which should |
119 | not exceed DST_BYTES. The number of bytes actually processed is | |
120 | returned as *CONSUMED. The return value is the length of the decoded | |
121 | text. Below is a template of these functions. */ | |
4ed46869 KH |
122 | #if 0 |
123 | decode_coding_XXX (coding, source, destination, src_bytes, dst_bytes, consumed) | |
124 | struct coding_system *coding; | |
125 | unsigned char *source, *destination; | |
126 | int src_bytes, dst_bytes; | |
127 | int *consumed; | |
128 | { | |
129 | ... | |
130 | } | |
131 | #endif | |
132 | ||
133 | /*** GENERAL NOTES on `encode_coding_XXX ()' functions *** | |
134 | ||
0ef69138 KH |
135 | These functions encode SRC_BYTES length text at SOURCE of Emacs' |
136 | internal format (emacs-mule) to CODING. The resulting text goes to | |
f4dee582 RS |
137 | a place pointed to by DESTINATION, the length of which should not |
138 | exceed DST_BYTES. The number of bytes actually processed is | |
139 | returned as *CONSUMED. The return value is the length of the | |
140 | encoded text. Below is a template of these functions. */ | |
4ed46869 KH |
141 | #if 0 |
142 | encode_coding_XXX (coding, source, destination, src_bytes, dst_bytes, consumed) | |
143 | struct coding_system *coding; | |
144 | unsigned char *source, *destination; | |
145 | int src_bytes, dst_bytes; | |
146 | int *consumed; | |
147 | { | |
148 | ... | |
149 | } | |
150 | #endif | |
151 | ||
152 | /*** COMMONLY USED MACROS ***/ | |
153 | ||
154 | /* The following three macros ONE_MORE_BYTE, TWO_MORE_BYTES, and | |
155 | THREE_MORE_BYTES safely get one, two, and three bytes from the | |
156 | source text respectively. If there are not enough bytes in the | |
157 | source, they jump to `label_end_of_loop'. The caller should set | |
158 | variables `src' and `src_end' to appropriate areas in advance. */ | |
159 | ||
160 | #define ONE_MORE_BYTE(c1) \ | |
161 | do { \ | |
162 | if (src < src_end) \ | |
163 | c1 = *src++; \ | |
164 | else \ | |
165 | goto label_end_of_loop; \ | |
166 | } while (0) | |
167 | ||
168 | #define TWO_MORE_BYTES(c1, c2) \ | |
169 | do { \ | |
170 | if (src + 1 < src_end) \ | |
171 | c1 = *src++, c2 = *src++; \ | |
172 | else \ | |
173 | goto label_end_of_loop; \ | |
174 | } while (0) | |
175 | ||
176 | #define THREE_MORE_BYTES(c1, c2, c3) \ | |
177 | do { \ | |
178 | if (src + 2 < src_end) \ | |
179 | c1 = *src++, c2 = *src++, c3 = *src++; \ | |
180 | else \ | |
181 | goto label_end_of_loop; \ | |
182 | } while (0) | |
183 | ||
184 | /* The following three macros DECODE_CHARACTER_ASCII, | |
185 | DECODE_CHARACTER_DIMENSION1, and DECODE_CHARACTER_DIMENSION2 put | |
186 | the multi-byte form of a character of each class at the place | |
187 | pointed by `dst'. The caller should set the variable `dst' to | |
188 | point to an appropriate area and the variable `coding' to point to | |
189 | the coding-system of the currently decoding text in advance. */ | |
190 | ||
191 | /* Decode one ASCII character C. */ | |
192 | ||
193 | #define DECODE_CHARACTER_ASCII(c) \ | |
194 | do { \ | |
195 | if (COMPOSING_P (coding->composing)) \ | |
196 | *dst++ = 0xA0, *dst++ = (c) | 0x80; \ | |
197 | else \ | |
198 | *dst++ = (c); \ | |
199 | } while (0) | |
200 | ||
f4dee582 | 201 | /* Decode one DIMENSION1 character whose charset is CHARSET and whose |
4ed46869 KH |
202 | position-code is C. */ |
203 | ||
204 | #define DECODE_CHARACTER_DIMENSION1(charset, c) \ | |
205 | do { \ | |
206 | unsigned char leading_code = CHARSET_LEADING_CODE_BASE (charset); \ | |
207 | if (COMPOSING_P (coding->composing)) \ | |
208 | *dst++ = leading_code + 0x20; \ | |
209 | else \ | |
210 | *dst++ = leading_code; \ | |
211 | if (leading_code = CHARSET_LEADING_CODE_EXT (charset)) \ | |
212 | *dst++ = leading_code; \ | |
213 | *dst++ = (c) | 0x80; \ | |
214 | } while (0) | |
215 | ||
f4dee582 | 216 | /* Decode one DIMENSION2 character whose charset is CHARSET and whose |
4ed46869 KH |
217 | position-codes are C1 and C2. */ |
218 | ||
219 | #define DECODE_CHARACTER_DIMENSION2(charset, c1, c2) \ | |
220 | do { \ | |
221 | DECODE_CHARACTER_DIMENSION1 (charset, c1); \ | |
222 | *dst++ = (c2) | 0x80; \ | |
223 | } while (0) | |
224 | ||
225 | \f | |
226 | /*** 1. Preamble ***/ | |
227 | ||
228 | #include <stdio.h> | |
229 | ||
230 | #ifdef emacs | |
231 | ||
232 | #include <config.h> | |
233 | #include "lisp.h" | |
234 | #include "buffer.h" | |
235 | #include "charset.h" | |
236 | #include "ccl.h" | |
237 | #include "coding.h" | |
238 | #include "window.h" | |
239 | ||
240 | #else /* not emacs */ | |
241 | ||
242 | #include "mulelib.h" | |
243 | ||
244 | #endif /* not emacs */ | |
245 | ||
246 | Lisp_Object Qcoding_system, Qeol_type; | |
247 | Lisp_Object Qbuffer_file_coding_system; | |
248 | Lisp_Object Qpost_read_conversion, Qpre_write_conversion; | |
249 | ||
250 | extern Lisp_Object Qinsert_file_contents, Qwrite_region; | |
251 | Lisp_Object Qcall_process, Qcall_process_region, Qprocess_argument; | |
252 | Lisp_Object Qstart_process, Qopen_network_stream; | |
253 | Lisp_Object Qtarget_idx; | |
254 | ||
255 | /* Mnemonic character of each format of end-of-line. */ | |
256 | int eol_mnemonic_unix, eol_mnemonic_dos, eol_mnemonic_mac; | |
257 | /* Mnemonic character to indicate format of end-of-line is not yet | |
258 | decided. */ | |
259 | int eol_mnemonic_undecided; | |
260 | ||
9ce27fde KH |
261 | /* Format of end-of-line decided by system. This is CODING_EOL_LF on |
262 | Unix, CODING_EOL_CRLF on DOS/Windows, and CODING_EOL_CR on Mac. */ | |
263 | int system_eol_type; | |
264 | ||
4ed46869 KH |
265 | #ifdef emacs |
266 | ||
02ba4723 | 267 | Lisp_Object Qcoding_system_spec, Qcoding_system_p, Qcoding_system_error; |
4ed46869 | 268 | |
9ce27fde KH |
269 | /* Coding system emacs-mule is for converting only end-of-line format. */ |
270 | Lisp_Object Qemacs_mule; | |
271 | ||
4ed46869 KH |
272 | /* Coding-systems are handed between Emacs Lisp programs and C internal |
273 | routines by the following three variables. */ | |
274 | /* Coding-system for reading files and receiving data from process. */ | |
275 | Lisp_Object Vcoding_system_for_read; | |
276 | /* Coding-system for writing files and sending data to process. */ | |
277 | Lisp_Object Vcoding_system_for_write; | |
278 | /* Coding-system actually used in the latest I/O. */ | |
279 | Lisp_Object Vlast_coding_system_used; | |
280 | ||
9ce27fde KH |
281 | /* Flag to inhibit code conversion of end-of-line format. */ |
282 | int inhibit_eol_conversion; | |
283 | ||
4ed46869 KH |
284 | /* Coding-system of what terminal accept for displaying. */ |
285 | struct coding_system terminal_coding; | |
286 | ||
287 | /* Coding-system of what is sent from terminal keyboard. */ | |
288 | struct coding_system keyboard_coding; | |
289 | ||
02ba4723 KH |
290 | Lisp_Object Vfile_coding_system_alist; |
291 | Lisp_Object Vprocess_coding_system_alist; | |
292 | Lisp_Object Vnetwork_coding_system_alist; | |
4ed46869 KH |
293 | |
294 | #endif /* emacs */ | |
295 | ||
296 | Lisp_Object Qcoding_category_index; | |
297 | ||
298 | /* List of symbols `coding-category-xxx' ordered by priority. */ | |
299 | Lisp_Object Vcoding_category_list; | |
300 | ||
301 | /* Table of coding-systems currently assigned to each coding-category. */ | |
302 | Lisp_Object coding_category_table[CODING_CATEGORY_IDX_MAX]; | |
303 | ||
304 | /* Table of names of symbol for each coding-category. */ | |
305 | char *coding_category_name[CODING_CATEGORY_IDX_MAX] = { | |
0ef69138 | 306 | "coding-category-emacs-mule", |
4ed46869 KH |
307 | "coding-category-sjis", |
308 | "coding-category-iso-7", | |
309 | "coding-category-iso-8-1", | |
310 | "coding-category-iso-8-2", | |
7717c392 KH |
311 | "coding-category-iso-7-else", |
312 | "coding-category-iso-8-else", | |
4ed46869 KH |
313 | "coding-category-big5", |
314 | "coding-category-binary" | |
315 | }; | |
316 | ||
bdd9fb48 KH |
317 | /* Flag to tell if we look up unification table on character code |
318 | conversion. */ | |
319 | Lisp_Object Venable_character_unification; | |
a5d301df KH |
320 | /* Standard unification table to look up on decoding (reading). */ |
321 | Lisp_Object Vstandard_character_unification_table_for_decode; | |
322 | /* Standard unification table to look up on encoding (writing). */ | |
323 | Lisp_Object Vstandard_character_unification_table_for_encode; | |
bdd9fb48 KH |
324 | |
325 | Lisp_Object Qcharacter_unification_table; | |
a5d301df KH |
326 | Lisp_Object Qcharacter_unification_table_for_decode; |
327 | Lisp_Object Qcharacter_unification_table_for_encode; | |
4ed46869 KH |
328 | |
329 | /* Alist of charsets vs revision number. */ | |
330 | Lisp_Object Vcharset_revision_alist; | |
331 | ||
02ba4723 KH |
332 | /* Default coding systems used for process I/O. */ |
333 | Lisp_Object Vdefault_process_coding_system; | |
334 | ||
4ed46869 | 335 | \f |
0ef69138 | 336 | /*** 2. Emacs internal format (emacs-mule) handlers ***/ |
4ed46869 KH |
337 | |
338 | /* Emacs' internal format for encoding multiple character sets is a | |
f4dee582 RS |
339 | kind of multi-byte encoding, i.e. characters are encoded by |
340 | variable-length sequences of one-byte codes. ASCII characters | |
341 | and control characters (e.g. `tab', `newline') are represented by | |
342 | one-byte sequences which are their ASCII codes, in the range 0x00 | |
343 | through 0x7F. The other characters are represented by a sequence | |
344 | of `base leading-code', optional `extended leading-code', and one | |
345 | or two `position-code's. The length of the sequence is determined | |
346 | by the base leading-code. Leading-code takes the range 0x80 | |
347 | through 0x9F, whereas extended leading-code and position-code take | |
348 | the range 0xA0 through 0xFF. See `charset.h' for more details | |
349 | about leading-code and position-code. | |
350 | ||
351 | There's one exception to this rule. Special leading-code | |
4ed46869 KH |
352 | `leading-code-composition' denotes that the following several |
353 | characters should be composed into one character. Leading-codes of | |
354 | components (except for ASCII) are added 0x20. An ASCII character | |
355 | component is represented by a 2-byte sequence of `0xA0' and | |
f4dee582 RS |
356 | `ASCII-code + 0x80'. See also the comments in `charset.h' for the |
357 | details of composite character. Hence, we can summarize the code | |
4ed46869 KH |
358 | range as follows: |
359 | ||
360 | --- CODE RANGE of Emacs' internal format --- | |
361 | (character set) (range) | |
362 | ASCII 0x00 .. 0x7F | |
363 | ELSE (1st byte) 0x80 .. 0x9F | |
364 | (rest bytes) 0xA0 .. 0xFF | |
365 | --------------------------------------------- | |
366 | ||
367 | */ | |
368 | ||
369 | enum emacs_code_class_type emacs_code_class[256]; | |
370 | ||
371 | /* Go to the next statement only if *SRC is accessible and the code is | |
372 | greater than 0xA0. */ | |
373 | #define CHECK_CODE_RANGE_A0_FF \ | |
374 | do { \ | |
375 | if (src >= src_end) \ | |
376 | goto label_end_of_switch; \ | |
377 | else if (*src++ < 0xA0) \ | |
378 | return 0; \ | |
379 | } while (0) | |
380 | ||
381 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". | |
382 | Check if a text is encoded in Emacs' internal format. If it is, | |
0ef69138 | 383 | return CODING_CATEGORY_MASK_EMASC_MULE, else return 0. */ |
4ed46869 KH |
384 | |
385 | int | |
0ef69138 | 386 | detect_coding_emacs_mule (src, src_end) |
4ed46869 KH |
387 | unsigned char *src, *src_end; |
388 | { | |
389 | unsigned char c; | |
390 | int composing = 0; | |
391 | ||
392 | while (src < src_end) | |
393 | { | |
394 | c = *src++; | |
395 | ||
396 | if (composing) | |
397 | { | |
398 | if (c < 0xA0) | |
399 | composing = 0; | |
400 | else | |
401 | c -= 0x20; | |
402 | } | |
403 | ||
404 | switch (emacs_code_class[c]) | |
405 | { | |
406 | case EMACS_ascii_code: | |
407 | case EMACS_linefeed_code: | |
408 | break; | |
409 | ||
410 | case EMACS_control_code: | |
411 | if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) | |
412 | return 0; | |
413 | break; | |
414 | ||
415 | case EMACS_invalid_code: | |
416 | return 0; | |
417 | ||
418 | case EMACS_leading_code_composition: /* c == 0x80 */ | |
419 | if (composing) | |
420 | CHECK_CODE_RANGE_A0_FF; | |
421 | else | |
422 | composing = 1; | |
423 | break; | |
424 | ||
425 | case EMACS_leading_code_4: | |
426 | CHECK_CODE_RANGE_A0_FF; | |
427 | /* fall down to check it two more times ... */ | |
428 | ||
429 | case EMACS_leading_code_3: | |
430 | CHECK_CODE_RANGE_A0_FF; | |
431 | /* fall down to check it one more time ... */ | |
432 | ||
433 | case EMACS_leading_code_2: | |
434 | CHECK_CODE_RANGE_A0_FF; | |
435 | break; | |
436 | ||
437 | default: | |
438 | label_end_of_switch: | |
439 | break; | |
440 | } | |
441 | } | |
0ef69138 | 442 | return CODING_CATEGORY_MASK_EMACS_MULE; |
4ed46869 KH |
443 | } |
444 | ||
445 | \f | |
446 | /*** 3. ISO2022 handlers ***/ | |
447 | ||
448 | /* The following note describes the coding system ISO2022 briefly. | |
f4dee582 RS |
449 | Since the intention of this note is to help in understanding of |
450 | the programs in this file, some parts are NOT ACCURATE or OVERLY | |
4ed46869 KH |
451 | SIMPLIFIED. For the thorough understanding, please refer to the |
452 | original document of ISO2022. | |
453 | ||
454 | ISO2022 provides many mechanisms to encode several character sets | |
f4dee582 | 455 | in 7-bit and 8-bit environment. If one chooses 7-bite environment, |
4ed46869 | 456 | all text is encoded by codes of less than 128. This may make the |
f4dee582 RS |
457 | encoded text a little bit longer, but the text gets more stability |
458 | to pass through several gateways (some of them strip off the MSB). | |
4ed46869 | 459 | |
f4dee582 | 460 | There are two kinds of character set: control character set and |
4ed46869 KH |
461 | graphic character set. The former contains control characters such |
462 | as `newline' and `escape' to provide control functions (control | |
f4dee582 | 463 | functions are provided also by escape sequences). The latter |
4ed46869 KH |
464 | contains graphic characters such as ' A' and '-'. Emacs recognizes |
465 | two control character sets and many graphic character sets. | |
466 | ||
467 | Graphic character sets are classified into one of the following | |
468 | four classes, DIMENSION1_CHARS94, DIMENSION1_CHARS96, | |
469 | DIMENSION2_CHARS94, DIMENSION2_CHARS96 according to the number of | |
470 | bytes (DIMENSION) and the number of characters in one dimension | |
471 | (CHARS) of the set. In addition, each character set is assigned an | |
472 | identification tag (called "final character" and denoted as <F> | |
473 | here after) which is unique in each class. <F> of each character | |
474 | set is decided by ECMA(*) when it is registered in ISO. Code range | |
475 | of <F> is 0x30..0x7F (0x30..0x3F are for private use only). | |
476 | ||
477 | Note (*): ECMA = European Computer Manufacturers Association | |
478 | ||
479 | Here are examples of graphic character set [NAME(<F>)]: | |
480 | o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ... | |
481 | o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ... | |
482 | o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ... | |
483 | o DIMENSION2_CHARS96 -- none for the moment | |
484 | ||
485 | A code area (1byte=8bits) is divided into 4 areas, C0, GL, C1, and GR. | |
486 | C0 [0x00..0x1F] -- control character plane 0 | |
487 | GL [0x20..0x7F] -- graphic character plane 0 | |
488 | C1 [0x80..0x9F] -- control character plane 1 | |
489 | GR [0xA0..0xFF] -- graphic character plane 1 | |
490 | ||
491 | A control character set is directly designated and invoked to C0 or | |
492 | C1 by an escape sequence. The most common case is that ISO646's | |
493 | control character set is designated/invoked to C0 and ISO6429's | |
494 | control character set is designated/invoked to C1, and usually | |
495 | these designations/invocations are omitted in a coded text. With | |
496 | 7-bit environment, only C0 can be used, and a control character for | |
497 | C1 is encoded by an appropriate escape sequence to fit in the | |
498 | environment. All control characters for C1 are defined the | |
499 | corresponding escape sequences. | |
500 | ||
501 | A graphic character set is at first designated to one of four | |
502 | graphic registers (G0 through G3), then these graphic registers are | |
503 | invoked to GL or GR. These designations and invocations can be | |
504 | done independently. The most common case is that G0 is invoked to | |
505 | GL, G1 is invoked to GR, and ASCII is designated to G0, and usually | |
506 | these invocations and designations are omitted in a coded text. | |
507 | With 7-bit environment, only GL can be used. | |
508 | ||
509 | When a graphic character set of CHARS94 is invoked to GL, code 0x20 | |
510 | and 0x7F of GL area work as control characters SPACE and DEL | |
511 | respectively, and code 0xA0 and 0xFF of GR area should not be used. | |
512 | ||
513 | There are two ways of invocation: locking-shift and single-shift. | |
514 | With locking-shift, the invocation lasts until the next different | |
515 | invocation, whereas with single-shift, the invocation works only | |
516 | for the following character and doesn't affect locking-shift. | |
517 | Invocations are done by the following control characters or escape | |
518 | sequences. | |
519 | ||
520 | ---------------------------------------------------------------------- | |
521 | function control char escape sequence description | |
522 | ---------------------------------------------------------------------- | |
523 | SI (shift-in) 0x0F none invoke G0 to GL | |
524 | SI (shift-out) 0x0E none invoke G1 to GL | |
525 | LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL | |
526 | LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL | |
527 | SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 into GL | |
528 | SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 into GL | |
529 | ---------------------------------------------------------------------- | |
530 | The first four are for locking-shift. Control characters for these | |
531 | functions are defined by macros ISO_CODE_XXX in `coding.h'. | |
532 | ||
533 | Designations are done by the following escape sequences. | |
534 | ---------------------------------------------------------------------- | |
535 | escape sequence description | |
536 | ---------------------------------------------------------------------- | |
537 | ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0 | |
538 | ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1 | |
539 | ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2 | |
540 | ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3 | |
541 | ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*) | |
542 | ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1 | |
543 | ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2 | |
544 | ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3 | |
545 | ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**) | |
546 | ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1 | |
547 | ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2 | |
548 | ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3 | |
549 | ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*) | |
550 | ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1 | |
551 | ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2 | |
552 | ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3 | |
553 | ---------------------------------------------------------------------- | |
554 | ||
555 | In this list, "DIMENSION1_CHARS94<F>" means a graphic character set | |
556 | of dimension 1, chars 94, and final character <F>, and etc. | |
557 | ||
558 | Note (*): Although these designations are not allowed in ISO2022, | |
559 | Emacs accepts them on decoding, and produces them on encoding | |
560 | CHARS96 character set in a coding system which is characterized as | |
561 | 7-bit environment, non-locking-shift, and non-single-shift. | |
562 | ||
563 | Note (**): If <F> is '@', 'A', or 'B', the intermediate character | |
564 | '(' can be omitted. We call this as "short-form" here after. | |
565 | ||
566 | Now you may notice that there are a lot of ways for encoding the | |
f4dee582 | 567 | same multilingual text in ISO2022. Actually, there exists many |
4ed46869 KH |
568 | coding systems such as Compound Text (used in X's inter client |
569 | communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR | |
570 | (used in Korean Internet), EUC (Extended UNIX Code, used in Asian | |
571 | localized platforms), and all of these are variants of ISO2022. | |
572 | ||
573 | In addition to the above, Emacs handles two more kinds of escape | |
574 | sequences: ISO6429's direction specification and Emacs' private | |
575 | sequence for specifying character composition. | |
576 | ||
577 | ISO6429's direction specification takes the following format: | |
578 | o CSI ']' -- end of the current direction | |
579 | o CSI '0' ']' -- end of the current direction | |
580 | o CSI '1' ']' -- start of left-to-right text | |
581 | o CSI '2' ']' -- start of right-to-left text | |
582 | The control character CSI (0x9B: control sequence introducer) is | |
583 | abbreviated to the escape sequence ESC '[' in 7-bit environment. | |
584 | ||
585 | Character composition specification takes the following format: | |
586 | o ESC '0' -- start character composition | |
587 | o ESC '1' -- end character composition | |
588 | Since these are not standard escape sequences of any ISO, the use | |
589 | of them for these meaning is restricted to Emacs only. */ | |
590 | ||
591 | enum iso_code_class_type iso_code_class[256]; | |
592 | ||
593 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". | |
594 | Check if a text is encoded in ISO2022. If it is, returns an | |
595 | integer in which appropriate flag bits any of: | |
596 | CODING_CATEGORY_MASK_ISO_7 | |
597 | CODING_CATEGORY_MASK_ISO_8_1 | |
598 | CODING_CATEGORY_MASK_ISO_8_2 | |
7717c392 KH |
599 | CODING_CATEGORY_MASK_ISO_7_ELSE |
600 | CODING_CATEGORY_MASK_ISO_8_ELSE | |
4ed46869 KH |
601 | are set. If a code which should never appear in ISO2022 is found, |
602 | returns 0. */ | |
603 | ||
604 | int | |
605 | detect_coding_iso2022 (src, src_end) | |
606 | unsigned char *src, *src_end; | |
607 | { | |
765a2ca5 KH |
608 | int mask = (CODING_CATEGORY_MASK_ISO_7 |
609 | | CODING_CATEGORY_MASK_ISO_8_1 | |
610 | | CODING_CATEGORY_MASK_ISO_8_2 | |
7717c392 KH |
611 | | CODING_CATEGORY_MASK_ISO_7_ELSE |
612 | | CODING_CATEGORY_MASK_ISO_8_ELSE | |
613 | ); | |
bcf26d6a KH |
614 | int g1 = 0; /* 1 iff designating to G1. */ |
615 | int c, i; | |
4ed46869 | 616 | |
e0e989f6 | 617 | while (src < src_end) |
4ed46869 KH |
618 | { |
619 | c = *src++; | |
620 | switch (c) | |
621 | { | |
622 | case ISO_CODE_ESC: | |
e0e989f6 | 623 | if (src >= src_end) |
4ed46869 KH |
624 | break; |
625 | c = *src++; | |
bcf26d6a | 626 | if (src < src_end |
e0e989f6 KH |
627 | && ((c >= '(' && c <= '/') |
628 | || c == '$' && ((*src >= '(' && *src <= '/') | |
629 | || (*src >= '@' && *src <= 'B')))) | |
4ed46869 | 630 | { |
e0e989f6 KH |
631 | /* Valid designation sequence. */ |
632 | if (c == ')' || (c == '$' && *src == ')')) | |
bcf26d6a KH |
633 | { |
634 | g1 = 1; | |
7717c392 KH |
635 | mask &= ~(CODING_CATEGORY_MASK_ISO_7 |
636 | | CODING_CATEGORY_MASK_ISO_7_ELSE); | |
bcf26d6a | 637 | } |
e0e989f6 KH |
638 | src++; |
639 | break; | |
4ed46869 | 640 | } |
4ed46869 | 641 | else if (c == 'N' || c == 'O' || c == 'n' || c == 'o') |
7717c392 KH |
642 | mask &= (CODING_CATEGORY_MASK_ISO_7_ELSE |
643 | | CODING_CATEGORY_MASK_ISO_8_ELSE); | |
4ed46869 KH |
644 | break; |
645 | ||
4ed46869 | 646 | case ISO_CODE_SO: |
e0e989f6 | 647 | if (g1) |
7717c392 KH |
648 | mask &= (CODING_CATEGORY_MASK_ISO_7_ELSE |
649 | | CODING_CATEGORY_MASK_ISO_8_ELSE); | |
e0e989f6 KH |
650 | break; |
651 | ||
4ed46869 KH |
652 | case ISO_CODE_CSI: |
653 | case ISO_CODE_SS2: | |
654 | case ISO_CODE_SS3: | |
7717c392 KH |
655 | mask &= ~(CODING_CATEGORY_MASK_ISO_7 |
656 | | CODING_CATEGORY_MASK_ISO_7_ELSE); | |
4ed46869 KH |
657 | break; |
658 | ||
659 | default: | |
660 | if (c < 0x80) | |
661 | break; | |
662 | else if (c < 0xA0) | |
663 | return 0; | |
664 | else | |
665 | { | |
7717c392 | 666 | unsigned char *src_begin = src; |
4ed46869 | 667 | |
7717c392 KH |
668 | mask &= ~(CODING_CATEGORY_MASK_ISO_7 |
669 | | CODING_CATEGORY_MASK_ISO_7_ELSE); | |
e0e989f6 | 670 | while (src < src_end && *src >= 0xA0) |
7717c392 KH |
671 | src++; |
672 | if ((src - src_begin - 1) & 1 && src < src_end) | |
4ed46869 KH |
673 | mask &= ~CODING_CATEGORY_MASK_ISO_8_2; |
674 | } | |
675 | break; | |
676 | } | |
677 | } | |
678 | ||
679 | return mask; | |
680 | } | |
681 | ||
682 | /* Decode a character of which charset is CHARSET and the 1st position | |
bdd9fb48 | 683 | code is C1. If dimension of CHARSET is 2, the 2nd position code is |
4ed46869 KH |
684 | fetched from SRC and set to C2. If CHARSET is negative, it means |
685 | that we are decoding ill formed text, and what we can do is just to | |
686 | read C1 as is. */ | |
687 | ||
bdd9fb48 KH |
688 | #define DECODE_ISO_CHARACTER(charset, c1) \ |
689 | do { \ | |
690 | int c_alt, charset_alt = (charset); \ | |
691 | if (COMPOSING_HEAD_P (coding->composing)) \ | |
692 | { \ | |
693 | *dst++ = LEADING_CODE_COMPOSITION; \ | |
694 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ | |
695 | /* To tell composition rules are embeded. */ \ | |
696 | *dst++ = 0xFF; \ | |
697 | coding->composing += 2; \ | |
698 | } \ | |
699 | if ((charset) >= 0) \ | |
700 | { \ | |
701 | if (CHARSET_DIMENSION (charset) == 2) \ | |
702 | ONE_MORE_BYTE (c2); \ | |
703 | if (!NILP (unification_table) \ | |
704 | && ((c_alt = unify_char (unification_table, \ | |
705 | -1, (charset), c1, c2)) >= 0)) \ | |
706 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
707 | } \ | |
708 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ | |
709 | DECODE_CHARACTER_ASCII (c1); \ | |
710 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
711 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
712 | else \ | |
713 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
714 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ | |
715 | /* To tell a composition rule follows. */ \ | |
716 | coding->composing = COMPOSING_WITH_RULE_RULE; \ | |
4ed46869 KH |
717 | } while (0) |
718 | ||
719 | /* Set designation state into CODING. */ | |
720 | #define DECODE_DESIGNATION(reg, dimension, chars, final_char) \ | |
721 | do { \ | |
2e34157c RS |
722 | int charset = ISO_CHARSET_TABLE (make_number (dimension), \ |
723 | make_number (chars), \ | |
724 | make_number (final_char)); \ | |
4ed46869 KH |
725 | if (charset >= 0) \ |
726 | { \ | |
727 | if (coding->direction == 1 \ | |
728 | && CHARSET_REVERSE_CHARSET (charset) >= 0) \ | |
729 | charset = CHARSET_REVERSE_CHARSET (charset); \ | |
730 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ | |
731 | } \ | |
732 | } while (0) | |
733 | ||
734 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */ | |
735 | ||
736 | int | |
737 | decode_coding_iso2022 (coding, source, destination, | |
738 | src_bytes, dst_bytes, consumed) | |
739 | struct coding_system *coding; | |
740 | unsigned char *source, *destination; | |
741 | int src_bytes, dst_bytes; | |
742 | int *consumed; | |
743 | { | |
744 | unsigned char *src = source; | |
745 | unsigned char *src_end = source + src_bytes; | |
746 | unsigned char *dst = destination; | |
747 | unsigned char *dst_end = destination + dst_bytes; | |
748 | /* Since the maximum bytes produced by each loop is 7, we subtract 6 | |
749 | from DST_END to assure that overflow checking is necessary only | |
750 | at the head of loop. */ | |
751 | unsigned char *adjusted_dst_end = dst_end - 6; | |
752 | int charset; | |
753 | /* Charsets invoked to graphic plane 0 and 1 respectively. */ | |
754 | int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
755 | int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); | |
a5d301df KH |
756 | Lisp_Object unification_table |
757 | = coding->character_unification_table_for_decode; | |
bdd9fb48 KH |
758 | |
759 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
a5d301df | 760 | unification_table = Vstandard_character_unification_table_for_decode; |
4ed46869 KH |
761 | |
762 | while (src < src_end && dst < adjusted_dst_end) | |
763 | { | |
764 | /* SRC_BASE remembers the start position in source in each loop. | |
765 | The loop will be exited when there's not enough source text | |
766 | to analyze long escape sequence or 2-byte code (within macros | |
767 | ONE_MORE_BYTE or TWO_MORE_BYTES). In that case, SRC is reset | |
768 | to SRC_BASE before exiting. */ | |
769 | unsigned char *src_base = src; | |
bdd9fb48 | 770 | int c1 = *src++, c2; |
4ed46869 KH |
771 | |
772 | switch (iso_code_class [c1]) | |
773 | { | |
774 | case ISO_0x20_or_0x7F: | |
775 | if (!coding->composing | |
776 | && (charset0 < 0 || CHARSET_CHARS (charset0) == 94)) | |
777 | { | |
778 | /* This is SPACE or DEL. */ | |
779 | *dst++ = c1; | |
780 | break; | |
781 | } | |
782 | /* This is a graphic character, we fall down ... */ | |
783 | ||
784 | case ISO_graphic_plane_0: | |
785 | if (coding->composing == COMPOSING_WITH_RULE_RULE) | |
786 | { | |
787 | /* This is a composition rule. */ | |
788 | *dst++ = c1 | 0x80; | |
789 | coding->composing = COMPOSING_WITH_RULE_TAIL; | |
790 | } | |
791 | else | |
792 | DECODE_ISO_CHARACTER (charset0, c1); | |
793 | break; | |
794 | ||
795 | case ISO_0xA0_or_0xFF: | |
796 | if (charset1 < 0 || CHARSET_CHARS (charset1) == 94) | |
797 | { | |
798 | /* Invalid code. */ | |
799 | *dst++ = c1; | |
800 | break; | |
801 | } | |
802 | /* This is a graphic character, we fall down ... */ | |
803 | ||
804 | case ISO_graphic_plane_1: | |
805 | DECODE_ISO_CHARACTER (charset1, c1); | |
806 | break; | |
807 | ||
808 | case ISO_control_code: | |
809 | /* All ISO2022 control characters in this class have the | |
810 | same representation in Emacs internal format. */ | |
811 | *dst++ = c1; | |
812 | break; | |
813 | ||
814 | case ISO_carriage_return: | |
815 | if (coding->eol_type == CODING_EOL_CR) | |
816 | { | |
817 | *dst++ = '\n'; | |
818 | } | |
819 | else if (coding->eol_type == CODING_EOL_CRLF) | |
820 | { | |
821 | ONE_MORE_BYTE (c1); | |
822 | if (c1 == ISO_CODE_LF) | |
823 | *dst++ = '\n'; | |
824 | else | |
825 | { | |
826 | src--; | |
827 | *dst++ = c1; | |
828 | } | |
829 | } | |
830 | else | |
831 | { | |
832 | *dst++ = c1; | |
833 | } | |
834 | break; | |
835 | ||
836 | case ISO_shift_out: | |
e0e989f6 KH |
837 | if (CODING_SPEC_ISO_DESIGNATION (coding, 1) < 0) |
838 | goto label_invalid_escape_sequence; | |
4ed46869 KH |
839 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; |
840 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
841 | break; | |
842 | ||
843 | case ISO_shift_in: | |
844 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; | |
845 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
846 | break; | |
847 | ||
848 | case ISO_single_shift_2_7: | |
849 | case ISO_single_shift_2: | |
850 | /* SS2 is handled as an escape sequence of ESC 'N' */ | |
851 | c1 = 'N'; | |
852 | goto label_escape_sequence; | |
853 | ||
854 | case ISO_single_shift_3: | |
855 | /* SS2 is handled as an escape sequence of ESC 'O' */ | |
856 | c1 = 'O'; | |
857 | goto label_escape_sequence; | |
858 | ||
859 | case ISO_control_sequence_introducer: | |
860 | /* CSI is handled as an escape sequence of ESC '[' ... */ | |
861 | c1 = '['; | |
862 | goto label_escape_sequence; | |
863 | ||
864 | case ISO_escape: | |
865 | ONE_MORE_BYTE (c1); | |
866 | label_escape_sequence: | |
867 | /* Escape sequences handled by Emacs are invocation, | |
868 | designation, direction specification, and character | |
869 | composition specification. */ | |
870 | switch (c1) | |
871 | { | |
872 | case '&': /* revision of following character set */ | |
873 | ONE_MORE_BYTE (c1); | |
874 | if (!(c1 >= '@' && c1 <= '~')) | |
e0e989f6 | 875 | goto label_invalid_escape_sequence; |
4ed46869 KH |
876 | ONE_MORE_BYTE (c1); |
877 | if (c1 != ISO_CODE_ESC) | |
e0e989f6 | 878 | goto label_invalid_escape_sequence; |
4ed46869 KH |
879 | ONE_MORE_BYTE (c1); |
880 | goto label_escape_sequence; | |
881 | ||
882 | case '$': /* designation of 2-byte character set */ | |
883 | ONE_MORE_BYTE (c1); | |
884 | if (c1 >= '@' && c1 <= 'B') | |
885 | { /* designation of JISX0208.1978, GB2312.1980, | |
886 | or JISX0208.1980 */ | |
887 | DECODE_DESIGNATION (0, 2, 94, c1); | |
888 | } | |
889 | else if (c1 >= 0x28 && c1 <= 0x2B) | |
890 | { /* designation of DIMENSION2_CHARS94 character set */ | |
891 | ONE_MORE_BYTE (c2); | |
892 | DECODE_DESIGNATION (c1 - 0x28, 2, 94, c2); | |
893 | } | |
894 | else if (c1 >= 0x2C && c1 <= 0x2F) | |
895 | { /* designation of DIMENSION2_CHARS96 character set */ | |
896 | ONE_MORE_BYTE (c2); | |
897 | DECODE_DESIGNATION (c1 - 0x2C, 2, 96, c2); | |
898 | } | |
899 | else | |
e0e989f6 | 900 | goto label_invalid_escape_sequence; |
4ed46869 KH |
901 | break; |
902 | ||
903 | case 'n': /* invocation of locking-shift-2 */ | |
e0e989f6 KH |
904 | if (CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) |
905 | goto label_invalid_escape_sequence; | |
4ed46869 | 906 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; |
e0e989f6 | 907 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
4ed46869 KH |
908 | break; |
909 | ||
910 | case 'o': /* invocation of locking-shift-3 */ | |
e0e989f6 KH |
911 | if (CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) |
912 | goto label_invalid_escape_sequence; | |
4ed46869 | 913 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; |
e0e989f6 | 914 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
4ed46869 KH |
915 | break; |
916 | ||
917 | case 'N': /* invocation of single-shift-2 */ | |
e0e989f6 KH |
918 | if (CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) |
919 | goto label_invalid_escape_sequence; | |
4ed46869 KH |
920 | ONE_MORE_BYTE (c1); |
921 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 2); | |
922 | DECODE_ISO_CHARACTER (charset, c1); | |
923 | break; | |
924 | ||
925 | case 'O': /* invocation of single-shift-3 */ | |
e0e989f6 KH |
926 | if (CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) |
927 | goto label_invalid_escape_sequence; | |
4ed46869 KH |
928 | ONE_MORE_BYTE (c1); |
929 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 3); | |
930 | DECODE_ISO_CHARACTER (charset, c1); | |
931 | break; | |
932 | ||
933 | case '0': /* start composing without embeded rules */ | |
934 | coding->composing = COMPOSING_NO_RULE_HEAD; | |
935 | break; | |
936 | ||
937 | case '1': /* end composing */ | |
938 | coding->composing = COMPOSING_NO; | |
939 | break; | |
940 | ||
941 | case '2': /* start composing with embeded rules */ | |
942 | coding->composing = COMPOSING_WITH_RULE_HEAD; | |
943 | break; | |
944 | ||
945 | case '[': /* specification of direction */ | |
946 | /* For the moment, nested direction is not supported. | |
947 | So, the value of `coding->direction' is 0 or 1: 0 | |
948 | means left-to-right, 1 means right-to-left. */ | |
949 | ONE_MORE_BYTE (c1); | |
950 | switch (c1) | |
951 | { | |
952 | case ']': /* end of the current direction */ | |
953 | coding->direction = 0; | |
954 | ||
955 | case '0': /* end of the current direction */ | |
956 | case '1': /* start of left-to-right direction */ | |
957 | ONE_MORE_BYTE (c1); | |
958 | if (c1 == ']') | |
959 | coding->direction = 0; | |
960 | else | |
961 | goto label_invalid_escape_sequence; | |
962 | break; | |
963 | ||
964 | case '2': /* start of right-to-left direction */ | |
965 | ONE_MORE_BYTE (c1); | |
966 | if (c1 == ']') | |
967 | coding->direction= 1; | |
968 | else | |
969 | goto label_invalid_escape_sequence; | |
970 | break; | |
971 | ||
972 | default: | |
973 | goto label_invalid_escape_sequence; | |
974 | } | |
975 | break; | |
976 | ||
977 | default: | |
978 | if (c1 >= 0x28 && c1 <= 0x2B) | |
979 | { /* designation of DIMENSION1_CHARS94 character set */ | |
980 | ONE_MORE_BYTE (c2); | |
981 | DECODE_DESIGNATION (c1 - 0x28, 1, 94, c2); | |
982 | } | |
983 | else if (c1 >= 0x2C && c1 <= 0x2F) | |
984 | { /* designation of DIMENSION1_CHARS96 character set */ | |
985 | ONE_MORE_BYTE (c2); | |
986 | DECODE_DESIGNATION (c1 - 0x2C, 1, 96, c2); | |
987 | } | |
988 | else | |
989 | { | |
990 | goto label_invalid_escape_sequence; | |
991 | } | |
992 | } | |
993 | /* We must update these variables now. */ | |
994 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); | |
995 | charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); | |
996 | break; | |
997 | ||
998 | label_invalid_escape_sequence: | |
999 | { | |
1000 | int length = src - src_base; | |
1001 | ||
1002 | bcopy (src_base, dst, length); | |
1003 | dst += length; | |
1004 | } | |
1005 | } | |
1006 | continue; | |
1007 | ||
1008 | label_end_of_loop: | |
1009 | coding->carryover_size = src - src_base; | |
1010 | bcopy (src_base, coding->carryover, coding->carryover_size); | |
1011 | src = src_base; | |
1012 | break; | |
1013 | } | |
1014 | ||
1015 | /* If this is the last block of the text to be decoded, we had | |
1016 | better just flush out all remaining codes in the text although | |
1017 | they are not valid characters. */ | |
1018 | if (coding->last_block) | |
1019 | { | |
1020 | bcopy (src, dst, src_end - src); | |
1021 | dst += (src_end - src); | |
1022 | src = src_end; | |
1023 | } | |
1024 | *consumed = src - source; | |
1025 | return dst - destination; | |
1026 | } | |
1027 | ||
f4dee582 | 1028 | /* ISO2022 encoding stuff. */ |
4ed46869 KH |
1029 | |
1030 | /* | |
f4dee582 | 1031 | It is not enough to say just "ISO2022" on encoding, we have to |
4ed46869 KH |
1032 | specify more details. In Emacs, each coding-system of ISO2022 |
1033 | variant has the following specifications: | |
1034 | 1. Initial designation to G0 thru G3. | |
1035 | 2. Allows short-form designation? | |
1036 | 3. ASCII should be designated to G0 before control characters? | |
1037 | 4. ASCII should be designated to G0 at end of line? | |
1038 | 5. 7-bit environment or 8-bit environment? | |
1039 | 6. Use locking-shift? | |
1040 | 7. Use Single-shift? | |
1041 | And the following two are only for Japanese: | |
1042 | 8. Use ASCII in place of JIS0201-1976-Roman? | |
1043 | 9. Use JISX0208-1983 in place of JISX0208-1978? | |
1044 | These specifications are encoded in `coding->flags' as flag bits | |
1045 | defined by macros CODING_FLAG_ISO_XXX. See `coding.h' for more | |
f4dee582 | 1046 | details. |
4ed46869 KH |
1047 | */ |
1048 | ||
1049 | /* Produce codes (escape sequence) for designating CHARSET to graphic | |
1050 | register REG. If <final-char> of CHARSET is '@', 'A', or 'B' and | |
1051 | the coding system CODING allows, produce designation sequence of | |
1052 | short-form. */ | |
1053 | ||
1054 | #define ENCODE_DESIGNATION(charset, reg, coding) \ | |
1055 | do { \ | |
1056 | unsigned char final_char = CHARSET_ISO_FINAL_CHAR (charset); \ | |
1057 | char *intermediate_char_94 = "()*+"; \ | |
1058 | char *intermediate_char_96 = ",-./"; \ | |
1059 | Lisp_Object temp \ | |
1060 | = Fassq (make_number (charset), Vcharset_revision_alist); \ | |
1061 | if (! NILP (temp)) \ | |
1062 | { \ | |
1063 | *dst++ = ISO_CODE_ESC; \ | |
1064 | *dst++ = '&'; \ | |
1065 | *dst++ = XINT (XCONS (temp)->cdr) + '@'; \ | |
1066 | } \ | |
1067 | *dst++ = ISO_CODE_ESC; \ | |
1068 | if (CHARSET_DIMENSION (charset) == 1) \ | |
1069 | { \ | |
1070 | if (CHARSET_CHARS (charset) == 94) \ | |
1071 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ | |
1072 | else \ | |
1073 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ | |
1074 | } \ | |
1075 | else \ | |
1076 | { \ | |
1077 | *dst++ = '$'; \ | |
1078 | if (CHARSET_CHARS (charset) == 94) \ | |
1079 | { \ | |
1080 | if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \ | |
1081 | || reg != 0 \ | |
1082 | || final_char < '@' || final_char > 'B') \ | |
1083 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ | |
1084 | } \ | |
1085 | else \ | |
1086 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ | |
1087 | } \ | |
1088 | *dst++ = final_char; \ | |
1089 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ | |
1090 | } while (0) | |
1091 | ||
1092 | /* The following two macros produce codes (control character or escape | |
1093 | sequence) for ISO2022 single-shift functions (single-shift-2 and | |
1094 | single-shift-3). */ | |
1095 | ||
1096 | #define ENCODE_SINGLE_SHIFT_2 \ | |
1097 | do { \ | |
1098 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1099 | *dst++ = ISO_CODE_ESC, *dst++ = 'N'; \ | |
1100 | else \ | |
1101 | *dst++ = ISO_CODE_SS2; \ | |
1102 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ | |
1103 | } while (0) | |
1104 | ||
1105 | #define ENCODE_SINGLE_SHIFT_3 \ | |
1106 | do { \ | |
1107 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1108 | *dst++ = ISO_CODE_ESC, *dst++ = 'O'; \ | |
1109 | else \ | |
1110 | *dst++ = ISO_CODE_SS3; \ | |
1111 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ | |
1112 | } while (0) | |
1113 | ||
1114 | /* The following four macros produce codes (control character or | |
1115 | escape sequence) for ISO2022 locking-shift functions (shift-in, | |
1116 | shift-out, locking-shift-2, and locking-shift-3). */ | |
1117 | ||
1118 | #define ENCODE_SHIFT_IN \ | |
1119 | do { \ | |
1120 | *dst++ = ISO_CODE_SI; \ | |
1121 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; \ | |
1122 | } while (0) | |
1123 | ||
1124 | #define ENCODE_SHIFT_OUT \ | |
1125 | do { \ | |
1126 | *dst++ = ISO_CODE_SO; \ | |
1127 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; \ | |
1128 | } while (0) | |
1129 | ||
1130 | #define ENCODE_LOCKING_SHIFT_2 \ | |
1131 | do { \ | |
1132 | *dst++ = ISO_CODE_ESC, *dst++ = 'n'; \ | |
1133 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; \ | |
1134 | } while (0) | |
1135 | ||
1136 | #define ENCODE_LOCKING_SHIFT_3 \ | |
1137 | do { \ | |
1138 | *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \ | |
1139 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; \ | |
1140 | } while (0) | |
1141 | ||
f4dee582 RS |
1142 | /* Produce codes for a DIMENSION1 character whose character set is |
1143 | CHARSET and whose position-code is C1. Designation and invocation | |
4ed46869 KH |
1144 | sequences are also produced in advance if necessary. */ |
1145 | ||
1146 | ||
1147 | #define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \ | |
1148 | do { \ | |
1149 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ | |
1150 | { \ | |
1151 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1152 | *dst++ = c1 & 0x7F; \ | |
1153 | else \ | |
1154 | *dst++ = c1 | 0x80; \ | |
1155 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ | |
1156 | break; \ | |
1157 | } \ | |
1158 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ | |
1159 | { \ | |
1160 | *dst++ = c1 & 0x7F; \ | |
1161 | break; \ | |
1162 | } \ | |
1163 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ | |
1164 | { \ | |
1165 | *dst++ = c1 | 0x80; \ | |
1166 | break; \ | |
1167 | } \ | |
1168 | else \ | |
1169 | /* Since CHARSET is not yet invoked to any graphic planes, we \ | |
1170 | must invoke it, or, at first, designate it to some graphic \ | |
1171 | register. Then repeat the loop to actually produce the \ | |
1172 | character. */ \ | |
1173 | dst = encode_invocation_designation (charset, coding, dst); \ | |
1174 | } while (1) | |
1175 | ||
f4dee582 RS |
1176 | /* Produce codes for a DIMENSION2 character whose character set is |
1177 | CHARSET and whose position-codes are C1 and C2. Designation and | |
4ed46869 KH |
1178 | invocation codes are also produced in advance if necessary. */ |
1179 | ||
1180 | #define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2) \ | |
1181 | do { \ | |
1182 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ | |
1183 | { \ | |
1184 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ | |
1185 | *dst++ = c1 & 0x7F, *dst++ = c2 & 0x7F; \ | |
1186 | else \ | |
1187 | *dst++ = c1 | 0x80, *dst++ = c2 | 0x80; \ | |
1188 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ | |
1189 | break; \ | |
1190 | } \ | |
1191 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ | |
1192 | { \ | |
1193 | *dst++ = c1 & 0x7F, *dst++= c2 & 0x7F; \ | |
1194 | break; \ | |
1195 | } \ | |
1196 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ | |
1197 | { \ | |
1198 | *dst++ = c1 | 0x80, *dst++= c2 | 0x80; \ | |
1199 | break; \ | |
1200 | } \ | |
1201 | else \ | |
1202 | /* Since CHARSET is not yet invoked to any graphic planes, we \ | |
1203 | must invoke it, or, at first, designate it to some graphic \ | |
1204 | register. Then repeat the loop to actually produce the \ | |
1205 | character. */ \ | |
1206 | dst = encode_invocation_designation (charset, coding, dst); \ | |
1207 | } while (1) | |
1208 | ||
bdd9fb48 KH |
1209 | #define ENCODE_ISO_CHARACTER(charset, c1, c2) \ |
1210 | do { \ | |
1211 | int c_alt, charset_alt; \ | |
1212 | if (!NILP (unification_table) \ | |
1213 | && ((c_alt = unify_char (unification_table, -1, charset, c1, c2)) \ | |
a5d301df | 1214 | >= 0)) \ |
bdd9fb48 KH |
1215 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
1216 | else \ | |
1217 | charset_alt = charset; \ | |
1218 | if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
1219 | ENCODE_ISO_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
1220 | else \ | |
1221 | ENCODE_ISO_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
1222 | } while (0) | |
1223 | ||
4ed46869 KH |
1224 | /* Produce designation and invocation codes at a place pointed by DST |
1225 | to use CHARSET. The element `spec.iso2022' of *CODING is updated. | |
1226 | Return new DST. */ | |
1227 | ||
1228 | unsigned char * | |
1229 | encode_invocation_designation (charset, coding, dst) | |
1230 | int charset; | |
1231 | struct coding_system *coding; | |
1232 | unsigned char *dst; | |
1233 | { | |
1234 | int reg; /* graphic register number */ | |
1235 | ||
1236 | /* At first, check designations. */ | |
1237 | for (reg = 0; reg < 4; reg++) | |
1238 | if (charset == CODING_SPEC_ISO_DESIGNATION (coding, reg)) | |
1239 | break; | |
1240 | ||
1241 | if (reg >= 4) | |
1242 | { | |
1243 | /* CHARSET is not yet designated to any graphic registers. */ | |
1244 | /* At first check the requested designation. */ | |
1245 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); | |
1ba9e4ab KH |
1246 | if (reg == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION) |
1247 | /* Since CHARSET requests no special designation, designate it | |
1248 | to graphic register 0. */ | |
4ed46869 KH |
1249 | reg = 0; |
1250 | ||
1251 | ENCODE_DESIGNATION (charset, reg, coding); | |
1252 | } | |
1253 | ||
1254 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != reg | |
1255 | && CODING_SPEC_ISO_INVOCATION (coding, 1) != reg) | |
1256 | { | |
1257 | /* Since the graphic register REG is not invoked to any graphic | |
1258 | planes, invoke it to graphic plane 0. */ | |
1259 | switch (reg) | |
1260 | { | |
1261 | case 0: /* graphic register 0 */ | |
1262 | ENCODE_SHIFT_IN; | |
1263 | break; | |
1264 | ||
1265 | case 1: /* graphic register 1 */ | |
1266 | ENCODE_SHIFT_OUT; | |
1267 | break; | |
1268 | ||
1269 | case 2: /* graphic register 2 */ | |
1270 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) | |
1271 | ENCODE_SINGLE_SHIFT_2; | |
1272 | else | |
1273 | ENCODE_LOCKING_SHIFT_2; | |
1274 | break; | |
1275 | ||
1276 | case 3: /* graphic register 3 */ | |
1277 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) | |
1278 | ENCODE_SINGLE_SHIFT_3; | |
1279 | else | |
1280 | ENCODE_LOCKING_SHIFT_3; | |
1281 | break; | |
1282 | } | |
1283 | } | |
1284 | return dst; | |
1285 | } | |
1286 | ||
1287 | /* The following two macros produce codes for indicating composition. */ | |
1288 | #define ENCODE_COMPOSITION_NO_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '0' | |
1289 | #define ENCODE_COMPOSITION_WITH_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '2' | |
1290 | #define ENCODE_COMPOSITION_END *dst++ = ISO_CODE_ESC, *dst++ = '1' | |
1291 | ||
1292 | /* The following three macros produce codes for indicating direction | |
1293 | of text. */ | |
1294 | #define ENCODE_CONTROL_SEQUENCE_INTRODUCER \ | |
1295 | do { \ | |
1296 | if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS) \ | |
1297 | *dst++ = ISO_CODE_ESC, *dst++ = '['; \ | |
1298 | else \ | |
1299 | *dst++ = ISO_CODE_CSI; \ | |
1300 | } while (0) | |
1301 | ||
1302 | #define ENCODE_DIRECTION_R2L \ | |
1303 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '2', *dst++ = ']' | |
1304 | ||
1305 | #define ENCODE_DIRECTION_L2R \ | |
1306 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '0', *dst++ = ']' | |
1307 | ||
1308 | /* Produce codes for designation and invocation to reset the graphic | |
1309 | planes and registers to initial state. */ | |
e0e989f6 KH |
1310 | #define ENCODE_RESET_PLANE_AND_REGISTER \ |
1311 | do { \ | |
1312 | int reg; \ | |
1313 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != 0) \ | |
1314 | ENCODE_SHIFT_IN; \ | |
1315 | for (reg = 0; reg < 4; reg++) \ | |
1316 | if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg) >= 0 \ | |
1317 | && (CODING_SPEC_ISO_DESIGNATION (coding, reg) \ | |
1318 | != CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg))) \ | |
1319 | ENCODE_DESIGNATION \ | |
1320 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg), reg, coding); \ | |
4ed46869 KH |
1321 | } while (0) |
1322 | ||
bdd9fb48 KH |
1323 | /* Produce designation sequences of charsets in the line started from |
1324 | *SRC to a place pointed by DSTP. | |
1325 | ||
1326 | If the current block ends before any end-of-line, we may fail to | |
1327 | find all the necessary *designations. */ | |
1328 | encode_designation_at_bol (coding, table, src, src_end, dstp) | |
e0e989f6 | 1329 | struct coding_system *coding; |
bdd9fb48 | 1330 | Lisp_Object table; |
e0e989f6 KH |
1331 | unsigned char *src, *src_end, **dstp; |
1332 | { | |
bdd9fb48 KH |
1333 | int charset, c, found = 0, reg; |
1334 | /* Table of charsets to be designated to each graphic register. */ | |
1335 | int r[4]; | |
1336 | unsigned char *dst = *dstp; | |
1337 | ||
1338 | for (reg = 0; reg < 4; reg++) | |
1339 | r[reg] = -1; | |
1340 | ||
1341 | while (src < src_end && *src != '\n' && found < 4) | |
e0e989f6 | 1342 | { |
bdd9fb48 KH |
1343 | int bytes = BYTES_BY_CHAR_HEAD (*src); |
1344 | ||
1345 | if (NILP (table)) | |
1346 | charset = CHARSET_AT (src); | |
1347 | else | |
e0e989f6 | 1348 | { |
bdd9fb48 KH |
1349 | int c_alt, c1, c2; |
1350 | ||
1351 | SPLIT_STRING(src, bytes, charset, c1, c2); | |
1352 | if ((c_alt = unify_char (table, -1, charset, c1, c2)) >= 0) | |
1353 | charset = CHAR_CHARSET (c_alt); | |
e0e989f6 | 1354 | } |
bdd9fb48 | 1355 | |
e0e989f6 | 1356 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); |
1ba9e4ab | 1357 | if (r[reg] == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION) |
bdd9fb48 KH |
1358 | { |
1359 | found++; | |
1360 | r[reg] = charset; | |
1361 | } | |
1362 | ||
1363 | src += bytes; | |
1364 | } | |
1365 | ||
1366 | if (found) | |
1367 | { | |
1368 | for (reg = 0; reg < 4; reg++) | |
1369 | if (r[reg] >= 0 | |
1370 | && CODING_SPEC_ISO_DESIGNATION (coding, reg) != r[reg]) | |
1371 | ENCODE_DESIGNATION (r[reg], reg, coding); | |
1372 | *dstp = dst; | |
e0e989f6 | 1373 | } |
e0e989f6 KH |
1374 | } |
1375 | ||
4ed46869 KH |
1376 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */ |
1377 | ||
1378 | int | |
1379 | encode_coding_iso2022 (coding, source, destination, | |
1380 | src_bytes, dst_bytes, consumed) | |
1381 | struct coding_system *coding; | |
1382 | unsigned char *source, *destination; | |
1383 | int src_bytes, dst_bytes; | |
1384 | int *consumed; | |
1385 | { | |
1386 | unsigned char *src = source; | |
1387 | unsigned char *src_end = source + src_bytes; | |
1388 | unsigned char *dst = destination; | |
1389 | unsigned char *dst_end = destination + dst_bytes; | |
e0e989f6 | 1390 | /* Since the maximum bytes produced by each loop is 20, we subtract 19 |
4ed46869 KH |
1391 | from DST_END to assure overflow checking is necessary only at the |
1392 | head of loop. */ | |
e0e989f6 | 1393 | unsigned char *adjusted_dst_end = dst_end - 19; |
a5d301df KH |
1394 | Lisp_Object unification_table |
1395 | = coding->character_unification_table_for_encode; | |
bdd9fb48 KH |
1396 | |
1397 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
a5d301df | 1398 | unification_table = Vstandard_character_unification_table_for_encode; |
4ed46869 KH |
1399 | |
1400 | while (src < src_end && dst < adjusted_dst_end) | |
1401 | { | |
1402 | /* SRC_BASE remembers the start position in source in each loop. | |
1403 | The loop will be exited when there's not enough source text | |
1404 | to analyze multi-byte codes (within macros ONE_MORE_BYTE, | |
1405 | TWO_MORE_BYTES, and THREE_MORE_BYTES). In that case, SRC is | |
1406 | reset to SRC_BASE before exiting. */ | |
1407 | unsigned char *src_base = src; | |
bdd9fb48 | 1408 | int charset, c1, c2, c3, c4; |
4ed46869 | 1409 | |
e0e989f6 KH |
1410 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL |
1411 | && CODING_SPEC_ISO_BOL (coding)) | |
1412 | { | |
bdd9fb48 KH |
1413 | /* We have to produce designation sequences if any now. */ |
1414 | encode_designation_at_bol (coding, unification_table, | |
1415 | src, src_end, &dst); | |
e0e989f6 KH |
1416 | CODING_SPEC_ISO_BOL (coding) = 0; |
1417 | } | |
1418 | ||
1419 | c1 = *src++; | |
4ed46869 KH |
1420 | /* If we are seeing a component of a composite character, we are |
1421 | seeing a leading-code specially encoded for composition, or a | |
1422 | composition rule if composing with rule. We must set C1 | |
1423 | to a normal leading-code or an ASCII code. If we are not at | |
1424 | a composed character, we must reset the composition state. */ | |
1425 | if (COMPOSING_P (coding->composing)) | |
1426 | { | |
1427 | if (c1 < 0xA0) | |
1428 | { | |
1429 | /* We are not in a composite character any longer. */ | |
1430 | coding->composing = COMPOSING_NO; | |
1431 | ENCODE_COMPOSITION_END; | |
1432 | } | |
1433 | else | |
1434 | { | |
1435 | if (coding->composing == COMPOSING_WITH_RULE_RULE) | |
1436 | { | |
1437 | *dst++ = c1 & 0x7F; | |
1438 | coding->composing = COMPOSING_WITH_RULE_HEAD; | |
1439 | continue; | |
1440 | } | |
1441 | else if (coding->composing == COMPOSING_WITH_RULE_HEAD) | |
1442 | coding->composing = COMPOSING_WITH_RULE_RULE; | |
1443 | if (c1 == 0xA0) | |
1444 | { | |
1445 | /* This is an ASCII component. */ | |
1446 | ONE_MORE_BYTE (c1); | |
1447 | c1 &= 0x7F; | |
1448 | } | |
1449 | else | |
1450 | /* This is a leading-code of non ASCII component. */ | |
1451 | c1 -= 0x20; | |
1452 | } | |
1453 | } | |
1454 | ||
1455 | /* Now encode one character. C1 is a control character, an | |
1456 | ASCII character, or a leading-code of multi-byte character. */ | |
1457 | switch (emacs_code_class[c1]) | |
1458 | { | |
1459 | case EMACS_ascii_code: | |
bdd9fb48 | 1460 | ENCODE_ISO_CHARACTER (CHARSET_ASCII, c1, /* dummy */ c2); |
4ed46869 KH |
1461 | break; |
1462 | ||
1463 | case EMACS_control_code: | |
1464 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) | |
e0e989f6 | 1465 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1466 | *dst++ = c1; |
1467 | break; | |
1468 | ||
1469 | case EMACS_carriage_return_code: | |
1470 | if (!coding->selective) | |
1471 | { | |
1472 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) | |
e0e989f6 | 1473 | ENCODE_RESET_PLANE_AND_REGISTER; |
4ed46869 KH |
1474 | *dst++ = c1; |
1475 | break; | |
1476 | } | |
1477 | /* fall down to treat '\r' as '\n' ... */ | |
1478 | ||
1479 | case EMACS_linefeed_code: | |
1480 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL) | |
e0e989f6 KH |
1481 | ENCODE_RESET_PLANE_AND_REGISTER; |
1482 | if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL) | |
1483 | bcopy (coding->spec.iso2022.initial_designation, | |
1484 | coding->spec.iso2022.current_designation, | |
1485 | sizeof coding->spec.iso2022.initial_designation); | |
4ed46869 | 1486 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 1487 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
1488 | *dst++ = ISO_CODE_LF; |
1489 | else if (coding->eol_type == CODING_EOL_CRLF) | |
1490 | *dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF; | |
1491 | else | |
1492 | *dst++ = ISO_CODE_CR; | |
e0e989f6 | 1493 | CODING_SPEC_ISO_BOL (coding) = 1; |
4ed46869 KH |
1494 | break; |
1495 | ||
1496 | case EMACS_leading_code_2: | |
1497 | ONE_MORE_BYTE (c2); | |
bdd9fb48 | 1498 | ENCODE_ISO_CHARACTER (c1, c2, /* dummy */ c3); |
4ed46869 KH |
1499 | break; |
1500 | ||
1501 | case EMACS_leading_code_3: | |
1502 | TWO_MORE_BYTES (c2, c3); | |
1503 | if (c1 < LEADING_CODE_PRIVATE_11) | |
bdd9fb48 | 1504 | ENCODE_ISO_CHARACTER (c1, c2, c3); |
4ed46869 | 1505 | else |
bdd9fb48 | 1506 | ENCODE_ISO_CHARACTER (c2, c3, /* dummy */ c4); |
4ed46869 KH |
1507 | break; |
1508 | ||
1509 | case EMACS_leading_code_4: | |
1510 | THREE_MORE_BYTES (c2, c3, c4); | |
bdd9fb48 | 1511 | ENCODE_ISO_CHARACTER (c2, c3, c4); |
4ed46869 KH |
1512 | break; |
1513 | ||
1514 | case EMACS_leading_code_composition: | |
1515 | ONE_MORE_BYTE (c1); | |
1516 | if (c1 == 0xFF) | |
1517 | { | |
1518 | coding->composing = COMPOSING_WITH_RULE_HEAD; | |
1519 | ENCODE_COMPOSITION_WITH_RULE_START; | |
1520 | } | |
1521 | else | |
1522 | { | |
1523 | /* Rewind one byte because it is a character code of | |
1524 | composition elements. */ | |
1525 | src--; | |
1526 | coding->composing = COMPOSING_NO_RULE_HEAD; | |
1527 | ENCODE_COMPOSITION_NO_RULE_START; | |
1528 | } | |
1529 | break; | |
1530 | ||
1531 | case EMACS_invalid_code: | |
1532 | *dst++ = c1; | |
1533 | break; | |
1534 | } | |
1535 | continue; | |
1536 | label_end_of_loop: | |
1537 | coding->carryover_size = src - src_base; | |
1538 | bcopy (src_base, coding->carryover, coding->carryover_size); | |
4ed46869 KH |
1539 | break; |
1540 | } | |
1541 | ||
1542 | /* If this is the last block of the text to be encoded, we must | |
bdd9fb48 KH |
1543 | reset graphic planes and registers to the initial state. */ |
1544 | if (src >= src_end && coding->last_block) | |
4ed46869 | 1545 | { |
e0e989f6 | 1546 | ENCODE_RESET_PLANE_AND_REGISTER; |
bdd9fb48 KH |
1547 | if (coding->carryover_size > 0 |
1548 | && coding->carryover_size < (dst_end - dst)) | |
1549 | { | |
1550 | bcopy (coding->carryover, dst, coding->carryover_size); | |
1551 | dst += coding->carryover_size; | |
1552 | coding->carryover_size = 0; | |
1553 | } | |
4ed46869 KH |
1554 | } |
1555 | *consumed = src - source; | |
1556 | return dst - destination; | |
1557 | } | |
1558 | ||
1559 | \f | |
1560 | /*** 4. SJIS and BIG5 handlers ***/ | |
1561 | ||
f4dee582 | 1562 | /* Although SJIS and BIG5 are not ISO's coding system, they are used |
4ed46869 KH |
1563 | quite widely. So, for the moment, Emacs supports them in the bare |
1564 | C code. But, in the future, they may be supported only by CCL. */ | |
1565 | ||
1566 | /* SJIS is a coding system encoding three character sets: ASCII, right | |
1567 | half of JISX0201-Kana, and JISX0208. An ASCII character is encoded | |
1568 | as is. A character of charset katakana-jisx0201 is encoded by | |
1569 | "position-code + 0x80". A character of charset japanese-jisx0208 | |
1570 | is encoded in 2-byte but two position-codes are divided and shifted | |
1571 | so that it fit in the range below. | |
1572 | ||
1573 | --- CODE RANGE of SJIS --- | |
1574 | (character set) (range) | |
1575 | ASCII 0x00 .. 0x7F | |
1576 | KATAKANA-JISX0201 0xA0 .. 0xDF | |
1577 | JISX0208 (1st byte) 0x80 .. 0x9F and 0xE0 .. 0xFF | |
1578 | (2nd byte) 0x40 .. 0xFF | |
1579 | ------------------------------- | |
1580 | ||
1581 | */ | |
1582 | ||
1583 | /* BIG5 is a coding system encoding two character sets: ASCII and | |
1584 | Big5. An ASCII character is encoded as is. Big5 is a two-byte | |
1585 | character set and is encoded in two-byte. | |
1586 | ||
1587 | --- CODE RANGE of BIG5 --- | |
1588 | (character set) (range) | |
1589 | ASCII 0x00 .. 0x7F | |
1590 | Big5 (1st byte) 0xA1 .. 0xFE | |
1591 | (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE | |
1592 | -------------------------- | |
1593 | ||
1594 | Since the number of characters in Big5 is larger than maximum | |
1595 | characters in Emacs' charset (96x96), it can't be handled as one | |
1596 | charset. So, in Emacs, Big5 is divided into two: `charset-big5-1' | |
1597 | and `charset-big5-2'. Both are DIMENSION2 and CHARS94. The former | |
1598 | contains frequently used characters and the latter contains less | |
1599 | frequently used characters. */ | |
1600 | ||
1601 | /* Macros to decode or encode a character of Big5 in BIG5. B1 and B2 | |
1602 | are the 1st and 2nd position-codes of Big5 in BIG5 coding system. | |
1603 | C1 and C2 are the 1st and 2nd position-codes of of Emacs' internal | |
1604 | format. CHARSET is `charset_big5_1' or `charset_big5_2'. */ | |
1605 | ||
1606 | /* Number of Big5 characters which have the same code in 1st byte. */ | |
1607 | #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) | |
1608 | ||
1609 | #define DECODE_BIG5(b1, b2, charset, c1, c2) \ | |
1610 | do { \ | |
1611 | unsigned int temp \ | |
1612 | = (b1 - 0xA1) * BIG5_SAME_ROW + b2 - (b2 < 0x7F ? 0x40 : 0x62); \ | |
1613 | if (b1 < 0xC9) \ | |
1614 | charset = charset_big5_1; \ | |
1615 | else \ | |
1616 | { \ | |
1617 | charset = charset_big5_2; \ | |
1618 | temp -= (0xC9 - 0xA1) * BIG5_SAME_ROW; \ | |
1619 | } \ | |
1620 | c1 = temp / (0xFF - 0xA1) + 0x21; \ | |
1621 | c2 = temp % (0xFF - 0xA1) + 0x21; \ | |
1622 | } while (0) | |
1623 | ||
1624 | #define ENCODE_BIG5(charset, c1, c2, b1, b2) \ | |
1625 | do { \ | |
1626 | unsigned int temp = (c1 - 0x21) * (0xFF - 0xA1) + (c2 - 0x21); \ | |
1627 | if (charset == charset_big5_2) \ | |
1628 | temp += BIG5_SAME_ROW * (0xC9 - 0xA1); \ | |
1629 | b1 = temp / BIG5_SAME_ROW + 0xA1; \ | |
1630 | b2 = temp % BIG5_SAME_ROW; \ | |
1631 | b2 += b2 < 0x3F ? 0x40 : 0x62; \ | |
1632 | } while (0) | |
1633 | ||
a5d301df KH |
1634 | #define DECODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ |
1635 | do { \ | |
1636 | int c_alt, charset_alt = (charset); \ | |
1637 | if (!NILP (unification_table) \ | |
1638 | && ((c_alt = unify_char (unification_table, \ | |
1639 | -1, (charset), c1, c2)) >= 0)) \ | |
1640 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
1641 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ | |
1642 | DECODE_CHARACTER_ASCII (c1); \ | |
1643 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
1644 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ | |
1645 | else \ | |
1646 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ | |
1647 | } while (0) | |
1648 | ||
1649 | #define ENCODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ | |
1650 | do { \ | |
1651 | int c_alt, charset_alt; \ | |
1652 | if (!NILP (unification_table) \ | |
1653 | && ((c_alt = unify_char (unification_table, -1, charset, c1, c2)) \ | |
1654 | >= 0)) \ | |
1655 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ | |
1656 | else \ | |
1657 | charset_alt = charset; \ | |
1658 | if (charset_alt == charset_ascii) \ | |
1659 | *dst++ = c1; \ | |
1660 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ | |
1661 | { \ | |
1662 | if (sjis_p && charset_alt == charset_katakana_jisx0201) \ | |
1663 | *dst++ = c1; \ | |
1664 | else \ | |
1665 | *dst++ = charset_alt, *dst++ = c1; \ | |
1666 | } \ | |
1667 | else \ | |
1668 | { \ | |
1669 | c1 &= 0x7F, c2 &= 0x7F; \ | |
1670 | if (sjis_p && charset_alt == charset_jisx0208) \ | |
1671 | { \ | |
1672 | unsigned char s1, s2; \ | |
1673 | \ | |
1674 | ENCODE_SJIS (c1, c2, s1, s2); \ | |
1675 | *dst++ = s1, *dst++ = s2; \ | |
1676 | } \ | |
1677 | else if (!sjis_p \ | |
1678 | && (charset_alt == charset_big5_1 \ | |
1679 | || charset_alt == charset_big5_2)) \ | |
1680 | { \ | |
1681 | unsigned char b1, b2; \ | |
1682 | \ | |
9ce27fde | 1683 | ENCODE_BIG5 (charset_alt, c1, c2, b1, b2); \ |
a5d301df KH |
1684 | *dst++ = b1, *dst++ = b2; \ |
1685 | } \ | |
1686 | else \ | |
1687 | *dst++ = charset_alt, *dst++ = c1, *dst++ = c2; \ | |
1688 | } \ | |
1689 | } while (0); | |
1690 | ||
4ed46869 KH |
1691 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
1692 | Check if a text is encoded in SJIS. If it is, return | |
1693 | CODING_CATEGORY_MASK_SJIS, else return 0. */ | |
1694 | ||
1695 | int | |
1696 | detect_coding_sjis (src, src_end) | |
1697 | unsigned char *src, *src_end; | |
1698 | { | |
1699 | unsigned char c; | |
1700 | ||
1701 | while (src < src_end) | |
1702 | { | |
1703 | c = *src++; | |
1704 | if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) | |
1705 | return 0; | |
1706 | if ((c >= 0x80 && c < 0xA0) || c >= 0xE0) | |
1707 | { | |
1708 | if (src < src_end && *src++ < 0x40) | |
1709 | return 0; | |
1710 | } | |
1711 | } | |
1712 | return CODING_CATEGORY_MASK_SJIS; | |
1713 | } | |
1714 | ||
1715 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". | |
1716 | Check if a text is encoded in BIG5. If it is, return | |
1717 | CODING_CATEGORY_MASK_BIG5, else return 0. */ | |
1718 | ||
1719 | int | |
1720 | detect_coding_big5 (src, src_end) | |
1721 | unsigned char *src, *src_end; | |
1722 | { | |
1723 | unsigned char c; | |
1724 | ||
1725 | while (src < src_end) | |
1726 | { | |
1727 | c = *src++; | |
1728 | if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) | |
1729 | return 0; | |
1730 | if (c >= 0xA1) | |
1731 | { | |
1732 | if (src >= src_end) | |
1733 | break; | |
1734 | c = *src++; | |
1735 | if (c < 0x40 || (c >= 0x7F && c <= 0xA0)) | |
1736 | return 0; | |
1737 | } | |
1738 | } | |
1739 | return CODING_CATEGORY_MASK_BIG5; | |
1740 | } | |
1741 | ||
1742 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". | |
1743 | If SJIS_P is 1, decode SJIS text, else decode BIG5 test. */ | |
1744 | ||
1745 | int | |
1746 | decode_coding_sjis_big5 (coding, source, destination, | |
1747 | src_bytes, dst_bytes, consumed, sjis_p) | |
1748 | struct coding_system *coding; | |
1749 | unsigned char *source, *destination; | |
1750 | int src_bytes, dst_bytes; | |
1751 | int *consumed; | |
1752 | int sjis_p; | |
1753 | { | |
1754 | unsigned char *src = source; | |
1755 | unsigned char *src_end = source + src_bytes; | |
1756 | unsigned char *dst = destination; | |
1757 | unsigned char *dst_end = destination + dst_bytes; | |
1758 | /* Since the maximum bytes produced by each loop is 4, we subtract 3 | |
1759 | from DST_END to assure overflow checking is necessary only at the | |
1760 | head of loop. */ | |
1761 | unsigned char *adjusted_dst_end = dst_end - 3; | |
a5d301df KH |
1762 | Lisp_Object unification_table |
1763 | = coding->character_unification_table_for_decode; | |
1764 | ||
1765 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
1766 | unification_table = Vstandard_character_unification_table_for_decode; | |
4ed46869 KH |
1767 | |
1768 | while (src < src_end && dst < adjusted_dst_end) | |
1769 | { | |
1770 | /* SRC_BASE remembers the start position in source in each loop. | |
1771 | The loop will be exited when there's not enough source text | |
1772 | to analyze two-byte character (within macro ONE_MORE_BYTE). | |
1773 | In that case, SRC is reset to SRC_BASE before exiting. */ | |
1774 | unsigned char *src_base = src; | |
1775 | unsigned char c1 = *src++, c2, c3, c4; | |
1776 | ||
1777 | if (c1 == '\r') | |
1778 | { | |
1779 | if (coding->eol_type == CODING_EOL_CRLF) | |
1780 | { | |
1781 | ONE_MORE_BYTE (c2); | |
1782 | if (c2 == '\n') | |
1783 | *dst++ = c2; | |
1784 | else | |
1785 | /* To process C2 again, SRC is subtracted by 1. */ | |
1786 | *dst++ = c1, src--; | |
1787 | } | |
1788 | else | |
1789 | *dst++ = c1; | |
1790 | } | |
a5d301df | 1791 | else if (c1 < 0x20) |
4ed46869 | 1792 | *dst++ = c1; |
a5d301df KH |
1793 | else if (c1 < 0x80) |
1794 | DECODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); | |
4ed46869 KH |
1795 | else if (c1 < 0xA0 || c1 >= 0xE0) |
1796 | { | |
1797 | /* SJIS -> JISX0208, BIG5 -> Big5 (only if 0xE0 <= c1 < 0xFF) */ | |
1798 | if (sjis_p) | |
1799 | { | |
1800 | ONE_MORE_BYTE (c2); | |
1801 | DECODE_SJIS (c1, c2, c3, c4); | |
a5d301df | 1802 | DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208, c3, c4); |
4ed46869 KH |
1803 | } |
1804 | else if (c1 >= 0xE0 && c1 < 0xFF) | |
1805 | { | |
1806 | int charset; | |
1807 | ||
1808 | ONE_MORE_BYTE (c2); | |
1809 | DECODE_BIG5 (c1, c2, charset, c3, c4); | |
a5d301df | 1810 | DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4); |
4ed46869 KH |
1811 | } |
1812 | else /* Invalid code */ | |
1813 | *dst++ = c1; | |
1814 | } | |
1815 | else | |
1816 | { | |
1817 | /* SJIS -> JISX0201-Kana, BIG5 -> Big5 */ | |
1818 | if (sjis_p) | |
a5d301df | 1819 | DECODE_SJIS_BIG5_CHARACTER (charset_katakana_jisx0201, c1, /* dummy */ c2); |
4ed46869 KH |
1820 | else |
1821 | { | |
1822 | int charset; | |
1823 | ||
1824 | ONE_MORE_BYTE (c2); | |
1825 | DECODE_BIG5 (c1, c2, charset, c3, c4); | |
a5d301df | 1826 | DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4); |
4ed46869 KH |
1827 | } |
1828 | } | |
1829 | continue; | |
1830 | ||
1831 | label_end_of_loop: | |
1832 | coding->carryover_size = src - src_base; | |
1833 | bcopy (src_base, coding->carryover, coding->carryover_size); | |
1834 | src = src_base; | |
1835 | break; | |
1836 | } | |
1837 | ||
1838 | *consumed = src - source; | |
1839 | return dst - destination; | |
1840 | } | |
1841 | ||
1842 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". | |
1843 | This function can encode `charset_ascii', `charset_katakana_jisx0201', | |
1844 | `charset_jisx0208', `charset_big5_1', and `charset_big5-2'. We are | |
1845 | sure that all these charsets are registered as official charset | |
1846 | (i.e. do not have extended leading-codes). Characters of other | |
1847 | charsets are produced without any encoding. If SJIS_P is 1, encode | |
1848 | SJIS text, else encode BIG5 text. */ | |
1849 | ||
1850 | int | |
1851 | encode_coding_sjis_big5 (coding, source, destination, | |
1852 | src_bytes, dst_bytes, consumed, sjis_p) | |
1853 | struct coding_system *coding; | |
1854 | unsigned char *source, *destination; | |
1855 | int src_bytes, dst_bytes; | |
1856 | int *consumed; | |
1857 | int sjis_p; | |
1858 | { | |
1859 | unsigned char *src = source; | |
1860 | unsigned char *src_end = source + src_bytes; | |
1861 | unsigned char *dst = destination; | |
1862 | unsigned char *dst_end = destination + dst_bytes; | |
1863 | /* Since the maximum bytes produced by each loop is 2, we subtract 1 | |
1864 | from DST_END to assure overflow checking is necessary only at the | |
1865 | head of loop. */ | |
1866 | unsigned char *adjusted_dst_end = dst_end - 1; | |
a5d301df KH |
1867 | Lisp_Object unification_table |
1868 | = coding->character_unification_table_for_encode; | |
1869 | ||
1870 | if (!NILP (Venable_character_unification) && NILP (unification_table)) | |
1871 | unification_table = Vstandard_character_unification_table_for_encode; | |
4ed46869 KH |
1872 | |
1873 | while (src < src_end && dst < adjusted_dst_end) | |
1874 | { | |
1875 | /* SRC_BASE remembers the start position in source in each loop. | |
1876 | The loop will be exited when there's not enough source text | |
1877 | to analyze multi-byte codes (within macros ONE_MORE_BYTE and | |
1878 | TWO_MORE_BYTES). In that case, SRC is reset to SRC_BASE | |
1879 | before exiting. */ | |
1880 | unsigned char *src_base = src; | |
1881 | unsigned char c1 = *src++, c2, c3, c4; | |
1882 | ||
1883 | if (coding->composing) | |
1884 | { | |
1885 | if (c1 == 0xA0) | |
1886 | { | |
1887 | ONE_MORE_BYTE (c1); | |
1888 | c1 &= 0x7F; | |
1889 | } | |
1890 | else if (c1 >= 0xA0) | |
1891 | c1 -= 0x20; | |
1892 | else | |
1893 | coding->composing = 0; | |
1894 | } | |
1895 | ||
1896 | switch (emacs_code_class[c1]) | |
1897 | { | |
1898 | case EMACS_ascii_code: | |
a5d301df KH |
1899 | ENCODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); |
1900 | break; | |
1901 | ||
4ed46869 KH |
1902 | case EMACS_control_code: |
1903 | *dst++ = c1; | |
1904 | break; | |
1905 | ||
1906 | case EMACS_carriage_return_code: | |
1907 | if (!coding->selective) | |
1908 | { | |
1909 | *dst++ = c1; | |
1910 | break; | |
1911 | } | |
1912 | /* fall down to treat '\r' as '\n' ... */ | |
1913 | ||
1914 | case EMACS_linefeed_code: | |
1915 | if (coding->eol_type == CODING_EOL_LF | |
0ef69138 | 1916 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
1917 | *dst++ = '\n'; |
1918 | else if (coding->eol_type == CODING_EOL_CRLF) | |
1919 | *dst++ = '\r', *dst++ = '\n'; | |
1920 | else | |
1921 | *dst++ = '\r'; | |
1922 | break; | |
1923 | ||
1924 | case EMACS_leading_code_2: | |
1925 | ONE_MORE_BYTE (c2); | |
a5d301df | 1926 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, /* dummy */ c3); |
4ed46869 KH |
1927 | break; |
1928 | ||
1929 | case EMACS_leading_code_3: | |
1930 | TWO_MORE_BYTES (c2, c3); | |
a5d301df | 1931 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, c3); |
4ed46869 KH |
1932 | break; |
1933 | ||
1934 | case EMACS_leading_code_4: | |
1935 | THREE_MORE_BYTES (c2, c3, c4); | |
a5d301df | 1936 | ENCODE_SJIS_BIG5_CHARACTER (c2, c3, c4); |
4ed46869 KH |
1937 | break; |
1938 | ||
1939 | case EMACS_leading_code_composition: | |
1940 | coding->composing = 1; | |
1941 | break; | |
1942 | ||
1943 | default: /* i.e. case EMACS_invalid_code: */ | |
1944 | *dst++ = c1; | |
1945 | } | |
1946 | continue; | |
1947 | ||
1948 | label_end_of_loop: | |
1949 | coding->carryover_size = src - src_base; | |
1950 | bcopy (src_base, coding->carryover, coding->carryover_size); | |
1951 | src = src_base; | |
1952 | break; | |
1953 | } | |
1954 | ||
1955 | *consumed = src - source; | |
1956 | return dst - destination; | |
1957 | } | |
1958 | ||
1959 | \f | |
1960 | /*** 5. End-of-line handlers ***/ | |
1961 | ||
1962 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". | |
1963 | This function is called only when `coding->eol_type' is | |
1964 | CODING_EOL_CRLF or CODING_EOL_CR. */ | |
1965 | ||
1966 | decode_eol (coding, source, destination, src_bytes, dst_bytes, consumed) | |
1967 | struct coding_system *coding; | |
1968 | unsigned char *source, *destination; | |
1969 | int src_bytes, dst_bytes; | |
1970 | int *consumed; | |
1971 | { | |
1972 | unsigned char *src = source; | |
1973 | unsigned char *src_end = source + src_bytes; | |
1974 | unsigned char *dst = destination; | |
1975 | unsigned char *dst_end = destination + dst_bytes; | |
1976 | int produced; | |
1977 | ||
1978 | switch (coding->eol_type) | |
1979 | { | |
1980 | case CODING_EOL_CRLF: | |
1981 | { | |
1982 | /* Since the maximum bytes produced by each loop is 2, we | |
1983 | subtract 1 from DST_END to assure overflow checking is | |
1984 | necessary only at the head of loop. */ | |
1985 | unsigned char *adjusted_dst_end = dst_end - 1; | |
1986 | ||
1987 | while (src < src_end && dst < adjusted_dst_end) | |
1988 | { | |
1989 | unsigned char *src_base = src; | |
1990 | unsigned char c = *src++; | |
1991 | if (c == '\r') | |
1992 | { | |
1993 | ONE_MORE_BYTE (c); | |
1994 | if (c != '\n') | |
1995 | *dst++ = '\r'; | |
bfd99048 | 1996 | *dst++ = c; |
4ed46869 KH |
1997 | } |
1998 | else | |
1999 | *dst++ = c; | |
2000 | continue; | |
2001 | ||
2002 | label_end_of_loop: | |
2003 | coding->carryover_size = src - src_base; | |
2004 | bcopy (src_base, coding->carryover, coding->carryover_size); | |
2005 | src = src_base; | |
2006 | break; | |
2007 | } | |
2008 | *consumed = src - source; | |
2009 | produced = dst - destination; | |
2010 | break; | |
2011 | } | |
2012 | ||
2013 | case CODING_EOL_CR: | |
2014 | produced = (src_bytes > dst_bytes) ? dst_bytes : src_bytes; | |
2015 | bcopy (source, destination, produced); | |
2016 | dst_end = destination + produced; | |
2017 | while (dst < dst_end) | |
2018 | if (*dst++ == '\r') dst[-1] = '\n'; | |
2019 | *consumed = produced; | |
2020 | break; | |
2021 | ||
2022 | default: /* i.e. case: CODING_EOL_LF */ | |
2023 | produced = (src_bytes > dst_bytes) ? dst_bytes : src_bytes; | |
2024 | bcopy (source, destination, produced); | |
2025 | *consumed = produced; | |
2026 | break; | |
2027 | } | |
2028 | ||
2029 | return produced; | |
2030 | } | |
2031 | ||
2032 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". Encode | |
2033 | format of end-of-line according to `coding->eol_type'. If | |
2034 | `coding->selective' is 1, code '\r' in source text also means | |
2035 | end-of-line. */ | |
2036 | ||
2037 | encode_eol (coding, source, destination, src_bytes, dst_bytes, consumed) | |
2038 | struct coding_system *coding; | |
2039 | unsigned char *source, *destination; | |
2040 | int src_bytes, dst_bytes; | |
2041 | int *consumed; | |
2042 | { | |
2043 | unsigned char *src = source; | |
2044 | unsigned char *dst = destination; | |
2045 | int produced; | |
2046 | ||
2047 | if (src_bytes <= 0) | |
2048 | return 0; | |
2049 | ||
2050 | switch (coding->eol_type) | |
2051 | { | |
2052 | case CODING_EOL_LF: | |
0ef69138 | 2053 | case CODING_EOL_UNDECIDED: |
4ed46869 KH |
2054 | produced = (src_bytes > dst_bytes) ? dst_bytes : src_bytes; |
2055 | bcopy (source, destination, produced); | |
2056 | if (coding->selective) | |
2057 | { | |
2058 | int i = produced; | |
2059 | while (i--) | |
2060 | if (*dst++ == '\r') dst[-1] = '\n'; | |
2061 | } | |
2062 | *consumed = produced; | |
2063 | ||
2064 | case CODING_EOL_CRLF: | |
2065 | { | |
2066 | unsigned char c; | |
2067 | unsigned char *src_end = source + src_bytes; | |
2068 | unsigned char *dst_end = destination + dst_bytes; | |
2069 | /* Since the maximum bytes produced by each loop is 2, we | |
2070 | subtract 1 from DST_END to assure overflow checking is | |
2071 | necessary only at the head of loop. */ | |
2072 | unsigned char *adjusted_dst_end = dst_end - 1; | |
2073 | ||
2074 | while (src < src_end && dst < adjusted_dst_end) | |
2075 | { | |
2076 | c = *src++; | |
2077 | if (c == '\n' || (c == '\r' && coding->selective)) | |
2078 | *dst++ = '\r', *dst++ = '\n'; | |
2079 | else | |
2080 | *dst++ = c; | |
2081 | } | |
2082 | produced = dst - destination; | |
2083 | *consumed = src - source; | |
2084 | break; | |
2085 | } | |
2086 | ||
2087 | default: /* i.e. case CODING_EOL_CR: */ | |
2088 | produced = (src_bytes > dst_bytes) ? dst_bytes : src_bytes; | |
2089 | bcopy (source, destination, produced); | |
2090 | { | |
2091 | int i = produced; | |
2092 | while (i--) | |
2093 | if (*dst++ == '\n') dst[-1] = '\r'; | |
2094 | } | |
2095 | *consumed = produced; | |
2096 | } | |
2097 | ||
2098 | return produced; | |
2099 | } | |
2100 | ||
2101 | \f | |
2102 | /*** 6. C library functions ***/ | |
2103 | ||
2104 | /* In Emacs Lisp, coding system is represented by a Lisp symbol which | |
2105 | has a property `coding-system'. The value of this property is a | |
2106 | vector of length 5 (called as coding-vector). Among elements of | |
2107 | this vector, the first (element[0]) and the fifth (element[4]) | |
2108 | carry important information for decoding/encoding. Before | |
2109 | decoding/encoding, this information should be set in fields of a | |
2110 | structure of type `coding_system'. | |
2111 | ||
2112 | A value of property `coding-system' can be a symbol of another | |
2113 | subsidiary coding-system. In that case, Emacs gets coding-vector | |
2114 | from that symbol. | |
2115 | ||
2116 | `element[0]' contains information to be set in `coding->type'. The | |
2117 | value and its meaning is as follows: | |
2118 | ||
0ef69138 KH |
2119 | 0 -- coding_type_emacs_mule |
2120 | 1 -- coding_type_sjis | |
2121 | 2 -- coding_type_iso2022 | |
2122 | 3 -- coding_type_big5 | |
2123 | 4 -- coding_type_ccl encoder/decoder written in CCL | |
2124 | nil -- coding_type_no_conversion | |
2125 | t -- coding_type_undecided (automatic conversion on decoding, | |
2126 | no-conversion on encoding) | |
4ed46869 KH |
2127 | |
2128 | `element[4]' contains information to be set in `coding->flags' and | |
2129 | `coding->spec'. The meaning varies by `coding->type'. | |
2130 | ||
2131 | If `coding->type' is `coding_type_iso2022', element[4] is a vector | |
2132 | of length 32 (of which the first 13 sub-elements are used now). | |
2133 | Meanings of these sub-elements are: | |
2134 | ||
2135 | sub-element[N] where N is 0 through 3: to be set in `coding->spec.iso2022' | |
2136 | If the value is an integer of valid charset, the charset is | |
2137 | assumed to be designated to graphic register N initially. | |
2138 | ||
2139 | If the value is minus, it is a minus value of charset which | |
2140 | reserves graphic register N, which means that the charset is | |
2141 | not designated initially but should be designated to graphic | |
2142 | register N just before encoding a character in that charset. | |
2143 | ||
2144 | If the value is nil, graphic register N is never used on | |
2145 | encoding. | |
2146 | ||
2147 | sub-element[N] where N is 4 through 11: to be set in `coding->flags' | |
2148 | Each value takes t or nil. See the section ISO2022 of | |
2149 | `coding.h' for more information. | |
2150 | ||
2151 | If `coding->type' is `coding_type_big5', element[4] is t to denote | |
2152 | BIG5-ETen or nil to denote BIG5-HKU. | |
2153 | ||
2154 | If `coding->type' takes the other value, element[4] is ignored. | |
2155 | ||
2156 | Emacs Lisp's coding system also carries information about format of | |
2157 | end-of-line in a value of property `eol-type'. If the value is | |
2158 | integer, 0 means CODING_EOL_LF, 1 means CODING_EOL_CRLF, and 2 | |
2159 | means CODING_EOL_CR. If it is not integer, it should be a vector | |
2160 | of subsidiary coding systems of which property `eol-type' has one | |
2161 | of above values. | |
2162 | ||
2163 | */ | |
2164 | ||
2165 | /* Extract information for decoding/encoding from CODING_SYSTEM_SYMBOL | |
2166 | and set it in CODING. If CODING_SYSTEM_SYMBOL is invalid, CODING | |
2167 | is setup so that no conversion is necessary and return -1, else | |
2168 | return 0. */ | |
2169 | ||
2170 | int | |
e0e989f6 KH |
2171 | setup_coding_system (coding_system, coding) |
2172 | Lisp_Object coding_system; | |
4ed46869 KH |
2173 | struct coding_system *coding; |
2174 | { | |
4ed46869 KH |
2175 | Lisp_Object type, eol_type; |
2176 | ||
f4dee582 | 2177 | /* At first, set several fields to default values. */ |
4ed46869 KH |
2178 | coding->require_flushing = 0; |
2179 | coding->last_block = 0; | |
2180 | coding->selective = 0; | |
2181 | coding->composing = 0; | |
2182 | coding->direction = 0; | |
2183 | coding->carryover_size = 0; | |
4ed46869 | 2184 | coding->post_read_conversion = coding->pre_write_conversion = Qnil; |
a5d301df KH |
2185 | coding->character_unification_table_for_decode = Qnil; |
2186 | coding->character_unification_table_for_encode = Qnil; | |
4ed46869 | 2187 | |
e0e989f6 KH |
2188 | Vlast_coding_system_used = coding->symbol = coding_system; |
2189 | eol_type = Qnil; | |
2190 | /* Get value of property `coding-system' until we get a vector. | |
2191 | While doing that, also get values of properties | |
a5d301df KH |
2192 | `post-read-conversion', `pre-write-conversion', |
2193 | `character-unification-table-for-decode', | |
2194 | `character-unification-table-for-encode' and `eol-type'. */ | |
e0e989f6 | 2195 | while (!NILP (coding_system) && SYMBOLP (coding_system)) |
4ed46869 | 2196 | { |
4ed46869 | 2197 | if (NILP (coding->post_read_conversion)) |
e0e989f6 | 2198 | coding->post_read_conversion = Fget (coding_system, |
4ed46869 | 2199 | Qpost_read_conversion); |
e0e989f6 KH |
2200 | if (NILP (coding->pre_write_conversion)) |
2201 | coding->pre_write_conversion = Fget (coding_system, | |
4ed46869 | 2202 | Qpre_write_conversion); |
9ce27fde | 2203 | if (!inhibit_eol_conversion && NILP (eol_type)) |
e0e989f6 | 2204 | eol_type = Fget (coding_system, Qeol_type); |
a5d301df KH |
2205 | |
2206 | if (NILP (coding->character_unification_table_for_decode)) | |
2207 | coding->character_unification_table_for_decode | |
2208 | = Fget (coding_system, Qcharacter_unification_table_for_decode); | |
2209 | ||
2210 | if (NILP (coding->character_unification_table_for_encode)) | |
2211 | coding->character_unification_table_for_encode | |
2212 | = Fget (coding_system, Qcharacter_unification_table_for_encode); | |
2213 | ||
e0e989f6 | 2214 | coding_system = Fget (coding_system, Qcoding_system); |
4ed46869 | 2215 | } |
a5d301df KH |
2216 | |
2217 | while (!NILP (coding->character_unification_table_for_decode) | |
2218 | && SYMBOLP (coding->character_unification_table_for_decode)) | |
2219 | coding->character_unification_table_for_decode | |
2220 | = Fget (coding->character_unification_table_for_decode, | |
2221 | Qcharacter_unification_table_for_decode); | |
2222 | if (!NILP (coding->character_unification_table_for_decode) | |
2223 | && !CHAR_TABLE_P (coding->character_unification_table_for_decode)) | |
2224 | coding->character_unification_table_for_decode = Qnil; | |
2225 | ||
2226 | while (!NILP (coding->character_unification_table_for_encode) | |
2227 | && SYMBOLP (coding->character_unification_table_for_encode)) | |
2228 | coding->character_unification_table_for_encode | |
2229 | = Fget (coding->character_unification_table_for_encode, | |
2230 | Qcharacter_unification_table_for_encode); | |
2231 | if (!NILP (coding->character_unification_table_for_encode) | |
2232 | && !CHAR_TABLE_P (coding->character_unification_table_for_encode)) | |
2233 | coding->character_unification_table_for_encode = Qnil; | |
2234 | ||
e0e989f6 KH |
2235 | if (!VECTORP (coding_system) |
2236 | || XVECTOR (coding_system)->size != 5) | |
4ed46869 KH |
2237 | goto label_invalid_coding_system; |
2238 | ||
4ed46869 | 2239 | if (VECTORP (eol_type)) |
0ef69138 | 2240 | coding->eol_type = CODING_EOL_UNDECIDED; |
4ed46869 KH |
2241 | else if (XFASTINT (eol_type) == 1) |
2242 | coding->eol_type = CODING_EOL_CRLF; | |
2243 | else if (XFASTINT (eol_type) == 2) | |
2244 | coding->eol_type = CODING_EOL_CR; | |
2245 | else | |
2246 | coding->eol_type = CODING_EOL_LF; | |
2247 | ||
e0e989f6 | 2248 | type = XVECTOR (coding_system)->contents[0]; |
4ed46869 KH |
2249 | switch (XFASTINT (type)) |
2250 | { | |
2251 | case 0: | |
0ef69138 | 2252 | coding->type = coding_type_emacs_mule; |
4ed46869 KH |
2253 | break; |
2254 | ||
2255 | case 1: | |
2256 | coding->type = coding_type_sjis; | |
2257 | break; | |
2258 | ||
2259 | case 2: | |
2260 | coding->type = coding_type_iso2022; | |
2261 | { | |
e0e989f6 | 2262 | Lisp_Object val = XVECTOR (coding_system)->contents[4]; |
4ed46869 KH |
2263 | Lisp_Object *flags; |
2264 | int i, charset, default_reg_bits = 0; | |
2265 | ||
2266 | if (!VECTORP (val) || XVECTOR (val)->size != 32) | |
2267 | goto label_invalid_coding_system; | |
2268 | ||
2269 | flags = XVECTOR (val)->contents; | |
2270 | coding->flags | |
2271 | = ((NILP (flags[4]) ? 0 : CODING_FLAG_ISO_SHORT_FORM) | |
2272 | | (NILP (flags[5]) ? 0 : CODING_FLAG_ISO_RESET_AT_EOL) | |
2273 | | (NILP (flags[6]) ? 0 : CODING_FLAG_ISO_RESET_AT_CNTL) | |
2274 | | (NILP (flags[7]) ? 0 : CODING_FLAG_ISO_SEVEN_BITS) | |
2275 | | (NILP (flags[8]) ? 0 : CODING_FLAG_ISO_LOCKING_SHIFT) | |
2276 | | (NILP (flags[9]) ? 0 : CODING_FLAG_ISO_SINGLE_SHIFT) | |
2277 | | (NILP (flags[10]) ? 0 : CODING_FLAG_ISO_USE_ROMAN) | |
2278 | | (NILP (flags[11]) ? 0 : CODING_FLAG_ISO_USE_OLDJIS) | |
e0e989f6 KH |
2279 | | (NILP (flags[12]) ? 0 : CODING_FLAG_ISO_NO_DIRECTION) |
2280 | | (NILP (flags[13]) ? 0 : CODING_FLAG_ISO_INIT_AT_BOL) | |
2281 | | (NILP (flags[14]) ? 0 : CODING_FLAG_ISO_DESIGNATE_AT_BOL)); | |
4ed46869 KH |
2282 | |
2283 | /* Invoke graphic register 0 to plane 0. */ | |
2284 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; | |
2285 | /* Invoke graphic register 1 to plane 1 if we can use full 8-bit. */ | |
2286 | CODING_SPEC_ISO_INVOCATION (coding, 1) | |
2287 | = (coding->flags & CODING_FLAG_ISO_SEVEN_BITS ? -1 : 1); | |
2288 | /* Not single shifting at first. */ | |
2289 | CODING_SPEC_ISO_SINGLE_SHIFTING(coding) = 0; | |
e0e989f6 KH |
2290 | /* Beginning of buffer should also be regarded as bol. */ |
2291 | CODING_SPEC_ISO_BOL(coding) = 1; | |
4ed46869 KH |
2292 | |
2293 | /* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations. | |
2294 | FLAGS[REG] can be one of below: | |
2295 | integer CHARSET: CHARSET occupies register I, | |
2296 | t: designate nothing to REG initially, but can be used | |
2297 | by any charsets, | |
2298 | list of integer, nil, or t: designate the first | |
2299 | element (if integer) to REG initially, the remaining | |
2300 | elements (if integer) is designated to REG on request, | |
2301 | if an element is t, REG can be used by any charset, | |
2302 | nil: REG is never used. */ | |
467e7675 | 2303 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
1ba9e4ab KH |
2304 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
2305 | = CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION; | |
4ed46869 KH |
2306 | for (i = 0; i < 4; i++) |
2307 | { | |
2308 | if (INTEGERP (flags[i]) | |
e0e989f6 KH |
2309 | && (charset = XINT (flags[i]), CHARSET_VALID_P (charset)) |
2310 | || (charset = get_charset_id (flags[i])) >= 0) | |
4ed46869 KH |
2311 | { |
2312 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; | |
2313 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) = i; | |
2314 | } | |
2315 | else if (EQ (flags[i], Qt)) | |
2316 | { | |
2317 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
2318 | default_reg_bits |= 1 << i; | |
2319 | } | |
2320 | else if (CONSP (flags[i])) | |
2321 | { | |
2322 | Lisp_Object tail = flags[i]; | |
2323 | ||
2324 | if (INTEGERP (XCONS (tail)->car) | |
2325 | && (charset = XINT (XCONS (tail)->car), | |
e0e989f6 KH |
2326 | CHARSET_VALID_P (charset)) |
2327 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) | |
4ed46869 KH |
2328 | { |
2329 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; | |
2330 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) =i; | |
2331 | } | |
2332 | else | |
2333 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
2334 | tail = XCONS (tail)->cdr; | |
2335 | while (CONSP (tail)) | |
2336 | { | |
2337 | if (INTEGERP (XCONS (tail)->car) | |
2338 | && (charset = XINT (XCONS (tail)->car), | |
e0e989f6 KH |
2339 | CHARSET_VALID_P (charset)) |
2340 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) | |
4ed46869 KH |
2341 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
2342 | = i; | |
2343 | else if (EQ (XCONS (tail)->car, Qt)) | |
2344 | default_reg_bits |= 1 << i; | |
2345 | tail = XCONS (tail)->cdr; | |
2346 | } | |
2347 | } | |
2348 | else | |
2349 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; | |
2350 | ||
2351 | CODING_SPEC_ISO_DESIGNATION (coding, i) | |
2352 | = CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i); | |
2353 | } | |
2354 | ||
2355 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) | |
2356 | { | |
2357 | /* REG 1 can be used only by locking shift in 7-bit env. */ | |
2358 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) | |
2359 | default_reg_bits &= ~2; | |
2360 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) | |
2361 | /* Without any shifting, only REG 0 and 1 can be used. */ | |
2362 | default_reg_bits &= 3; | |
2363 | } | |
2364 | ||
467e7675 | 2365 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
4ed46869 | 2366 | if (CHARSET_VALID_P (charset) |
1ba9e4ab KH |
2367 | && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
2368 | == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)) | |
4ed46869 KH |
2369 | { |
2370 | /* We have not yet decided where to designate CHARSET. */ | |
2371 | int reg_bits = default_reg_bits; | |
2372 | ||
2373 | if (CHARSET_CHARS (charset) == 96) | |
2374 | /* A charset of CHARS96 can't be designated to REG 0. */ | |
2375 | reg_bits &= ~1; | |
2376 | ||
2377 | if (reg_bits) | |
2378 | /* There exist some default graphic register. */ | |
2379 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) | |
2380 | = (reg_bits & 1 | |
2381 | ? 0 : (reg_bits & 2 ? 1 : (reg_bits & 4 ? 2 : 3))); | |
2382 | else | |
2383 | /* We anyway have to designate CHARSET to somewhere. */ | |
2384 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) | |
2385 | = (CHARSET_CHARS (charset) == 94 | |
2386 | ? 0 | |
2387 | : ((coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT | |
2388 | || ! coding->flags & CODING_FLAG_ISO_SEVEN_BITS) | |
2389 | ? 1 | |
2390 | : (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT | |
2391 | ? 2 : 0))); | |
2392 | } | |
2393 | } | |
2394 | coding->require_flushing = 1; | |
2395 | break; | |
2396 | ||
2397 | case 3: | |
2398 | coding->type = coding_type_big5; | |
2399 | coding->flags | |
e0e989f6 | 2400 | = (NILP (XVECTOR (coding_system)->contents[4]) |
4ed46869 KH |
2401 | ? CODING_FLAG_BIG5_HKU |
2402 | : CODING_FLAG_BIG5_ETEN); | |
2403 | break; | |
2404 | ||
2405 | case 4: | |
2406 | coding->type = coding_type_ccl; | |
2407 | { | |
e0e989f6 | 2408 | Lisp_Object val = XVECTOR (coding_system)->contents[4]; |
4ed46869 KH |
2409 | if (CONSP (val) |
2410 | && VECTORP (XCONS (val)->car) | |
2411 | && VECTORP (XCONS (val)->cdr)) | |
2412 | { | |
2413 | setup_ccl_program (&(coding->spec.ccl.decoder), XCONS (val)->car); | |
2414 | setup_ccl_program (&(coding->spec.ccl.encoder), XCONS (val)->cdr); | |
2415 | } | |
2416 | else | |
2417 | goto label_invalid_coding_system; | |
2418 | } | |
2419 | coding->require_flushing = 1; | |
2420 | break; | |
2421 | ||
2422 | default: | |
2423 | if (EQ (type, Qt)) | |
0ef69138 | 2424 | coding->type = coding_type_undecided; |
4ed46869 KH |
2425 | else |
2426 | coding->type = coding_type_no_conversion; | |
2427 | break; | |
2428 | } | |
2429 | return 0; | |
2430 | ||
2431 | label_invalid_coding_system: | |
2432 | coding->type = coding_type_no_conversion; | |
dec137e5 | 2433 | coding->eol_type = CODING_EOL_LF; |
e0e989f6 KH |
2434 | coding->symbol = coding->pre_write_conversion = coding->post_read_conversion |
2435 | = Qnil; | |
4ed46869 KH |
2436 | return -1; |
2437 | } | |
2438 | ||
2439 | /* Emacs has a mechanism to automatically detect a coding system if it | |
2440 | is one of Emacs' internal format, ISO2022, SJIS, and BIG5. But, | |
2441 | it's impossible to distinguish some coding systems accurately | |
2442 | because they use the same range of codes. So, at first, coding | |
2443 | systems are categorized into 7, those are: | |
2444 | ||
0ef69138 | 2445 | o coding-category-emacs-mule |
4ed46869 KH |
2446 | |
2447 | The category for a coding system which has the same code range | |
2448 | as Emacs' internal format. Assigned the coding-system (Lisp | |
0ef69138 | 2449 | symbol) `emacs-mule' by default. |
4ed46869 KH |
2450 | |
2451 | o coding-category-sjis | |
2452 | ||
2453 | The category for a coding system which has the same code range | |
2454 | as SJIS. Assigned the coding-system (Lisp | |
7717c392 | 2455 | symbol) `japanese-shift-jis' by default. |
4ed46869 KH |
2456 | |
2457 | o coding-category-iso-7 | |
2458 | ||
2459 | The category for a coding system which has the same code range | |
7717c392 KH |
2460 | as ISO2022 of 7-bit environment. This doesn't use any locking |
2461 | shift and single shift functions. Assigned the coding-system | |
2462 | (Lisp symbol) `iso-2022-7bit' by default. | |
4ed46869 KH |
2463 | |
2464 | o coding-category-iso-8-1 | |
2465 | ||
2466 | The category for a coding system which has the same code range | |
2467 | as ISO2022 of 8-bit environment and graphic plane 1 used only | |
7717c392 KH |
2468 | for DIMENSION1 charset. This doesn't use any locking shift |
2469 | and single shift functions. Assigned the coding-system (Lisp | |
2470 | symbol) `iso-latin-1' by default. | |
4ed46869 KH |
2471 | |
2472 | o coding-category-iso-8-2 | |
2473 | ||
2474 | The category for a coding system which has the same code range | |
2475 | as ISO2022 of 8-bit environment and graphic plane 1 used only | |
7717c392 KH |
2476 | for DIMENSION2 charset. This doesn't use any locking shift |
2477 | and single shift functions. Assigned the coding-system (Lisp | |
2478 | symbol) `japanese-iso-8bit' by default. | |
4ed46869 | 2479 | |
7717c392 | 2480 | o coding-category-iso-7-else |
4ed46869 KH |
2481 | |
2482 | The category for a coding system which has the same code range | |
7717c392 KH |
2483 | as ISO2022 of 7-bit environemnt but uses locking shift or |
2484 | single shift functions. Assigned the coding-system (Lisp | |
2485 | symbol) `iso-2022-7bit-lock' by default. | |
2486 | ||
2487 | o coding-category-iso-8-else | |
2488 | ||
2489 | The category for a coding system which has the same code range | |
2490 | as ISO2022 of 8-bit environemnt but uses locking shift or | |
2491 | single shift functions. Assigned the coding-system (Lisp | |
2492 | symbol) `iso-2022-8bit-ss2' by default. | |
4ed46869 KH |
2493 | |
2494 | o coding-category-big5 | |
2495 | ||
2496 | The category for a coding system which has the same code range | |
2497 | as BIG5. Assigned the coding-system (Lisp symbol) | |
e0e989f6 | 2498 | `cn-big5' by default. |
4ed46869 KH |
2499 | |
2500 | o coding-category-binary | |
2501 | ||
2502 | The category for a coding system not categorized in any of the | |
2503 | above. Assigned the coding-system (Lisp symbol) | |
e0e989f6 | 2504 | `no-conversion' by default. |
4ed46869 KH |
2505 | |
2506 | Each of them is a Lisp symbol and the value is an actual | |
2507 | `coding-system's (this is also a Lisp symbol) assigned by a user. | |
2508 | What Emacs does actually is to detect a category of coding system. | |
2509 | Then, it uses a `coding-system' assigned to it. If Emacs can't | |
2510 | decide only one possible category, it selects a category of the | |
2511 | highest priority. Priorities of categories are also specified by a | |
2512 | user in a Lisp variable `coding-category-list'. | |
2513 | ||
2514 | */ | |
2515 | ||
2516 | /* Detect how a text of length SRC_BYTES pointed by SRC is encoded. | |
2517 | If it detects possible coding systems, return an integer in which | |
2518 | appropriate flag bits are set. Flag bits are defined by macros | |
2519 | CODING_CATEGORY_MASK_XXX in `coding.h'. */ | |
2520 | ||
2521 | int | |
2522 | detect_coding_mask (src, src_bytes) | |
2523 | unsigned char *src; | |
2524 | int src_bytes; | |
2525 | { | |
2526 | register unsigned char c; | |
2527 | unsigned char *src_end = src + src_bytes; | |
2528 | int mask; | |
2529 | ||
2530 | /* At first, skip all ASCII characters and control characters except | |
2531 | for three ISO2022 specific control characters. */ | |
bcf26d6a | 2532 | label_loop_detect_coding: |
4ed46869 KH |
2533 | while (src < src_end) |
2534 | { | |
2535 | c = *src; | |
2536 | if (c >= 0x80 | |
2537 | || (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO)) | |
2538 | break; | |
2539 | src++; | |
2540 | } | |
2541 | ||
2542 | if (src >= src_end) | |
2543 | /* We found nothing other than ASCII. There's nothing to do. */ | |
2544 | return CODING_CATEGORY_MASK_ANY; | |
2545 | ||
2546 | /* The text seems to be encoded in some multilingual coding system. | |
2547 | Now, try to find in which coding system the text is encoded. */ | |
2548 | if (c < 0x80) | |
bcf26d6a KH |
2549 | { |
2550 | /* i.e. (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) */ | |
2551 | /* C is an ISO2022 specific control code of C0. */ | |
2552 | mask = detect_coding_iso2022 (src, src_end); | |
2553 | src++; | |
2554 | if (mask == CODING_CATEGORY_MASK_ANY) | |
2555 | /* No valid ISO2022 code follows C. Try again. */ | |
2556 | goto label_loop_detect_coding; | |
2557 | } | |
4ed46869 KH |
2558 | else if (c == ISO_CODE_SS2 || c == ISO_CODE_SS3 || c == ISO_CODE_CSI) |
2559 | /* C is an ISO2022 specific control code of C1, | |
2560 | or the first byte of SJIS's 2-byte character code, | |
2561 | or a leading code of Emacs. */ | |
2562 | mask = (detect_coding_iso2022 (src, src_end) | |
2563 | | detect_coding_sjis (src, src_end) | |
0ef69138 | 2564 | | detect_coding_emacs_mule (src, src_end)); |
4ed46869 KH |
2565 | |
2566 | else if (c < 0xA0) | |
2567 | /* C is the first byte of SJIS character code, | |
2568 | or a leading-code of Emacs. */ | |
2569 | mask = (detect_coding_sjis (src, src_end) | |
0ef69138 | 2570 | | detect_coding_emacs_mule (src, src_end)); |
4ed46869 KH |
2571 | |
2572 | else | |
2573 | /* C is a character of ISO2022 in graphic plane right, | |
2574 | or a SJIS's 1-byte character code (i.e. JISX0201), | |
2575 | or the first byte of BIG5's 2-byte code. */ | |
2576 | mask = (detect_coding_iso2022 (src, src_end) | |
2577 | | detect_coding_sjis (src, src_end) | |
2578 | | detect_coding_big5 (src, src_end)); | |
2579 | ||
2580 | return mask; | |
2581 | } | |
2582 | ||
2583 | /* Detect how a text of length SRC_BYTES pointed by SRC is encoded. | |
2584 | The information of the detected coding system is set in CODING. */ | |
2585 | ||
2586 | void | |
2587 | detect_coding (coding, src, src_bytes) | |
2588 | struct coding_system *coding; | |
2589 | unsigned char *src; | |
2590 | int src_bytes; | |
2591 | { | |
2592 | int mask = detect_coding_mask (src, src_bytes); | |
2593 | int idx; | |
2594 | ||
2595 | if (mask == CODING_CATEGORY_MASK_ANY) | |
2596 | /* We found nothing other than ASCII. There's nothing to do. */ | |
2597 | return; | |
2598 | ||
2599 | if (!mask) | |
2600 | /* The source text seems to be encoded in unknown coding system. | |
2601 | Emacs regards the category of such a kind of coding system as | |
2602 | `coding-category-binary'. We assume that a user has assigned | |
2603 | an appropriate coding system for a `coding-category-binary'. */ | |
2604 | idx = CODING_CATEGORY_IDX_BINARY; | |
2605 | else | |
2606 | { | |
2607 | /* We found some plausible coding systems. Let's use a coding | |
2608 | system of the highest priority. */ | |
2609 | Lisp_Object val = Vcoding_category_list; | |
2610 | ||
2611 | if (CONSP (val)) | |
2612 | while (!NILP (val)) | |
2613 | { | |
2614 | idx = XFASTINT (Fget (XCONS (val)->car, Qcoding_category_index)); | |
2615 | if ((idx < CODING_CATEGORY_IDX_MAX) && (mask & (1 << idx))) | |
2616 | break; | |
2617 | val = XCONS (val)->cdr; | |
2618 | } | |
2619 | else | |
2620 | val = Qnil; | |
2621 | ||
2622 | if (NILP (val)) | |
2623 | { | |
2624 | /* For unknown reason, `Vcoding_category_list' contains none | |
2625 | of found categories. Let's use any of them. */ | |
2626 | for (idx = 0; idx < CODING_CATEGORY_IDX_MAX; idx++) | |
2627 | if (mask & (1 << idx)) | |
2628 | break; | |
2629 | } | |
2630 | } | |
2631 | setup_coding_system (XSYMBOL (coding_category_table[idx])->value, coding); | |
2632 | } | |
2633 | ||
2634 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC | |
2635 | is encoded. Return one of CODING_EOL_LF, CODING_EOL_CRLF, | |
0ef69138 | 2636 | CODING_EOL_CR, and CODING_EOL_UNDECIDED. */ |
4ed46869 KH |
2637 | |
2638 | int | |
2639 | detect_eol_type (src, src_bytes) | |
2640 | unsigned char *src; | |
2641 | int src_bytes; | |
2642 | { | |
2643 | unsigned char *src_end = src + src_bytes; | |
2644 | unsigned char c; | |
2645 | ||
2646 | while (src < src_end) | |
2647 | { | |
2648 | c = *src++; | |
2649 | if (c == '\n') | |
2650 | return CODING_EOL_LF; | |
2651 | else if (c == '\r') | |
2652 | { | |
2653 | if (src < src_end && *src == '\n') | |
2654 | return CODING_EOL_CRLF; | |
2655 | else | |
2656 | return CODING_EOL_CR; | |
2657 | } | |
2658 | } | |
0ef69138 | 2659 | return CODING_EOL_UNDECIDED; |
4ed46869 KH |
2660 | } |
2661 | ||
2662 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC | |
2663 | is encoded. If it detects an appropriate format of end-of-line, it | |
2664 | sets the information in *CODING. */ | |
2665 | ||
2666 | void | |
2667 | detect_eol (coding, src, src_bytes) | |
2668 | struct coding_system *coding; | |
2669 | unsigned char *src; | |
2670 | int src_bytes; | |
2671 | { | |
2672 | Lisp_Object val; | |
2673 | int eol_type = detect_eol_type (src, src_bytes); | |
2674 | ||
0ef69138 | 2675 | if (eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
2676 | /* We found no end-of-line in the source text. */ |
2677 | return; | |
2678 | ||
2679 | val = Fget (coding->symbol, Qeol_type); | |
2680 | if (VECTORP (val) && XVECTOR (val)->size == 3) | |
2681 | setup_coding_system (XVECTOR (val)->contents[eol_type], coding); | |
2682 | } | |
2683 | ||
2684 | /* See "GENERAL NOTES about `decode_coding_XXX ()' functions". Before | |
2685 | decoding, it may detect coding system and format of end-of-line if | |
2686 | those are not yet decided. */ | |
2687 | ||
2688 | int | |
2689 | decode_coding (coding, source, destination, src_bytes, dst_bytes, consumed) | |
2690 | struct coding_system *coding; | |
2691 | unsigned char *source, *destination; | |
2692 | int src_bytes, dst_bytes; | |
2693 | int *consumed; | |
2694 | { | |
2695 | int produced; | |
2696 | ||
2697 | if (src_bytes <= 0) | |
2698 | { | |
2699 | *consumed = 0; | |
2700 | return 0; | |
2701 | } | |
2702 | ||
0ef69138 | 2703 | if (coding->type == coding_type_undecided) |
4ed46869 KH |
2704 | detect_coding (coding, source, src_bytes); |
2705 | ||
0ef69138 | 2706 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
2707 | detect_eol (coding, source, src_bytes); |
2708 | ||
2709 | coding->carryover_size = 0; | |
2710 | switch (coding->type) | |
2711 | { | |
2712 | case coding_type_no_conversion: | |
2713 | label_no_conversion: | |
2714 | produced = (src_bytes > dst_bytes) ? dst_bytes : src_bytes; | |
2715 | bcopy (source, destination, produced); | |
2716 | *consumed = produced; | |
2717 | break; | |
2718 | ||
0ef69138 KH |
2719 | case coding_type_emacs_mule: |
2720 | case coding_type_undecided: | |
4ed46869 | 2721 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 2722 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
2723 | goto label_no_conversion; |
2724 | produced = decode_eol (coding, source, destination, | |
2725 | src_bytes, dst_bytes, consumed); | |
2726 | break; | |
2727 | ||
2728 | case coding_type_sjis: | |
2729 | produced = decode_coding_sjis_big5 (coding, source, destination, | |
2730 | src_bytes, dst_bytes, consumed, | |
2731 | 1); | |
2732 | break; | |
2733 | ||
2734 | case coding_type_iso2022: | |
2735 | produced = decode_coding_iso2022 (coding, source, destination, | |
2736 | src_bytes, dst_bytes, consumed); | |
2737 | break; | |
2738 | ||
2739 | case coding_type_big5: | |
2740 | produced = decode_coding_sjis_big5 (coding, source, destination, | |
2741 | src_bytes, dst_bytes, consumed, | |
2742 | 0); | |
2743 | break; | |
2744 | ||
2745 | case coding_type_ccl: | |
2746 | produced = ccl_driver (&coding->spec.ccl.decoder, source, destination, | |
2747 | src_bytes, dst_bytes, consumed); | |
2748 | break; | |
2749 | } | |
2750 | ||
2751 | return produced; | |
2752 | } | |
2753 | ||
2754 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". */ | |
2755 | ||
2756 | int | |
2757 | encode_coding (coding, source, destination, src_bytes, dst_bytes, consumed) | |
2758 | struct coding_system *coding; | |
2759 | unsigned char *source, *destination; | |
2760 | int src_bytes, dst_bytes; | |
2761 | int *consumed; | |
2762 | { | |
2763 | int produced; | |
2764 | ||
2765 | coding->carryover_size = 0; | |
2766 | switch (coding->type) | |
2767 | { | |
2768 | case coding_type_no_conversion: | |
2769 | label_no_conversion: | |
2770 | produced = (src_bytes > dst_bytes) ? dst_bytes : src_bytes; | |
2771 | if (produced > 0) | |
2772 | { | |
2773 | bcopy (source, destination, produced); | |
2774 | if (coding->selective) | |
2775 | { | |
2776 | unsigned char *p = destination, *pend = destination + produced; | |
2777 | while (p < pend) | |
e0e989f6 | 2778 | if (*p++ == '\015') p[-1] = '\n'; |
4ed46869 KH |
2779 | } |
2780 | } | |
2781 | *consumed = produced; | |
2782 | break; | |
2783 | ||
0ef69138 KH |
2784 | case coding_type_emacs_mule: |
2785 | case coding_type_undecided: | |
4ed46869 | 2786 | if (coding->eol_type == CODING_EOL_LF |
0ef69138 | 2787 | || coding->eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
2788 | goto label_no_conversion; |
2789 | produced = encode_eol (coding, source, destination, | |
2790 | src_bytes, dst_bytes, consumed); | |
2791 | break; | |
2792 | ||
2793 | case coding_type_sjis: | |
2794 | produced = encode_coding_sjis_big5 (coding, source, destination, | |
2795 | src_bytes, dst_bytes, consumed, | |
2796 | 1); | |
2797 | break; | |
2798 | ||
2799 | case coding_type_iso2022: | |
2800 | produced = encode_coding_iso2022 (coding, source, destination, | |
2801 | src_bytes, dst_bytes, consumed); | |
2802 | break; | |
2803 | ||
2804 | case coding_type_big5: | |
2805 | produced = encode_coding_sjis_big5 (coding, source, destination, | |
2806 | src_bytes, dst_bytes, consumed, | |
2807 | 0); | |
2808 | break; | |
2809 | ||
2810 | case coding_type_ccl: | |
2811 | produced = ccl_driver (&coding->spec.ccl.encoder, source, destination, | |
2812 | src_bytes, dst_bytes, consumed); | |
2813 | break; | |
2814 | } | |
2815 | ||
2816 | return produced; | |
2817 | } | |
2818 | ||
2819 | #define CONVERSION_BUFFER_EXTRA_ROOM 256 | |
2820 | ||
2821 | /* Return maximum size (bytes) of a buffer enough for decoding | |
2822 | SRC_BYTES of text encoded in CODING. */ | |
2823 | ||
2824 | int | |
2825 | decoding_buffer_size (coding, src_bytes) | |
2826 | struct coding_system *coding; | |
2827 | int src_bytes; | |
2828 | { | |
2829 | int magnification; | |
2830 | ||
2831 | if (coding->type == coding_type_iso2022) | |
2832 | magnification = 3; | |
2833 | else if (coding->type == coding_type_ccl) | |
2834 | magnification = coding->spec.ccl.decoder.buf_magnification; | |
2835 | else | |
2836 | magnification = 2; | |
2837 | ||
2838 | return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM); | |
2839 | } | |
2840 | ||
2841 | /* Return maximum size (bytes) of a buffer enough for encoding | |
2842 | SRC_BYTES of text to CODING. */ | |
2843 | ||
2844 | int | |
2845 | encoding_buffer_size (coding, src_bytes) | |
2846 | struct coding_system *coding; | |
2847 | int src_bytes; | |
2848 | { | |
2849 | int magnification; | |
2850 | ||
2851 | if (coding->type == coding_type_ccl) | |
2852 | magnification = coding->spec.ccl.encoder.buf_magnification; | |
2853 | else | |
2854 | magnification = 3; | |
2855 | ||
2856 | return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM); | |
2857 | } | |
2858 | ||
2859 | #ifndef MINIMUM_CONVERSION_BUFFER_SIZE | |
2860 | #define MINIMUM_CONVERSION_BUFFER_SIZE 1024 | |
2861 | #endif | |
2862 | ||
2863 | char *conversion_buffer; | |
2864 | int conversion_buffer_size; | |
2865 | ||
2866 | /* Return a pointer to a SIZE bytes of buffer to be used for encoding | |
2867 | or decoding. Sufficient memory is allocated automatically. If we | |
2868 | run out of memory, return NULL. */ | |
2869 | ||
2870 | char * | |
2871 | get_conversion_buffer (size) | |
2872 | int size; | |
2873 | { | |
2874 | if (size > conversion_buffer_size) | |
2875 | { | |
2876 | char *buf; | |
2877 | int real_size = conversion_buffer_size * 2; | |
2878 | ||
2879 | while (real_size < size) real_size *= 2; | |
2880 | buf = (char *) xmalloc (real_size); | |
2881 | xfree (conversion_buffer); | |
2882 | conversion_buffer = buf; | |
2883 | conversion_buffer_size = real_size; | |
2884 | } | |
2885 | return conversion_buffer; | |
2886 | } | |
2887 | ||
2888 | \f | |
2889 | #ifdef emacs | |
2890 | /*** 7. Emacs Lisp library functions ***/ | |
2891 | ||
02ba4723 | 2892 | DEFUN ("coding-system-spec", Fcoding_system_spec, Scoding_system_spec, |
4ed46869 | 2893 | 1, 1, 0, |
02ba4723 | 2894 | "Return coding-spec of CODING-SYSTEM.\n\ |
4ed46869 KH |
2895 | If CODING-SYSTEM is not a valid coding-system, return nil.") |
2896 | (obj) | |
2897 | Lisp_Object obj; | |
2898 | { | |
2899 | while (SYMBOLP (obj) && !NILP (obj)) | |
2900 | obj = Fget (obj, Qcoding_system); | |
2901 | return ((NILP (obj) || !VECTORP (obj) || XVECTOR (obj)->size != 5) | |
2902 | ? Qnil : obj); | |
2903 | } | |
2904 | ||
2905 | DEFUN ("coding-system-p", Fcoding_system_p, Scoding_system_p, 1, 1, 0, | |
2906 | "Return t if OBJECT is nil or a coding-system.\n\ | |
2907 | See document of make-coding-system for coding-system object.") | |
2908 | (obj) | |
2909 | Lisp_Object obj; | |
2910 | { | |
02ba4723 | 2911 | return ((NILP (obj) || !NILP (Fcoding_system_spec (obj))) ? Qt : Qnil); |
4ed46869 KH |
2912 | } |
2913 | ||
9d991de8 RS |
2914 | DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system, |
2915 | Sread_non_nil_coding_system, 1, 1, 0, | |
e0e989f6 | 2916 | "Read a coding system from the minibuffer, prompting with string PROMPT.") |
4ed46869 KH |
2917 | (prompt) |
2918 | Lisp_Object prompt; | |
2919 | { | |
e0e989f6 | 2920 | Lisp_Object val; |
9d991de8 RS |
2921 | do |
2922 | { | |
02ba4723 | 2923 | val = Fcompleting_read (prompt, Vobarray, Qcoding_system_spec, |
9d991de8 RS |
2924 | Qt, Qnil, Qnil, Qnil); |
2925 | } | |
2926 | while (XSTRING (val)->size == 0); | |
e0e989f6 | 2927 | return (Fintern (val, Qnil)); |
4ed46869 KH |
2928 | } |
2929 | ||
2930 | DEFUN ("read-coding-system", Fread_coding_system, Sread_coding_system, 1, 1, 0, | |
e0e989f6 | 2931 | "Read a coding system or nil from the minibuffer, prompting with string PROMPT.") |
4ed46869 KH |
2932 | (prompt) |
2933 | Lisp_Object prompt; | |
2934 | { | |
e0e989f6 | 2935 | Lisp_Object val = Fcompleting_read (prompt, Vobarray, Qcoding_system_p, |
9d991de8 | 2936 | Qt, Qnil, Qnil, Qnil); |
e0e989f6 | 2937 | return (XSTRING (val)->size == 0 ? Qnil : Fintern (val, Qnil)); |
4ed46869 KH |
2938 | } |
2939 | ||
2940 | DEFUN ("check-coding-system", Fcheck_coding_system, Scheck_coding_system, | |
2941 | 1, 1, 0, | |
2942 | "Check validity of CODING-SYSTEM.\n\ | |
2943 | If valid, return CODING-SYSTEM, else `coding-system-error' is signaled.\n\ | |
2944 | CODING-SYSTEM is valid if it is a symbol and has \"coding-system\" property.\n\ | |
2945 | The value of property should be a vector of length 5.") | |
2946 | (coding_system) | |
2947 | Lisp_Object coding_system; | |
2948 | { | |
2949 | CHECK_SYMBOL (coding_system, 0); | |
2950 | if (!NILP (Fcoding_system_p (coding_system))) | |
2951 | return coding_system; | |
2952 | while (1) | |
02ba4723 | 2953 | Fsignal (Qcoding_system_error, Fcons (coding_system, Qnil)); |
4ed46869 KH |
2954 | } |
2955 | ||
2956 | DEFUN ("detect-coding-region", Fdetect_coding_region, Sdetect_coding_region, | |
2957 | 2, 2, 0, | |
2958 | "Detect coding-system of the text in the region between START and END.\n\ | |
2959 | Return a list of possible coding-systems ordered by priority.\n\ | |
0ef69138 | 2960 | If only ASCII characters are found, it returns `undecided'\n\ |
4ed46869 KH |
2961 | or its subsidiary coding-system according to a detected end-of-line format.") |
2962 | (b, e) | |
2963 | Lisp_Object b, e; | |
2964 | { | |
2965 | int coding_mask, eol_type; | |
2966 | Lisp_Object val; | |
2967 | int beg, end; | |
2968 | ||
2969 | validate_region (&b, &e); | |
2970 | beg = XINT (b), end = XINT (e); | |
2971 | if (beg < GPT && end >= GPT) move_gap (end); | |
2972 | ||
2973 | coding_mask = detect_coding_mask (POS_ADDR (beg), end - beg); | |
2974 | eol_type = detect_eol_type (POS_ADDR (beg), end - beg); | |
2975 | ||
2976 | if (coding_mask == CODING_CATEGORY_MASK_ANY) | |
2977 | { | |
0ef69138 KH |
2978 | val = intern ("undecided"); |
2979 | if (eol_type != CODING_EOL_UNDECIDED) | |
4ed46869 KH |
2980 | { |
2981 | Lisp_Object val2 = Fget (val, Qeol_type); | |
2982 | if (VECTORP (val2)) | |
2983 | val = XVECTOR (val2)->contents[eol_type]; | |
2984 | } | |
2985 | } | |
2986 | else | |
2987 | { | |
2988 | Lisp_Object val2; | |
2989 | ||
2990 | /* At first, gather possible coding-systems in VAL in a reverse | |
2991 | order. */ | |
2992 | val = Qnil; | |
2993 | for (val2 = Vcoding_category_list; | |
2994 | !NILP (val2); | |
2995 | val2 = XCONS (val2)->cdr) | |
2996 | { | |
2997 | int idx | |
2998 | = XFASTINT (Fget (XCONS (val2)->car, Qcoding_category_index)); | |
2999 | if (coding_mask & (1 << idx)) | |
3000 | val = Fcons (Fsymbol_value (XCONS (val2)->car), val); | |
3001 | } | |
3002 | ||
3003 | /* Then, change the order of the list, while getting subsidiary | |
3004 | coding-systems. */ | |
3005 | val2 = val; | |
3006 | val = Qnil; | |
3007 | for (; !NILP (val2); val2 = XCONS (val2)->cdr) | |
3008 | { | |
0ef69138 | 3009 | if (eol_type == CODING_EOL_UNDECIDED) |
4ed46869 KH |
3010 | val = Fcons (XCONS (val2)->car, val); |
3011 | else | |
3012 | { | |
3013 | Lisp_Object val3 = Fget (XCONS (val2)->car, Qeol_type); | |
3014 | if (VECTORP (val3)) | |
3015 | val = Fcons (XVECTOR (val3)->contents[eol_type], val); | |
3016 | else | |
3017 | val = Fcons (XCONS (val2)->car, val); | |
3018 | } | |
3019 | } | |
3020 | } | |
3021 | ||
3022 | return val; | |
3023 | } | |
3024 | ||
3025 | /* Scan text in the region between *BEGP and *ENDP, skip characters | |
3026 | which we never have to encode to (iff ENCODEP is 1) or decode from | |
3027 | coding system CODING at the head and tail, then set BEGP and ENDP | |
3028 | to the addresses of start and end of the text we actually convert. */ | |
3029 | ||
3030 | void | |
3031 | shrink_conversion_area (begp, endp, coding, encodep) | |
3032 | unsigned char **begp, **endp; | |
3033 | struct coding_system *coding; | |
3034 | int encodep; | |
3035 | { | |
3036 | register unsigned char *beg_addr = *begp, *end_addr = *endp; | |
3037 | ||
3038 | if (coding->eol_type != CODING_EOL_LF | |
0ef69138 | 3039 | && coding->eol_type != CODING_EOL_UNDECIDED) |
4ed46869 KH |
3040 | /* Since we anyway have to convert end-of-line format, it is not |
3041 | worth skipping at most 100 bytes or so. */ | |
3042 | return; | |
3043 | ||
3044 | if (encodep) /* for encoding */ | |
3045 | { | |
3046 | switch (coding->type) | |
3047 | { | |
3048 | case coding_type_no_conversion: | |
0ef69138 KH |
3049 | case coding_type_emacs_mule: |
3050 | case coding_type_undecided: | |
4ed46869 KH |
3051 | /* We need no conversion. */ |
3052 | *begp = *endp; | |
3053 | return; | |
3054 | case coding_type_ccl: | |
3055 | /* We can't skip any data. */ | |
3056 | return; | |
e0e989f6 KH |
3057 | case coding_type_iso2022: |
3058 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL) | |
3059 | { | |
3060 | unsigned char *bol = beg_addr; | |
3061 | while (beg_addr < end_addr && *beg_addr < 0x80) | |
3062 | { | |
3063 | beg_addr++; | |
3064 | if (*(beg_addr - 1) == '\n') | |
3065 | bol = beg_addr; | |
3066 | } | |
3067 | beg_addr = bol; | |
3068 | goto label_skip_tail; | |
3069 | } | |
3070 | /* fall down ... */ | |
4ed46869 KH |
3071 | default: |
3072 | /* We can skip all ASCII characters at the head and tail. */ | |
3073 | while (beg_addr < end_addr && *beg_addr < 0x80) beg_addr++; | |
e0e989f6 | 3074 | label_skip_tail: |
4ed46869 KH |
3075 | while (beg_addr < end_addr && *(end_addr - 1) < 0x80) end_addr--; |
3076 | break; | |
3077 | } | |
3078 | } | |
3079 | else /* for decoding */ | |
3080 | { | |
3081 | switch (coding->type) | |
3082 | { | |
3083 | case coding_type_no_conversion: | |
3084 | /* We need no conversion. */ | |
3085 | *begp = *endp; | |
3086 | return; | |
0ef69138 | 3087 | case coding_type_emacs_mule: |
4ed46869 KH |
3088 | if (coding->eol_type == CODING_EOL_LF) |
3089 | { | |
3090 | /* We need no conversion. */ | |
3091 | *begp = *endp; | |
3092 | return; | |
3093 | } | |
3094 | /* We can skip all but carriage-return. */ | |
3095 | while (beg_addr < end_addr && *beg_addr != '\r') beg_addr++; | |
3096 | while (beg_addr < end_addr && *(end_addr - 1) != '\r') end_addr--; | |
3097 | break; | |
3098 | case coding_type_sjis: | |
3099 | case coding_type_big5: | |
3100 | /* We can skip all ASCII characters at the head. */ | |
3101 | while (beg_addr < end_addr && *beg_addr < 0x80) beg_addr++; | |
3102 | /* We can skip all ASCII characters at the tail except for | |
3103 | the second byte of SJIS or BIG5 code. */ | |
3104 | while (beg_addr < end_addr && *(end_addr - 1) < 0x80) end_addr--; | |
3105 | if (end_addr != *endp) | |
3106 | end_addr++; | |
3107 | break; | |
3108 | case coding_type_ccl: | |
3109 | /* We can't skip any data. */ | |
3110 | return; | |
3111 | default: /* i.e. case coding_type_iso2022: */ | |
3112 | { | |
3113 | unsigned char c; | |
3114 | ||
3115 | /* We can skip all ASCII characters except for a few | |
3116 | control codes at the head. */ | |
3117 | while (beg_addr < end_addr && (c = *beg_addr) < 0x80 | |
3118 | && c != ISO_CODE_CR && c != ISO_CODE_SO | |
3119 | && c != ISO_CODE_SI && c != ISO_CODE_ESC) | |
3120 | beg_addr++; | |
3121 | } | |
3122 | break; | |
3123 | } | |
3124 | } | |
3125 | *begp = beg_addr; | |
3126 | *endp = end_addr; | |
3127 | return; | |
3128 | } | |
3129 | ||
3130 | /* Encode to (iff ENCODEP is 1) or decode form coding system CODING a | |
3131 | text between B and E. B and E are buffer position. */ | |
3132 | ||
3133 | Lisp_Object | |
3134 | code_convert_region (b, e, coding, encodep) | |
3135 | Lisp_Object b, e; | |
3136 | struct coding_system *coding; | |
3137 | int encodep; | |
3138 | { | |
3139 | int beg, end, len, consumed, produced; | |
3140 | char *buf; | |
3141 | unsigned char *begp, *endp; | |
3142 | int pos = PT; | |
3143 | ||
3144 | validate_region (&b, &e); | |
3145 | beg = XINT (b), end = XINT (e); | |
3146 | if (beg < GPT && end >= GPT) | |
3147 | move_gap (end); | |
3148 | ||
3149 | if (encodep && !NILP (coding->pre_write_conversion)) | |
3150 | { | |
3151 | /* We must call a pre-conversion function which may put a new | |
3152 | text to be converted in a new buffer. */ | |
3153 | struct buffer *old = current_buffer, *new; | |
3154 | ||
3155 | TEMP_SET_PT (beg); | |
3156 | call2 (coding->pre_write_conversion, b, e); | |
3157 | if (old != current_buffer) | |
3158 | { | |
3159 | /* Replace the original text by the text just generated. */ | |
3160 | len = ZV - BEGV; | |
3161 | new = current_buffer; | |
3162 | set_buffer_internal (old); | |
3163 | del_range (beg, end); | |
3164 | insert_from_buffer (new, 1, len, 0); | |
3165 | end = beg + len; | |
3166 | } | |
3167 | } | |
3168 | ||
3169 | /* We may be able to shrink the conversion region. */ | |
3170 | begp = POS_ADDR (beg); endp = begp + (end - beg); | |
3171 | shrink_conversion_area (&begp, &endp, coding, encodep); | |
3172 | ||
3173 | if (begp == endp) | |
3174 | /* We need no conversion. */ | |
3175 | len = end - beg; | |
3176 | else | |
3177 | { | |
3178 | beg += begp - POS_ADDR (beg); | |
3179 | end = beg + (endp - begp); | |
3180 | ||
3181 | if (encodep) | |
3182 | len = encoding_buffer_size (coding, end - beg); | |
3183 | else | |
3184 | len = decoding_buffer_size (coding, end - beg); | |
3185 | buf = get_conversion_buffer (len); | |
3186 | ||
3187 | coding->last_block = 1; | |
3188 | produced = (encodep | |
3189 | ? encode_coding (coding, POS_ADDR (beg), buf, end - beg, len, | |
3190 | &consumed) | |
3191 | : decode_coding (coding, POS_ADDR (beg), buf, end - beg, len, | |
3192 | &consumed)); | |
3193 | ||
3194 | len = produced + (beg - XINT (b)) + (XINT (e) - end); | |
3195 | ||
3196 | TEMP_SET_PT (beg); | |
3197 | insert (buf, produced); | |
3198 | del_range (PT, PT + end - beg); | |
3199 | if (pos >= end) | |
3200 | pos = PT + (pos - end); | |
3201 | else if (pos > beg) | |
3202 | pos = beg; | |
3203 | TEMP_SET_PT (pos); | |
3204 | } | |
3205 | ||
3206 | if (!encodep && !NILP (coding->post_read_conversion)) | |
3207 | { | |
3208 | /* We must call a post-conversion function which may alter | |
3209 | the text just converted. */ | |
3210 | Lisp_Object insval; | |
3211 | ||
3212 | beg = XINT (b); | |
3213 | TEMP_SET_PT (beg); | |
3214 | insval = call1 (coding->post_read_conversion, make_number (len)); | |
3215 | CHECK_NUMBER (insval, 0); | |
3216 | len = XINT (insval); | |
3217 | } | |
3218 | ||
3219 | return make_number (len); | |
3220 | } | |
3221 | ||
3222 | Lisp_Object | |
e0e989f6 KH |
3223 | code_convert_string (str, coding, encodep, nocopy) |
3224 | Lisp_Object str, nocopy; | |
4ed46869 KH |
3225 | struct coding_system *coding; |
3226 | int encodep; | |
3227 | { | |
3228 | int len, consumed, produced; | |
3229 | char *buf; | |
3230 | unsigned char *begp, *endp; | |
3231 | int head_skip, tail_skip; | |
3232 | struct gcpro gcpro1; | |
3233 | ||
3234 | if (encodep && !NILP (coding->pre_write_conversion) | |
3235 | || !encodep && !NILP (coding->post_read_conversion)) | |
3236 | { | |
3237 | /* Since we have to call Lisp functions which assume target text | |
3238 | is in a buffer, after setting a temporary buffer, call | |
3239 | code_convert_region. */ | |
3240 | int count = specpdl_ptr - specpdl; | |
3241 | int len = XSTRING (str)->size; | |
3242 | Lisp_Object result; | |
3243 | struct buffer *old = current_buffer; | |
3244 | ||
3245 | record_unwind_protect (Fset_buffer, Fcurrent_buffer ()); | |
3246 | temp_output_buffer_setup (" *code-converting-work*"); | |
3247 | set_buffer_internal (XBUFFER (Vstandard_output)); | |
3248 | insert_from_string (str, 0, len, 0); | |
3249 | code_convert_region (make_number (BEGV), make_number (ZV), | |
3250 | coding, encodep); | |
3251 | result = make_buffer_string (BEGV, ZV, 0); | |
3252 | set_buffer_internal (old); | |
3253 | return unbind_to (count, result); | |
3254 | } | |
3255 | ||
3256 | /* We may be able to shrink the conversion region. */ | |
3257 | begp = XSTRING (str)->data; | |
3258 | endp = begp + XSTRING (str)->size; | |
3259 | shrink_conversion_area (&begp, &endp, coding, encodep); | |
3260 | ||
3261 | if (begp == endp) | |
3262 | /* We need no conversion. */ | |
e0e989f6 | 3263 | return (NILP (nocopy) ? Fcopy_sequence (str) : str); |
4ed46869 KH |
3264 | |
3265 | head_skip = begp - XSTRING (str)->data; | |
3266 | tail_skip = XSTRING (str)->size - head_skip - (endp - begp); | |
3267 | ||
3268 | GCPRO1 (str); | |
3269 | ||
3270 | if (encodep) | |
3271 | len = encoding_buffer_size (coding, endp - begp); | |
3272 | else | |
3273 | len = decoding_buffer_size (coding, endp - begp); | |
3274 | buf = get_conversion_buffer (len + head_skip + tail_skip); | |
3275 | ||
3276 | bcopy (XSTRING (str)->data, buf, head_skip); | |
3277 | coding->last_block = 1; | |
3278 | produced = (encodep | |
3279 | ? encode_coding (coding, XSTRING (str)->data + head_skip, | |
3280 | buf + head_skip, endp - begp, len, &consumed) | |
3281 | : decode_coding (coding, XSTRING (str)->data + head_skip, | |
3282 | buf + head_skip, endp - begp, len, &consumed)); | |
3283 | bcopy (XSTRING (str)->data + head_skip + (endp - begp), | |
3284 | buf + head_skip + produced, | |
3285 | tail_skip); | |
3286 | ||
3287 | UNGCPRO; | |
3288 | ||
3289 | return make_string (buf, head_skip + produced + tail_skip); | |
3290 | } | |
3291 | ||
3292 | DEFUN ("decode-coding-region", Fdecode_coding_region, Sdecode_coding_region, | |
e0e989f6 KH |
3293 | 3, 3, "r\nzCoding system: ", |
3294 | "Decode current region by specified coding system.\n\ | |
3295 | When called from a program, takes three arguments:\n\ | |
3296 | START, END, and CODING-SYSTEM. START END are buffer positions.\n\ | |
4ed46869 KH |
3297 | Return length of decoded text.") |
3298 | (b, e, coding_system) | |
3299 | Lisp_Object b, e, coding_system; | |
3300 | { | |
3301 | struct coding_system coding; | |
3302 | ||
3303 | CHECK_NUMBER_COERCE_MARKER (b, 0); | |
3304 | CHECK_NUMBER_COERCE_MARKER (e, 1); | |
3305 | CHECK_SYMBOL (coding_system, 2); | |
3306 | ||
e0e989f6 KH |
3307 | if (NILP (coding_system)) |
3308 | return make_number (XFASTINT (e) - XFASTINT (b)); | |
4ed46869 KH |
3309 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
3310 | error ("Invalid coding-system: %s", XSYMBOL (coding_system)->name->data); | |
3311 | ||
3312 | return code_convert_region (b, e, &coding, 0); | |
3313 | } | |
3314 | ||
3315 | DEFUN ("encode-coding-region", Fencode_coding_region, Sencode_coding_region, | |
e0e989f6 KH |
3316 | 3, 3, "r\nzCoding system: ", |
3317 | "Encode current region by specified coding system.\n\ | |
3318 | When called from a program, takes three arguments:\n\ | |
3319 | START, END, and CODING-SYSTEM. START END are buffer positions.\n\ | |
4ed46869 KH |
3320 | Return length of encoded text.") |
3321 | (b, e, coding_system) | |
3322 | Lisp_Object b, e, coding_system; | |
3323 | { | |
3324 | struct coding_system coding; | |
3325 | ||
3326 | CHECK_NUMBER_COERCE_MARKER (b, 0); | |
3327 | CHECK_NUMBER_COERCE_MARKER (e, 1); | |
3328 | CHECK_SYMBOL (coding_system, 2); | |
3329 | ||
e0e989f6 KH |
3330 | if (NILP (coding_system)) |
3331 | return make_number (XFASTINT (e) - XFASTINT (b)); | |
4ed46869 KH |
3332 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
3333 | error ("Invalid coding-system: %s", XSYMBOL (coding_system)->name->data); | |
3334 | ||
3335 | return code_convert_region (b, e, &coding, 1); | |
3336 | } | |
3337 | ||
3338 | DEFUN ("decode-coding-string", Fdecode_coding_string, Sdecode_coding_string, | |
e0e989f6 KH |
3339 | 2, 3, 0, |
3340 | "Decode STRING which is encoded in CODING-SYSTEM, and return the result.\n\ | |
3341 | Optional arg NOCOPY non-nil means return STRING itself if there's no need\n\ | |
3342 | of decoding.") | |
3343 | (string, coding_system, nocopy) | |
3344 | Lisp_Object string, coding_system, nocopy; | |
4ed46869 KH |
3345 | { |
3346 | struct coding_system coding; | |
3347 | ||
3348 | CHECK_STRING (string, 0); | |
3349 | CHECK_SYMBOL (coding_system, 1); | |
3350 | ||
e0e989f6 KH |
3351 | if (NILP (coding_system)) |
3352 | return (NILP (nocopy) ? Fcopy_sequence (string) : string); | |
4ed46869 KH |
3353 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
3354 | error ("Invalid coding-system: %s", XSYMBOL (coding_system)->name->data); | |
3355 | ||
e0e989f6 | 3356 | return code_convert_string (string, &coding, 0, nocopy); |
4ed46869 KH |
3357 | } |
3358 | ||
3359 | DEFUN ("encode-coding-string", Fencode_coding_string, Sencode_coding_string, | |
e0e989f6 KH |
3360 | 2, 3, 0, |
3361 | "Encode STRING to CODING-SYSTEM, and return the result.\n\ | |
3362 | Optional arg NOCOPY non-nil means return STRING itself if there's no need\n\ | |
3363 | of encoding.") | |
3364 | (string, coding_system, nocopy) | |
3365 | Lisp_Object string, coding_system, nocopy; | |
4ed46869 KH |
3366 | { |
3367 | struct coding_system coding; | |
3368 | ||
3369 | CHECK_STRING (string, 0); | |
3370 | CHECK_SYMBOL (coding_system, 1); | |
3371 | ||
e0e989f6 KH |
3372 | if (NILP (coding_system)) |
3373 | return (NILP (nocopy) ? Fcopy_sequence (string) : string); | |
4ed46869 KH |
3374 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
3375 | error ("Invalid coding-system: %s", XSYMBOL (coding_system)->name->data); | |
3376 | ||
e0e989f6 | 3377 | return code_convert_string (string, &coding, 1, nocopy); |
4ed46869 KH |
3378 | } |
3379 | ||
3380 | DEFUN ("decode-sjis-char", Fdecode_sjis_char, Sdecode_sjis_char, 1, 1, 0, | |
e0e989f6 | 3381 | "Decode a JISX0208 character of shift-jis encoding.\n\ |
4ed46869 KH |
3382 | CODE is the character code in SJIS.\n\ |
3383 | Return the corresponding character.") | |
3384 | (code) | |
3385 | Lisp_Object code; | |
3386 | { | |
3387 | unsigned char c1, c2, s1, s2; | |
3388 | Lisp_Object val; | |
3389 | ||
3390 | CHECK_NUMBER (code, 0); | |
3391 | s1 = (XFASTINT (code)) >> 8, s2 = (XFASTINT (code)) & 0xFF; | |
3392 | DECODE_SJIS (s1, s2, c1, c2); | |
3393 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset_jisx0208, c1, c2)); | |
3394 | return val; | |
3395 | } | |
3396 | ||
3397 | DEFUN ("encode-sjis-char", Fencode_sjis_char, Sencode_sjis_char, 1, 1, 0, | |
3398 | "Encode a JISX0208 character CHAR to SJIS coding-system.\n\ | |
3399 | Return the corresponding character code in SJIS.") | |
3400 | (ch) | |
3401 | Lisp_Object ch; | |
3402 | { | |
bcf26d6a | 3403 | int charset, c1, c2, s1, s2; |
4ed46869 KH |
3404 | Lisp_Object val; |
3405 | ||
3406 | CHECK_NUMBER (ch, 0); | |
3407 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
3408 | if (charset == charset_jisx0208) | |
3409 | { | |
3410 | ENCODE_SJIS (c1, c2, s1, s2); | |
bcf26d6a | 3411 | XSETFASTINT (val, (s1 << 8) | s2); |
4ed46869 KH |
3412 | } |
3413 | else | |
3414 | XSETFASTINT (val, 0); | |
3415 | return val; | |
3416 | } | |
3417 | ||
3418 | DEFUN ("decode-big5-char", Fdecode_big5_char, Sdecode_big5_char, 1, 1, 0, | |
3419 | "Decode a Big5 character CODE of BIG5 coding-system.\n\ | |
3420 | CODE is the character code in BIG5.\n\ | |
3421 | Return the corresponding character.") | |
3422 | (code) | |
3423 | Lisp_Object code; | |
3424 | { | |
3425 | int charset; | |
3426 | unsigned char b1, b2, c1, c2; | |
3427 | Lisp_Object val; | |
3428 | ||
3429 | CHECK_NUMBER (code, 0); | |
3430 | b1 = (XFASTINT (code)) >> 8, b2 = (XFASTINT (code)) & 0xFF; | |
3431 | DECODE_BIG5 (b1, b2, charset, c1, c2); | |
3432 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset, c1, c2)); | |
3433 | return val; | |
3434 | } | |
3435 | ||
3436 | DEFUN ("encode-big5-char", Fencode_big5_char, Sencode_big5_char, 1, 1, 0, | |
3437 | "Encode the Big5 character CHAR to BIG5 coding-system.\n\ | |
3438 | Return the corresponding character code in Big5.") | |
3439 | (ch) | |
3440 | Lisp_Object ch; | |
3441 | { | |
bcf26d6a | 3442 | int charset, c1, c2, b1, b2; |
4ed46869 KH |
3443 | Lisp_Object val; |
3444 | ||
3445 | CHECK_NUMBER (ch, 0); | |
3446 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
3447 | if (charset == charset_big5_1 || charset == charset_big5_2) | |
3448 | { | |
3449 | ENCODE_BIG5 (charset, c1, c2, b1, b2); | |
bcf26d6a | 3450 | XSETFASTINT (val, (b1 << 8) | b2); |
4ed46869 KH |
3451 | } |
3452 | else | |
3453 | XSETFASTINT (val, 0); | |
3454 | return val; | |
3455 | } | |
3456 | ||
1ba9e4ab KH |
3457 | DEFUN ("set-terminal-coding-system-internal", |
3458 | Fset_terminal_coding_system_internal, | |
3459 | Sset_terminal_coding_system_internal, 1, 1, 0, "") | |
4ed46869 KH |
3460 | (coding_system) |
3461 | Lisp_Object coding_system; | |
3462 | { | |
3463 | CHECK_SYMBOL (coding_system, 0); | |
3464 | setup_coding_system (Fcheck_coding_system (coding_system), &terminal_coding); | |
4ed46869 KH |
3465 | return Qnil; |
3466 | } | |
3467 | ||
3468 | DEFUN ("terminal-coding-system", | |
3469 | Fterminal_coding_system, Sterminal_coding_system, 0, 0, 0, | |
3470 | "Return coding-system of your terminal.") | |
3471 | () | |
3472 | { | |
3473 | return terminal_coding.symbol; | |
3474 | } | |
3475 | ||
1ba9e4ab KH |
3476 | DEFUN ("set-keyboard-coding-system-internal", |
3477 | Fset_keyboard_coding_system_internal, | |
3478 | Sset_keyboard_coding_system_internal, 1, 1, 0, "") | |
4ed46869 KH |
3479 | (coding_system) |
3480 | Lisp_Object coding_system; | |
3481 | { | |
3482 | CHECK_SYMBOL (coding_system, 0); | |
3483 | setup_coding_system (Fcheck_coding_system (coding_system), &keyboard_coding); | |
3484 | return Qnil; | |
3485 | } | |
3486 | ||
3487 | DEFUN ("keyboard-coding-system", | |
3488 | Fkeyboard_coding_system, Skeyboard_coding_system, 0, 0, 0, | |
3489 | "Return coding-system of what is sent from terminal keyboard.") | |
3490 | () | |
3491 | { | |
3492 | return keyboard_coding.symbol; | |
3493 | } | |
3494 | ||
3495 | \f | |
a5d301df KH |
3496 | DEFUN ("find-operation-coding-system", Ffind_operation_coding_system, |
3497 | Sfind_operation_coding_system, 1, MANY, 0, | |
3498 | "Choose a coding system for an operation based on the target name.\n\ | |
9ce27fde KH |
3499 | The value names a pair of coding systems: (DECODING-SYSTEM ENCODING-SYSTEM).\n\ |
3500 | DECODING-SYSTEM is the coding system to use for decoding\n\ | |
3501 | \(in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system\n\ | |
3502 | for encoding (in case OPERATION does encoding).\n\ | |
ccdb79f5 RS |
3503 | \n\ |
3504 | The first argument OPERATION specifies an I/O primitive:\n\ | |
3505 | For file I/O, `insert-file-contents' or `write-region'.\n\ | |
3506 | For process I/O, `call-process', `call-process-region', or `start-process'.\n\ | |
3507 | For network I/O, `open-network-stream'.\n\ | |
3508 | \n\ | |
3509 | The remaining arguments should be the same arguments that were passed\n\ | |
3510 | to the primitive. Depending on which primitive, one of those arguments\n\ | |
3511 | is selected as the TARGET. For example, if OPERATION does file I/O,\n\ | |
3512 | whichever argument specifies the file name is TARGET.\n\ | |
3513 | \n\ | |
3514 | TARGET has a meaning which depends on OPERATION:\n\ | |
4ed46869 KH |
3515 | For file I/O, TARGET is a file name.\n\ |
3516 | For process I/O, TARGET is a process name.\n\ | |
3517 | For network I/O, TARGET is a service name or a port number\n\ | |
3518 | \n\ | |
02ba4723 KH |
3519 | This function looks up what specified for TARGET in,\n\ |
3520 | `file-coding-system-alist', `process-coding-system-alist',\n\ | |
3521 | or `network-coding-system-alist' depending on OPERATION.\n\ | |
3522 | They may specify a coding system, a cons of coding systems,\n\ | |
3523 | or a function symbol to call.\n\ | |
3524 | In the last case, we call the function with one argument,\n\ | |
9ce27fde | 3525 | which is a list of all the arguments given to this function.") |
4ed46869 KH |
3526 | (nargs, args) |
3527 | int nargs; | |
3528 | Lisp_Object *args; | |
3529 | { | |
3530 | Lisp_Object operation, target_idx, target, val; | |
3531 | register Lisp_Object chain; | |
3532 | ||
3533 | if (nargs < 2) | |
3534 | error ("Too few arguments"); | |
3535 | operation = args[0]; | |
3536 | if (!SYMBOLP (operation) | |
3537 | || !INTEGERP (target_idx = Fget (operation, Qtarget_idx))) | |
3538 | error ("Invalid first arguement"); | |
3539 | if (nargs < 1 + XINT (target_idx)) | |
3540 | error ("Too few arguments for operation: %s", | |
3541 | XSYMBOL (operation)->name->data); | |
3542 | target = args[XINT (target_idx) + 1]; | |
3543 | if (!(STRINGP (target) | |
3544 | || (EQ (operation, Qopen_network_stream) && INTEGERP (target)))) | |
3545 | error ("Invalid %dth argument", XINT (target_idx) + 1); | |
3546 | ||
2e34157c RS |
3547 | chain = ((EQ (operation, Qinsert_file_contents) |
3548 | || EQ (operation, Qwrite_region)) | |
02ba4723 | 3549 | ? Vfile_coding_system_alist |
2e34157c | 3550 | : (EQ (operation, Qopen_network_stream) |
02ba4723 KH |
3551 | ? Vnetwork_coding_system_alist |
3552 | : Vprocess_coding_system_alist)); | |
4ed46869 KH |
3553 | if (NILP (chain)) |
3554 | return Qnil; | |
3555 | ||
02ba4723 | 3556 | for (; CONSP (chain); chain = XCONS (chain)->cdr) |
4ed46869 KH |
3557 | { |
3558 | Lisp_Object elt = XCONS (chain)->car; | |
3559 | ||
3560 | if (CONSP (elt) | |
3561 | && ((STRINGP (target) | |
3562 | && STRINGP (XCONS (elt)->car) | |
3563 | && fast_string_match (XCONS (elt)->car, target) >= 0) | |
3564 | || (INTEGERP (target) && EQ (target, XCONS (elt)->car)))) | |
02ba4723 KH |
3565 | { |
3566 | val = XCONS (elt)->cdr; | |
3567 | if (CONSP (val)) | |
3568 | return val; | |
3569 | if (! SYMBOLP (val)) | |
3570 | return Qnil; | |
3571 | if (! NILP (Fcoding_system_p (val))) | |
3572 | return Fcons (val, val); | |
465edc86 | 3573 | if (!NILP (Ffboundp (val))) |
5d632ccf | 3574 | return call1 (val, Flist (nargs, args)); |
02ba4723 KH |
3575 | return Qnil; |
3576 | } | |
4ed46869 KH |
3577 | } |
3578 | return Qnil; | |
3579 | } | |
3580 | ||
3581 | #endif /* emacs */ | |
3582 | ||
3583 | \f | |
3584 | /*** 8. Post-amble ***/ | |
3585 | ||
3586 | init_coding_once () | |
3587 | { | |
3588 | int i; | |
3589 | ||
0ef69138 | 3590 | /* Emacs' internal format specific initialize routine. */ |
4ed46869 KH |
3591 | for (i = 0; i <= 0x20; i++) |
3592 | emacs_code_class[i] = EMACS_control_code; | |
3593 | emacs_code_class[0x0A] = EMACS_linefeed_code; | |
3594 | emacs_code_class[0x0D] = EMACS_carriage_return_code; | |
3595 | for (i = 0x21 ; i < 0x7F; i++) | |
3596 | emacs_code_class[i] = EMACS_ascii_code; | |
3597 | emacs_code_class[0x7F] = EMACS_control_code; | |
3598 | emacs_code_class[0x80] = EMACS_leading_code_composition; | |
3599 | for (i = 0x81; i < 0xFF; i++) | |
3600 | emacs_code_class[i] = EMACS_invalid_code; | |
3601 | emacs_code_class[LEADING_CODE_PRIVATE_11] = EMACS_leading_code_3; | |
3602 | emacs_code_class[LEADING_CODE_PRIVATE_12] = EMACS_leading_code_3; | |
3603 | emacs_code_class[LEADING_CODE_PRIVATE_21] = EMACS_leading_code_4; | |
3604 | emacs_code_class[LEADING_CODE_PRIVATE_22] = EMACS_leading_code_4; | |
3605 | ||
3606 | /* ISO2022 specific initialize routine. */ | |
3607 | for (i = 0; i < 0x20; i++) | |
3608 | iso_code_class[i] = ISO_control_code; | |
3609 | for (i = 0x21; i < 0x7F; i++) | |
3610 | iso_code_class[i] = ISO_graphic_plane_0; | |
3611 | for (i = 0x80; i < 0xA0; i++) | |
3612 | iso_code_class[i] = ISO_control_code; | |
3613 | for (i = 0xA1; i < 0xFF; i++) | |
3614 | iso_code_class[i] = ISO_graphic_plane_1; | |
3615 | iso_code_class[0x20] = iso_code_class[0x7F] = ISO_0x20_or_0x7F; | |
3616 | iso_code_class[0xA0] = iso_code_class[0xFF] = ISO_0xA0_or_0xFF; | |
3617 | iso_code_class[ISO_CODE_CR] = ISO_carriage_return; | |
3618 | iso_code_class[ISO_CODE_SO] = ISO_shift_out; | |
3619 | iso_code_class[ISO_CODE_SI] = ISO_shift_in; | |
3620 | iso_code_class[ISO_CODE_SS2_7] = ISO_single_shift_2_7; | |
3621 | iso_code_class[ISO_CODE_ESC] = ISO_escape; | |
3622 | iso_code_class[ISO_CODE_SS2] = ISO_single_shift_2; | |
3623 | iso_code_class[ISO_CODE_SS3] = ISO_single_shift_3; | |
3624 | iso_code_class[ISO_CODE_CSI] = ISO_control_sequence_introducer; | |
3625 | ||
e0e989f6 KH |
3626 | conversion_buffer_size = MINIMUM_CONVERSION_BUFFER_SIZE; |
3627 | conversion_buffer = (char *) xmalloc (MINIMUM_CONVERSION_BUFFER_SIZE); | |
3628 | ||
3629 | setup_coding_system (Qnil, &keyboard_coding); | |
3630 | setup_coding_system (Qnil, &terminal_coding); | |
9ce27fde KH |
3631 | |
3632 | #if defined (MSDOS) || defined (WINDOWSNT) | |
3633 | system_eol_type = CODING_EOL_CRLF; | |
3634 | #else | |
3635 | system_eol_type = CODING_EOL_LF; | |
3636 | #endif | |
e0e989f6 KH |
3637 | } |
3638 | ||
3639 | #ifdef emacs | |
3640 | ||
3641 | syms_of_coding () | |
3642 | { | |
3643 | Qtarget_idx = intern ("target-idx"); | |
3644 | staticpro (&Qtarget_idx); | |
3645 | ||
9ce27fde | 3646 | /* Target FILENAME is the first argument. */ |
e0e989f6 | 3647 | Fput (Qinsert_file_contents, Qtarget_idx, make_number (0)); |
9ce27fde | 3648 | /* Target FILENAME is the third argument. */ |
e0e989f6 KH |
3649 | Fput (Qwrite_region, Qtarget_idx, make_number (2)); |
3650 | ||
3651 | Qcall_process = intern ("call-process"); | |
3652 | staticpro (&Qcall_process); | |
9ce27fde | 3653 | /* Target PROGRAM is the first argument. */ |
e0e989f6 KH |
3654 | Fput (Qcall_process, Qtarget_idx, make_number (0)); |
3655 | ||
3656 | Qcall_process_region = intern ("call-process-region"); | |
3657 | staticpro (&Qcall_process_region); | |
9ce27fde | 3658 | /* Target PROGRAM is the third argument. */ |
e0e989f6 KH |
3659 | Fput (Qcall_process_region, Qtarget_idx, make_number (2)); |
3660 | ||
3661 | Qstart_process = intern ("start-process"); | |
3662 | staticpro (&Qstart_process); | |
9ce27fde | 3663 | /* Target PROGRAM is the third argument. */ |
e0e989f6 KH |
3664 | Fput (Qstart_process, Qtarget_idx, make_number (2)); |
3665 | ||
3666 | Qopen_network_stream = intern ("open-network-stream"); | |
3667 | staticpro (&Qopen_network_stream); | |
9ce27fde | 3668 | /* Target SERVICE is the fourth argument. */ |
e0e989f6 KH |
3669 | Fput (Qopen_network_stream, Qtarget_idx, make_number (3)); |
3670 | ||
4ed46869 KH |
3671 | Qcoding_system = intern ("coding-system"); |
3672 | staticpro (&Qcoding_system); | |
3673 | ||
3674 | Qeol_type = intern ("eol-type"); | |
3675 | staticpro (&Qeol_type); | |
3676 | ||
3677 | Qbuffer_file_coding_system = intern ("buffer-file-coding-system"); | |
3678 | staticpro (&Qbuffer_file_coding_system); | |
3679 | ||
3680 | Qpost_read_conversion = intern ("post-read-conversion"); | |
3681 | staticpro (&Qpost_read_conversion); | |
3682 | ||
3683 | Qpre_write_conversion = intern ("pre-write-conversion"); | |
3684 | staticpro (&Qpre_write_conversion); | |
3685 | ||
02ba4723 KH |
3686 | Qcoding_system_spec = intern ("coding-system-spec"); |
3687 | staticpro (&Qcoding_system_spec); | |
4ed46869 KH |
3688 | |
3689 | Qcoding_system_p = intern ("coding-system-p"); | |
3690 | staticpro (&Qcoding_system_p); | |
3691 | ||
3692 | Qcoding_system_error = intern ("coding-system-error"); | |
3693 | staticpro (&Qcoding_system_error); | |
3694 | ||
3695 | Fput (Qcoding_system_error, Qerror_conditions, | |
3696 | Fcons (Qcoding_system_error, Fcons (Qerror, Qnil))); | |
3697 | Fput (Qcoding_system_error, Qerror_message, | |
9ce27fde | 3698 | build_string ("Invalid coding system")); |
4ed46869 KH |
3699 | |
3700 | Qcoding_category_index = intern ("coding-category-index"); | |
3701 | staticpro (&Qcoding_category_index); | |
3702 | ||
3703 | { | |
3704 | int i; | |
3705 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) | |
3706 | { | |
3707 | coding_category_table[i] = intern (coding_category_name[i]); | |
3708 | staticpro (&coding_category_table[i]); | |
3709 | Fput (coding_category_table[i], Qcoding_category_index, | |
3710 | make_number (i)); | |
3711 | } | |
3712 | } | |
3713 | ||
bdd9fb48 KH |
3714 | Qcharacter_unification_table = intern ("character-unification-table"); |
3715 | staticpro (&Qcharacter_unification_table); | |
3716 | Fput (Qcharacter_unification_table, Qchar_table_extra_slots, | |
3717 | make_number (0)); | |
3718 | ||
a5d301df KH |
3719 | Qcharacter_unification_table_for_decode |
3720 | = intern ("character-unification-table-for-decode"); | |
3721 | staticpro (&Qcharacter_unification_table_for_decode); | |
3722 | ||
3723 | Qcharacter_unification_table_for_encode | |
3724 | = intern ("character-unification-table-for-encode"); | |
3725 | staticpro (&Qcharacter_unification_table_for_encode); | |
3726 | ||
9ce27fde KH |
3727 | Qemacs_mule = intern ("emacs-mule"); |
3728 | staticpro (&Qemacs_mule); | |
3729 | ||
02ba4723 | 3730 | defsubr (&Scoding_system_spec); |
4ed46869 KH |
3731 | defsubr (&Scoding_system_p); |
3732 | defsubr (&Sread_coding_system); | |
3733 | defsubr (&Sread_non_nil_coding_system); | |
3734 | defsubr (&Scheck_coding_system); | |
3735 | defsubr (&Sdetect_coding_region); | |
3736 | defsubr (&Sdecode_coding_region); | |
3737 | defsubr (&Sencode_coding_region); | |
3738 | defsubr (&Sdecode_coding_string); | |
3739 | defsubr (&Sencode_coding_string); | |
3740 | defsubr (&Sdecode_sjis_char); | |
3741 | defsubr (&Sencode_sjis_char); | |
3742 | defsubr (&Sdecode_big5_char); | |
3743 | defsubr (&Sencode_big5_char); | |
1ba9e4ab | 3744 | defsubr (&Sset_terminal_coding_system_internal); |
4ed46869 | 3745 | defsubr (&Sterminal_coding_system); |
1ba9e4ab | 3746 | defsubr (&Sset_keyboard_coding_system_internal); |
4ed46869 | 3747 | defsubr (&Skeyboard_coding_system); |
a5d301df | 3748 | defsubr (&Sfind_operation_coding_system); |
4ed46869 KH |
3749 | |
3750 | DEFVAR_LISP ("coding-category-list", &Vcoding_category_list, | |
3751 | "List of coding-categories (symbols) ordered by priority."); | |
3752 | { | |
3753 | int i; | |
3754 | ||
3755 | Vcoding_category_list = Qnil; | |
3756 | for (i = CODING_CATEGORY_IDX_MAX - 1; i >= 0; i--) | |
3757 | Vcoding_category_list | |
3758 | = Fcons (coding_category_table[i], Vcoding_category_list); | |
3759 | } | |
3760 | ||
3761 | DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read, | |
3762 | "A variable of internal use only.\n\ | |
3763 | If the value is a coding system, it is used for decoding on read operation.\n\ | |
3764 | If not, an appropriate element in `coding-system-alist' (which see) is used."); | |
3765 | Vcoding_system_for_read = Qnil; | |
3766 | ||
3767 | DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write, | |
3768 | "A variable of internal use only.\n\ | |
3769 | If the value is a coding system, it is used for encoding on write operation.\n\ | |
3770 | If not, an appropriate element in `coding-system-alist' (which see) is used."); | |
3771 | Vcoding_system_for_write = Qnil; | |
3772 | ||
3773 | DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used, | |
3774 | "Coding-system used in the latest file or process I/O."); | |
3775 | Vlast_coding_system_used = Qnil; | |
3776 | ||
9ce27fde KH |
3777 | DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion, |
3778 | "*Non-nil inhibit code conversion of end-of-line format in any cases."); | |
3779 | inhibit_eol_conversion = 0; | |
3780 | ||
02ba4723 KH |
3781 | DEFVAR_LISP ("file-coding-system-alist", &Vfile_coding_system_alist, |
3782 | "Alist to decide a coding system to use for a file I/O operation.\n\ | |
3783 | The format is ((PATTERN . VAL) ...),\n\ | |
3784 | where PATTERN is a regular expression matching a file name,\n\ | |
3785 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
3786 | If VAL is a coding system, it is used for both decoding and encoding\n\ | |
3787 | the file contents.\n\ | |
3788 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
3789 | and the cdr part is used for encoding.\n\ | |
3790 | If VAL is a function symbol, the function must return a coding system\n\ | |
3791 | or a cons of coding systems which are used as above.\n\ | |
e0e989f6 | 3792 | \n\ |
9ce27fde | 3793 | See also the function `find-operation-coding-system'."); |
02ba4723 KH |
3794 | Vfile_coding_system_alist = Qnil; |
3795 | ||
3796 | DEFVAR_LISP ("process-coding-system-alist", &Vprocess_coding_system_alist, | |
3797 | "Alist to decide a coding system to use for a process I/O operation.\n\ | |
3798 | The format is ((PATTERN . VAL) ...),\n\ | |
3799 | where PATTERN is a regular expression matching a program name,\n\ | |
3800 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
3801 | If VAL is a coding system, it is used for both decoding what received\n\ | |
3802 | from the program and encoding what sent to the program.\n\ | |
3803 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
3804 | and the cdr part is used for encoding.\n\ | |
3805 | If VAL is a function symbol, the function must return a coding system\n\ | |
3806 | or a cons of coding systems which are used as above.\n\ | |
4ed46869 | 3807 | \n\ |
9ce27fde | 3808 | See also the function `find-operation-coding-system'."); |
02ba4723 KH |
3809 | Vprocess_coding_system_alist = Qnil; |
3810 | ||
3811 | DEFVAR_LISP ("network-coding-system-alist", &Vnetwork_coding_system_alist, | |
3812 | "Alist to decide a coding system to use for a network I/O operation.\n\ | |
3813 | The format is ((PATTERN . VAL) ...),\n\ | |
3814 | where PATTERN is a regular expression matching a network service name\n\ | |
3815 | or is a port number to connect to,\n\ | |
3816 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ | |
3817 | If VAL is a coding system, it is used for both decoding what received\n\ | |
3818 | from the network stream and encoding what sent to the network stream.\n\ | |
3819 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ | |
3820 | and the cdr part is used for encoding.\n\ | |
3821 | If VAL is a function symbol, the function must return a coding system\n\ | |
3822 | or a cons of coding systems which are used as above.\n\ | |
4ed46869 | 3823 | \n\ |
9ce27fde | 3824 | See also the function `find-operation-coding-system'."); |
02ba4723 | 3825 | Vnetwork_coding_system_alist = Qnil; |
4ed46869 KH |
3826 | |
3827 | DEFVAR_INT ("eol-mnemonic-unix", &eol_mnemonic_unix, | |
3828 | "Mnemonic character indicating UNIX-like end-of-line format (i.e. LF) ."); | |
458822a0 | 3829 | eol_mnemonic_unix = ':'; |
4ed46869 KH |
3830 | |
3831 | DEFVAR_INT ("eol-mnemonic-dos", &eol_mnemonic_dos, | |
3832 | "Mnemonic character indicating DOS-like end-of-line format (i.e. CRLF)."); | |
458822a0 | 3833 | eol_mnemonic_dos = '\\'; |
4ed46869 KH |
3834 | |
3835 | DEFVAR_INT ("eol-mnemonic-mac", &eol_mnemonic_mac, | |
3836 | "Mnemonic character indicating MAC-like end-of-line format (i.e. CR)."); | |
458822a0 | 3837 | eol_mnemonic_mac = '/'; |
4ed46869 KH |
3838 | |
3839 | DEFVAR_INT ("eol-mnemonic-undecided", &eol_mnemonic_undecided, | |
3840 | "Mnemonic character indicating end-of-line format is not yet decided."); | |
458822a0 | 3841 | eol_mnemonic_undecided = ':'; |
4ed46869 | 3842 | |
bdd9fb48 KH |
3843 | DEFVAR_LISP ("enable-character-unification", &Venable_character_unification, |
3844 | "Non-nil means ISO 2022 encoder/decoder do character unification."); | |
3845 | Venable_character_unification = Qt; | |
3846 | ||
a5d301df KH |
3847 | DEFVAR_LISP ("standard-character-unification-table-for-decode", |
3848 | &Vstandard_character_unification_table_for_decode, | |
bdd9fb48 | 3849 | "Table for unifying characters when reading."); |
a5d301df | 3850 | Vstandard_character_unification_table_for_decode = Qnil; |
bdd9fb48 | 3851 | |
a5d301df KH |
3852 | DEFVAR_LISP ("standard-character-unification-table-for-encode", |
3853 | &Vstandard_character_unification_table_for_encode, | |
bdd9fb48 | 3854 | "Table for unifying characters when writing."); |
a5d301df | 3855 | Vstandard_character_unification_table_for_encode = Qnil; |
4ed46869 KH |
3856 | |
3857 | DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist, | |
3858 | "Alist of charsets vs revision numbers.\n\ | |
3859 | While encoding, if a charset (car part of an element) is found,\n\ | |
3860 | designate it with the escape sequence identifing revision (cdr part of the element)."); | |
3861 | Vcharset_revision_alist = Qnil; | |
02ba4723 KH |
3862 | |
3863 | DEFVAR_LISP ("default-process-coding-system", | |
3864 | &Vdefault_process_coding_system, | |
3865 | "Cons of coding systems used for process I/O by default.\n\ | |
3866 | The car part is used for decoding a process output,\n\ | |
3867 | the cdr part is used for encoding a text to be sent to a process."); | |
3868 | Vdefault_process_coding_system = Qnil; | |
4ed46869 KH |
3869 | } |
3870 | ||
3871 | #endif /* emacs */ |