| 1 | /* Coding system handler (conversion, detection, and etc). |
| 2 | Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN. |
| 3 | Licensed to the Free Software Foundation. |
| 4 | |
| 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. |
| 11 | |
| 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. |
| 16 | |
| 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. */ |
| 21 | |
| 22 | /*** TABLE OF CONTENTS *** |
| 23 | |
| 24 | 1. Preamble |
| 25 | 2. Emacs' internal format (emacs-mule) handlers |
| 26 | 3. ISO2022 handlers |
| 27 | 4. Shift-JIS and BIG5 handlers |
| 28 | 5. CCL handlers |
| 29 | 6. End-of-line handlers |
| 30 | 7. C library functions |
| 31 | 8. Emacs Lisp library functions |
| 32 | 9. Post-amble |
| 33 | |
| 34 | */ |
| 35 | |
| 36 | /*** GENERAL NOTE on CODING SYSTEM *** |
| 37 | |
| 38 | Coding system is an encoding mechanism of one or more character |
| 39 | sets. Here's a list of coding systems which Emacs can handle. When |
| 40 | we say "decode", it means converting some other coding system to |
| 41 | Emacs' internal format (emacs-internal), and when we say "encode", |
| 42 | it means converting the coding system emacs-mule to some other |
| 43 | coding system. |
| 44 | |
| 45 | 0. Emacs' internal format (emacs-mule) |
| 46 | |
| 47 | Emacs itself holds a multi-lingual character in a buffer and a string |
| 48 | in a special format. Details are described in section 2. |
| 49 | |
| 50 | 1. ISO2022 |
| 51 | |
| 52 | The most famous coding system for multiple character sets. X's |
| 53 | Compound Text, various EUCs (Extended Unix Code), and coding |
| 54 | systems used in Internet communication such as ISO-2022-JP are |
| 55 | all variants of ISO2022. Details are described in section 3. |
| 56 | |
| 57 | 2. SJIS (or Shift-JIS or MS-Kanji-Code) |
| 58 | |
| 59 | A coding system to encode character sets: ASCII, JISX0201, and |
| 60 | JISX0208. Widely used for PC's in Japan. Details are described in |
| 61 | section 4. |
| 62 | |
| 63 | 3. BIG5 |
| 64 | |
| 65 | A coding system to encode character sets: ASCII and Big5. Widely |
| 66 | used by Chinese (mainly in Taiwan and Hong Kong). Details are |
| 67 | described in section 4. In this file, when we write "BIG5" |
| 68 | (all uppercase), we mean the coding system, and when we write |
| 69 | "Big5" (capitalized), we mean the character set. |
| 70 | |
| 71 | 4. Raw text |
| 72 | |
| 73 | A coding system for a text containing random 8-bit code. Emacs does |
| 74 | no code conversion on such a text except for end-of-line format. |
| 75 | |
| 76 | 5. Other |
| 77 | |
| 78 | If a user wants to read/write a text encoded in a coding system not |
| 79 | listed above, he can supply a decoder and an encoder for it in CCL |
| 80 | (Code Conversion Language) programs. Emacs executes the CCL program |
| 81 | while reading/writing. |
| 82 | |
| 83 | Emacs represents a coding system by a Lisp symbol that has a property |
| 84 | `coding-system'. But, before actually using the coding system, the |
| 85 | information about it is set in a structure of type `struct |
| 86 | coding_system' for rapid processing. See section 6 for more details. |
| 87 | |
| 88 | */ |
| 89 | |
| 90 | /*** GENERAL NOTES on END-OF-LINE FORMAT *** |
| 91 | |
| 92 | How end-of-line of a text is encoded depends on a system. For |
| 93 | instance, Unix's format is just one byte of `line-feed' code, |
| 94 | whereas DOS's format is two-byte sequence of `carriage-return' and |
| 95 | `line-feed' codes. MacOS's format is usually one byte of |
| 96 | `carriage-return'. |
| 97 | |
| 98 | Since text characters encoding and end-of-line encoding are |
| 99 | independent, any coding system described above can take |
| 100 | any format of end-of-line. So, Emacs has information of format of |
| 101 | end-of-line in each coding-system. See section 6 for more details. |
| 102 | |
| 103 | */ |
| 104 | |
| 105 | /*** GENERAL NOTES on `detect_coding_XXX ()' functions *** |
| 106 | |
| 107 | These functions check if a text between SRC and SRC_END is encoded |
| 108 | in the coding system category XXX. Each returns an integer value in |
| 109 | which appropriate flag bits for the category XXX is set. The flag |
| 110 | bits are defined in macros CODING_CATEGORY_MASK_XXX. Below is the |
| 111 | template of these functions. */ |
| 112 | #if 0 |
| 113 | int |
| 114 | detect_coding_emacs_mule (src, src_end) |
| 115 | unsigned char *src, *src_end; |
| 116 | { |
| 117 | ... |
| 118 | } |
| 119 | #endif |
| 120 | |
| 121 | /*** GENERAL NOTES on `decode_coding_XXX ()' functions *** |
| 122 | |
| 123 | These functions decode SRC_BYTES length text at SOURCE encoded in |
| 124 | CODING to Emacs' internal format (emacs-mule). The resulting text |
| 125 | goes to a place pointed to by DESTINATION, the length of which |
| 126 | should not exceed DST_BYTES. These functions set the information of |
| 127 | original and decoded texts in the members produced, produced_char, |
| 128 | consumed, and consumed_char of the structure *CODING. |
| 129 | |
| 130 | The return value is an integer (CODING_FINISH_XXX) indicating how |
| 131 | the decoding finished. |
| 132 | |
| 133 | DST_BYTES zero means that source area and destination area are |
| 134 | overlapped, which means that we can produce a decoded text until it |
| 135 | reaches at the head of not-yet-decoded source text. |
| 136 | |
| 137 | Below is a template of these functions. */ |
| 138 | #if 0 |
| 139 | decode_coding_XXX (coding, source, destination, src_bytes, dst_bytes) |
| 140 | struct coding_system *coding; |
| 141 | unsigned char *source, *destination; |
| 142 | int src_bytes, dst_bytes; |
| 143 | { |
| 144 | ... |
| 145 | } |
| 146 | #endif |
| 147 | |
| 148 | /*** GENERAL NOTES on `encode_coding_XXX ()' functions *** |
| 149 | |
| 150 | These functions encode SRC_BYTES length text at SOURCE of Emacs' |
| 151 | internal format (emacs-mule) to CODING. The resulting text goes to |
| 152 | a place pointed to by DESTINATION, the length of which should not |
| 153 | exceed DST_BYTES. These functions set the information of |
| 154 | original and encoded texts in the members produced, produced_char, |
| 155 | consumed, and consumed_char of the structure *CODING. |
| 156 | |
| 157 | The return value is an integer (CODING_FINISH_XXX) indicating how |
| 158 | the encoding finished. |
| 159 | |
| 160 | DST_BYTES zero means that source area and destination area are |
| 161 | overlapped, which means that we can produce a decoded text until it |
| 162 | reaches at the head of not-yet-decoded source text. |
| 163 | |
| 164 | Below is a template of these functions. */ |
| 165 | #if 0 |
| 166 | encode_coding_XXX (coding, source, destination, src_bytes, dst_bytes) |
| 167 | struct coding_system *coding; |
| 168 | unsigned char *source, *destination; |
| 169 | int src_bytes, dst_bytes; |
| 170 | { |
| 171 | ... |
| 172 | } |
| 173 | #endif |
| 174 | |
| 175 | /*** COMMONLY USED MACROS ***/ |
| 176 | |
| 177 | /* The following three macros ONE_MORE_BYTE, TWO_MORE_BYTES, and |
| 178 | THREE_MORE_BYTES safely get one, two, and three bytes from the |
| 179 | source text respectively. If there are not enough bytes in the |
| 180 | source, they jump to `label_end_of_loop'. The caller should set |
| 181 | variables `src' and `src_end' to appropriate areas in advance. */ |
| 182 | |
| 183 | #define ONE_MORE_BYTE(c1) \ |
| 184 | do { \ |
| 185 | if (src < src_end) \ |
| 186 | c1 = *src++; \ |
| 187 | else \ |
| 188 | goto label_end_of_loop; \ |
| 189 | } while (0) |
| 190 | |
| 191 | #define TWO_MORE_BYTES(c1, c2) \ |
| 192 | do { \ |
| 193 | if (src + 1 < src_end) \ |
| 194 | c1 = *src++, c2 = *src++; \ |
| 195 | else \ |
| 196 | goto label_end_of_loop; \ |
| 197 | } while (0) |
| 198 | |
| 199 | #define THREE_MORE_BYTES(c1, c2, c3) \ |
| 200 | do { \ |
| 201 | if (src + 2 < src_end) \ |
| 202 | c1 = *src++, c2 = *src++, c3 = *src++; \ |
| 203 | else \ |
| 204 | goto label_end_of_loop; \ |
| 205 | } while (0) |
| 206 | |
| 207 | /* The following three macros DECODE_CHARACTER_ASCII, |
| 208 | DECODE_CHARACTER_DIMENSION1, and DECODE_CHARACTER_DIMENSION2 put |
| 209 | the multi-byte form of a character of each class at the place |
| 210 | pointed by `dst'. The caller should set the variable `dst' to |
| 211 | point to an appropriate area and the variable `coding' to point to |
| 212 | the coding-system of the currently decoding text in advance. */ |
| 213 | |
| 214 | /* Decode one ASCII character C. */ |
| 215 | |
| 216 | #define DECODE_CHARACTER_ASCII(c) \ |
| 217 | do { \ |
| 218 | if (COMPOSING_P (coding->composing)) \ |
| 219 | *dst++ = 0xA0, *dst++ = (c) | 0x80; \ |
| 220 | else \ |
| 221 | { \ |
| 222 | *dst++ = (c); \ |
| 223 | coding->produced_char++; \ |
| 224 | } \ |
| 225 | } while (0) |
| 226 | |
| 227 | /* Decode one DIMENSION1 character whose charset is CHARSET and whose |
| 228 | position-code is C. */ |
| 229 | |
| 230 | #define DECODE_CHARACTER_DIMENSION1(charset, c) \ |
| 231 | do { \ |
| 232 | unsigned char leading_code = CHARSET_LEADING_CODE_BASE (charset); \ |
| 233 | if (COMPOSING_P (coding->composing)) \ |
| 234 | *dst++ = leading_code + 0x20; \ |
| 235 | else \ |
| 236 | { \ |
| 237 | *dst++ = leading_code; \ |
| 238 | coding->produced_char++; \ |
| 239 | } \ |
| 240 | if (leading_code = CHARSET_LEADING_CODE_EXT (charset)) \ |
| 241 | *dst++ = leading_code; \ |
| 242 | *dst++ = (c) | 0x80; \ |
| 243 | } while (0) |
| 244 | |
| 245 | /* Decode one DIMENSION2 character whose charset is CHARSET and whose |
| 246 | position-codes are C1 and C2. */ |
| 247 | |
| 248 | #define DECODE_CHARACTER_DIMENSION2(charset, c1, c2) \ |
| 249 | do { \ |
| 250 | DECODE_CHARACTER_DIMENSION1 (charset, c1); \ |
| 251 | *dst++ = (c2) | 0x80; \ |
| 252 | } while (0) |
| 253 | |
| 254 | \f |
| 255 | /*** 1. Preamble ***/ |
| 256 | |
| 257 | #include <stdio.h> |
| 258 | |
| 259 | #ifdef emacs |
| 260 | |
| 261 | #include <config.h> |
| 262 | #include "lisp.h" |
| 263 | #include "buffer.h" |
| 264 | #include "charset.h" |
| 265 | #include "ccl.h" |
| 266 | #include "coding.h" |
| 267 | #include "window.h" |
| 268 | |
| 269 | #else /* not emacs */ |
| 270 | |
| 271 | #include "mulelib.h" |
| 272 | |
| 273 | #endif /* not emacs */ |
| 274 | |
| 275 | Lisp_Object Qcoding_system, Qeol_type; |
| 276 | Lisp_Object Qbuffer_file_coding_system; |
| 277 | Lisp_Object Qpost_read_conversion, Qpre_write_conversion; |
| 278 | Lisp_Object Qno_conversion, Qundecided; |
| 279 | Lisp_Object Qcoding_system_history; |
| 280 | Lisp_Object Qsafe_charsets; |
| 281 | Lisp_Object Qvalid_codes; |
| 282 | |
| 283 | extern Lisp_Object Qinsert_file_contents, Qwrite_region; |
| 284 | Lisp_Object Qcall_process, Qcall_process_region, Qprocess_argument; |
| 285 | Lisp_Object Qstart_process, Qopen_network_stream; |
| 286 | Lisp_Object Qtarget_idx; |
| 287 | |
| 288 | Lisp_Object Vselect_safe_coding_system_function; |
| 289 | |
| 290 | /* Mnemonic character of each format of end-of-line. */ |
| 291 | int eol_mnemonic_unix, eol_mnemonic_dos, eol_mnemonic_mac; |
| 292 | /* Mnemonic character to indicate format of end-of-line is not yet |
| 293 | decided. */ |
| 294 | int eol_mnemonic_undecided; |
| 295 | |
| 296 | /* Format of end-of-line decided by system. This is CODING_EOL_LF on |
| 297 | Unix, CODING_EOL_CRLF on DOS/Windows, and CODING_EOL_CR on Mac. */ |
| 298 | int system_eol_type; |
| 299 | |
| 300 | #ifdef emacs |
| 301 | |
| 302 | Lisp_Object Vcoding_system_list, Vcoding_system_alist; |
| 303 | |
| 304 | Lisp_Object Qcoding_system_p, Qcoding_system_error; |
| 305 | |
| 306 | /* Coding system emacs-mule and raw-text are for converting only |
| 307 | end-of-line format. */ |
| 308 | Lisp_Object Qemacs_mule, Qraw_text; |
| 309 | |
| 310 | /* Coding-systems are handed between Emacs Lisp programs and C internal |
| 311 | routines by the following three variables. */ |
| 312 | /* Coding-system for reading files and receiving data from process. */ |
| 313 | Lisp_Object Vcoding_system_for_read; |
| 314 | /* Coding-system for writing files and sending data to process. */ |
| 315 | Lisp_Object Vcoding_system_for_write; |
| 316 | /* Coding-system actually used in the latest I/O. */ |
| 317 | Lisp_Object Vlast_coding_system_used; |
| 318 | |
| 319 | /* A vector of length 256 which contains information about special |
| 320 | Latin codes (especially for dealing with Microsoft codes). */ |
| 321 | Lisp_Object Vlatin_extra_code_table; |
| 322 | |
| 323 | /* Flag to inhibit code conversion of end-of-line format. */ |
| 324 | int inhibit_eol_conversion; |
| 325 | |
| 326 | /* Flag to make buffer-file-coding-system inherit from process-coding. */ |
| 327 | int inherit_process_coding_system; |
| 328 | |
| 329 | /* Coding system to be used to encode text for terminal display. */ |
| 330 | struct coding_system terminal_coding; |
| 331 | |
| 332 | /* Coding system to be used to encode text for terminal display when |
| 333 | terminal coding system is nil. */ |
| 334 | struct coding_system safe_terminal_coding; |
| 335 | |
| 336 | /* Coding system of what is sent from terminal keyboard. */ |
| 337 | struct coding_system keyboard_coding; |
| 338 | |
| 339 | /* Default coding system to be used to write a file. */ |
| 340 | struct coding_system default_buffer_file_coding; |
| 341 | |
| 342 | Lisp_Object Vfile_coding_system_alist; |
| 343 | Lisp_Object Vprocess_coding_system_alist; |
| 344 | Lisp_Object Vnetwork_coding_system_alist; |
| 345 | |
| 346 | #endif /* emacs */ |
| 347 | |
| 348 | Lisp_Object Qcoding_category, Qcoding_category_index; |
| 349 | |
| 350 | /* List of symbols `coding-category-xxx' ordered by priority. */ |
| 351 | Lisp_Object Vcoding_category_list; |
| 352 | |
| 353 | /* Table of coding categories (Lisp symbols). */ |
| 354 | Lisp_Object Vcoding_category_table; |
| 355 | |
| 356 | /* Table of names of symbol for each coding-category. */ |
| 357 | char *coding_category_name[CODING_CATEGORY_IDX_MAX] = { |
| 358 | "coding-category-emacs-mule", |
| 359 | "coding-category-sjis", |
| 360 | "coding-category-iso-7", |
| 361 | "coding-category-iso-7-tight", |
| 362 | "coding-category-iso-8-1", |
| 363 | "coding-category-iso-8-2", |
| 364 | "coding-category-iso-7-else", |
| 365 | "coding-category-iso-8-else", |
| 366 | "coding-category-ccl", |
| 367 | "coding-category-big5", |
| 368 | "coding-category-raw-text", |
| 369 | "coding-category-binary" |
| 370 | }; |
| 371 | |
| 372 | /* Table of pointers to coding systems corresponding to each coding |
| 373 | categories. */ |
| 374 | struct coding_system *coding_system_table[CODING_CATEGORY_IDX_MAX]; |
| 375 | |
| 376 | /* Table of coding category masks. Nth element is a mask for a coding |
| 377 | cateogry of which priority is Nth. */ |
| 378 | static |
| 379 | int coding_priorities[CODING_CATEGORY_IDX_MAX]; |
| 380 | |
| 381 | /* Flag to tell if we look up translation table on character code |
| 382 | conversion. */ |
| 383 | Lisp_Object Venable_character_translation; |
| 384 | /* Standard translation table to look up on decoding (reading). */ |
| 385 | Lisp_Object Vstandard_translation_table_for_decode; |
| 386 | /* Standard translation table to look up on encoding (writing). */ |
| 387 | Lisp_Object Vstandard_translation_table_for_encode; |
| 388 | |
| 389 | Lisp_Object Qtranslation_table; |
| 390 | Lisp_Object Qtranslation_table_id; |
| 391 | Lisp_Object Qtranslation_table_for_decode; |
| 392 | Lisp_Object Qtranslation_table_for_encode; |
| 393 | |
| 394 | /* Alist of charsets vs revision number. */ |
| 395 | Lisp_Object Vcharset_revision_alist; |
| 396 | |
| 397 | /* Default coding systems used for process I/O. */ |
| 398 | Lisp_Object Vdefault_process_coding_system; |
| 399 | |
| 400 | \f |
| 401 | /*** 2. Emacs internal format (emacs-mule) handlers ***/ |
| 402 | |
| 403 | /* Emacs' internal format for encoding multiple character sets is a |
| 404 | kind of multi-byte encoding, i.e. characters are encoded by |
| 405 | variable-length sequences of one-byte codes. ASCII characters |
| 406 | and control characters (e.g. `tab', `newline') are represented by |
| 407 | one-byte sequences which are their ASCII codes, in the range 0x00 |
| 408 | through 0x7F. The other characters are represented by a sequence |
| 409 | of `base leading-code', optional `extended leading-code', and one |
| 410 | or two `position-code's. The length of the sequence is determined |
| 411 | by the base leading-code. Leading-code takes the range 0x80 |
| 412 | through 0x9F, whereas extended leading-code and position-code take |
| 413 | the range 0xA0 through 0xFF. See `charset.h' for more details |
| 414 | about leading-code and position-code. |
| 415 | |
| 416 | There's one exception to this rule. Special leading-code |
| 417 | `leading-code-composition' denotes that the following several |
| 418 | characters should be composed into one character. Leading-codes of |
| 419 | components (except for ASCII) are added 0x20. An ASCII character |
| 420 | component is represented by a 2-byte sequence of `0xA0' and |
| 421 | `ASCII-code + 0x80'. See also the comments in `charset.h' for the |
| 422 | details of composite character. Hence, we can summarize the code |
| 423 | range as follows: |
| 424 | |
| 425 | --- CODE RANGE of Emacs' internal format --- |
| 426 | (character set) (range) |
| 427 | ASCII 0x00 .. 0x7F |
| 428 | ELSE (1st byte) 0x80 .. 0x9F |
| 429 | (rest bytes) 0xA0 .. 0xFF |
| 430 | --------------------------------------------- |
| 431 | |
| 432 | */ |
| 433 | |
| 434 | enum emacs_code_class_type emacs_code_class[256]; |
| 435 | |
| 436 | /* Go to the next statement only if *SRC is accessible and the code is |
| 437 | greater than 0xA0. */ |
| 438 | #define CHECK_CODE_RANGE_A0_FF \ |
| 439 | do { \ |
| 440 | if (src >= src_end) \ |
| 441 | goto label_end_of_switch; \ |
| 442 | else if (*src++ < 0xA0) \ |
| 443 | return 0; \ |
| 444 | } while (0) |
| 445 | |
| 446 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 447 | Check if a text is encoded in Emacs' internal format. If it is, |
| 448 | return CODING_CATEGORY_MASK_EMACS_MULE, else return 0. */ |
| 449 | |
| 450 | int |
| 451 | detect_coding_emacs_mule (src, src_end) |
| 452 | unsigned char *src, *src_end; |
| 453 | { |
| 454 | unsigned char c; |
| 455 | int composing = 0; |
| 456 | |
| 457 | while (src < src_end) |
| 458 | { |
| 459 | c = *src++; |
| 460 | |
| 461 | if (composing) |
| 462 | { |
| 463 | if (c < 0xA0) |
| 464 | composing = 0; |
| 465 | else |
| 466 | c -= 0x20; |
| 467 | } |
| 468 | |
| 469 | switch (emacs_code_class[c]) |
| 470 | { |
| 471 | case EMACS_ascii_code: |
| 472 | case EMACS_linefeed_code: |
| 473 | break; |
| 474 | |
| 475 | case EMACS_control_code: |
| 476 | if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) |
| 477 | return 0; |
| 478 | break; |
| 479 | |
| 480 | case EMACS_invalid_code: |
| 481 | return 0; |
| 482 | |
| 483 | case EMACS_leading_code_composition: /* c == 0x80 */ |
| 484 | if (composing) |
| 485 | CHECK_CODE_RANGE_A0_FF; |
| 486 | else |
| 487 | composing = 1; |
| 488 | break; |
| 489 | |
| 490 | case EMACS_leading_code_4: |
| 491 | CHECK_CODE_RANGE_A0_FF; |
| 492 | /* fall down to check it two more times ... */ |
| 493 | |
| 494 | case EMACS_leading_code_3: |
| 495 | CHECK_CODE_RANGE_A0_FF; |
| 496 | /* fall down to check it one more time ... */ |
| 497 | |
| 498 | case EMACS_leading_code_2: |
| 499 | CHECK_CODE_RANGE_A0_FF; |
| 500 | break; |
| 501 | |
| 502 | default: |
| 503 | label_end_of_switch: |
| 504 | break; |
| 505 | } |
| 506 | } |
| 507 | return CODING_CATEGORY_MASK_EMACS_MULE; |
| 508 | } |
| 509 | |
| 510 | \f |
| 511 | /*** 3. ISO2022 handlers ***/ |
| 512 | |
| 513 | /* The following note describes the coding system ISO2022 briefly. |
| 514 | Since the intention of this note is to help in understanding of |
| 515 | the programs in this file, some parts are NOT ACCURATE or OVERLY |
| 516 | SIMPLIFIED. For the thorough understanding, please refer to the |
| 517 | original document of ISO2022. |
| 518 | |
| 519 | ISO2022 provides many mechanisms to encode several character sets |
| 520 | in 7-bit and 8-bit environment. If one chooses 7-bite environment, |
| 521 | all text is encoded by codes of less than 128. This may make the |
| 522 | encoded text a little bit longer, but the text gets more stability |
| 523 | to pass through several gateways (some of them strip off the MSB). |
| 524 | |
| 525 | There are two kinds of character set: control character set and |
| 526 | graphic character set. The former contains control characters such |
| 527 | as `newline' and `escape' to provide control functions (control |
| 528 | functions are provided also by escape sequences). The latter |
| 529 | contains graphic characters such as ' A' and '-'. Emacs recognizes |
| 530 | two control character sets and many graphic character sets. |
| 531 | |
| 532 | Graphic character sets are classified into one of the following |
| 533 | four classes, DIMENSION1_CHARS94, DIMENSION1_CHARS96, |
| 534 | DIMENSION2_CHARS94, DIMENSION2_CHARS96 according to the number of |
| 535 | bytes (DIMENSION) and the number of characters in one dimension |
| 536 | (CHARS) of the set. In addition, each character set is assigned an |
| 537 | identification tag (called "final character" and denoted as <F> |
| 538 | here after) which is unique in each class. <F> of each character |
| 539 | set is decided by ECMA(*) when it is registered in ISO. Code range |
| 540 | of <F> is 0x30..0x7F (0x30..0x3F are for private use only). |
| 541 | |
| 542 | Note (*): ECMA = European Computer Manufacturers Association |
| 543 | |
| 544 | Here are examples of graphic character set [NAME(<F>)]: |
| 545 | o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ... |
| 546 | o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ... |
| 547 | o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ... |
| 548 | o DIMENSION2_CHARS96 -- none for the moment |
| 549 | |
| 550 | A code area (1byte=8bits) is divided into 4 areas, C0, GL, C1, and GR. |
| 551 | C0 [0x00..0x1F] -- control character plane 0 |
| 552 | GL [0x20..0x7F] -- graphic character plane 0 |
| 553 | C1 [0x80..0x9F] -- control character plane 1 |
| 554 | GR [0xA0..0xFF] -- graphic character plane 1 |
| 555 | |
| 556 | A control character set is directly designated and invoked to C0 or |
| 557 | C1 by an escape sequence. The most common case is that ISO646's |
| 558 | control character set is designated/invoked to C0 and ISO6429's |
| 559 | control character set is designated/invoked to C1, and usually |
| 560 | these designations/invocations are omitted in a coded text. With |
| 561 | 7-bit environment, only C0 can be used, and a control character for |
| 562 | C1 is encoded by an appropriate escape sequence to fit in the |
| 563 | environment. All control characters for C1 are defined the |
| 564 | corresponding escape sequences. |
| 565 | |
| 566 | A graphic character set is at first designated to one of four |
| 567 | graphic registers (G0 through G3), then these graphic registers are |
| 568 | invoked to GL or GR. These designations and invocations can be |
| 569 | done independently. The most common case is that G0 is invoked to |
| 570 | GL, G1 is invoked to GR, and ASCII is designated to G0, and usually |
| 571 | these invocations and designations are omitted in a coded text. |
| 572 | With 7-bit environment, only GL can be used. |
| 573 | |
| 574 | When a graphic character set of CHARS94 is invoked to GL, code 0x20 |
| 575 | and 0x7F of GL area work as control characters SPACE and DEL |
| 576 | respectively, and code 0xA0 and 0xFF of GR area should not be used. |
| 577 | |
| 578 | There are two ways of invocation: locking-shift and single-shift. |
| 579 | With locking-shift, the invocation lasts until the next different |
| 580 | invocation, whereas with single-shift, the invocation works only |
| 581 | for the following character and doesn't affect locking-shift. |
| 582 | Invocations are done by the following control characters or escape |
| 583 | sequences. |
| 584 | |
| 585 | ---------------------------------------------------------------------- |
| 586 | function control char escape sequence description |
| 587 | ---------------------------------------------------------------------- |
| 588 | SI (shift-in) 0x0F none invoke G0 to GL |
| 589 | SO (shift-out) 0x0E none invoke G1 to GL |
| 590 | LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL |
| 591 | LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL |
| 592 | SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 into GL |
| 593 | SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 into GL |
| 594 | ---------------------------------------------------------------------- |
| 595 | The first four are for locking-shift. Control characters for these |
| 596 | functions are defined by macros ISO_CODE_XXX in `coding.h'. |
| 597 | |
| 598 | Designations are done by the following escape sequences. |
| 599 | ---------------------------------------------------------------------- |
| 600 | escape sequence description |
| 601 | ---------------------------------------------------------------------- |
| 602 | ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0 |
| 603 | ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1 |
| 604 | ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2 |
| 605 | ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3 |
| 606 | ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*) |
| 607 | ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1 |
| 608 | ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2 |
| 609 | ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3 |
| 610 | ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**) |
| 611 | ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1 |
| 612 | ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2 |
| 613 | ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3 |
| 614 | ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*) |
| 615 | ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1 |
| 616 | ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2 |
| 617 | ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3 |
| 618 | ---------------------------------------------------------------------- |
| 619 | |
| 620 | In this list, "DIMENSION1_CHARS94<F>" means a graphic character set |
| 621 | of dimension 1, chars 94, and final character <F>, and etc. |
| 622 | |
| 623 | Note (*): Although these designations are not allowed in ISO2022, |
| 624 | Emacs accepts them on decoding, and produces them on encoding |
| 625 | CHARS96 character set in a coding system which is characterized as |
| 626 | 7-bit environment, non-locking-shift, and non-single-shift. |
| 627 | |
| 628 | Note (**): If <F> is '@', 'A', or 'B', the intermediate character |
| 629 | '(' can be omitted. We call this as "short-form" here after. |
| 630 | |
| 631 | Now you may notice that there are a lot of ways for encoding the |
| 632 | same multilingual text in ISO2022. Actually, there exists many |
| 633 | coding systems such as Compound Text (used in X's inter client |
| 634 | communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR |
| 635 | (used in Korean Internet), EUC (Extended UNIX Code, used in Asian |
| 636 | localized platforms), and all of these are variants of ISO2022. |
| 637 | |
| 638 | In addition to the above, Emacs handles two more kinds of escape |
| 639 | sequences: ISO6429's direction specification and Emacs' private |
| 640 | sequence for specifying character composition. |
| 641 | |
| 642 | ISO6429's direction specification takes the following format: |
| 643 | o CSI ']' -- end of the current direction |
| 644 | o CSI '0' ']' -- end of the current direction |
| 645 | o CSI '1' ']' -- start of left-to-right text |
| 646 | o CSI '2' ']' -- start of right-to-left text |
| 647 | The control character CSI (0x9B: control sequence introducer) is |
| 648 | abbreviated to the escape sequence ESC '[' in 7-bit environment. |
| 649 | |
| 650 | Character composition specification takes the following format: |
| 651 | o ESC '0' -- start character composition |
| 652 | o ESC '1' -- end character composition |
| 653 | Since these are not standard escape sequences of any ISO, the use |
| 654 | of them for these meaning is restricted to Emacs only. */ |
| 655 | |
| 656 | enum iso_code_class_type iso_code_class[256]; |
| 657 | |
| 658 | #define CHARSET_OK(idx, charset) \ |
| 659 | (coding_system_table[idx] \ |
| 660 | && (coding_system_table[idx]->safe_charsets[charset] \ |
| 661 | || (CODING_SPEC_ISO_REQUESTED_DESIGNATION \ |
| 662 | (coding_system_table[idx], charset) \ |
| 663 | != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION))) |
| 664 | |
| 665 | #define SHIFT_OUT_OK(idx) \ |
| 666 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding_system_table[idx], 1) >= 0) |
| 667 | |
| 668 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 669 | Check if a text is encoded in ISO2022. If it is, returns an |
| 670 | integer in which appropriate flag bits any of: |
| 671 | CODING_CATEGORY_MASK_ISO_7 |
| 672 | CODING_CATEGORY_MASK_ISO_7_TIGHT |
| 673 | CODING_CATEGORY_MASK_ISO_8_1 |
| 674 | CODING_CATEGORY_MASK_ISO_8_2 |
| 675 | CODING_CATEGORY_MASK_ISO_7_ELSE |
| 676 | CODING_CATEGORY_MASK_ISO_8_ELSE |
| 677 | are set. If a code which should never appear in ISO2022 is found, |
| 678 | returns 0. */ |
| 679 | |
| 680 | int |
| 681 | detect_coding_iso2022 (src, src_end) |
| 682 | unsigned char *src, *src_end; |
| 683 | { |
| 684 | int mask = CODING_CATEGORY_MASK_ISO; |
| 685 | int mask_found = 0; |
| 686 | int reg[4], shift_out = 0, single_shifting = 0; |
| 687 | int c, c1, i, charset; |
| 688 | |
| 689 | reg[0] = CHARSET_ASCII, reg[1] = reg[2] = reg[3] = -1; |
| 690 | while (mask && src < src_end) |
| 691 | { |
| 692 | c = *src++; |
| 693 | switch (c) |
| 694 | { |
| 695 | case ISO_CODE_ESC: |
| 696 | single_shifting = 0; |
| 697 | if (src >= src_end) |
| 698 | break; |
| 699 | c = *src++; |
| 700 | if (c >= '(' && c <= '/') |
| 701 | { |
| 702 | /* Designation sequence for a charset of dimension 1. */ |
| 703 | if (src >= src_end) |
| 704 | break; |
| 705 | c1 = *src++; |
| 706 | if (c1 < ' ' || c1 >= 0x80 |
| 707 | || (charset = iso_charset_table[0][c >= ','][c1]) < 0) |
| 708 | /* Invalid designation sequence. Just ignore. */ |
| 709 | break; |
| 710 | reg[(c - '(') % 4] = charset; |
| 711 | } |
| 712 | else if (c == '$') |
| 713 | { |
| 714 | /* Designation sequence for a charset of dimension 2. */ |
| 715 | if (src >= src_end) |
| 716 | break; |
| 717 | c = *src++; |
| 718 | if (c >= '@' && c <= 'B') |
| 719 | /* Designation for JISX0208.1978, GB2312, or JISX0208. */ |
| 720 | reg[0] = charset = iso_charset_table[1][0][c]; |
| 721 | else if (c >= '(' && c <= '/') |
| 722 | { |
| 723 | if (src >= src_end) |
| 724 | break; |
| 725 | c1 = *src++; |
| 726 | if (c1 < ' ' || c1 >= 0x80 |
| 727 | || (charset = iso_charset_table[1][c >= ','][c1]) < 0) |
| 728 | /* Invalid designation sequence. Just ignore. */ |
| 729 | break; |
| 730 | reg[(c - '(') % 4] = charset; |
| 731 | } |
| 732 | else |
| 733 | /* Invalid designation sequence. Just ignore. */ |
| 734 | break; |
| 735 | } |
| 736 | else if (c == 'N' || c == 'n') |
| 737 | { |
| 738 | if (shift_out == 0 |
| 739 | && (reg[1] >= 0 |
| 740 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE) |
| 741 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))) |
| 742 | { |
| 743 | /* Locking shift out. */ |
| 744 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; |
| 745 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; |
| 746 | shift_out = 1; |
| 747 | } |
| 748 | break; |
| 749 | } |
| 750 | else if (c == 'O' || c == 'o') |
| 751 | { |
| 752 | if (shift_out == 1) |
| 753 | { |
| 754 | /* Locking shift in. */ |
| 755 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; |
| 756 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; |
| 757 | shift_out = 0; |
| 758 | } |
| 759 | break; |
| 760 | } |
| 761 | else if (c == '0' || c == '1' || c == '2') |
| 762 | /* Start/end composition. Just ignore. */ |
| 763 | break; |
| 764 | else |
| 765 | /* Invalid escape sequence. Just ignore. */ |
| 766 | break; |
| 767 | |
| 768 | /* We found a valid designation sequence for CHARSET. */ |
| 769 | mask &= ~CODING_CATEGORY_MASK_ISO_8BIT; |
| 770 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7, charset)) |
| 771 | mask_found |= CODING_CATEGORY_MASK_ISO_7; |
| 772 | else |
| 773 | mask &= ~CODING_CATEGORY_MASK_ISO_7; |
| 774 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT, charset)) |
| 775 | mask_found |= CODING_CATEGORY_MASK_ISO_7_TIGHT; |
| 776 | else |
| 777 | mask &= ~CODING_CATEGORY_MASK_ISO_7_TIGHT; |
| 778 | if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE, charset)) |
| 779 | mask &= ~CODING_CATEGORY_MASK_ISO_7_ELSE; |
| 780 | if (! CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE, charset)) |
| 781 | mask &= ~CODING_CATEGORY_MASK_ISO_8_ELSE; |
| 782 | break; |
| 783 | |
| 784 | case ISO_CODE_SO: |
| 785 | single_shifting = 0; |
| 786 | if (shift_out == 0 |
| 787 | && (reg[1] >= 0 |
| 788 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE) |
| 789 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))) |
| 790 | { |
| 791 | /* Locking shift out. */ |
| 792 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; |
| 793 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; |
| 794 | } |
| 795 | break; |
| 796 | |
| 797 | case ISO_CODE_SI: |
| 798 | single_shifting = 0; |
| 799 | if (shift_out == 1) |
| 800 | { |
| 801 | /* Locking shift in. */ |
| 802 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; |
| 803 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; |
| 804 | } |
| 805 | break; |
| 806 | |
| 807 | case ISO_CODE_CSI: |
| 808 | single_shifting = 0; |
| 809 | case ISO_CODE_SS2: |
| 810 | case ISO_CODE_SS3: |
| 811 | { |
| 812 | int newmask = CODING_CATEGORY_MASK_ISO_8_ELSE; |
| 813 | |
| 814 | if (c != ISO_CODE_CSI) |
| 815 | { |
| 816 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
| 817 | & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 818 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
| 819 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
| 820 | & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 821 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
| 822 | single_shifting = 1; |
| 823 | } |
| 824 | if (VECTORP (Vlatin_extra_code_table) |
| 825 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) |
| 826 | { |
| 827 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
| 828 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 829 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
| 830 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
| 831 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 832 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
| 833 | } |
| 834 | mask &= newmask; |
| 835 | mask_found |= newmask; |
| 836 | } |
| 837 | break; |
| 838 | |
| 839 | default: |
| 840 | if (c < 0x80) |
| 841 | { |
| 842 | single_shifting = 0; |
| 843 | break; |
| 844 | } |
| 845 | else if (c < 0xA0) |
| 846 | { |
| 847 | single_shifting = 0; |
| 848 | if (VECTORP (Vlatin_extra_code_table) |
| 849 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) |
| 850 | { |
| 851 | int newmask = 0; |
| 852 | |
| 853 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
| 854 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 855 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
| 856 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
| 857 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 858 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
| 859 | mask &= newmask; |
| 860 | mask_found |= newmask; |
| 861 | } |
| 862 | else |
| 863 | return 0; |
| 864 | } |
| 865 | else |
| 866 | { |
| 867 | unsigned char *src_begin = src; |
| 868 | |
| 869 | mask &= ~(CODING_CATEGORY_MASK_ISO_7BIT |
| 870 | | CODING_CATEGORY_MASK_ISO_7_ELSE); |
| 871 | mask_found |= CODING_CATEGORY_MASK_ISO_8_1; |
| 872 | /* Check the length of succeeding codes of the range |
| 873 | 0xA0..0FF. If the byte length is odd, we exclude |
| 874 | CODING_CATEGORY_MASK_ISO_8_2. We can check this only |
| 875 | when we are not single shifting. */ |
| 876 | if (!single_shifting) |
| 877 | { |
| 878 | while (src < src_end && *src >= 0xA0) |
| 879 | src++; |
| 880 | if ((src - src_begin - 1) & 1 && src < src_end) |
| 881 | mask &= ~CODING_CATEGORY_MASK_ISO_8_2; |
| 882 | else |
| 883 | mask_found |= CODING_CATEGORY_MASK_ISO_8_2; |
| 884 | } |
| 885 | } |
| 886 | break; |
| 887 | } |
| 888 | } |
| 889 | |
| 890 | return (mask & mask_found); |
| 891 | } |
| 892 | |
| 893 | /* Decode a character of which charset is CHARSET and the 1st position |
| 894 | code is C1. If dimension of CHARSET is 2, the 2nd position code is |
| 895 | fetched from SRC and set to C2. If CHARSET is negative, it means |
| 896 | that we are decoding ill formed text, and what we can do is just to |
| 897 | read C1 as is. */ |
| 898 | |
| 899 | #define DECODE_ISO_CHARACTER(charset, c1) \ |
| 900 | do { \ |
| 901 | int c_alt, charset_alt = (charset); \ |
| 902 | if (COMPOSING_HEAD_P (coding->composing)) \ |
| 903 | { \ |
| 904 | *dst++ = LEADING_CODE_COMPOSITION; \ |
| 905 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ |
| 906 | /* To tell composition rules are embeded. */ \ |
| 907 | *dst++ = 0xFF; \ |
| 908 | coding->composing += 2; \ |
| 909 | } \ |
| 910 | if (charset_alt >= 0) \ |
| 911 | { \ |
| 912 | if (CHARSET_DIMENSION (charset_alt) == 2) \ |
| 913 | { \ |
| 914 | ONE_MORE_BYTE (c2); \ |
| 915 | if (iso_code_class[(c2) & 0x7F] != ISO_0x20_or_0x7F \ |
| 916 | && iso_code_class[(c2) & 0x7F] != ISO_graphic_plane_0) \ |
| 917 | { \ |
| 918 | src--; \ |
| 919 | charset_alt = CHARSET_ASCII; \ |
| 920 | } \ |
| 921 | } \ |
| 922 | if (!NILP (translation_table) \ |
| 923 | && ((c_alt = translate_char (translation_table, \ |
| 924 | -1, charset_alt, c1, c2)) >= 0)) \ |
| 925 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
| 926 | } \ |
| 927 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ |
| 928 | DECODE_CHARACTER_ASCII (c1); \ |
| 929 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ |
| 930 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ |
| 931 | else \ |
| 932 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ |
| 933 | if (COMPOSING_WITH_RULE_P (coding->composing)) \ |
| 934 | /* To tell a composition rule follows. */ \ |
| 935 | coding->composing = COMPOSING_WITH_RULE_RULE; \ |
| 936 | } while (0) |
| 937 | |
| 938 | /* Set designation state into CODING. */ |
| 939 | #define DECODE_DESIGNATION(reg, dimension, chars, final_char) \ |
| 940 | do { \ |
| 941 | int charset = ISO_CHARSET_TABLE (make_number (dimension), \ |
| 942 | make_number (chars), \ |
| 943 | make_number (final_char)); \ |
| 944 | if (charset >= 0 \ |
| 945 | && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) == reg \ |
| 946 | || coding->safe_charsets[charset])) \ |
| 947 | { \ |
| 948 | if (coding->spec.iso2022.last_invalid_designation_register == 0 \ |
| 949 | && reg == 0 \ |
| 950 | && charset == CHARSET_ASCII) \ |
| 951 | { \ |
| 952 | /* We should insert this designation sequence as is so \ |
| 953 | that it is surely written back to a file. */ \ |
| 954 | coding->spec.iso2022.last_invalid_designation_register = -1; \ |
| 955 | goto label_invalid_code; \ |
| 956 | } \ |
| 957 | coding->spec.iso2022.last_invalid_designation_register = -1; \ |
| 958 | if ((coding->mode & CODING_MODE_DIRECTION) \ |
| 959 | && CHARSET_REVERSE_CHARSET (charset) >= 0) \ |
| 960 | charset = CHARSET_REVERSE_CHARSET (charset); \ |
| 961 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ |
| 962 | } \ |
| 963 | else \ |
| 964 | { \ |
| 965 | coding->spec.iso2022.last_invalid_designation_register = reg; \ |
| 966 | goto label_invalid_code; \ |
| 967 | } \ |
| 968 | } while (0) |
| 969 | |
| 970 | /* Check if the current composing sequence contains only valid codes. |
| 971 | If the composing sequence doesn't end before SRC_END, return -1. |
| 972 | Else, if it contains only valid codes, return 0. |
| 973 | Else return the length of the composing sequence. */ |
| 974 | |
| 975 | int |
| 976 | check_composing_code (coding, src, src_end) |
| 977 | struct coding_system *coding; |
| 978 | unsigned char *src, *src_end; |
| 979 | { |
| 980 | unsigned char *src_start = src; |
| 981 | int invalid_code_found = 0; |
| 982 | int charset, c, c1, dim; |
| 983 | |
| 984 | while (src < src_end) |
| 985 | { |
| 986 | if (*src++ != ISO_CODE_ESC) continue; |
| 987 | if (src >= src_end) break; |
| 988 | if ((c = *src++) == '1') /* end of compsition */ |
| 989 | return (invalid_code_found ? src - src_start : 0); |
| 990 | if (src + 2 >= src_end) break; |
| 991 | if (!coding->flags & CODING_FLAG_ISO_DESIGNATION) |
| 992 | invalid_code_found = 1; |
| 993 | else |
| 994 | { |
| 995 | dim = 0; |
| 996 | if (c == '$') |
| 997 | { |
| 998 | dim = 1; |
| 999 | c = (*src >= '@' && *src <= 'B') ? '(' : *src++; |
| 1000 | } |
| 1001 | if (c >= '(' && c <= '/') |
| 1002 | { |
| 1003 | c1 = *src++; |
| 1004 | if ((c1 < ' ' || c1 >= 0x80) |
| 1005 | || (charset = iso_charset_table[dim][c >= ','][c1]) < 0 |
| 1006 | || ! coding->safe_charsets[charset] |
| 1007 | || (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 1008 | == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)) |
| 1009 | invalid_code_found = 1; |
| 1010 | } |
| 1011 | else |
| 1012 | invalid_code_found = 1; |
| 1013 | } |
| 1014 | } |
| 1015 | return (invalid_code_found |
| 1016 | ? src - src_start |
| 1017 | : (coding->mode & CODING_MODE_LAST_BLOCK ? 0 : -1)); |
| 1018 | } |
| 1019 | |
| 1020 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */ |
| 1021 | |
| 1022 | int |
| 1023 | decode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
| 1024 | struct coding_system *coding; |
| 1025 | unsigned char *source, *destination; |
| 1026 | int src_bytes, dst_bytes; |
| 1027 | { |
| 1028 | unsigned char *src = source; |
| 1029 | unsigned char *src_end = source + src_bytes; |
| 1030 | unsigned char *dst = destination; |
| 1031 | unsigned char *dst_end = destination + dst_bytes; |
| 1032 | /* Since the maximum bytes produced by each loop is 7, we subtract 6 |
| 1033 | from DST_END to assure that overflow checking is necessary only |
| 1034 | at the head of loop. */ |
| 1035 | unsigned char *adjusted_dst_end = dst_end - 6; |
| 1036 | int charset; |
| 1037 | /* Charsets invoked to graphic plane 0 and 1 respectively. */ |
| 1038 | int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1039 | int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); |
| 1040 | Lisp_Object translation_table |
| 1041 | = coding->translation_table_for_decode; |
| 1042 | int result = CODING_FINISH_NORMAL; |
| 1043 | |
| 1044 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
| 1045 | translation_table = Vstandard_translation_table_for_decode; |
| 1046 | |
| 1047 | coding->produced_char = 0; |
| 1048 | coding->fake_multibyte = 0; |
| 1049 | while (src < src_end && (dst_bytes |
| 1050 | ? (dst < adjusted_dst_end) |
| 1051 | : (dst < src - 6))) |
| 1052 | { |
| 1053 | /* SRC_BASE remembers the start position in source in each loop. |
| 1054 | The loop will be exited when there's not enough source text |
| 1055 | to analyze long escape sequence or 2-byte code (within macros |
| 1056 | ONE_MORE_BYTE or TWO_MORE_BYTES). In that case, SRC is reset |
| 1057 | to SRC_BASE before exiting. */ |
| 1058 | unsigned char *src_base = src; |
| 1059 | int c1 = *src++, c2; |
| 1060 | |
| 1061 | switch (iso_code_class [c1]) |
| 1062 | { |
| 1063 | case ISO_0x20_or_0x7F: |
| 1064 | if (!coding->composing |
| 1065 | && (charset0 < 0 || CHARSET_CHARS (charset0) == 94)) |
| 1066 | { |
| 1067 | /* This is SPACE or DEL. */ |
| 1068 | *dst++ = c1; |
| 1069 | coding->produced_char++; |
| 1070 | break; |
| 1071 | } |
| 1072 | /* This is a graphic character, we fall down ... */ |
| 1073 | |
| 1074 | case ISO_graphic_plane_0: |
| 1075 | if (coding->composing == COMPOSING_WITH_RULE_RULE) |
| 1076 | { |
| 1077 | /* This is a composition rule. */ |
| 1078 | *dst++ = c1 | 0x80; |
| 1079 | coding->composing = COMPOSING_WITH_RULE_TAIL; |
| 1080 | } |
| 1081 | else |
| 1082 | DECODE_ISO_CHARACTER (charset0, c1); |
| 1083 | break; |
| 1084 | |
| 1085 | case ISO_0xA0_or_0xFF: |
| 1086 | if (charset1 < 0 || CHARSET_CHARS (charset1) == 94 |
| 1087 | || coding->flags & CODING_FLAG_ISO_SEVEN_BITS) |
| 1088 | goto label_invalid_code; |
| 1089 | /* This is a graphic character, we fall down ... */ |
| 1090 | |
| 1091 | case ISO_graphic_plane_1: |
| 1092 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) |
| 1093 | goto label_invalid_code; |
| 1094 | else |
| 1095 | DECODE_ISO_CHARACTER (charset1, c1); |
| 1096 | break; |
| 1097 | |
| 1098 | case ISO_control_code: |
| 1099 | /* All ISO2022 control characters in this class have the |
| 1100 | same representation in Emacs internal format. */ |
| 1101 | if (c1 == '\n' |
| 1102 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 1103 | && (coding->eol_type == CODING_EOL_CR |
| 1104 | || coding->eol_type == CODING_EOL_CRLF)) |
| 1105 | { |
| 1106 | result = CODING_FINISH_INCONSISTENT_EOL; |
| 1107 | goto label_end_of_loop_2; |
| 1108 | } |
| 1109 | *dst++ = c1; |
| 1110 | coding->produced_char++; |
| 1111 | break; |
| 1112 | |
| 1113 | case ISO_carriage_return: |
| 1114 | if (coding->eol_type == CODING_EOL_CR) |
| 1115 | *dst++ = '\n'; |
| 1116 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 1117 | { |
| 1118 | ONE_MORE_BYTE (c1); |
| 1119 | if (c1 == ISO_CODE_LF) |
| 1120 | *dst++ = '\n'; |
| 1121 | else |
| 1122 | { |
| 1123 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 1124 | { |
| 1125 | result = CODING_FINISH_INCONSISTENT_EOL; |
| 1126 | goto label_end_of_loop_2; |
| 1127 | } |
| 1128 | src--; |
| 1129 | *dst++ = '\r'; |
| 1130 | } |
| 1131 | } |
| 1132 | else |
| 1133 | *dst++ = c1; |
| 1134 | coding->produced_char++; |
| 1135 | break; |
| 1136 | |
| 1137 | case ISO_shift_out: |
| 1138 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
| 1139 | || CODING_SPEC_ISO_DESIGNATION (coding, 1) < 0) |
| 1140 | goto label_invalid_code; |
| 1141 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; |
| 1142 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1143 | break; |
| 1144 | |
| 1145 | case ISO_shift_in: |
| 1146 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
| 1147 | goto label_invalid_code; |
| 1148 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; |
| 1149 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1150 | break; |
| 1151 | |
| 1152 | case ISO_single_shift_2_7: |
| 1153 | case ISO_single_shift_2: |
| 1154 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
| 1155 | goto label_invalid_code; |
| 1156 | /* SS2 is handled as an escape sequence of ESC 'N' */ |
| 1157 | c1 = 'N'; |
| 1158 | goto label_escape_sequence; |
| 1159 | |
| 1160 | case ISO_single_shift_3: |
| 1161 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
| 1162 | goto label_invalid_code; |
| 1163 | /* SS2 is handled as an escape sequence of ESC 'O' */ |
| 1164 | c1 = 'O'; |
| 1165 | goto label_escape_sequence; |
| 1166 | |
| 1167 | case ISO_control_sequence_introducer: |
| 1168 | /* CSI is handled as an escape sequence of ESC '[' ... */ |
| 1169 | c1 = '['; |
| 1170 | goto label_escape_sequence; |
| 1171 | |
| 1172 | case ISO_escape: |
| 1173 | ONE_MORE_BYTE (c1); |
| 1174 | label_escape_sequence: |
| 1175 | /* Escape sequences handled by Emacs are invocation, |
| 1176 | designation, direction specification, and character |
| 1177 | composition specification. */ |
| 1178 | switch (c1) |
| 1179 | { |
| 1180 | case '&': /* revision of following character set */ |
| 1181 | ONE_MORE_BYTE (c1); |
| 1182 | if (!(c1 >= '@' && c1 <= '~')) |
| 1183 | goto label_invalid_code; |
| 1184 | ONE_MORE_BYTE (c1); |
| 1185 | if (c1 != ISO_CODE_ESC) |
| 1186 | goto label_invalid_code; |
| 1187 | ONE_MORE_BYTE (c1); |
| 1188 | goto label_escape_sequence; |
| 1189 | |
| 1190 | case '$': /* designation of 2-byte character set */ |
| 1191 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
| 1192 | goto label_invalid_code; |
| 1193 | ONE_MORE_BYTE (c1); |
| 1194 | if (c1 >= '@' && c1 <= 'B') |
| 1195 | { /* designation of JISX0208.1978, GB2312.1980, |
| 1196 | or JISX0208.1980 */ |
| 1197 | DECODE_DESIGNATION (0, 2, 94, c1); |
| 1198 | } |
| 1199 | else if (c1 >= 0x28 && c1 <= 0x2B) |
| 1200 | { /* designation of DIMENSION2_CHARS94 character set */ |
| 1201 | ONE_MORE_BYTE (c2); |
| 1202 | DECODE_DESIGNATION (c1 - 0x28, 2, 94, c2); |
| 1203 | } |
| 1204 | else if (c1 >= 0x2C && c1 <= 0x2F) |
| 1205 | { /* designation of DIMENSION2_CHARS96 character set */ |
| 1206 | ONE_MORE_BYTE (c2); |
| 1207 | DECODE_DESIGNATION (c1 - 0x2C, 2, 96, c2); |
| 1208 | } |
| 1209 | else |
| 1210 | goto label_invalid_code; |
| 1211 | break; |
| 1212 | |
| 1213 | case 'n': /* invocation of locking-shift-2 */ |
| 1214 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
| 1215 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) |
| 1216 | goto label_invalid_code; |
| 1217 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; |
| 1218 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1219 | break; |
| 1220 | |
| 1221 | case 'o': /* invocation of locking-shift-3 */ |
| 1222 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
| 1223 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) |
| 1224 | goto label_invalid_code; |
| 1225 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; |
| 1226 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1227 | break; |
| 1228 | |
| 1229 | case 'N': /* invocation of single-shift-2 */ |
| 1230 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1231 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) |
| 1232 | goto label_invalid_code; |
| 1233 | ONE_MORE_BYTE (c1); |
| 1234 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 2); |
| 1235 | DECODE_ISO_CHARACTER (charset, c1); |
| 1236 | break; |
| 1237 | |
| 1238 | case 'O': /* invocation of single-shift-3 */ |
| 1239 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1240 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) |
| 1241 | goto label_invalid_code; |
| 1242 | ONE_MORE_BYTE (c1); |
| 1243 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 3); |
| 1244 | DECODE_ISO_CHARACTER (charset, c1); |
| 1245 | break; |
| 1246 | |
| 1247 | case '0': case '2': /* start composing */ |
| 1248 | /* Before processing composing, we must be sure that all |
| 1249 | characters being composed are supported by CODING. |
| 1250 | If not, we must give up composing and insert the |
| 1251 | bunch of codes for composing as is without decoding. */ |
| 1252 | { |
| 1253 | int result1; |
| 1254 | |
| 1255 | result1 = check_composing_code (coding, src, src_end); |
| 1256 | if (result1 == 0) |
| 1257 | { |
| 1258 | coding->composing = (c1 == '0' |
| 1259 | ? COMPOSING_NO_RULE_HEAD |
| 1260 | : COMPOSING_WITH_RULE_HEAD); |
| 1261 | coding->produced_char++; |
| 1262 | } |
| 1263 | else if (result1 > 0) |
| 1264 | { |
| 1265 | if (result1 + 2 < (dst_bytes ? dst_end : src_base) - dst) |
| 1266 | { |
| 1267 | bcopy (src_base, dst, result1 + 2); |
| 1268 | src += result1; |
| 1269 | dst += result1 + 2; |
| 1270 | coding->produced_char += result1 + 2; |
| 1271 | } |
| 1272 | else |
| 1273 | { |
| 1274 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 1275 | goto label_end_of_loop_2; |
| 1276 | } |
| 1277 | } |
| 1278 | else |
| 1279 | goto label_end_of_loop; |
| 1280 | } |
| 1281 | break; |
| 1282 | |
| 1283 | case '1': /* end composing */ |
| 1284 | coding->composing = COMPOSING_NO; |
| 1285 | break; |
| 1286 | |
| 1287 | case '[': /* specification of direction */ |
| 1288 | if (coding->flags & CODING_FLAG_ISO_NO_DIRECTION) |
| 1289 | goto label_invalid_code; |
| 1290 | /* For the moment, nested direction is not supported. |
| 1291 | So, `coding->mode & CODING_MODE_DIRECTION' zero means |
| 1292 | left-to-right, and nozero means right-to-left. */ |
| 1293 | ONE_MORE_BYTE (c1); |
| 1294 | switch (c1) |
| 1295 | { |
| 1296 | case ']': /* end of the current direction */ |
| 1297 | coding->mode &= ~CODING_MODE_DIRECTION; |
| 1298 | |
| 1299 | case '0': /* end of the current direction */ |
| 1300 | case '1': /* start of left-to-right direction */ |
| 1301 | ONE_MORE_BYTE (c1); |
| 1302 | if (c1 == ']') |
| 1303 | coding->mode &= ~CODING_MODE_DIRECTION; |
| 1304 | else |
| 1305 | goto label_invalid_code; |
| 1306 | break; |
| 1307 | |
| 1308 | case '2': /* start of right-to-left direction */ |
| 1309 | ONE_MORE_BYTE (c1); |
| 1310 | if (c1 == ']') |
| 1311 | coding->mode |= CODING_MODE_DIRECTION; |
| 1312 | else |
| 1313 | goto label_invalid_code; |
| 1314 | break; |
| 1315 | |
| 1316 | default: |
| 1317 | goto label_invalid_code; |
| 1318 | } |
| 1319 | break; |
| 1320 | |
| 1321 | default: |
| 1322 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
| 1323 | goto label_invalid_code; |
| 1324 | if (c1 >= 0x28 && c1 <= 0x2B) |
| 1325 | { /* designation of DIMENSION1_CHARS94 character set */ |
| 1326 | ONE_MORE_BYTE (c2); |
| 1327 | DECODE_DESIGNATION (c1 - 0x28, 1, 94, c2); |
| 1328 | } |
| 1329 | else if (c1 >= 0x2C && c1 <= 0x2F) |
| 1330 | { /* designation of DIMENSION1_CHARS96 character set */ |
| 1331 | ONE_MORE_BYTE (c2); |
| 1332 | DECODE_DESIGNATION (c1 - 0x2C, 1, 96, c2); |
| 1333 | } |
| 1334 | else |
| 1335 | { |
| 1336 | goto label_invalid_code; |
| 1337 | } |
| 1338 | } |
| 1339 | /* We must update these variables now. */ |
| 1340 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1341 | charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); |
| 1342 | break; |
| 1343 | |
| 1344 | label_invalid_code: |
| 1345 | while (src_base < src) |
| 1346 | *dst++ = *src_base++; |
| 1347 | coding->fake_multibyte = 1; |
| 1348 | } |
| 1349 | continue; |
| 1350 | |
| 1351 | label_end_of_loop: |
| 1352 | result = CODING_FINISH_INSUFFICIENT_SRC; |
| 1353 | label_end_of_loop_2: |
| 1354 | src = src_base; |
| 1355 | break; |
| 1356 | } |
| 1357 | |
| 1358 | if (src < src_end) |
| 1359 | { |
| 1360 | if (result == CODING_FINISH_NORMAL) |
| 1361 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 1362 | else if (result != CODING_FINISH_INCONSISTENT_EOL |
| 1363 | && coding->mode & CODING_MODE_LAST_BLOCK) |
| 1364 | { |
| 1365 | /* This is the last block of the text to be decoded. We had |
| 1366 | better just flush out all remaining codes in the text |
| 1367 | although they are not valid characters. */ |
| 1368 | src_bytes = src_end - src; |
| 1369 | if (dst_bytes && (dst_end - dst < src_bytes)) |
| 1370 | src_bytes = dst_end - dst; |
| 1371 | bcopy (src, dst, src_bytes); |
| 1372 | dst += src_bytes; |
| 1373 | src += src_bytes; |
| 1374 | coding->fake_multibyte = 1; |
| 1375 | } |
| 1376 | } |
| 1377 | |
| 1378 | coding->consumed = coding->consumed_char = src - source; |
| 1379 | coding->produced = dst - destination; |
| 1380 | return result; |
| 1381 | } |
| 1382 | |
| 1383 | /* ISO2022 encoding stuff. */ |
| 1384 | |
| 1385 | /* |
| 1386 | It is not enough to say just "ISO2022" on encoding, we have to |
| 1387 | specify more details. In Emacs, each coding system of ISO2022 |
| 1388 | variant has the following specifications: |
| 1389 | 1. Initial designation to G0 thru G3. |
| 1390 | 2. Allows short-form designation? |
| 1391 | 3. ASCII should be designated to G0 before control characters? |
| 1392 | 4. ASCII should be designated to G0 at end of line? |
| 1393 | 5. 7-bit environment or 8-bit environment? |
| 1394 | 6. Use locking-shift? |
| 1395 | 7. Use Single-shift? |
| 1396 | And the following two are only for Japanese: |
| 1397 | 8. Use ASCII in place of JIS0201-1976-Roman? |
| 1398 | 9. Use JISX0208-1983 in place of JISX0208-1978? |
| 1399 | These specifications are encoded in `coding->flags' as flag bits |
| 1400 | defined by macros CODING_FLAG_ISO_XXX. See `coding.h' for more |
| 1401 | details. |
| 1402 | */ |
| 1403 | |
| 1404 | /* Produce codes (escape sequence) for designating CHARSET to graphic |
| 1405 | register REG. If <final-char> of CHARSET is '@', 'A', or 'B' and |
| 1406 | the coding system CODING allows, produce designation sequence of |
| 1407 | short-form. */ |
| 1408 | |
| 1409 | #define ENCODE_DESIGNATION(charset, reg, coding) \ |
| 1410 | do { \ |
| 1411 | unsigned char final_char = CHARSET_ISO_FINAL_CHAR (charset); \ |
| 1412 | char *intermediate_char_94 = "()*+"; \ |
| 1413 | char *intermediate_char_96 = ",-./"; \ |
| 1414 | int revision = CODING_SPEC_ISO_REVISION_NUMBER(coding, charset); \ |
| 1415 | if (revision < 255) \ |
| 1416 | { \ |
| 1417 | *dst++ = ISO_CODE_ESC; \ |
| 1418 | *dst++ = '&'; \ |
| 1419 | *dst++ = '@' + revision; \ |
| 1420 | } \ |
| 1421 | *dst++ = ISO_CODE_ESC; \ |
| 1422 | if (CHARSET_DIMENSION (charset) == 1) \ |
| 1423 | { \ |
| 1424 | if (CHARSET_CHARS (charset) == 94) \ |
| 1425 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ |
| 1426 | else \ |
| 1427 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ |
| 1428 | } \ |
| 1429 | else \ |
| 1430 | { \ |
| 1431 | *dst++ = '$'; \ |
| 1432 | if (CHARSET_CHARS (charset) == 94) \ |
| 1433 | { \ |
| 1434 | if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \ |
| 1435 | || reg != 0 \ |
| 1436 | || final_char < '@' || final_char > 'B') \ |
| 1437 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ |
| 1438 | } \ |
| 1439 | else \ |
| 1440 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ |
| 1441 | } \ |
| 1442 | *dst++ = final_char; \ |
| 1443 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ |
| 1444 | } while (0) |
| 1445 | |
| 1446 | /* The following two macros produce codes (control character or escape |
| 1447 | sequence) for ISO2022 single-shift functions (single-shift-2 and |
| 1448 | single-shift-3). */ |
| 1449 | |
| 1450 | #define ENCODE_SINGLE_SHIFT_2 \ |
| 1451 | do { \ |
| 1452 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 1453 | *dst++ = ISO_CODE_ESC, *dst++ = 'N'; \ |
| 1454 | else \ |
| 1455 | { \ |
| 1456 | *dst++ = ISO_CODE_SS2; \ |
| 1457 | coding->fake_multibyte = 1; \ |
| 1458 | } \ |
| 1459 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
| 1460 | } while (0) |
| 1461 | |
| 1462 | #define ENCODE_SINGLE_SHIFT_3 \ |
| 1463 | do { \ |
| 1464 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 1465 | *dst++ = ISO_CODE_ESC, *dst++ = 'O'; \ |
| 1466 | else \ |
| 1467 | { \ |
| 1468 | *dst++ = ISO_CODE_SS3; \ |
| 1469 | coding->fake_multibyte = 1; \ |
| 1470 | } \ |
| 1471 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
| 1472 | } while (0) |
| 1473 | |
| 1474 | /* The following four macros produce codes (control character or |
| 1475 | escape sequence) for ISO2022 locking-shift functions (shift-in, |
| 1476 | shift-out, locking-shift-2, and locking-shift-3). */ |
| 1477 | |
| 1478 | #define ENCODE_SHIFT_IN \ |
| 1479 | do { \ |
| 1480 | *dst++ = ISO_CODE_SI; \ |
| 1481 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; \ |
| 1482 | } while (0) |
| 1483 | |
| 1484 | #define ENCODE_SHIFT_OUT \ |
| 1485 | do { \ |
| 1486 | *dst++ = ISO_CODE_SO; \ |
| 1487 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; \ |
| 1488 | } while (0) |
| 1489 | |
| 1490 | #define ENCODE_LOCKING_SHIFT_2 \ |
| 1491 | do { \ |
| 1492 | *dst++ = ISO_CODE_ESC, *dst++ = 'n'; \ |
| 1493 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; \ |
| 1494 | } while (0) |
| 1495 | |
| 1496 | #define ENCODE_LOCKING_SHIFT_3 \ |
| 1497 | do { \ |
| 1498 | *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \ |
| 1499 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; \ |
| 1500 | } while (0) |
| 1501 | |
| 1502 | /* Produce codes for a DIMENSION1 character whose character set is |
| 1503 | CHARSET and whose position-code is C1. Designation and invocation |
| 1504 | sequences are also produced in advance if necessary. */ |
| 1505 | |
| 1506 | |
| 1507 | #define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \ |
| 1508 | do { \ |
| 1509 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ |
| 1510 | { \ |
| 1511 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 1512 | *dst++ = c1 & 0x7F; \ |
| 1513 | else \ |
| 1514 | *dst++ = c1 | 0x80; \ |
| 1515 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ |
| 1516 | break; \ |
| 1517 | } \ |
| 1518 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ |
| 1519 | { \ |
| 1520 | *dst++ = c1 & 0x7F; \ |
| 1521 | break; \ |
| 1522 | } \ |
| 1523 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ |
| 1524 | { \ |
| 1525 | *dst++ = c1 | 0x80; \ |
| 1526 | break; \ |
| 1527 | } \ |
| 1528 | else if (coding->flags & CODING_FLAG_ISO_SAFE \ |
| 1529 | && !coding->safe_charsets[charset]) \ |
| 1530 | { \ |
| 1531 | /* We should not encode this character, instead produce one or \ |
| 1532 | two `?'s. */ \ |
| 1533 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ |
| 1534 | if (CHARSET_WIDTH (charset) == 2) \ |
| 1535 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ |
| 1536 | break; \ |
| 1537 | } \ |
| 1538 | else \ |
| 1539 | /* Since CHARSET is not yet invoked to any graphic planes, we \ |
| 1540 | must invoke it, or, at first, designate it to some graphic \ |
| 1541 | register. Then repeat the loop to actually produce the \ |
| 1542 | character. */ \ |
| 1543 | dst = encode_invocation_designation (charset, coding, dst); \ |
| 1544 | } while (1) |
| 1545 | |
| 1546 | /* Produce codes for a DIMENSION2 character whose character set is |
| 1547 | CHARSET and whose position-codes are C1 and C2. Designation and |
| 1548 | invocation codes are also produced in advance if necessary. */ |
| 1549 | |
| 1550 | #define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2) \ |
| 1551 | do { \ |
| 1552 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ |
| 1553 | { \ |
| 1554 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 1555 | *dst++ = c1 & 0x7F, *dst++ = c2 & 0x7F; \ |
| 1556 | else \ |
| 1557 | *dst++ = c1 | 0x80, *dst++ = c2 | 0x80; \ |
| 1558 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ |
| 1559 | break; \ |
| 1560 | } \ |
| 1561 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ |
| 1562 | { \ |
| 1563 | *dst++ = c1 & 0x7F, *dst++= c2 & 0x7F; \ |
| 1564 | break; \ |
| 1565 | } \ |
| 1566 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ |
| 1567 | { \ |
| 1568 | *dst++ = c1 | 0x80, *dst++= c2 | 0x80; \ |
| 1569 | break; \ |
| 1570 | } \ |
| 1571 | else if (coding->flags & CODING_FLAG_ISO_SAFE \ |
| 1572 | && !coding->safe_charsets[charset]) \ |
| 1573 | { \ |
| 1574 | /* We should not encode this character, instead produce one or \ |
| 1575 | two `?'s. */ \ |
| 1576 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ |
| 1577 | if (CHARSET_WIDTH (charset) == 2) \ |
| 1578 | *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \ |
| 1579 | break; \ |
| 1580 | } \ |
| 1581 | else \ |
| 1582 | /* Since CHARSET is not yet invoked to any graphic planes, we \ |
| 1583 | must invoke it, or, at first, designate it to some graphic \ |
| 1584 | register. Then repeat the loop to actually produce the \ |
| 1585 | character. */ \ |
| 1586 | dst = encode_invocation_designation (charset, coding, dst); \ |
| 1587 | } while (1) |
| 1588 | |
| 1589 | #define ENCODE_ISO_CHARACTER(charset, c1, c2) \ |
| 1590 | do { \ |
| 1591 | int c_alt, charset_alt; \ |
| 1592 | if (!NILP (translation_table) \ |
| 1593 | && ((c_alt = translate_char (translation_table, -1, \ |
| 1594 | charset, c1, c2)) \ |
| 1595 | >= 0)) \ |
| 1596 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
| 1597 | else \ |
| 1598 | charset_alt = charset; \ |
| 1599 | if (CHARSET_DIMENSION (charset_alt) == 1) \ |
| 1600 | { \ |
| 1601 | if (charset == CHARSET_ASCII \ |
| 1602 | && coding->flags & CODING_FLAG_ISO_USE_ROMAN) \ |
| 1603 | charset_alt = charset_latin_jisx0201; \ |
| 1604 | ENCODE_ISO_CHARACTER_DIMENSION1 (charset_alt, c1); \ |
| 1605 | } \ |
| 1606 | else \ |
| 1607 | { \ |
| 1608 | if (charset == charset_jisx0208 \ |
| 1609 | && coding->flags & CODING_FLAG_ISO_USE_OLDJIS) \ |
| 1610 | charset_alt = charset_jisx0208_1978; \ |
| 1611 | ENCODE_ISO_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ |
| 1612 | } \ |
| 1613 | if (! COMPOSING_P (coding->composing)) \ |
| 1614 | coding->consumed_char++; \ |
| 1615 | } while (0) |
| 1616 | |
| 1617 | /* Produce designation and invocation codes at a place pointed by DST |
| 1618 | to use CHARSET. The element `spec.iso2022' of *CODING is updated. |
| 1619 | Return new DST. */ |
| 1620 | |
| 1621 | unsigned char * |
| 1622 | encode_invocation_designation (charset, coding, dst) |
| 1623 | int charset; |
| 1624 | struct coding_system *coding; |
| 1625 | unsigned char *dst; |
| 1626 | { |
| 1627 | int reg; /* graphic register number */ |
| 1628 | |
| 1629 | /* At first, check designations. */ |
| 1630 | for (reg = 0; reg < 4; reg++) |
| 1631 | if (charset == CODING_SPEC_ISO_DESIGNATION (coding, reg)) |
| 1632 | break; |
| 1633 | |
| 1634 | if (reg >= 4) |
| 1635 | { |
| 1636 | /* CHARSET is not yet designated to any graphic registers. */ |
| 1637 | /* At first check the requested designation. */ |
| 1638 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); |
| 1639 | if (reg == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION) |
| 1640 | /* Since CHARSET requests no special designation, designate it |
| 1641 | to graphic register 0. */ |
| 1642 | reg = 0; |
| 1643 | |
| 1644 | ENCODE_DESIGNATION (charset, reg, coding); |
| 1645 | } |
| 1646 | |
| 1647 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != reg |
| 1648 | && CODING_SPEC_ISO_INVOCATION (coding, 1) != reg) |
| 1649 | { |
| 1650 | /* Since the graphic register REG is not invoked to any graphic |
| 1651 | planes, invoke it to graphic plane 0. */ |
| 1652 | switch (reg) |
| 1653 | { |
| 1654 | case 0: /* graphic register 0 */ |
| 1655 | ENCODE_SHIFT_IN; |
| 1656 | break; |
| 1657 | |
| 1658 | case 1: /* graphic register 1 */ |
| 1659 | ENCODE_SHIFT_OUT; |
| 1660 | break; |
| 1661 | |
| 1662 | case 2: /* graphic register 2 */ |
| 1663 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1664 | ENCODE_SINGLE_SHIFT_2; |
| 1665 | else |
| 1666 | ENCODE_LOCKING_SHIFT_2; |
| 1667 | break; |
| 1668 | |
| 1669 | case 3: /* graphic register 3 */ |
| 1670 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1671 | ENCODE_SINGLE_SHIFT_3; |
| 1672 | else |
| 1673 | ENCODE_LOCKING_SHIFT_3; |
| 1674 | break; |
| 1675 | } |
| 1676 | } |
| 1677 | return dst; |
| 1678 | } |
| 1679 | |
| 1680 | /* The following two macros produce codes for indicating composition. */ |
| 1681 | #define ENCODE_COMPOSITION_NO_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '0' |
| 1682 | #define ENCODE_COMPOSITION_WITH_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '2' |
| 1683 | #define ENCODE_COMPOSITION_END *dst++ = ISO_CODE_ESC, *dst++ = '1' |
| 1684 | |
| 1685 | /* The following three macros produce codes for indicating direction |
| 1686 | of text. */ |
| 1687 | #define ENCODE_CONTROL_SEQUENCE_INTRODUCER \ |
| 1688 | do { \ |
| 1689 | if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS) \ |
| 1690 | *dst++ = ISO_CODE_ESC, *dst++ = '['; \ |
| 1691 | else \ |
| 1692 | *dst++ = ISO_CODE_CSI; \ |
| 1693 | } while (0) |
| 1694 | |
| 1695 | #define ENCODE_DIRECTION_R2L \ |
| 1696 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '2', *dst++ = ']' |
| 1697 | |
| 1698 | #define ENCODE_DIRECTION_L2R \ |
| 1699 | ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '0', *dst++ = ']' |
| 1700 | |
| 1701 | /* Produce codes for designation and invocation to reset the graphic |
| 1702 | planes and registers to initial state. */ |
| 1703 | #define ENCODE_RESET_PLANE_AND_REGISTER \ |
| 1704 | do { \ |
| 1705 | int reg; \ |
| 1706 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != 0) \ |
| 1707 | ENCODE_SHIFT_IN; \ |
| 1708 | for (reg = 0; reg < 4; reg++) \ |
| 1709 | if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg) >= 0 \ |
| 1710 | && (CODING_SPEC_ISO_DESIGNATION (coding, reg) \ |
| 1711 | != CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg))) \ |
| 1712 | ENCODE_DESIGNATION \ |
| 1713 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg), reg, coding); \ |
| 1714 | } while (0) |
| 1715 | |
| 1716 | /* Produce designation sequences of charsets in the line started from |
| 1717 | SRC to a place pointed by *DSTP, and update DSTP. |
| 1718 | |
| 1719 | If the current block ends before any end-of-line, we may fail to |
| 1720 | find all the necessary designations. */ |
| 1721 | |
| 1722 | void |
| 1723 | encode_designation_at_bol (coding, table, src, src_end, dstp) |
| 1724 | struct coding_system *coding; |
| 1725 | Lisp_Object table; |
| 1726 | unsigned char *src, *src_end, **dstp; |
| 1727 | { |
| 1728 | int charset, c, found = 0, reg; |
| 1729 | /* Table of charsets to be designated to each graphic register. */ |
| 1730 | int r[4]; |
| 1731 | unsigned char *dst = *dstp; |
| 1732 | |
| 1733 | for (reg = 0; reg < 4; reg++) |
| 1734 | r[reg] = -1; |
| 1735 | |
| 1736 | while (src < src_end && *src != '\n' && found < 4) |
| 1737 | { |
| 1738 | int bytes = BYTES_BY_CHAR_HEAD (*src); |
| 1739 | |
| 1740 | if (NILP (table)) |
| 1741 | charset = CHARSET_AT (src); |
| 1742 | else |
| 1743 | { |
| 1744 | int c_alt; |
| 1745 | unsigned char c1, c2; |
| 1746 | |
| 1747 | SPLIT_STRING(src, bytes, charset, c1, c2); |
| 1748 | if ((c_alt = translate_char (table, -1, charset, c1, c2)) >= 0) |
| 1749 | charset = CHAR_CHARSET (c_alt); |
| 1750 | } |
| 1751 | |
| 1752 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); |
| 1753 | if (reg != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION && r[reg] < 0) |
| 1754 | { |
| 1755 | found++; |
| 1756 | r[reg] = charset; |
| 1757 | } |
| 1758 | |
| 1759 | src += bytes; |
| 1760 | } |
| 1761 | |
| 1762 | if (found) |
| 1763 | { |
| 1764 | for (reg = 0; reg < 4; reg++) |
| 1765 | if (r[reg] >= 0 |
| 1766 | && CODING_SPEC_ISO_DESIGNATION (coding, reg) != r[reg]) |
| 1767 | ENCODE_DESIGNATION (r[reg], reg, coding); |
| 1768 | *dstp = dst; |
| 1769 | } |
| 1770 | } |
| 1771 | |
| 1772 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */ |
| 1773 | |
| 1774 | int |
| 1775 | encode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
| 1776 | struct coding_system *coding; |
| 1777 | unsigned char *source, *destination; |
| 1778 | int src_bytes, dst_bytes; |
| 1779 | { |
| 1780 | unsigned char *src = source; |
| 1781 | unsigned char *src_end = source + src_bytes; |
| 1782 | unsigned char *dst = destination; |
| 1783 | unsigned char *dst_end = destination + dst_bytes; |
| 1784 | /* Since the maximum bytes produced by each loop is 20, we subtract 19 |
| 1785 | from DST_END to assure overflow checking is necessary only at the |
| 1786 | head of loop. */ |
| 1787 | unsigned char *adjusted_dst_end = dst_end - 19; |
| 1788 | Lisp_Object translation_table |
| 1789 | = coding->translation_table_for_encode; |
| 1790 | int result = CODING_FINISH_NORMAL; |
| 1791 | |
| 1792 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
| 1793 | translation_table = Vstandard_translation_table_for_encode; |
| 1794 | |
| 1795 | coding->consumed_char = 0; |
| 1796 | coding->fake_multibyte = 0; |
| 1797 | while (src < src_end && (dst_bytes |
| 1798 | ? (dst < adjusted_dst_end) |
| 1799 | : (dst < src - 19))) |
| 1800 | { |
| 1801 | /* SRC_BASE remembers the start position in source in each loop. |
| 1802 | The loop will be exited when there's not enough source text |
| 1803 | to analyze multi-byte codes (within macros ONE_MORE_BYTE, |
| 1804 | TWO_MORE_BYTES, and THREE_MORE_BYTES). In that case, SRC is |
| 1805 | reset to SRC_BASE before exiting. */ |
| 1806 | unsigned char *src_base = src; |
| 1807 | int charset, c1, c2, c3, c4; |
| 1808 | |
| 1809 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL |
| 1810 | && CODING_SPEC_ISO_BOL (coding)) |
| 1811 | { |
| 1812 | /* We have to produce designation sequences if any now. */ |
| 1813 | encode_designation_at_bol (coding, translation_table, |
| 1814 | src, src_end, &dst); |
| 1815 | CODING_SPEC_ISO_BOL (coding) = 0; |
| 1816 | } |
| 1817 | |
| 1818 | c1 = *src++; |
| 1819 | /* If we are seeing a component of a composite character, we are |
| 1820 | seeing a leading-code encoded irregularly for composition, or |
| 1821 | a composition rule if composing with rule. We must set C1 to |
| 1822 | a normal leading-code or an ASCII code. If we are not seeing |
| 1823 | a composite character, we must reset composition, |
| 1824 | designation, and invocation states. */ |
| 1825 | if (COMPOSING_P (coding->composing)) |
| 1826 | { |
| 1827 | if (c1 < 0xA0) |
| 1828 | { |
| 1829 | /* We are not in a composite character any longer. */ |
| 1830 | coding->composing = COMPOSING_NO; |
| 1831 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 1832 | ENCODE_COMPOSITION_END; |
| 1833 | } |
| 1834 | else |
| 1835 | { |
| 1836 | if (coding->composing == COMPOSING_WITH_RULE_RULE) |
| 1837 | { |
| 1838 | *dst++ = c1 & 0x7F; |
| 1839 | coding->composing = COMPOSING_WITH_RULE_HEAD; |
| 1840 | continue; |
| 1841 | } |
| 1842 | else if (coding->composing == COMPOSING_WITH_RULE_HEAD) |
| 1843 | coding->composing = COMPOSING_WITH_RULE_RULE; |
| 1844 | if (c1 == 0xA0) |
| 1845 | { |
| 1846 | /* This is an ASCII component. */ |
| 1847 | ONE_MORE_BYTE (c1); |
| 1848 | c1 &= 0x7F; |
| 1849 | } |
| 1850 | else |
| 1851 | /* This is a leading-code of non ASCII component. */ |
| 1852 | c1 -= 0x20; |
| 1853 | } |
| 1854 | } |
| 1855 | |
| 1856 | /* Now encode one character. C1 is a control character, an |
| 1857 | ASCII character, or a leading-code of multi-byte character. */ |
| 1858 | switch (emacs_code_class[c1]) |
| 1859 | { |
| 1860 | case EMACS_ascii_code: |
| 1861 | ENCODE_ISO_CHARACTER (CHARSET_ASCII, c1, /* dummy */ c2); |
| 1862 | break; |
| 1863 | |
| 1864 | case EMACS_control_code: |
| 1865 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) |
| 1866 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 1867 | *dst++ = c1; |
| 1868 | coding->consumed_char++; |
| 1869 | break; |
| 1870 | |
| 1871 | case EMACS_carriage_return_code: |
| 1872 | if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
| 1873 | { |
| 1874 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) |
| 1875 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 1876 | *dst++ = c1; |
| 1877 | coding->consumed_char++; |
| 1878 | break; |
| 1879 | } |
| 1880 | /* fall down to treat '\r' as '\n' ... */ |
| 1881 | |
| 1882 | case EMACS_linefeed_code: |
| 1883 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL) |
| 1884 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 1885 | if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL) |
| 1886 | bcopy (coding->spec.iso2022.initial_designation, |
| 1887 | coding->spec.iso2022.current_designation, |
| 1888 | sizeof coding->spec.iso2022.initial_designation); |
| 1889 | if (coding->eol_type == CODING_EOL_LF |
| 1890 | || coding->eol_type == CODING_EOL_UNDECIDED) |
| 1891 | *dst++ = ISO_CODE_LF; |
| 1892 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 1893 | *dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF; |
| 1894 | else |
| 1895 | *dst++ = ISO_CODE_CR; |
| 1896 | CODING_SPEC_ISO_BOL (coding) = 1; |
| 1897 | coding->consumed_char++; |
| 1898 | break; |
| 1899 | |
| 1900 | case EMACS_leading_code_2: |
| 1901 | ONE_MORE_BYTE (c2); |
| 1902 | if (c2 < 0xA0) |
| 1903 | { |
| 1904 | /* invalid sequence */ |
| 1905 | *dst++ = c1; |
| 1906 | src--; |
| 1907 | coding->consumed_char++; |
| 1908 | } |
| 1909 | else |
| 1910 | ENCODE_ISO_CHARACTER (c1, c2, /* dummy */ c3); |
| 1911 | break; |
| 1912 | |
| 1913 | case EMACS_leading_code_3: |
| 1914 | TWO_MORE_BYTES (c2, c3); |
| 1915 | if (c2 < 0xA0 || c3 < 0xA0) |
| 1916 | { |
| 1917 | /* invalid sequence */ |
| 1918 | *dst++ = c1; |
| 1919 | src -= 2; |
| 1920 | coding->consumed_char++; |
| 1921 | } |
| 1922 | else if (c1 < LEADING_CODE_PRIVATE_11) |
| 1923 | ENCODE_ISO_CHARACTER (c1, c2, c3); |
| 1924 | else |
| 1925 | ENCODE_ISO_CHARACTER (c2, c3, /* dummy */ c4); |
| 1926 | break; |
| 1927 | |
| 1928 | case EMACS_leading_code_4: |
| 1929 | THREE_MORE_BYTES (c2, c3, c4); |
| 1930 | if (c2 < 0xA0 || c3 < 0xA0 || c4 < 0xA0) |
| 1931 | { |
| 1932 | /* invalid sequence */ |
| 1933 | *dst++ = c1; |
| 1934 | src -= 3; |
| 1935 | coding->consumed_char++; |
| 1936 | } |
| 1937 | else |
| 1938 | ENCODE_ISO_CHARACTER (c2, c3, c4); |
| 1939 | break; |
| 1940 | |
| 1941 | case EMACS_leading_code_composition: |
| 1942 | ONE_MORE_BYTE (c2); |
| 1943 | if (c2 < 0xA0) |
| 1944 | { |
| 1945 | /* invalid sequence */ |
| 1946 | *dst++ = c1; |
| 1947 | src--; |
| 1948 | coding->consumed_char++; |
| 1949 | } |
| 1950 | else if (c2 == 0xFF) |
| 1951 | { |
| 1952 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 1953 | coding->composing = COMPOSING_WITH_RULE_HEAD; |
| 1954 | ENCODE_COMPOSITION_WITH_RULE_START; |
| 1955 | coding->consumed_char++; |
| 1956 | } |
| 1957 | else |
| 1958 | { |
| 1959 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 1960 | /* Rewind one byte because it is a character code of |
| 1961 | composition elements. */ |
| 1962 | src--; |
| 1963 | coding->composing = COMPOSING_NO_RULE_HEAD; |
| 1964 | ENCODE_COMPOSITION_NO_RULE_START; |
| 1965 | coding->consumed_char++; |
| 1966 | } |
| 1967 | break; |
| 1968 | |
| 1969 | case EMACS_invalid_code: |
| 1970 | *dst++ = c1; |
| 1971 | coding->consumed_char++; |
| 1972 | break; |
| 1973 | } |
| 1974 | continue; |
| 1975 | label_end_of_loop: |
| 1976 | result = CODING_FINISH_INSUFFICIENT_SRC; |
| 1977 | src = src_base; |
| 1978 | break; |
| 1979 | } |
| 1980 | |
| 1981 | if (src < src_end && result == CODING_FINISH_NORMAL) |
| 1982 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 1983 | |
| 1984 | /* If this is the last block of the text to be encoded, we must |
| 1985 | reset graphic planes and registers to the initial state, and |
| 1986 | flush out the carryover if any. */ |
| 1987 | if (coding->mode & CODING_MODE_LAST_BLOCK) |
| 1988 | { |
| 1989 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 1990 | if (COMPOSING_P (coding->composing)) |
| 1991 | ENCODE_COMPOSITION_END; |
| 1992 | } |
| 1993 | coding->consumed = src - source; |
| 1994 | coding->produced = coding->produced_char = dst - destination; |
| 1995 | return result; |
| 1996 | } |
| 1997 | |
| 1998 | \f |
| 1999 | /*** 4. SJIS and BIG5 handlers ***/ |
| 2000 | |
| 2001 | /* Although SJIS and BIG5 are not ISO's coding system, they are used |
| 2002 | quite widely. So, for the moment, Emacs supports them in the bare |
| 2003 | C code. But, in the future, they may be supported only by CCL. */ |
| 2004 | |
| 2005 | /* SJIS is a coding system encoding three character sets: ASCII, right |
| 2006 | half of JISX0201-Kana, and JISX0208. An ASCII character is encoded |
| 2007 | as is. A character of charset katakana-jisx0201 is encoded by |
| 2008 | "position-code + 0x80". A character of charset japanese-jisx0208 |
| 2009 | is encoded in 2-byte but two position-codes are divided and shifted |
| 2010 | so that it fit in the range below. |
| 2011 | |
| 2012 | --- CODE RANGE of SJIS --- |
| 2013 | (character set) (range) |
| 2014 | ASCII 0x00 .. 0x7F |
| 2015 | KATAKANA-JISX0201 0xA0 .. 0xDF |
| 2016 | JISX0208 (1st byte) 0x80 .. 0x9F and 0xE0 .. 0xEF |
| 2017 | (2nd byte) 0x40 .. 0xFF |
| 2018 | ------------------------------- |
| 2019 | |
| 2020 | */ |
| 2021 | |
| 2022 | /* BIG5 is a coding system encoding two character sets: ASCII and |
| 2023 | Big5. An ASCII character is encoded as is. Big5 is a two-byte |
| 2024 | character set and is encoded in two-byte. |
| 2025 | |
| 2026 | --- CODE RANGE of BIG5 --- |
| 2027 | (character set) (range) |
| 2028 | ASCII 0x00 .. 0x7F |
| 2029 | Big5 (1st byte) 0xA1 .. 0xFE |
| 2030 | (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE |
| 2031 | -------------------------- |
| 2032 | |
| 2033 | Since the number of characters in Big5 is larger than maximum |
| 2034 | characters in Emacs' charset (96x96), it can't be handled as one |
| 2035 | charset. So, in Emacs, Big5 is divided into two: `charset-big5-1' |
| 2036 | and `charset-big5-2'. Both are DIMENSION2 and CHARS94. The former |
| 2037 | contains frequently used characters and the latter contains less |
| 2038 | frequently used characters. */ |
| 2039 | |
| 2040 | /* Macros to decode or encode a character of Big5 in BIG5. B1 and B2 |
| 2041 | are the 1st and 2nd position-codes of Big5 in BIG5 coding system. |
| 2042 | C1 and C2 are the 1st and 2nd position-codes of of Emacs' internal |
| 2043 | format. CHARSET is `charset_big5_1' or `charset_big5_2'. */ |
| 2044 | |
| 2045 | /* Number of Big5 characters which have the same code in 1st byte. */ |
| 2046 | #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) |
| 2047 | |
| 2048 | #define DECODE_BIG5(b1, b2, charset, c1, c2) \ |
| 2049 | do { \ |
| 2050 | unsigned int temp \ |
| 2051 | = (b1 - 0xA1) * BIG5_SAME_ROW + b2 - (b2 < 0x7F ? 0x40 : 0x62); \ |
| 2052 | if (b1 < 0xC9) \ |
| 2053 | charset = charset_big5_1; \ |
| 2054 | else \ |
| 2055 | { \ |
| 2056 | charset = charset_big5_2; \ |
| 2057 | temp -= (0xC9 - 0xA1) * BIG5_SAME_ROW; \ |
| 2058 | } \ |
| 2059 | c1 = temp / (0xFF - 0xA1) + 0x21; \ |
| 2060 | c2 = temp % (0xFF - 0xA1) + 0x21; \ |
| 2061 | } while (0) |
| 2062 | |
| 2063 | #define ENCODE_BIG5(charset, c1, c2, b1, b2) \ |
| 2064 | do { \ |
| 2065 | unsigned int temp = (c1 - 0x21) * (0xFF - 0xA1) + (c2 - 0x21); \ |
| 2066 | if (charset == charset_big5_2) \ |
| 2067 | temp += BIG5_SAME_ROW * (0xC9 - 0xA1); \ |
| 2068 | b1 = temp / BIG5_SAME_ROW + 0xA1; \ |
| 2069 | b2 = temp % BIG5_SAME_ROW; \ |
| 2070 | b2 += b2 < 0x3F ? 0x40 : 0x62; \ |
| 2071 | } while (0) |
| 2072 | |
| 2073 | #define DECODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ |
| 2074 | do { \ |
| 2075 | int c_alt, charset_alt = (charset); \ |
| 2076 | if (!NILP (translation_table) \ |
| 2077 | && ((c_alt = translate_char (translation_table, \ |
| 2078 | -1, (charset), c1, c2)) >= 0)) \ |
| 2079 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
| 2080 | if (charset_alt == CHARSET_ASCII || charset_alt < 0) \ |
| 2081 | DECODE_CHARACTER_ASCII (c1); \ |
| 2082 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ |
| 2083 | DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \ |
| 2084 | else \ |
| 2085 | DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \ |
| 2086 | } while (0) |
| 2087 | |
| 2088 | #define ENCODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \ |
| 2089 | do { \ |
| 2090 | int c_alt, charset_alt; \ |
| 2091 | if (!NILP (translation_table) \ |
| 2092 | && ((c_alt = translate_char (translation_table, -1, \ |
| 2093 | charset, c1, c2)) \ |
| 2094 | >= 0)) \ |
| 2095 | SPLIT_CHAR (c_alt, charset_alt, c1, c2); \ |
| 2096 | else \ |
| 2097 | charset_alt = charset; \ |
| 2098 | if (charset_alt == charset_ascii) \ |
| 2099 | *dst++ = c1; \ |
| 2100 | else if (CHARSET_DIMENSION (charset_alt) == 1) \ |
| 2101 | { \ |
| 2102 | if (sjis_p && charset_alt == charset_katakana_jisx0201) \ |
| 2103 | *dst++ = c1; \ |
| 2104 | else \ |
| 2105 | { \ |
| 2106 | *dst++ = charset_alt, *dst++ = c1; \ |
| 2107 | coding->fake_multibyte = 1; \ |
| 2108 | } \ |
| 2109 | } \ |
| 2110 | else \ |
| 2111 | { \ |
| 2112 | c1 &= 0x7F, c2 &= 0x7F; \ |
| 2113 | if (sjis_p && charset_alt == charset_jisx0208) \ |
| 2114 | { \ |
| 2115 | unsigned char s1, s2; \ |
| 2116 | \ |
| 2117 | ENCODE_SJIS (c1, c2, s1, s2); \ |
| 2118 | *dst++ = s1, *dst++ = s2; \ |
| 2119 | coding->fake_multibyte = 1; \ |
| 2120 | } \ |
| 2121 | else if (!sjis_p \ |
| 2122 | && (charset_alt == charset_big5_1 \ |
| 2123 | || charset_alt == charset_big5_2)) \ |
| 2124 | { \ |
| 2125 | unsigned char b1, b2; \ |
| 2126 | \ |
| 2127 | ENCODE_BIG5 (charset_alt, c1, c2, b1, b2); \ |
| 2128 | *dst++ = b1, *dst++ = b2; \ |
| 2129 | } \ |
| 2130 | else \ |
| 2131 | { \ |
| 2132 | *dst++ = charset_alt, *dst++ = c1, *dst++ = c2; \ |
| 2133 | coding->fake_multibyte = 1; \ |
| 2134 | } \ |
| 2135 | } \ |
| 2136 | coding->consumed_char++; \ |
| 2137 | } while (0); |
| 2138 | |
| 2139 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 2140 | Check if a text is encoded in SJIS. If it is, return |
| 2141 | CODING_CATEGORY_MASK_SJIS, else return 0. */ |
| 2142 | |
| 2143 | int |
| 2144 | detect_coding_sjis (src, src_end) |
| 2145 | unsigned char *src, *src_end; |
| 2146 | { |
| 2147 | unsigned char c; |
| 2148 | |
| 2149 | while (src < src_end) |
| 2150 | { |
| 2151 | c = *src++; |
| 2152 | if ((c >= 0x80 && c < 0xA0) || c >= 0xE0) |
| 2153 | { |
| 2154 | if (src < src_end && *src++ < 0x40) |
| 2155 | return 0; |
| 2156 | } |
| 2157 | } |
| 2158 | return CODING_CATEGORY_MASK_SJIS; |
| 2159 | } |
| 2160 | |
| 2161 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 2162 | Check if a text is encoded in BIG5. If it is, return |
| 2163 | CODING_CATEGORY_MASK_BIG5, else return 0. */ |
| 2164 | |
| 2165 | int |
| 2166 | detect_coding_big5 (src, src_end) |
| 2167 | unsigned char *src, *src_end; |
| 2168 | { |
| 2169 | unsigned char c; |
| 2170 | |
| 2171 | while (src < src_end) |
| 2172 | { |
| 2173 | c = *src++; |
| 2174 | if (c >= 0xA1) |
| 2175 | { |
| 2176 | if (src >= src_end) |
| 2177 | break; |
| 2178 | c = *src++; |
| 2179 | if (c < 0x40 || (c >= 0x7F && c <= 0xA0)) |
| 2180 | return 0; |
| 2181 | } |
| 2182 | } |
| 2183 | return CODING_CATEGORY_MASK_BIG5; |
| 2184 | } |
| 2185 | |
| 2186 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". |
| 2187 | If SJIS_P is 1, decode SJIS text, else decode BIG5 test. */ |
| 2188 | |
| 2189 | int |
| 2190 | decode_coding_sjis_big5 (coding, source, destination, |
| 2191 | src_bytes, dst_bytes, sjis_p) |
| 2192 | struct coding_system *coding; |
| 2193 | unsigned char *source, *destination; |
| 2194 | int src_bytes, dst_bytes; |
| 2195 | int sjis_p; |
| 2196 | { |
| 2197 | unsigned char *src = source; |
| 2198 | unsigned char *src_end = source + src_bytes; |
| 2199 | unsigned char *dst = destination; |
| 2200 | unsigned char *dst_end = destination + dst_bytes; |
| 2201 | /* Since the maximum bytes produced by each loop is 4, we subtract 3 |
| 2202 | from DST_END to assure overflow checking is necessary only at the |
| 2203 | head of loop. */ |
| 2204 | unsigned char *adjusted_dst_end = dst_end - 3; |
| 2205 | Lisp_Object translation_table |
| 2206 | = coding->translation_table_for_decode; |
| 2207 | int result = CODING_FINISH_NORMAL; |
| 2208 | |
| 2209 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
| 2210 | translation_table = Vstandard_translation_table_for_decode; |
| 2211 | |
| 2212 | coding->produced_char = 0; |
| 2213 | coding->fake_multibyte = 0; |
| 2214 | while (src < src_end && (dst_bytes |
| 2215 | ? (dst < adjusted_dst_end) |
| 2216 | : (dst < src - 3))) |
| 2217 | { |
| 2218 | /* SRC_BASE remembers the start position in source in each loop. |
| 2219 | The loop will be exited when there's not enough source text |
| 2220 | to analyze two-byte character (within macro ONE_MORE_BYTE). |
| 2221 | In that case, SRC is reset to SRC_BASE before exiting. */ |
| 2222 | unsigned char *src_base = src; |
| 2223 | unsigned char c1 = *src++, c2, c3, c4; |
| 2224 | |
| 2225 | if (c1 < 0x20) |
| 2226 | { |
| 2227 | if (c1 == '\r') |
| 2228 | { |
| 2229 | if (coding->eol_type == CODING_EOL_CRLF) |
| 2230 | { |
| 2231 | ONE_MORE_BYTE (c2); |
| 2232 | if (c2 == '\n') |
| 2233 | *dst++ = c2; |
| 2234 | else if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 2235 | { |
| 2236 | result = CODING_FINISH_INCONSISTENT_EOL; |
| 2237 | goto label_end_of_loop_2; |
| 2238 | } |
| 2239 | else |
| 2240 | /* To process C2 again, SRC is subtracted by 1. */ |
| 2241 | *dst++ = c1, src--; |
| 2242 | } |
| 2243 | else if (coding->eol_type == CODING_EOL_CR) |
| 2244 | *dst++ = '\n'; |
| 2245 | else |
| 2246 | *dst++ = c1; |
| 2247 | } |
| 2248 | else if (c1 == '\n' |
| 2249 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 2250 | && (coding->eol_type == CODING_EOL_CR |
| 2251 | || coding->eol_type == CODING_EOL_CRLF)) |
| 2252 | { |
| 2253 | result = CODING_FINISH_INCONSISTENT_EOL; |
| 2254 | goto label_end_of_loop_2; |
| 2255 | } |
| 2256 | else |
| 2257 | *dst++ = c1; |
| 2258 | coding->produced_char++; |
| 2259 | } |
| 2260 | else if (c1 < 0x80) |
| 2261 | DECODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); |
| 2262 | else |
| 2263 | { |
| 2264 | if (sjis_p) |
| 2265 | { |
| 2266 | if (c1 < 0xA0 || (c1 >= 0xE0 && c1 < 0xF0)) |
| 2267 | { |
| 2268 | /* SJIS -> JISX0208 */ |
| 2269 | ONE_MORE_BYTE (c2); |
| 2270 | if (c2 >= 0x40) |
| 2271 | { |
| 2272 | DECODE_SJIS (c1, c2, c3, c4); |
| 2273 | DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208, c3, c4); |
| 2274 | } |
| 2275 | else |
| 2276 | goto label_invalid_code_2; |
| 2277 | } |
| 2278 | else if (c1 < 0xE0) |
| 2279 | /* SJIS -> JISX0201-Kana */ |
| 2280 | DECODE_SJIS_BIG5_CHARACTER (charset_katakana_jisx0201, c1, |
| 2281 | /* dummy */ c2); |
| 2282 | else |
| 2283 | goto label_invalid_code_1; |
| 2284 | } |
| 2285 | else |
| 2286 | { |
| 2287 | /* BIG5 -> Big5 */ |
| 2288 | if (c1 >= 0xA1 && c1 <= 0xFE) |
| 2289 | { |
| 2290 | ONE_MORE_BYTE (c2); |
| 2291 | if ((c2 >= 0x40 && c2 <= 0x7E) || (c2 >= 0xA1 && c2 <= 0xFE)) |
| 2292 | { |
| 2293 | int charset; |
| 2294 | |
| 2295 | DECODE_BIG5 (c1, c2, charset, c3, c4); |
| 2296 | DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4); |
| 2297 | } |
| 2298 | else |
| 2299 | goto label_invalid_code_2; |
| 2300 | } |
| 2301 | else |
| 2302 | goto label_invalid_code_1; |
| 2303 | } |
| 2304 | } |
| 2305 | continue; |
| 2306 | |
| 2307 | label_invalid_code_1: |
| 2308 | *dst++ = c1; |
| 2309 | coding->produced_char++; |
| 2310 | coding->fake_multibyte = 1; |
| 2311 | continue; |
| 2312 | |
| 2313 | label_invalid_code_2: |
| 2314 | *dst++ = c1; *dst++= c2; |
| 2315 | coding->produced_char += 2; |
| 2316 | coding->fake_multibyte = 1; |
| 2317 | continue; |
| 2318 | |
| 2319 | label_end_of_loop: |
| 2320 | result = CODING_FINISH_INSUFFICIENT_SRC; |
| 2321 | label_end_of_loop_2: |
| 2322 | src = src_base; |
| 2323 | break; |
| 2324 | } |
| 2325 | |
| 2326 | if (src < src_end) |
| 2327 | { |
| 2328 | if (result == CODING_FINISH_NORMAL) |
| 2329 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 2330 | else if (result != CODING_FINISH_INCONSISTENT_EOL |
| 2331 | && coding->mode & CODING_MODE_LAST_BLOCK) |
| 2332 | { |
| 2333 | src_bytes = src_end - src; |
| 2334 | if (dst_bytes && (dst_end - dst < src_bytes)) |
| 2335 | src_bytes = dst_end - dst; |
| 2336 | bcopy (dst, src, src_bytes); |
| 2337 | src += src_bytes; |
| 2338 | dst += src_bytes; |
| 2339 | coding->fake_multibyte = 1; |
| 2340 | } |
| 2341 | } |
| 2342 | |
| 2343 | coding->consumed = coding->consumed_char = src - source; |
| 2344 | coding->produced = dst - destination; |
| 2345 | return result; |
| 2346 | } |
| 2347 | |
| 2348 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". |
| 2349 | This function can encode `charset_ascii', `charset_katakana_jisx0201', |
| 2350 | `charset_jisx0208', `charset_big5_1', and `charset_big5-2'. We are |
| 2351 | sure that all these charsets are registered as official charset |
| 2352 | (i.e. do not have extended leading-codes). Characters of other |
| 2353 | charsets are produced without any encoding. If SJIS_P is 1, encode |
| 2354 | SJIS text, else encode BIG5 text. */ |
| 2355 | |
| 2356 | int |
| 2357 | encode_coding_sjis_big5 (coding, source, destination, |
| 2358 | src_bytes, dst_bytes, sjis_p) |
| 2359 | struct coding_system *coding; |
| 2360 | unsigned char *source, *destination; |
| 2361 | int src_bytes, dst_bytes; |
| 2362 | int sjis_p; |
| 2363 | { |
| 2364 | unsigned char *src = source; |
| 2365 | unsigned char *src_end = source + src_bytes; |
| 2366 | unsigned char *dst = destination; |
| 2367 | unsigned char *dst_end = destination + dst_bytes; |
| 2368 | /* Since the maximum bytes produced by each loop is 2, we subtract 1 |
| 2369 | from DST_END to assure overflow checking is necessary only at the |
| 2370 | head of loop. */ |
| 2371 | unsigned char *adjusted_dst_end = dst_end - 1; |
| 2372 | Lisp_Object translation_table |
| 2373 | = coding->translation_table_for_encode; |
| 2374 | int result = CODING_FINISH_NORMAL; |
| 2375 | |
| 2376 | if (!NILP (Venable_character_translation) && NILP (translation_table)) |
| 2377 | translation_table = Vstandard_translation_table_for_encode; |
| 2378 | |
| 2379 | coding->consumed_char = 0; |
| 2380 | coding->fake_multibyte = 0; |
| 2381 | while (src < src_end && (dst_bytes |
| 2382 | ? (dst < adjusted_dst_end) |
| 2383 | : (dst < src - 1))) |
| 2384 | { |
| 2385 | /* SRC_BASE remembers the start position in source in each loop. |
| 2386 | The loop will be exited when there's not enough source text |
| 2387 | to analyze multi-byte codes (within macros ONE_MORE_BYTE and |
| 2388 | TWO_MORE_BYTES). In that case, SRC is reset to SRC_BASE |
| 2389 | before exiting. */ |
| 2390 | unsigned char *src_base = src; |
| 2391 | unsigned char c1 = *src++, c2, c3, c4; |
| 2392 | |
| 2393 | if (coding->composing) |
| 2394 | { |
| 2395 | if (c1 == 0xA0) |
| 2396 | { |
| 2397 | ONE_MORE_BYTE (c1); |
| 2398 | c1 &= 0x7F; |
| 2399 | } |
| 2400 | else if (c1 >= 0xA0) |
| 2401 | c1 -= 0x20; |
| 2402 | else |
| 2403 | coding->composing = 0; |
| 2404 | } |
| 2405 | |
| 2406 | switch (emacs_code_class[c1]) |
| 2407 | { |
| 2408 | case EMACS_ascii_code: |
| 2409 | ENCODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2); |
| 2410 | break; |
| 2411 | |
| 2412 | case EMACS_control_code: |
| 2413 | *dst++ = c1; |
| 2414 | coding->consumed_char++; |
| 2415 | break; |
| 2416 | |
| 2417 | case EMACS_carriage_return_code: |
| 2418 | if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
| 2419 | { |
| 2420 | *dst++ = c1; |
| 2421 | coding->consumed_char++; |
| 2422 | break; |
| 2423 | } |
| 2424 | /* fall down to treat '\r' as '\n' ... */ |
| 2425 | |
| 2426 | case EMACS_linefeed_code: |
| 2427 | if (coding->eol_type == CODING_EOL_LF |
| 2428 | || coding->eol_type == CODING_EOL_UNDECIDED) |
| 2429 | *dst++ = '\n'; |
| 2430 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 2431 | *dst++ = '\r', *dst++ = '\n'; |
| 2432 | else |
| 2433 | *dst++ = '\r'; |
| 2434 | coding->consumed_char++; |
| 2435 | break; |
| 2436 | |
| 2437 | case EMACS_leading_code_2: |
| 2438 | ONE_MORE_BYTE (c2); |
| 2439 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, /* dummy */ c3); |
| 2440 | break; |
| 2441 | |
| 2442 | case EMACS_leading_code_3: |
| 2443 | TWO_MORE_BYTES (c2, c3); |
| 2444 | ENCODE_SJIS_BIG5_CHARACTER (c1, c2, c3); |
| 2445 | break; |
| 2446 | |
| 2447 | case EMACS_leading_code_4: |
| 2448 | THREE_MORE_BYTES (c2, c3, c4); |
| 2449 | ENCODE_SJIS_BIG5_CHARACTER (c2, c3, c4); |
| 2450 | break; |
| 2451 | |
| 2452 | case EMACS_leading_code_composition: |
| 2453 | coding->composing = 1; |
| 2454 | break; |
| 2455 | |
| 2456 | default: /* i.e. case EMACS_invalid_code: */ |
| 2457 | *dst++ = c1; |
| 2458 | coding->consumed_char++; |
| 2459 | } |
| 2460 | continue; |
| 2461 | |
| 2462 | label_end_of_loop: |
| 2463 | result = CODING_FINISH_INSUFFICIENT_SRC; |
| 2464 | src = src_base; |
| 2465 | break; |
| 2466 | } |
| 2467 | |
| 2468 | if (result == CODING_FINISH_NORMAL |
| 2469 | && src < src_end) |
| 2470 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 2471 | coding->consumed = src - source; |
| 2472 | coding->produced = coding->produced_char = dst - destination; |
| 2473 | return result; |
| 2474 | } |
| 2475 | |
| 2476 | \f |
| 2477 | /*** 5. CCL handlers ***/ |
| 2478 | |
| 2479 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 2480 | Check if a text is encoded in a coding system of which |
| 2481 | encoder/decoder are written in CCL program. If it is, return |
| 2482 | CODING_CATEGORY_MASK_CCL, else return 0. */ |
| 2483 | |
| 2484 | int |
| 2485 | detect_coding_ccl (src, src_end) |
| 2486 | unsigned char *src, *src_end; |
| 2487 | { |
| 2488 | unsigned char *valid; |
| 2489 | |
| 2490 | /* No coding system is assigned to coding-category-ccl. */ |
| 2491 | if (!coding_system_table[CODING_CATEGORY_IDX_CCL]) |
| 2492 | return 0; |
| 2493 | |
| 2494 | valid = coding_system_table[CODING_CATEGORY_IDX_CCL]->spec.ccl.valid_codes; |
| 2495 | while (src < src_end) |
| 2496 | { |
| 2497 | if (! valid[*src]) return 0; |
| 2498 | src++; |
| 2499 | } |
| 2500 | return CODING_CATEGORY_MASK_CCL; |
| 2501 | } |
| 2502 | |
| 2503 | \f |
| 2504 | /*** 6. End-of-line handlers ***/ |
| 2505 | |
| 2506 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". |
| 2507 | This function is called only when `coding->eol_type' is |
| 2508 | CODING_EOL_CRLF or CODING_EOL_CR. */ |
| 2509 | |
| 2510 | int |
| 2511 | decode_eol (coding, source, destination, src_bytes, dst_bytes) |
| 2512 | struct coding_system *coding; |
| 2513 | unsigned char *source, *destination; |
| 2514 | int src_bytes, dst_bytes; |
| 2515 | { |
| 2516 | unsigned char *src = source; |
| 2517 | unsigned char *src_end = source + src_bytes; |
| 2518 | unsigned char *dst = destination; |
| 2519 | unsigned char *dst_end = destination + dst_bytes; |
| 2520 | unsigned char c; |
| 2521 | int result = CODING_FINISH_NORMAL; |
| 2522 | |
| 2523 | coding->fake_multibyte = 0; |
| 2524 | |
| 2525 | if (src_bytes <= 0) |
| 2526 | return result; |
| 2527 | |
| 2528 | switch (coding->eol_type) |
| 2529 | { |
| 2530 | case CODING_EOL_CRLF: |
| 2531 | { |
| 2532 | /* Since the maximum bytes produced by each loop is 2, we |
| 2533 | subtract 1 from DST_END to assure overflow checking is |
| 2534 | necessary only at the head of loop. */ |
| 2535 | unsigned char *adjusted_dst_end = dst_end - 1; |
| 2536 | |
| 2537 | while (src < src_end && (dst_bytes |
| 2538 | ? (dst < adjusted_dst_end) |
| 2539 | : (dst < src - 1))) |
| 2540 | { |
| 2541 | unsigned char *src_base = src; |
| 2542 | |
| 2543 | c = *src++; |
| 2544 | if (c == '\r') |
| 2545 | { |
| 2546 | ONE_MORE_BYTE (c); |
| 2547 | if (c != '\n') |
| 2548 | { |
| 2549 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 2550 | { |
| 2551 | result = CODING_FINISH_INCONSISTENT_EOL; |
| 2552 | goto label_end_of_loop_2; |
| 2553 | } |
| 2554 | *dst++ = '\r'; |
| 2555 | if (BASE_LEADING_CODE_P (c)) |
| 2556 | coding->fake_multibyte = 1; |
| 2557 | } |
| 2558 | *dst++ = c; |
| 2559 | } |
| 2560 | else if (c == '\n' |
| 2561 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)) |
| 2562 | { |
| 2563 | result = CODING_FINISH_INCONSISTENT_EOL; |
| 2564 | goto label_end_of_loop_2; |
| 2565 | } |
| 2566 | else |
| 2567 | { |
| 2568 | *dst++ = c; |
| 2569 | if (BASE_LEADING_CODE_P (c)) |
| 2570 | coding->fake_multibyte = 1; |
| 2571 | } |
| 2572 | continue; |
| 2573 | |
| 2574 | label_end_of_loop: |
| 2575 | result = CODING_FINISH_INSUFFICIENT_SRC; |
| 2576 | label_end_of_loop_2: |
| 2577 | src = src_base; |
| 2578 | break; |
| 2579 | } |
| 2580 | if (result == CODING_FINISH_NORMAL |
| 2581 | && src < src_end) |
| 2582 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 2583 | } |
| 2584 | break; |
| 2585 | |
| 2586 | case CODING_EOL_CR: |
| 2587 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 2588 | { |
| 2589 | while (src < src_end) |
| 2590 | { |
| 2591 | if ((c = *src++) == '\n') |
| 2592 | break; |
| 2593 | if (BASE_LEADING_CODE_P (c)) |
| 2594 | coding->fake_multibyte = 1; |
| 2595 | } |
| 2596 | if (*--src == '\n') |
| 2597 | { |
| 2598 | src_bytes = src - source; |
| 2599 | result = CODING_FINISH_INCONSISTENT_EOL; |
| 2600 | } |
| 2601 | } |
| 2602 | if (dst_bytes && src_bytes > dst_bytes) |
| 2603 | { |
| 2604 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 2605 | src_bytes = dst_bytes; |
| 2606 | } |
| 2607 | if (dst_bytes) |
| 2608 | bcopy (source, destination, src_bytes); |
| 2609 | else |
| 2610 | safe_bcopy (source, destination, src_bytes); |
| 2611 | src = source + src_bytes; |
| 2612 | while (src_bytes--) if (*dst++ == '\r') dst[-1] = '\n'; |
| 2613 | break; |
| 2614 | |
| 2615 | default: /* i.e. case: CODING_EOL_LF */ |
| 2616 | if (dst_bytes && src_bytes > dst_bytes) |
| 2617 | { |
| 2618 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 2619 | src_bytes = dst_bytes; |
| 2620 | } |
| 2621 | if (dst_bytes) |
| 2622 | bcopy (source, destination, src_bytes); |
| 2623 | else |
| 2624 | safe_bcopy (source, destination, src_bytes); |
| 2625 | src += src_bytes; |
| 2626 | dst += src_bytes; |
| 2627 | coding->fake_multibyte = 1; |
| 2628 | break; |
| 2629 | } |
| 2630 | |
| 2631 | coding->consumed = coding->consumed_char = src - source; |
| 2632 | coding->produced = coding->produced_char = dst - destination; |
| 2633 | return result; |
| 2634 | } |
| 2635 | |
| 2636 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". Encode |
| 2637 | format of end-of-line according to `coding->eol_type'. If |
| 2638 | `coding->mode & CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code |
| 2639 | '\r' in source text also means end-of-line. */ |
| 2640 | |
| 2641 | int |
| 2642 | encode_eol (coding, source, destination, src_bytes, dst_bytes) |
| 2643 | struct coding_system *coding; |
| 2644 | unsigned char *source, *destination; |
| 2645 | int src_bytes, dst_bytes; |
| 2646 | { |
| 2647 | unsigned char *src = source; |
| 2648 | unsigned char *dst = destination; |
| 2649 | int result = CODING_FINISH_NORMAL; |
| 2650 | |
| 2651 | coding->fake_multibyte = 0; |
| 2652 | |
| 2653 | if (coding->eol_type == CODING_EOL_CRLF) |
| 2654 | { |
| 2655 | unsigned char c; |
| 2656 | unsigned char *src_end = source + src_bytes; |
| 2657 | unsigned char *dst_end = destination + dst_bytes; |
| 2658 | /* Since the maximum bytes produced by each loop is 2, we |
| 2659 | subtract 1 from DST_END to assure overflow checking is |
| 2660 | necessary only at the head of loop. */ |
| 2661 | unsigned char *adjusted_dst_end = dst_end - 1; |
| 2662 | |
| 2663 | while (src < src_end && (dst_bytes |
| 2664 | ? (dst < adjusted_dst_end) |
| 2665 | : (dst < src - 1))) |
| 2666 | { |
| 2667 | c = *src++; |
| 2668 | if (c == '\n' |
| 2669 | || (c == '\r' && (coding->mode & CODING_MODE_SELECTIVE_DISPLAY))) |
| 2670 | *dst++ = '\r', *dst++ = '\n'; |
| 2671 | else |
| 2672 | { |
| 2673 | *dst++ = c; |
| 2674 | if (BASE_LEADING_CODE_P (c)) |
| 2675 | coding->fake_multibyte = 1; |
| 2676 | } |
| 2677 | } |
| 2678 | if (src < src_end) |
| 2679 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 2680 | } |
| 2681 | else |
| 2682 | { |
| 2683 | unsigned char c; |
| 2684 | |
| 2685 | if (dst_bytes && src_bytes > dst_bytes) |
| 2686 | { |
| 2687 | src_bytes = dst_bytes; |
| 2688 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 2689 | } |
| 2690 | if (dst_bytes) |
| 2691 | bcopy (source, destination, src_bytes); |
| 2692 | else |
| 2693 | safe_bcopy (source, destination, src_bytes); |
| 2694 | dst_bytes = src_bytes; |
| 2695 | if (coding->eol_type == CODING_EOL_CR) |
| 2696 | { |
| 2697 | while (src_bytes--) |
| 2698 | { |
| 2699 | if ((c = *dst++) == '\n') |
| 2700 | dst[-1] = '\r'; |
| 2701 | else if (BASE_LEADING_CODE_P (c)) |
| 2702 | coding->fake_multibyte = 1; |
| 2703 | } |
| 2704 | } |
| 2705 | else |
| 2706 | { |
| 2707 | if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY) |
| 2708 | { |
| 2709 | while (src_bytes--) |
| 2710 | if (*dst++ == '\r') dst[-1] = '\n'; |
| 2711 | } |
| 2712 | coding->fake_multibyte = 1; |
| 2713 | } |
| 2714 | src = source + dst_bytes; |
| 2715 | dst = destination + dst_bytes; |
| 2716 | } |
| 2717 | |
| 2718 | coding->consumed = coding->consumed_char = src - source; |
| 2719 | coding->produced = coding->produced_char = dst - destination; |
| 2720 | return result; |
| 2721 | } |
| 2722 | |
| 2723 | \f |
| 2724 | /*** 7. C library functions ***/ |
| 2725 | |
| 2726 | /* In Emacs Lisp, coding system is represented by a Lisp symbol which |
| 2727 | has a property `coding-system'. The value of this property is a |
| 2728 | vector of length 5 (called as coding-vector). Among elements of |
| 2729 | this vector, the first (element[0]) and the fifth (element[4]) |
| 2730 | carry important information for decoding/encoding. Before |
| 2731 | decoding/encoding, this information should be set in fields of a |
| 2732 | structure of type `coding_system'. |
| 2733 | |
| 2734 | A value of property `coding-system' can be a symbol of another |
| 2735 | subsidiary coding-system. In that case, Emacs gets coding-vector |
| 2736 | from that symbol. |
| 2737 | |
| 2738 | `element[0]' contains information to be set in `coding->type'. The |
| 2739 | value and its meaning is as follows: |
| 2740 | |
| 2741 | 0 -- coding_type_emacs_mule |
| 2742 | 1 -- coding_type_sjis |
| 2743 | 2 -- coding_type_iso2022 |
| 2744 | 3 -- coding_type_big5 |
| 2745 | 4 -- coding_type_ccl encoder/decoder written in CCL |
| 2746 | nil -- coding_type_no_conversion |
| 2747 | t -- coding_type_undecided (automatic conversion on decoding, |
| 2748 | no-conversion on encoding) |
| 2749 | |
| 2750 | `element[4]' contains information to be set in `coding->flags' and |
| 2751 | `coding->spec'. The meaning varies by `coding->type'. |
| 2752 | |
| 2753 | If `coding->type' is `coding_type_iso2022', element[4] is a vector |
| 2754 | of length 32 (of which the first 13 sub-elements are used now). |
| 2755 | Meanings of these sub-elements are: |
| 2756 | |
| 2757 | sub-element[N] where N is 0 through 3: to be set in `coding->spec.iso2022' |
| 2758 | If the value is an integer of valid charset, the charset is |
| 2759 | assumed to be designated to graphic register N initially. |
| 2760 | |
| 2761 | If the value is minus, it is a minus value of charset which |
| 2762 | reserves graphic register N, which means that the charset is |
| 2763 | not designated initially but should be designated to graphic |
| 2764 | register N just before encoding a character in that charset. |
| 2765 | |
| 2766 | If the value is nil, graphic register N is never used on |
| 2767 | encoding. |
| 2768 | |
| 2769 | sub-element[N] where N is 4 through 11: to be set in `coding->flags' |
| 2770 | Each value takes t or nil. See the section ISO2022 of |
| 2771 | `coding.h' for more information. |
| 2772 | |
| 2773 | If `coding->type' is `coding_type_big5', element[4] is t to denote |
| 2774 | BIG5-ETen or nil to denote BIG5-HKU. |
| 2775 | |
| 2776 | If `coding->type' takes the other value, element[4] is ignored. |
| 2777 | |
| 2778 | Emacs Lisp's coding system also carries information about format of |
| 2779 | end-of-line in a value of property `eol-type'. If the value is |
| 2780 | integer, 0 means CODING_EOL_LF, 1 means CODING_EOL_CRLF, and 2 |
| 2781 | means CODING_EOL_CR. If it is not integer, it should be a vector |
| 2782 | of subsidiary coding systems of which property `eol-type' has one |
| 2783 | of above values. |
| 2784 | |
| 2785 | */ |
| 2786 | |
| 2787 | /* Extract information for decoding/encoding from CODING_SYSTEM_SYMBOL |
| 2788 | and set it in CODING. If CODING_SYSTEM_SYMBOL is invalid, CODING |
| 2789 | is setup so that no conversion is necessary and return -1, else |
| 2790 | return 0. */ |
| 2791 | |
| 2792 | int |
| 2793 | setup_coding_system (coding_system, coding) |
| 2794 | Lisp_Object coding_system; |
| 2795 | struct coding_system *coding; |
| 2796 | { |
| 2797 | Lisp_Object coding_spec, coding_type, eol_type, plist; |
| 2798 | Lisp_Object val; |
| 2799 | int i; |
| 2800 | |
| 2801 | /* Initialize some fields required for all kinds of coding systems. */ |
| 2802 | coding->symbol = coding_system; |
| 2803 | coding->common_flags = 0; |
| 2804 | coding->mode = 0; |
| 2805 | coding->heading_ascii = -1; |
| 2806 | coding->post_read_conversion = coding->pre_write_conversion = Qnil; |
| 2807 | coding_spec = Fget (coding_system, Qcoding_system); |
| 2808 | if (!VECTORP (coding_spec) |
| 2809 | || XVECTOR (coding_spec)->size != 5 |
| 2810 | || !CONSP (XVECTOR (coding_spec)->contents[3])) |
| 2811 | goto label_invalid_coding_system; |
| 2812 | |
| 2813 | eol_type = inhibit_eol_conversion ? Qnil : Fget (coding_system, Qeol_type); |
| 2814 | if (VECTORP (eol_type)) |
| 2815 | { |
| 2816 | coding->eol_type = CODING_EOL_UNDECIDED; |
| 2817 | coding->common_flags = CODING_REQUIRE_DETECTION_MASK; |
| 2818 | } |
| 2819 | else if (XFASTINT (eol_type) == 1) |
| 2820 | { |
| 2821 | coding->eol_type = CODING_EOL_CRLF; |
| 2822 | coding->common_flags |
| 2823 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 2824 | } |
| 2825 | else if (XFASTINT (eol_type) == 2) |
| 2826 | { |
| 2827 | coding->eol_type = CODING_EOL_CR; |
| 2828 | coding->common_flags |
| 2829 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 2830 | } |
| 2831 | else |
| 2832 | coding->eol_type = CODING_EOL_LF; |
| 2833 | |
| 2834 | coding_type = XVECTOR (coding_spec)->contents[0]; |
| 2835 | /* Try short cut. */ |
| 2836 | if (SYMBOLP (coding_type)) |
| 2837 | { |
| 2838 | if (EQ (coding_type, Qt)) |
| 2839 | { |
| 2840 | coding->type = coding_type_undecided; |
| 2841 | coding->common_flags |= CODING_REQUIRE_DETECTION_MASK; |
| 2842 | } |
| 2843 | else |
| 2844 | coding->type = coding_type_no_conversion; |
| 2845 | return 0; |
| 2846 | } |
| 2847 | |
| 2848 | /* Initialize remaining fields. */ |
| 2849 | coding->composing = 0; |
| 2850 | coding->translation_table_for_decode = Qnil; |
| 2851 | coding->translation_table_for_encode = Qnil; |
| 2852 | |
| 2853 | /* Get values of coding system properties: |
| 2854 | `post-read-conversion', `pre-write-conversion', |
| 2855 | `translation-table-for-decode', `translation-table-for-encode'. */ |
| 2856 | plist = XVECTOR (coding_spec)->contents[3]; |
| 2857 | coding->post_read_conversion = Fplist_get (plist, Qpost_read_conversion); |
| 2858 | coding->pre_write_conversion = Fplist_get (plist, Qpre_write_conversion); |
| 2859 | val = Fplist_get (plist, Qtranslation_table_for_decode); |
| 2860 | if (SYMBOLP (val)) |
| 2861 | val = Fget (val, Qtranslation_table_for_decode); |
| 2862 | coding->translation_table_for_decode = CHAR_TABLE_P (val) ? val : Qnil; |
| 2863 | val = Fplist_get (plist, Qtranslation_table_for_encode); |
| 2864 | if (SYMBOLP (val)) |
| 2865 | val = Fget (val, Qtranslation_table_for_encode); |
| 2866 | coding->translation_table_for_encode = CHAR_TABLE_P (val) ? val : Qnil; |
| 2867 | val = Fplist_get (plist, Qcoding_category); |
| 2868 | if (!NILP (val)) |
| 2869 | { |
| 2870 | val = Fget (val, Qcoding_category_index); |
| 2871 | if (INTEGERP (val)) |
| 2872 | coding->category_idx = XINT (val); |
| 2873 | else |
| 2874 | goto label_invalid_coding_system; |
| 2875 | } |
| 2876 | else |
| 2877 | goto label_invalid_coding_system; |
| 2878 | |
| 2879 | val = Fplist_get (plist, Qsafe_charsets); |
| 2880 | if (EQ (val, Qt)) |
| 2881 | { |
| 2882 | for (i = 0; i <= MAX_CHARSET; i++) |
| 2883 | coding->safe_charsets[i] = 1; |
| 2884 | } |
| 2885 | else |
| 2886 | { |
| 2887 | bzero (coding->safe_charsets, MAX_CHARSET + 1); |
| 2888 | while (CONSP (val)) |
| 2889 | { |
| 2890 | if ((i = get_charset_id (XCONS (val)->car)) >= 0) |
| 2891 | coding->safe_charsets[i] = 1; |
| 2892 | val = XCONS (val)->cdr; |
| 2893 | } |
| 2894 | } |
| 2895 | |
| 2896 | switch (XFASTINT (coding_type)) |
| 2897 | { |
| 2898 | case 0: |
| 2899 | coding->type = coding_type_emacs_mule; |
| 2900 | if (!NILP (coding->post_read_conversion)) |
| 2901 | coding->common_flags |= CODING_REQUIRE_DECODING_MASK; |
| 2902 | if (!NILP (coding->pre_write_conversion)) |
| 2903 | coding->common_flags |= CODING_REQUIRE_ENCODING_MASK; |
| 2904 | break; |
| 2905 | |
| 2906 | case 1: |
| 2907 | coding->type = coding_type_sjis; |
| 2908 | coding->common_flags |
| 2909 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 2910 | break; |
| 2911 | |
| 2912 | case 2: |
| 2913 | coding->type = coding_type_iso2022; |
| 2914 | coding->common_flags |
| 2915 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 2916 | { |
| 2917 | Lisp_Object val, temp; |
| 2918 | Lisp_Object *flags; |
| 2919 | int i, charset, reg_bits = 0; |
| 2920 | |
| 2921 | val = XVECTOR (coding_spec)->contents[4]; |
| 2922 | |
| 2923 | if (!VECTORP (val) || XVECTOR (val)->size != 32) |
| 2924 | goto label_invalid_coding_system; |
| 2925 | |
| 2926 | flags = XVECTOR (val)->contents; |
| 2927 | coding->flags |
| 2928 | = ((NILP (flags[4]) ? 0 : CODING_FLAG_ISO_SHORT_FORM) |
| 2929 | | (NILP (flags[5]) ? 0 : CODING_FLAG_ISO_RESET_AT_EOL) |
| 2930 | | (NILP (flags[6]) ? 0 : CODING_FLAG_ISO_RESET_AT_CNTL) |
| 2931 | | (NILP (flags[7]) ? 0 : CODING_FLAG_ISO_SEVEN_BITS) |
| 2932 | | (NILP (flags[8]) ? 0 : CODING_FLAG_ISO_LOCKING_SHIFT) |
| 2933 | | (NILP (flags[9]) ? 0 : CODING_FLAG_ISO_SINGLE_SHIFT) |
| 2934 | | (NILP (flags[10]) ? 0 : CODING_FLAG_ISO_USE_ROMAN) |
| 2935 | | (NILP (flags[11]) ? 0 : CODING_FLAG_ISO_USE_OLDJIS) |
| 2936 | | (NILP (flags[12]) ? 0 : CODING_FLAG_ISO_NO_DIRECTION) |
| 2937 | | (NILP (flags[13]) ? 0 : CODING_FLAG_ISO_INIT_AT_BOL) |
| 2938 | | (NILP (flags[14]) ? 0 : CODING_FLAG_ISO_DESIGNATE_AT_BOL) |
| 2939 | | (NILP (flags[15]) ? 0 : CODING_FLAG_ISO_SAFE) |
| 2940 | | (NILP (flags[16]) ? 0 : CODING_FLAG_ISO_LATIN_EXTRA) |
| 2941 | ); |
| 2942 | |
| 2943 | /* Invoke graphic register 0 to plane 0. */ |
| 2944 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; |
| 2945 | /* Invoke graphic register 1 to plane 1 if we can use full 8-bit. */ |
| 2946 | CODING_SPEC_ISO_INVOCATION (coding, 1) |
| 2947 | = (coding->flags & CODING_FLAG_ISO_SEVEN_BITS ? -1 : 1); |
| 2948 | /* Not single shifting at first. */ |
| 2949 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; |
| 2950 | /* Beginning of buffer should also be regarded as bol. */ |
| 2951 | CODING_SPEC_ISO_BOL (coding) = 1; |
| 2952 | |
| 2953 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
| 2954 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = 255; |
| 2955 | val = Vcharset_revision_alist; |
| 2956 | while (CONSP (val)) |
| 2957 | { |
| 2958 | charset = get_charset_id (Fcar_safe (XCONS (val)->car)); |
| 2959 | if (charset >= 0 |
| 2960 | && (temp = Fcdr_safe (XCONS (val)->car), INTEGERP (temp)) |
| 2961 | && (i = XINT (temp), (i >= 0 && (i + '@') < 128))) |
| 2962 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = i; |
| 2963 | val = XCONS (val)->cdr; |
| 2964 | } |
| 2965 | |
| 2966 | /* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations. |
| 2967 | FLAGS[REG] can be one of below: |
| 2968 | integer CHARSET: CHARSET occupies register I, |
| 2969 | t: designate nothing to REG initially, but can be used |
| 2970 | by any charsets, |
| 2971 | list of integer, nil, or t: designate the first |
| 2972 | element (if integer) to REG initially, the remaining |
| 2973 | elements (if integer) is designated to REG on request, |
| 2974 | if an element is t, REG can be used by any charsets, |
| 2975 | nil: REG is never used. */ |
| 2976 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
| 2977 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 2978 | = CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION; |
| 2979 | for (i = 0; i < 4; i++) |
| 2980 | { |
| 2981 | if (INTEGERP (flags[i]) |
| 2982 | && (charset = XINT (flags[i]), CHARSET_VALID_P (charset)) |
| 2983 | || (charset = get_charset_id (flags[i])) >= 0) |
| 2984 | { |
| 2985 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; |
| 2986 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) = i; |
| 2987 | } |
| 2988 | else if (EQ (flags[i], Qt)) |
| 2989 | { |
| 2990 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; |
| 2991 | reg_bits |= 1 << i; |
| 2992 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; |
| 2993 | } |
| 2994 | else if (CONSP (flags[i])) |
| 2995 | { |
| 2996 | Lisp_Object tail; |
| 2997 | tail = flags[i]; |
| 2998 | |
| 2999 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; |
| 3000 | if (INTEGERP (XCONS (tail)->car) |
| 3001 | && (charset = XINT (XCONS (tail)->car), |
| 3002 | CHARSET_VALID_P (charset)) |
| 3003 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) |
| 3004 | { |
| 3005 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; |
| 3006 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) =i; |
| 3007 | } |
| 3008 | else |
| 3009 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; |
| 3010 | tail = XCONS (tail)->cdr; |
| 3011 | while (CONSP (tail)) |
| 3012 | { |
| 3013 | if (INTEGERP (XCONS (tail)->car) |
| 3014 | && (charset = XINT (XCONS (tail)->car), |
| 3015 | CHARSET_VALID_P (charset)) |
| 3016 | || (charset = get_charset_id (XCONS (tail)->car)) >= 0) |
| 3017 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3018 | = i; |
| 3019 | else if (EQ (XCONS (tail)->car, Qt)) |
| 3020 | reg_bits |= 1 << i; |
| 3021 | tail = XCONS (tail)->cdr; |
| 3022 | } |
| 3023 | } |
| 3024 | else |
| 3025 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; |
| 3026 | |
| 3027 | CODING_SPEC_ISO_DESIGNATION (coding, i) |
| 3028 | = CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i); |
| 3029 | } |
| 3030 | |
| 3031 | if (reg_bits && ! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
| 3032 | { |
| 3033 | /* REG 1 can be used only by locking shift in 7-bit env. */ |
| 3034 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) |
| 3035 | reg_bits &= ~2; |
| 3036 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
| 3037 | /* Without any shifting, only REG 0 and 1 can be used. */ |
| 3038 | reg_bits &= 3; |
| 3039 | } |
| 3040 | |
| 3041 | if (reg_bits) |
| 3042 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
| 3043 | { |
| 3044 | if (CHARSET_VALID_P (charset)) |
| 3045 | { |
| 3046 | /* There exist some default graphic registers to be |
| 3047 | used CHARSET. */ |
| 3048 | |
| 3049 | /* We had better avoid designating a charset of |
| 3050 | CHARS96 to REG 0 as far as possible. */ |
| 3051 | if (CHARSET_CHARS (charset) == 96) |
| 3052 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3053 | = (reg_bits & 2 |
| 3054 | ? 1 : (reg_bits & 4 ? 2 : (reg_bits & 8 ? 3 : 0))); |
| 3055 | else |
| 3056 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3057 | = (reg_bits & 1 |
| 3058 | ? 0 : (reg_bits & 2 ? 1 : (reg_bits & 4 ? 2 : 3))); |
| 3059 | } |
| 3060 | } |
| 3061 | } |
| 3062 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
| 3063 | coding->spec.iso2022.last_invalid_designation_register = -1; |
| 3064 | break; |
| 3065 | |
| 3066 | case 3: |
| 3067 | coding->type = coding_type_big5; |
| 3068 | coding->common_flags |
| 3069 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3070 | coding->flags |
| 3071 | = (NILP (XVECTOR (coding_spec)->contents[4]) |
| 3072 | ? CODING_FLAG_BIG5_HKU |
| 3073 | : CODING_FLAG_BIG5_ETEN); |
| 3074 | break; |
| 3075 | |
| 3076 | case 4: |
| 3077 | coding->type = coding_type_ccl; |
| 3078 | coding->common_flags |
| 3079 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3080 | { |
| 3081 | Lisp_Object val; |
| 3082 | Lisp_Object decoder, encoder; |
| 3083 | |
| 3084 | val = XVECTOR (coding_spec)->contents[4]; |
| 3085 | if (CONSP (val) |
| 3086 | && SYMBOLP (XCONS (val)->car) |
| 3087 | && !NILP (decoder = Fget (XCONS (val)->car, Qccl_program_idx)) |
| 3088 | && !NILP (decoder = Fcdr (Faref (Vccl_program_table, decoder))) |
| 3089 | && SYMBOLP (XCONS (val)->cdr) |
| 3090 | && !NILP (encoder = Fget (XCONS (val)->cdr, Qccl_program_idx)) |
| 3091 | && !NILP (encoder = Fcdr (Faref (Vccl_program_table, encoder)))) |
| 3092 | { |
| 3093 | setup_ccl_program (&(coding->spec.ccl.decoder), decoder); |
| 3094 | setup_ccl_program (&(coding->spec.ccl.encoder), encoder); |
| 3095 | } |
| 3096 | else |
| 3097 | goto label_invalid_coding_system; |
| 3098 | |
| 3099 | bzero (coding->spec.ccl.valid_codes, 256); |
| 3100 | val = Fplist_get (plist, Qvalid_codes); |
| 3101 | if (CONSP (val)) |
| 3102 | { |
| 3103 | Lisp_Object this; |
| 3104 | |
| 3105 | for (; CONSP (val); val = XCONS (val)->cdr) |
| 3106 | { |
| 3107 | this = XCONS (val)->car; |
| 3108 | if (INTEGERP (this) |
| 3109 | && XINT (this) >= 0 && XINT (this) < 256) |
| 3110 | coding->spec.ccl.valid_codes[XINT (this)] = 1; |
| 3111 | else if (CONSP (this) |
| 3112 | && INTEGERP (XCONS (this)->car) |
| 3113 | && INTEGERP (XCONS (this)->cdr)) |
| 3114 | { |
| 3115 | int start = XINT (XCONS (this)->car); |
| 3116 | int end = XINT (XCONS (this)->cdr); |
| 3117 | |
| 3118 | if (start >= 0 && start <= end && end < 256) |
| 3119 | while (start < end) |
| 3120 | coding->spec.ccl.valid_codes[start++] = 1; |
| 3121 | } |
| 3122 | } |
| 3123 | } |
| 3124 | } |
| 3125 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
| 3126 | break; |
| 3127 | |
| 3128 | case 5: |
| 3129 | coding->type = coding_type_raw_text; |
| 3130 | break; |
| 3131 | |
| 3132 | default: |
| 3133 | goto label_invalid_coding_system; |
| 3134 | } |
| 3135 | return 0; |
| 3136 | |
| 3137 | label_invalid_coding_system: |
| 3138 | coding->type = coding_type_no_conversion; |
| 3139 | coding->category_idx = CODING_CATEGORY_IDX_BINARY; |
| 3140 | coding->common_flags = 0; |
| 3141 | coding->eol_type = CODING_EOL_LF; |
| 3142 | coding->pre_write_conversion = coding->post_read_conversion = Qnil; |
| 3143 | return -1; |
| 3144 | } |
| 3145 | |
| 3146 | /* Setup raw-text or one of its subsidiaries in the structure |
| 3147 | coding_system CODING according to the already setup value eol_type |
| 3148 | in CODING. CODING should be setup for some coding system in |
| 3149 | advance. */ |
| 3150 | |
| 3151 | void |
| 3152 | setup_raw_text_coding_system (coding) |
| 3153 | struct coding_system *coding; |
| 3154 | { |
| 3155 | if (coding->type != coding_type_raw_text) |
| 3156 | { |
| 3157 | coding->symbol = Qraw_text; |
| 3158 | coding->type = coding_type_raw_text; |
| 3159 | if (coding->eol_type != CODING_EOL_UNDECIDED) |
| 3160 | { |
| 3161 | Lisp_Object subsidiaries; |
| 3162 | subsidiaries = Fget (Qraw_text, Qeol_type); |
| 3163 | |
| 3164 | if (VECTORP (subsidiaries) |
| 3165 | && XVECTOR (subsidiaries)->size == 3) |
| 3166 | coding->symbol |
| 3167 | = XVECTOR (subsidiaries)->contents[coding->eol_type]; |
| 3168 | } |
| 3169 | } |
| 3170 | return; |
| 3171 | } |
| 3172 | |
| 3173 | /* Emacs has a mechanism to automatically detect a coding system if it |
| 3174 | is one of Emacs' internal format, ISO2022, SJIS, and BIG5. But, |
| 3175 | it's impossible to distinguish some coding systems accurately |
| 3176 | because they use the same range of codes. So, at first, coding |
| 3177 | systems are categorized into 7, those are: |
| 3178 | |
| 3179 | o coding-category-emacs-mule |
| 3180 | |
| 3181 | The category for a coding system which has the same code range |
| 3182 | as Emacs' internal format. Assigned the coding-system (Lisp |
| 3183 | symbol) `emacs-mule' by default. |
| 3184 | |
| 3185 | o coding-category-sjis |
| 3186 | |
| 3187 | The category for a coding system which has the same code range |
| 3188 | as SJIS. Assigned the coding-system (Lisp |
| 3189 | symbol) `japanese-shift-jis' by default. |
| 3190 | |
| 3191 | o coding-category-iso-7 |
| 3192 | |
| 3193 | The category for a coding system which has the same code range |
| 3194 | as ISO2022 of 7-bit environment. This doesn't use any locking |
| 3195 | shift and single shift functions. This can encode/decode all |
| 3196 | charsets. Assigned the coding-system (Lisp symbol) |
| 3197 | `iso-2022-7bit' by default. |
| 3198 | |
| 3199 | o coding-category-iso-7-tight |
| 3200 | |
| 3201 | Same as coding-category-iso-7 except that this can |
| 3202 | encode/decode only the specified charsets. |
| 3203 | |
| 3204 | o coding-category-iso-8-1 |
| 3205 | |
| 3206 | The category for a coding system which has the same code range |
| 3207 | as ISO2022 of 8-bit environment and graphic plane 1 used only |
| 3208 | for DIMENSION1 charset. This doesn't use any locking shift |
| 3209 | and single shift functions. Assigned the coding-system (Lisp |
| 3210 | symbol) `iso-latin-1' by default. |
| 3211 | |
| 3212 | o coding-category-iso-8-2 |
| 3213 | |
| 3214 | The category for a coding system which has the same code range |
| 3215 | as ISO2022 of 8-bit environment and graphic plane 1 used only |
| 3216 | for DIMENSION2 charset. This doesn't use any locking shift |
| 3217 | and single shift functions. Assigned the coding-system (Lisp |
| 3218 | symbol) `japanese-iso-8bit' by default. |
| 3219 | |
| 3220 | o coding-category-iso-7-else |
| 3221 | |
| 3222 | The category for a coding system which has the same code range |
| 3223 | as ISO2022 of 7-bit environemnt but uses locking shift or |
| 3224 | single shift functions. Assigned the coding-system (Lisp |
| 3225 | symbol) `iso-2022-7bit-lock' by default. |
| 3226 | |
| 3227 | o coding-category-iso-8-else |
| 3228 | |
| 3229 | The category for a coding system which has the same code range |
| 3230 | as ISO2022 of 8-bit environemnt but uses locking shift or |
| 3231 | single shift functions. Assigned the coding-system (Lisp |
| 3232 | symbol) `iso-2022-8bit-ss2' by default. |
| 3233 | |
| 3234 | o coding-category-big5 |
| 3235 | |
| 3236 | The category for a coding system which has the same code range |
| 3237 | as BIG5. Assigned the coding-system (Lisp symbol) |
| 3238 | `cn-big5' by default. |
| 3239 | |
| 3240 | o coding-category-ccl |
| 3241 | |
| 3242 | The category for a coding system of which encoder/decoder is |
| 3243 | written in CCL programs. The default value is nil, i.e., no |
| 3244 | coding system is assigned. |
| 3245 | |
| 3246 | o coding-category-binary |
| 3247 | |
| 3248 | The category for a coding system not categorized in any of the |
| 3249 | above. Assigned the coding-system (Lisp symbol) |
| 3250 | `no-conversion' by default. |
| 3251 | |
| 3252 | Each of them is a Lisp symbol and the value is an actual |
| 3253 | `coding-system's (this is also a Lisp symbol) assigned by a user. |
| 3254 | What Emacs does actually is to detect a category of coding system. |
| 3255 | Then, it uses a `coding-system' assigned to it. If Emacs can't |
| 3256 | decide only one possible category, it selects a category of the |
| 3257 | highest priority. Priorities of categories are also specified by a |
| 3258 | user in a Lisp variable `coding-category-list'. |
| 3259 | |
| 3260 | */ |
| 3261 | |
| 3262 | static |
| 3263 | int ascii_skip_code[256]; |
| 3264 | |
| 3265 | /* Detect how a text of length SRC_BYTES pointed by SOURCE is encoded. |
| 3266 | If it detects possible coding systems, return an integer in which |
| 3267 | appropriate flag bits are set. Flag bits are defined by macros |
| 3268 | CODING_CATEGORY_MASK_XXX in `coding.h'. |
| 3269 | |
| 3270 | How many ASCII characters are at the head is returned as *SKIP. */ |
| 3271 | |
| 3272 | static int |
| 3273 | detect_coding_mask (source, src_bytes, priorities, skip) |
| 3274 | unsigned char *source; |
| 3275 | int src_bytes, *priorities, *skip; |
| 3276 | { |
| 3277 | register unsigned char c; |
| 3278 | unsigned char *src = source, *src_end = source + src_bytes; |
| 3279 | unsigned int mask; |
| 3280 | int i; |
| 3281 | |
| 3282 | /* At first, skip all ASCII characters and control characters except |
| 3283 | for three ISO2022 specific control characters. */ |
| 3284 | ascii_skip_code[ISO_CODE_SO] = 0; |
| 3285 | ascii_skip_code[ISO_CODE_SI] = 0; |
| 3286 | ascii_skip_code[ISO_CODE_ESC] = 0; |
| 3287 | |
| 3288 | label_loop_detect_coding: |
| 3289 | while (src < src_end && ascii_skip_code[*src]) src++; |
| 3290 | *skip = src - source; |
| 3291 | |
| 3292 | if (src >= src_end) |
| 3293 | /* We found nothing other than ASCII. There's nothing to do. */ |
| 3294 | return 0; |
| 3295 | |
| 3296 | c = *src; |
| 3297 | /* The text seems to be encoded in some multilingual coding system. |
| 3298 | Now, try to find in which coding system the text is encoded. */ |
| 3299 | if (c < 0x80) |
| 3300 | { |
| 3301 | /* i.e. (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) */ |
| 3302 | /* C is an ISO2022 specific control code of C0. */ |
| 3303 | mask = detect_coding_iso2022 (src, src_end); |
| 3304 | if (mask == 0) |
| 3305 | { |
| 3306 | /* No valid ISO2022 code follows C. Try again. */ |
| 3307 | src++; |
| 3308 | if (c == ISO_CODE_ESC) |
| 3309 | ascii_skip_code[ISO_CODE_ESC] = 1; |
| 3310 | else |
| 3311 | ascii_skip_code[ISO_CODE_SO] = ascii_skip_code[ISO_CODE_SI] = 1; |
| 3312 | goto label_loop_detect_coding; |
| 3313 | } |
| 3314 | if (priorities) |
| 3315 | goto label_return_highest_only; |
| 3316 | } |
| 3317 | else |
| 3318 | { |
| 3319 | int try; |
| 3320 | |
| 3321 | if (c < 0xA0) |
| 3322 | { |
| 3323 | /* C is the first byte of SJIS character code, |
| 3324 | or a leading-code of Emacs' internal format (emacs-mule). */ |
| 3325 | try = CODING_CATEGORY_MASK_SJIS | CODING_CATEGORY_MASK_EMACS_MULE; |
| 3326 | |
| 3327 | /* Or, if C is a special latin extra code, |
| 3328 | or is an ISO2022 specific control code of C1 (SS2 or SS3), |
| 3329 | or is an ISO2022 control-sequence-introducer (CSI), |
| 3330 | we should also consider the possibility of ISO2022 codings. */ |
| 3331 | if ((VECTORP (Vlatin_extra_code_table) |
| 3332 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) |
| 3333 | || (c == ISO_CODE_SS2 || c == ISO_CODE_SS3) |
| 3334 | || (c == ISO_CODE_CSI |
| 3335 | && (src < src_end |
| 3336 | && (*src == ']' |
| 3337 | || ((*src == '0' || *src == '1' || *src == '2') |
| 3338 | && src + 1 < src_end |
| 3339 | && src[1] == ']'))))) |
| 3340 | try |= (CODING_CATEGORY_MASK_ISO_8_ELSE |
| 3341 | | CODING_CATEGORY_MASK_ISO_8BIT); |
| 3342 | } |
| 3343 | else |
| 3344 | /* C is a character of ISO2022 in graphic plane right, |
| 3345 | or a SJIS's 1-byte character code (i.e. JISX0201), |
| 3346 | or the first byte of BIG5's 2-byte code. */ |
| 3347 | try = (CODING_CATEGORY_MASK_ISO_8_ELSE |
| 3348 | | CODING_CATEGORY_MASK_ISO_8BIT |
| 3349 | | CODING_CATEGORY_MASK_SJIS |
| 3350 | | CODING_CATEGORY_MASK_BIG5); |
| 3351 | |
| 3352 | /* Or, we may have to consider the possibility of CCL. */ |
| 3353 | if (coding_system_table[CODING_CATEGORY_IDX_CCL] |
| 3354 | && (coding_system_table[CODING_CATEGORY_IDX_CCL] |
| 3355 | ->spec.ccl.valid_codes)[c]) |
| 3356 | try |= CODING_CATEGORY_MASK_CCL; |
| 3357 | |
| 3358 | mask = 0; |
| 3359 | if (priorities) |
| 3360 | { |
| 3361 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) |
| 3362 | { |
| 3363 | if (priorities[i] & try & CODING_CATEGORY_MASK_ISO) |
| 3364 | mask = detect_coding_iso2022 (src, src_end); |
| 3365 | else if (priorities[i] & try & CODING_CATEGORY_MASK_SJIS) |
| 3366 | mask = detect_coding_sjis (src, src_end); |
| 3367 | else if (priorities[i] & try & CODING_CATEGORY_MASK_BIG5) |
| 3368 | mask = detect_coding_big5 (src, src_end); |
| 3369 | else if (priorities[i] & try & CODING_CATEGORY_MASK_EMACS_MULE) |
| 3370 | mask = detect_coding_emacs_mule (src, src_end); |
| 3371 | else if (priorities[i] & try & CODING_CATEGORY_MASK_CCL) |
| 3372 | mask = detect_coding_ccl (src, src_end); |
| 3373 | else if (priorities[i] & CODING_CATEGORY_MASK_RAW_TEXT) |
| 3374 | mask = CODING_CATEGORY_MASK_RAW_TEXT; |
| 3375 | else if (priorities[i] & CODING_CATEGORY_MASK_BINARY) |
| 3376 | mask = CODING_CATEGORY_MASK_BINARY; |
| 3377 | if (mask) |
| 3378 | goto label_return_highest_only; |
| 3379 | } |
| 3380 | return CODING_CATEGORY_MASK_RAW_TEXT; |
| 3381 | } |
| 3382 | if (try & CODING_CATEGORY_MASK_ISO) |
| 3383 | mask |= detect_coding_iso2022 (src, src_end); |
| 3384 | if (try & CODING_CATEGORY_MASK_SJIS) |
| 3385 | mask |= detect_coding_sjis (src, src_end); |
| 3386 | if (try & CODING_CATEGORY_MASK_BIG5) |
| 3387 | mask |= detect_coding_big5 (src, src_end); |
| 3388 | if (try & CODING_CATEGORY_MASK_EMACS_MULE) |
| 3389 | mask |= detect_coding_emacs_mule (src, src_end); |
| 3390 | if (try & CODING_CATEGORY_MASK_CCL) |
| 3391 | mask |= detect_coding_ccl (src, src_end); |
| 3392 | } |
| 3393 | return (mask | CODING_CATEGORY_MASK_RAW_TEXT | CODING_CATEGORY_MASK_BINARY); |
| 3394 | |
| 3395 | label_return_highest_only: |
| 3396 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) |
| 3397 | { |
| 3398 | if (mask & priorities[i]) |
| 3399 | return priorities[i]; |
| 3400 | } |
| 3401 | return CODING_CATEGORY_MASK_RAW_TEXT; |
| 3402 | } |
| 3403 | |
| 3404 | /* Detect how a text of length SRC_BYTES pointed by SRC is encoded. |
| 3405 | The information of the detected coding system is set in CODING. */ |
| 3406 | |
| 3407 | void |
| 3408 | detect_coding (coding, src, src_bytes) |
| 3409 | struct coding_system *coding; |
| 3410 | unsigned char *src; |
| 3411 | int src_bytes; |
| 3412 | { |
| 3413 | unsigned int idx; |
| 3414 | int skip, mask, i; |
| 3415 | Lisp_Object val; |
| 3416 | |
| 3417 | val = Vcoding_category_list; |
| 3418 | mask = detect_coding_mask (src, src_bytes, coding_priorities, &skip); |
| 3419 | coding->heading_ascii = skip; |
| 3420 | |
| 3421 | if (!mask) return; |
| 3422 | |
| 3423 | /* We found a single coding system of the highest priority in MASK. */ |
| 3424 | idx = 0; |
| 3425 | while (mask && ! (mask & 1)) mask >>= 1, idx++; |
| 3426 | if (! mask) |
| 3427 | idx = CODING_CATEGORY_IDX_RAW_TEXT; |
| 3428 | |
| 3429 | val = XSYMBOL (XVECTOR (Vcoding_category_table)->contents[idx])->value; |
| 3430 | |
| 3431 | if (coding->eol_type != CODING_EOL_UNDECIDED) |
| 3432 | { |
| 3433 | Lisp_Object tmp; |
| 3434 | |
| 3435 | tmp = Fget (val, Qeol_type); |
| 3436 | if (VECTORP (tmp)) |
| 3437 | val = XVECTOR (tmp)->contents[coding->eol_type]; |
| 3438 | } |
| 3439 | setup_coding_system (val, coding); |
| 3440 | /* Set this again because setup_coding_system reset this member. */ |
| 3441 | coding->heading_ascii = skip; |
| 3442 | } |
| 3443 | |
| 3444 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by |
| 3445 | SOURCE is encoded. Return one of CODING_EOL_LF, CODING_EOL_CRLF, |
| 3446 | CODING_EOL_CR, and CODING_EOL_UNDECIDED. |
| 3447 | |
| 3448 | How many non-eol characters are at the head is returned as *SKIP. */ |
| 3449 | |
| 3450 | #define MAX_EOL_CHECK_COUNT 3 |
| 3451 | |
| 3452 | static int |
| 3453 | detect_eol_type (source, src_bytes, skip) |
| 3454 | unsigned char *source; |
| 3455 | int src_bytes, *skip; |
| 3456 | { |
| 3457 | unsigned char *src = source, *src_end = src + src_bytes; |
| 3458 | unsigned char c; |
| 3459 | int total = 0; /* How many end-of-lines are found so far. */ |
| 3460 | int eol_type = CODING_EOL_UNDECIDED; |
| 3461 | int this_eol_type; |
| 3462 | |
| 3463 | *skip = 0; |
| 3464 | |
| 3465 | while (src < src_end && total < MAX_EOL_CHECK_COUNT) |
| 3466 | { |
| 3467 | c = *src++; |
| 3468 | if (c == '\n' || c == '\r') |
| 3469 | { |
| 3470 | if (*skip == 0) |
| 3471 | *skip = src - 1 - source; |
| 3472 | total++; |
| 3473 | if (c == '\n') |
| 3474 | this_eol_type = CODING_EOL_LF; |
| 3475 | else if (src >= src_end || *src != '\n') |
| 3476 | this_eol_type = CODING_EOL_CR; |
| 3477 | else |
| 3478 | this_eol_type = CODING_EOL_CRLF, src++; |
| 3479 | |
| 3480 | if (eol_type == CODING_EOL_UNDECIDED) |
| 3481 | /* This is the first end-of-line. */ |
| 3482 | eol_type = this_eol_type; |
| 3483 | else if (eol_type != this_eol_type) |
| 3484 | { |
| 3485 | /* The found type is different from what found before. */ |
| 3486 | eol_type = CODING_EOL_INCONSISTENT; |
| 3487 | break; |
| 3488 | } |
| 3489 | } |
| 3490 | } |
| 3491 | |
| 3492 | if (*skip == 0) |
| 3493 | *skip = src_end - source; |
| 3494 | return eol_type; |
| 3495 | } |
| 3496 | |
| 3497 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC |
| 3498 | is encoded. If it detects an appropriate format of end-of-line, it |
| 3499 | sets the information in *CODING. */ |
| 3500 | |
| 3501 | void |
| 3502 | detect_eol (coding, src, src_bytes) |
| 3503 | struct coding_system *coding; |
| 3504 | unsigned char *src; |
| 3505 | int src_bytes; |
| 3506 | { |
| 3507 | Lisp_Object val; |
| 3508 | int skip; |
| 3509 | int eol_type = detect_eol_type (src, src_bytes, &skip); |
| 3510 | |
| 3511 | if (coding->heading_ascii > skip) |
| 3512 | coding->heading_ascii = skip; |
| 3513 | else |
| 3514 | skip = coding->heading_ascii; |
| 3515 | |
| 3516 | if (eol_type == CODING_EOL_UNDECIDED) |
| 3517 | return; |
| 3518 | if (eol_type == CODING_EOL_INCONSISTENT) |
| 3519 | { |
| 3520 | #if 0 |
| 3521 | /* This code is suppressed until we find a better way to |
| 3522 | distinguish raw text file and binary file. */ |
| 3523 | |
| 3524 | /* If we have already detected that the coding is raw-text, the |
| 3525 | coding should actually be no-conversion. */ |
| 3526 | if (coding->type == coding_type_raw_text) |
| 3527 | { |
| 3528 | setup_coding_system (Qno_conversion, coding); |
| 3529 | return; |
| 3530 | } |
| 3531 | /* Else, let's decode only text code anyway. */ |
| 3532 | #endif /* 0 */ |
| 3533 | eol_type = CODING_EOL_LF; |
| 3534 | } |
| 3535 | |
| 3536 | val = Fget (coding->symbol, Qeol_type); |
| 3537 | if (VECTORP (val) && XVECTOR (val)->size == 3) |
| 3538 | { |
| 3539 | setup_coding_system (XVECTOR (val)->contents[eol_type], coding); |
| 3540 | coding->heading_ascii = skip; |
| 3541 | } |
| 3542 | } |
| 3543 | |
| 3544 | #define CONVERSION_BUFFER_EXTRA_ROOM 256 |
| 3545 | |
| 3546 | #define DECODING_BUFFER_MAG(coding) \ |
| 3547 | (coding->type == coding_type_iso2022 \ |
| 3548 | ? 3 \ |
| 3549 | : ((coding->type == coding_type_sjis || coding->type == coding_type_big5) \ |
| 3550 | ? 2 \ |
| 3551 | : (coding->type == coding_type_raw_text \ |
| 3552 | ? 1 \ |
| 3553 | : (coding->type == coding_type_ccl \ |
| 3554 | ? coding->spec.ccl.decoder.buf_magnification \ |
| 3555 | : 2)))) |
| 3556 | |
| 3557 | /* Return maximum size (bytes) of a buffer enough for decoding |
| 3558 | SRC_BYTES of text encoded in CODING. */ |
| 3559 | |
| 3560 | int |
| 3561 | decoding_buffer_size (coding, src_bytes) |
| 3562 | struct coding_system *coding; |
| 3563 | int src_bytes; |
| 3564 | { |
| 3565 | return (src_bytes * DECODING_BUFFER_MAG (coding) |
| 3566 | + CONVERSION_BUFFER_EXTRA_ROOM); |
| 3567 | } |
| 3568 | |
| 3569 | /* Return maximum size (bytes) of a buffer enough for encoding |
| 3570 | SRC_BYTES of text to CODING. */ |
| 3571 | |
| 3572 | int |
| 3573 | encoding_buffer_size (coding, src_bytes) |
| 3574 | struct coding_system *coding; |
| 3575 | int src_bytes; |
| 3576 | { |
| 3577 | int magnification; |
| 3578 | |
| 3579 | if (coding->type == coding_type_ccl) |
| 3580 | magnification = coding->spec.ccl.encoder.buf_magnification; |
| 3581 | else |
| 3582 | magnification = 3; |
| 3583 | |
| 3584 | return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM); |
| 3585 | } |
| 3586 | |
| 3587 | #ifndef MINIMUM_CONVERSION_BUFFER_SIZE |
| 3588 | #define MINIMUM_CONVERSION_BUFFER_SIZE 1024 |
| 3589 | #endif |
| 3590 | |
| 3591 | char *conversion_buffer; |
| 3592 | int conversion_buffer_size; |
| 3593 | |
| 3594 | /* Return a pointer to a SIZE bytes of buffer to be used for encoding |
| 3595 | or decoding. Sufficient memory is allocated automatically. If we |
| 3596 | run out of memory, return NULL. */ |
| 3597 | |
| 3598 | char * |
| 3599 | get_conversion_buffer (size) |
| 3600 | int size; |
| 3601 | { |
| 3602 | if (size > conversion_buffer_size) |
| 3603 | { |
| 3604 | char *buf; |
| 3605 | int real_size = conversion_buffer_size * 2; |
| 3606 | |
| 3607 | while (real_size < size) real_size *= 2; |
| 3608 | buf = (char *) xmalloc (real_size); |
| 3609 | xfree (conversion_buffer); |
| 3610 | conversion_buffer = buf; |
| 3611 | conversion_buffer_size = real_size; |
| 3612 | } |
| 3613 | return conversion_buffer; |
| 3614 | } |
| 3615 | |
| 3616 | int |
| 3617 | ccl_coding_driver (coding, source, destination, src_bytes, dst_bytes, encodep) |
| 3618 | struct coding_system *coding; |
| 3619 | unsigned char *source, *destination; |
| 3620 | int src_bytes, dst_bytes, encodep; |
| 3621 | { |
| 3622 | struct ccl_program *ccl |
| 3623 | = encodep ? &coding->spec.ccl.encoder : &coding->spec.ccl.decoder; |
| 3624 | int result; |
| 3625 | |
| 3626 | if (encodep) |
| 3627 | ccl->last_block = coding->mode & CODING_MODE_LAST_BLOCK; |
| 3628 | |
| 3629 | coding->produced = ccl_driver (ccl, source, destination, |
| 3630 | src_bytes, dst_bytes, &(coding->consumed)); |
| 3631 | if (encodep) |
| 3632 | { |
| 3633 | coding->produced_char = coding->produced; |
| 3634 | coding->consumed_char |
| 3635 | = multibyte_chars_in_text (source, coding->consumed); |
| 3636 | } |
| 3637 | else |
| 3638 | { |
| 3639 | coding->produced_char |
| 3640 | = multibyte_chars_in_text (destination, coding->produced); |
| 3641 | coding->consumed_char = coding->consumed; |
| 3642 | } |
| 3643 | switch (ccl->status) |
| 3644 | { |
| 3645 | case CCL_STAT_SUSPEND_BY_SRC: |
| 3646 | result = CODING_FINISH_INSUFFICIENT_SRC; |
| 3647 | break; |
| 3648 | case CCL_STAT_SUSPEND_BY_DST: |
| 3649 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 3650 | break; |
| 3651 | default: |
| 3652 | result = CODING_FINISH_NORMAL; |
| 3653 | break; |
| 3654 | } |
| 3655 | return result; |
| 3656 | } |
| 3657 | |
| 3658 | /* See "GENERAL NOTES about `decode_coding_XXX ()' functions". Before |
| 3659 | decoding, it may detect coding system and format of end-of-line if |
| 3660 | those are not yet decided. */ |
| 3661 | |
| 3662 | int |
| 3663 | decode_coding (coding, source, destination, src_bytes, dst_bytes) |
| 3664 | struct coding_system *coding; |
| 3665 | unsigned char *source, *destination; |
| 3666 | int src_bytes, dst_bytes; |
| 3667 | { |
| 3668 | int result; |
| 3669 | |
| 3670 | if (src_bytes <= 0) |
| 3671 | { |
| 3672 | coding->produced = coding->produced_char = 0; |
| 3673 | coding->consumed = coding->consumed_char = 0; |
| 3674 | coding->fake_multibyte = 0; |
| 3675 | return CODING_FINISH_NORMAL; |
| 3676 | } |
| 3677 | |
| 3678 | if (coding->type == coding_type_undecided) |
| 3679 | detect_coding (coding, source, src_bytes); |
| 3680 | |
| 3681 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 3682 | detect_eol (coding, source, src_bytes); |
| 3683 | |
| 3684 | switch (coding->type) |
| 3685 | { |
| 3686 | case coding_type_emacs_mule: |
| 3687 | case coding_type_undecided: |
| 3688 | case coding_type_raw_text: |
| 3689 | if (coding->eol_type == CODING_EOL_LF |
| 3690 | || coding->eol_type == CODING_EOL_UNDECIDED) |
| 3691 | goto label_no_conversion; |
| 3692 | result = decode_eol (coding, source, destination, src_bytes, dst_bytes); |
| 3693 | break; |
| 3694 | |
| 3695 | case coding_type_sjis: |
| 3696 | result = decode_coding_sjis_big5 (coding, source, destination, |
| 3697 | src_bytes, dst_bytes, 1); |
| 3698 | break; |
| 3699 | |
| 3700 | case coding_type_iso2022: |
| 3701 | result = decode_coding_iso2022 (coding, source, destination, |
| 3702 | src_bytes, dst_bytes); |
| 3703 | break; |
| 3704 | |
| 3705 | case coding_type_big5: |
| 3706 | result = decode_coding_sjis_big5 (coding, source, destination, |
| 3707 | src_bytes, dst_bytes, 0); |
| 3708 | break; |
| 3709 | |
| 3710 | case coding_type_ccl: |
| 3711 | result = ccl_coding_driver (coding, source, destination, |
| 3712 | src_bytes, dst_bytes, 0); |
| 3713 | break; |
| 3714 | |
| 3715 | default: /* i.e. case coding_type_no_conversion: */ |
| 3716 | label_no_conversion: |
| 3717 | if (dst_bytes && src_bytes > dst_bytes) |
| 3718 | { |
| 3719 | coding->produced = dst_bytes; |
| 3720 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 3721 | } |
| 3722 | else |
| 3723 | { |
| 3724 | coding->produced = src_bytes; |
| 3725 | result = CODING_FINISH_NORMAL; |
| 3726 | } |
| 3727 | if (dst_bytes) |
| 3728 | bcopy (source, destination, coding->produced); |
| 3729 | else |
| 3730 | safe_bcopy (source, destination, coding->produced); |
| 3731 | coding->fake_multibyte = 1; |
| 3732 | coding->consumed |
| 3733 | = coding->consumed_char = coding->produced_char = coding->produced; |
| 3734 | break; |
| 3735 | } |
| 3736 | |
| 3737 | return result; |
| 3738 | } |
| 3739 | |
| 3740 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". */ |
| 3741 | |
| 3742 | int |
| 3743 | encode_coding (coding, source, destination, src_bytes, dst_bytes) |
| 3744 | struct coding_system *coding; |
| 3745 | unsigned char *source, *destination; |
| 3746 | int src_bytes, dst_bytes; |
| 3747 | { |
| 3748 | int result; |
| 3749 | |
| 3750 | if (src_bytes <= 0) |
| 3751 | { |
| 3752 | coding->produced = coding->produced_char = 0; |
| 3753 | coding->consumed = coding->consumed_char = 0; |
| 3754 | coding->fake_multibyte = 0; |
| 3755 | return CODING_FINISH_NORMAL; |
| 3756 | } |
| 3757 | |
| 3758 | switch (coding->type) |
| 3759 | { |
| 3760 | case coding_type_emacs_mule: |
| 3761 | case coding_type_undecided: |
| 3762 | case coding_type_raw_text: |
| 3763 | if (coding->eol_type == CODING_EOL_LF |
| 3764 | || coding->eol_type == CODING_EOL_UNDECIDED) |
| 3765 | goto label_no_conversion; |
| 3766 | result = encode_eol (coding, source, destination, src_bytes, dst_bytes); |
| 3767 | break; |
| 3768 | |
| 3769 | case coding_type_sjis: |
| 3770 | result = encode_coding_sjis_big5 (coding, source, destination, |
| 3771 | src_bytes, dst_bytes, 1); |
| 3772 | break; |
| 3773 | |
| 3774 | case coding_type_iso2022: |
| 3775 | result = encode_coding_iso2022 (coding, source, destination, |
| 3776 | src_bytes, dst_bytes); |
| 3777 | break; |
| 3778 | |
| 3779 | case coding_type_big5: |
| 3780 | result = encode_coding_sjis_big5 (coding, source, destination, |
| 3781 | src_bytes, dst_bytes, 0); |
| 3782 | break; |
| 3783 | |
| 3784 | case coding_type_ccl: |
| 3785 | result = ccl_coding_driver (coding, source, destination, |
| 3786 | src_bytes, dst_bytes, 1); |
| 3787 | break; |
| 3788 | |
| 3789 | default: /* i.e. case coding_type_no_conversion: */ |
| 3790 | label_no_conversion: |
| 3791 | if (dst_bytes && src_bytes > dst_bytes) |
| 3792 | { |
| 3793 | coding->produced = dst_bytes; |
| 3794 | result = CODING_FINISH_INSUFFICIENT_DST; |
| 3795 | } |
| 3796 | else |
| 3797 | { |
| 3798 | coding->produced = src_bytes; |
| 3799 | result = CODING_FINISH_NORMAL; |
| 3800 | } |
| 3801 | if (dst_bytes) |
| 3802 | bcopy (source, destination, coding->produced); |
| 3803 | else |
| 3804 | safe_bcopy (source, destination, coding->produced); |
| 3805 | if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY) |
| 3806 | { |
| 3807 | unsigned char *p = destination, *pend = p + coding->produced; |
| 3808 | while (p < pend) |
| 3809 | if (*p++ == '\015') p[-1] = '\n'; |
| 3810 | } |
| 3811 | coding->fake_multibyte = 1; |
| 3812 | coding->consumed |
| 3813 | = coding->consumed_char = coding->produced_char = coding->produced; |
| 3814 | break; |
| 3815 | } |
| 3816 | |
| 3817 | return result; |
| 3818 | } |
| 3819 | |
| 3820 | /* Scan text in the region between *BEG and *END (byte positions), |
| 3821 | skip characters which we don't have to decode by coding system |
| 3822 | CODING at the head and tail, then set *BEG and *END to the region |
| 3823 | of the text we actually have to convert. The caller should move |
| 3824 | the gap out of the region in advance. |
| 3825 | |
| 3826 | If STR is not NULL, *BEG and *END are indices into STR. */ |
| 3827 | |
| 3828 | static void |
| 3829 | shrink_decoding_region (beg, end, coding, str) |
| 3830 | int *beg, *end; |
| 3831 | struct coding_system *coding; |
| 3832 | unsigned char *str; |
| 3833 | { |
| 3834 | unsigned char *begp_orig, *begp, *endp_orig, *endp, c; |
| 3835 | int eol_conversion; |
| 3836 | |
| 3837 | if (coding->type == coding_type_ccl |
| 3838 | || coding->type == coding_type_undecided |
| 3839 | || !NILP (coding->post_read_conversion)) |
| 3840 | { |
| 3841 | /* We can't skip any data. */ |
| 3842 | return; |
| 3843 | } |
| 3844 | else if (coding->type == coding_type_no_conversion) |
| 3845 | { |
| 3846 | /* We need no conversion, but don't have to skip any data here. |
| 3847 | Decoding routine handles them effectively anyway. */ |
| 3848 | return; |
| 3849 | } |
| 3850 | |
| 3851 | eol_conversion = (coding->eol_type != CODING_EOL_LF); |
| 3852 | |
| 3853 | if ((! eol_conversion) && (coding->heading_ascii >= 0)) |
| 3854 | /* Detection routine has already found how much we can skip at the |
| 3855 | head. */ |
| 3856 | *beg += coding->heading_ascii; |
| 3857 | |
| 3858 | if (str) |
| 3859 | { |
| 3860 | begp_orig = begp = str + *beg; |
| 3861 | endp_orig = endp = str + *end; |
| 3862 | } |
| 3863 | else |
| 3864 | { |
| 3865 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
| 3866 | endp_orig = endp = begp + *end - *beg; |
| 3867 | } |
| 3868 | |
| 3869 | switch (coding->type) |
| 3870 | { |
| 3871 | case coding_type_emacs_mule: |
| 3872 | case coding_type_raw_text: |
| 3873 | if (eol_conversion) |
| 3874 | { |
| 3875 | if (coding->heading_ascii < 0) |
| 3876 | while (begp < endp && *begp != '\r' && *begp < 0x80) begp++; |
| 3877 | while (begp < endp && endp[-1] != '\r' && endp[-1] < 0x80) |
| 3878 | endp--; |
| 3879 | /* Do not consider LF as ascii if preceded by CR, since that |
| 3880 | confuses eol decoding. */ |
| 3881 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') |
| 3882 | endp++; |
| 3883 | } |
| 3884 | else |
| 3885 | begp = endp; |
| 3886 | break; |
| 3887 | |
| 3888 | case coding_type_sjis: |
| 3889 | case coding_type_big5: |
| 3890 | /* We can skip all ASCII characters at the head. */ |
| 3891 | if (coding->heading_ascii < 0) |
| 3892 | { |
| 3893 | if (eol_conversion) |
| 3894 | while (begp < endp && *begp < 0x80 && *begp != '\r') begp++; |
| 3895 | else |
| 3896 | while (begp < endp && *begp < 0x80) begp++; |
| 3897 | } |
| 3898 | /* We can skip all ASCII characters at the tail except for the |
| 3899 | second byte of SJIS or BIG5 code. */ |
| 3900 | if (eol_conversion) |
| 3901 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\r') endp--; |
| 3902 | else |
| 3903 | while (begp < endp && endp[-1] < 0x80) endp--; |
| 3904 | /* Do not consider LF as ascii if preceded by CR, since that |
| 3905 | confuses eol decoding. */ |
| 3906 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') |
| 3907 | endp++; |
| 3908 | if (begp < endp && endp < endp_orig && endp[-1] >= 0x80) |
| 3909 | endp++; |
| 3910 | break; |
| 3911 | |
| 3912 | default: /* i.e. case coding_type_iso2022: */ |
| 3913 | if (coding->heading_ascii < 0) |
| 3914 | { |
| 3915 | /* We can skip all ASCII characters at the head except for a |
| 3916 | few control codes. */ |
| 3917 | while (begp < endp && (c = *begp) < 0x80 |
| 3918 | && c != ISO_CODE_CR && c != ISO_CODE_SO |
| 3919 | && c != ISO_CODE_SI && c != ISO_CODE_ESC |
| 3920 | && (!eol_conversion || c != ISO_CODE_LF)) |
| 3921 | begp++; |
| 3922 | } |
| 3923 | switch (coding->category_idx) |
| 3924 | { |
| 3925 | case CODING_CATEGORY_IDX_ISO_8_1: |
| 3926 | case CODING_CATEGORY_IDX_ISO_8_2: |
| 3927 | /* We can skip all ASCII characters at the tail. */ |
| 3928 | if (eol_conversion) |
| 3929 | while (begp < endp && (c = endp[-1]) < 0x80 && c != '\r') endp--; |
| 3930 | else |
| 3931 | while (begp < endp && endp[-1] < 0x80) endp--; |
| 3932 | /* Do not consider LF as ascii if preceded by CR, since that |
| 3933 | confuses eol decoding. */ |
| 3934 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') |
| 3935 | endp++; |
| 3936 | break; |
| 3937 | |
| 3938 | case CODING_CATEGORY_IDX_ISO_7: |
| 3939 | case CODING_CATEGORY_IDX_ISO_7_TIGHT: |
| 3940 | /* We can skip all charactes at the tail except for ESC and |
| 3941 | the following 2-byte at the tail. */ |
| 3942 | if (eol_conversion) |
| 3943 | while (begp < endp |
| 3944 | && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC && c != '\r') |
| 3945 | endp--; |
| 3946 | else |
| 3947 | while (begp < endp |
| 3948 | && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC) |
| 3949 | endp--; |
| 3950 | /* Do not consider LF as ascii if preceded by CR, since that |
| 3951 | confuses eol decoding. */ |
| 3952 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') |
| 3953 | endp++; |
| 3954 | if (begp < endp && endp[-1] == ISO_CODE_ESC) |
| 3955 | { |
| 3956 | if (endp + 1 < endp_orig && end[0] == '(' && end[1] == 'B') |
| 3957 | /* This is an ASCII designation sequence. We can |
| 3958 | surely skip the tail. */ |
| 3959 | endp += 2; |
| 3960 | else |
| 3961 | /* Hmmm, we can't skip the tail. */ |
| 3962 | endp = endp_orig; |
| 3963 | } |
| 3964 | } |
| 3965 | } |
| 3966 | *beg += begp - begp_orig; |
| 3967 | *end += endp - endp_orig; |
| 3968 | return; |
| 3969 | } |
| 3970 | |
| 3971 | /* Like shrink_decoding_region but for encoding. */ |
| 3972 | |
| 3973 | static void |
| 3974 | shrink_encoding_region (beg, end, coding, str) |
| 3975 | int *beg, *end; |
| 3976 | struct coding_system *coding; |
| 3977 | unsigned char *str; |
| 3978 | { |
| 3979 | unsigned char *begp_orig, *begp, *endp_orig, *endp; |
| 3980 | int eol_conversion; |
| 3981 | |
| 3982 | if (coding->type == coding_type_ccl) |
| 3983 | /* We can't skip any data. */ |
| 3984 | return; |
| 3985 | else if (coding->type == coding_type_no_conversion) |
| 3986 | { |
| 3987 | /* We need no conversion. */ |
| 3988 | *beg = *end; |
| 3989 | return; |
| 3990 | } |
| 3991 | |
| 3992 | if (str) |
| 3993 | { |
| 3994 | begp_orig = begp = str + *beg; |
| 3995 | endp_orig = endp = str + *end; |
| 3996 | } |
| 3997 | else |
| 3998 | { |
| 3999 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
| 4000 | endp_orig = endp = begp + *end - *beg; |
| 4001 | } |
| 4002 | |
| 4003 | eol_conversion = (coding->eol_type == CODING_EOL_CR |
| 4004 | || coding->eol_type == CODING_EOL_CRLF); |
| 4005 | |
| 4006 | /* Here, we don't have to check coding->pre_write_conversion because |
| 4007 | the caller is expected to have handled it already. */ |
| 4008 | switch (coding->type) |
| 4009 | { |
| 4010 | case coding_type_undecided: |
| 4011 | case coding_type_emacs_mule: |
| 4012 | case coding_type_raw_text: |
| 4013 | if (eol_conversion) |
| 4014 | { |
| 4015 | while (begp < endp && *begp != '\n') begp++; |
| 4016 | while (begp < endp && endp[-1] != '\n') endp--; |
| 4017 | } |
| 4018 | else |
| 4019 | begp = endp; |
| 4020 | break; |
| 4021 | |
| 4022 | case coding_type_iso2022: |
| 4023 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL) |
| 4024 | { |
| 4025 | unsigned char *bol = begp; |
| 4026 | while (begp < endp && *begp < 0x80) |
| 4027 | { |
| 4028 | begp++; |
| 4029 | if (begp[-1] == '\n') |
| 4030 | bol = begp; |
| 4031 | } |
| 4032 | begp = bol; |
| 4033 | goto label_skip_tail; |
| 4034 | } |
| 4035 | /* fall down ... */ |
| 4036 | |
| 4037 | default: |
| 4038 | /* We can skip all ASCII characters at the head and tail. */ |
| 4039 | if (eol_conversion) |
| 4040 | while (begp < endp && *begp < 0x80 && *begp != '\n') begp++; |
| 4041 | else |
| 4042 | while (begp < endp && *begp < 0x80) begp++; |
| 4043 | label_skip_tail: |
| 4044 | if (eol_conversion) |
| 4045 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\n') endp--; |
| 4046 | else |
| 4047 | while (begp < endp && *(endp - 1) < 0x80) endp--; |
| 4048 | break; |
| 4049 | } |
| 4050 | |
| 4051 | *beg += begp - begp_orig; |
| 4052 | *end += endp - endp_orig; |
| 4053 | return; |
| 4054 | } |
| 4055 | |
| 4056 | /* Decode (if ENCODEP is zero) or encode (if ENCODEP is nonzero) the |
| 4057 | text from FROM to TO (byte positions are FROM_BYTE and TO_BYTE) by |
| 4058 | coding system CODING, and return the status code of code conversion |
| 4059 | (currently, this value has no meaning). |
| 4060 | |
| 4061 | How many characters (and bytes) are converted to how many |
| 4062 | characters (and bytes) are recorded in members of the structure |
| 4063 | CODING. |
| 4064 | |
| 4065 | If REPLACE is nonzero, we do various things as if the original text |
| 4066 | is deleted and a new text is inserted. See the comments in |
| 4067 | replace_range (insdel.c) to know what we are doing. */ |
| 4068 | |
| 4069 | int |
| 4070 | code_convert_region (from, from_byte, to, to_byte, coding, encodep, replace) |
| 4071 | int from, from_byte, to, to_byte, encodep, replace; |
| 4072 | struct coding_system *coding; |
| 4073 | { |
| 4074 | int len = to - from, len_byte = to_byte - from_byte; |
| 4075 | int require, inserted, inserted_byte; |
| 4076 | int head_skip, tail_skip, total_skip; |
| 4077 | Lisp_Object saved_coding_symbol; |
| 4078 | int multibyte = !NILP (current_buffer->enable_multibyte_characters); |
| 4079 | int first = 1; |
| 4080 | int fake_multibyte = 0; |
| 4081 | unsigned char *src, *dst; |
| 4082 | Lisp_Object deletion; |
| 4083 | |
| 4084 | deletion = Qnil; |
| 4085 | saved_coding_symbol = Qnil; |
| 4086 | |
| 4087 | if (from < PT && PT < to) |
| 4088 | SET_PT_BOTH (from, from_byte); |
| 4089 | |
| 4090 | if (replace) |
| 4091 | { |
| 4092 | int saved_from = from; |
| 4093 | |
| 4094 | prepare_to_modify_buffer (from, to, &from); |
| 4095 | if (saved_from != from) |
| 4096 | { |
| 4097 | to = from + len; |
| 4098 | if (multibyte) |
| 4099 | from_byte = CHAR_TO_BYTE (from), to_byte = CHAR_TO_BYTE (to); |
| 4100 | else |
| 4101 | from_byte = from, to_byte = to; |
| 4102 | len_byte = to_byte - from_byte; |
| 4103 | } |
| 4104 | } |
| 4105 | |
| 4106 | if (! encodep && CODING_REQUIRE_DETECTION (coding)) |
| 4107 | { |
| 4108 | /* We must detect encoding of text and eol format. */ |
| 4109 | |
| 4110 | if (from < GPT && to > GPT) |
| 4111 | move_gap_both (from, from_byte); |
| 4112 | if (coding->type == coding_type_undecided) |
| 4113 | { |
| 4114 | detect_coding (coding, BYTE_POS_ADDR (from_byte), len_byte); |
| 4115 | if (coding->type == coding_type_undecided) |
| 4116 | /* It seems that the text contains only ASCII, but we |
| 4117 | should not left it undecided because the deeper |
| 4118 | decoding routine (decode_coding) tries to detect the |
| 4119 | encodings again in vain. */ |
| 4120 | coding->type = coding_type_emacs_mule; |
| 4121 | } |
| 4122 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 4123 | { |
| 4124 | saved_coding_symbol = coding->symbol; |
| 4125 | detect_eol (coding, BYTE_POS_ADDR (from_byte), len_byte); |
| 4126 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 4127 | coding->eol_type = CODING_EOL_LF; |
| 4128 | /* We had better recover the original eol format if we |
| 4129 | encounter an inconsitent eol format while decoding. */ |
| 4130 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; |
| 4131 | } |
| 4132 | } |
| 4133 | |
| 4134 | coding->consumed_char = len, coding->consumed = len_byte; |
| 4135 | |
| 4136 | if (encodep |
| 4137 | ? ! CODING_REQUIRE_ENCODING (coding) |
| 4138 | : ! CODING_REQUIRE_DECODING (coding)) |
| 4139 | { |
| 4140 | coding->produced = len_byte; |
| 4141 | if (multibyte |
| 4142 | && ! replace |
| 4143 | /* See the comment of the member heading_ascii in coding.h. */ |
| 4144 | && coding->heading_ascii < len_byte) |
| 4145 | { |
| 4146 | /* We still may have to combine byte at the head and the |
| 4147 | tail of the text in the region. */ |
| 4148 | if (from < GPT && GPT < to) |
| 4149 | move_gap_both (to, to_byte); |
| 4150 | len = multibyte_chars_in_text (BYTE_POS_ADDR (from_byte), len_byte); |
| 4151 | adjust_after_insert (from, from_byte, to, to_byte, len); |
| 4152 | coding->produced_char = len; |
| 4153 | } |
| 4154 | else |
| 4155 | { |
| 4156 | if (!replace) |
| 4157 | adjust_after_insert (from, from_byte, to, to_byte, len_byte); |
| 4158 | coding->produced_char = len_byte; |
| 4159 | } |
| 4160 | return 0; |
| 4161 | } |
| 4162 | |
| 4163 | /* Now we convert the text. */ |
| 4164 | |
| 4165 | /* For encoding, we must process pre-write-conversion in advance. */ |
| 4166 | if (encodep |
| 4167 | && ! NILP (coding->pre_write_conversion) |
| 4168 | && SYMBOLP (coding->pre_write_conversion) |
| 4169 | && ! NILP (Ffboundp (coding->pre_write_conversion))) |
| 4170 | { |
| 4171 | /* The function in pre-write-conversion may put a new text in a |
| 4172 | new buffer. */ |
| 4173 | struct buffer *prev = current_buffer, *new; |
| 4174 | |
| 4175 | call2 (coding->pre_write_conversion, |
| 4176 | make_number (from), make_number (to)); |
| 4177 | if (current_buffer != prev) |
| 4178 | { |
| 4179 | len = ZV - BEGV; |
| 4180 | new = current_buffer; |
| 4181 | set_buffer_internal_1 (prev); |
| 4182 | del_range_2 (from, from_byte, to, to_byte); |
| 4183 | insert_from_buffer (new, BEG, len, 0); |
| 4184 | to = from + len; |
| 4185 | to_byte = multibyte ? CHAR_TO_BYTE (to) : to; |
| 4186 | len_byte = to_byte - from_byte; |
| 4187 | } |
| 4188 | } |
| 4189 | |
| 4190 | if (replace) |
| 4191 | deletion = make_buffer_string_both (from, from_byte, to, to_byte, 1); |
| 4192 | |
| 4193 | /* Try to skip the heading and tailing ASCIIs. */ |
| 4194 | { |
| 4195 | int from_byte_orig = from_byte, to_byte_orig = to_byte; |
| 4196 | |
| 4197 | if (from < GPT && GPT < to) |
| 4198 | move_gap_both (from, from_byte); |
| 4199 | if (encodep) |
| 4200 | shrink_encoding_region (&from_byte, &to_byte, coding, NULL); |
| 4201 | else |
| 4202 | shrink_decoding_region (&from_byte, &to_byte, coding, NULL); |
| 4203 | if (from_byte == to_byte) |
| 4204 | { |
| 4205 | coding->produced = len_byte; |
| 4206 | coding->produced_char = multibyte ? len : len_byte; |
| 4207 | if (!replace) |
| 4208 | /* We must record and adjust for this new text now. */ |
| 4209 | adjust_after_insert (from, from_byte_orig, to, to_byte_orig, len); |
| 4210 | return 0; |
| 4211 | } |
| 4212 | |
| 4213 | head_skip = from_byte - from_byte_orig; |
| 4214 | tail_skip = to_byte_orig - to_byte; |
| 4215 | total_skip = head_skip + tail_skip; |
| 4216 | from += head_skip; |
| 4217 | to -= tail_skip; |
| 4218 | len -= total_skip; len_byte -= total_skip; |
| 4219 | } |
| 4220 | |
| 4221 | /* For converion, we must put the gap before the text in addition to |
| 4222 | making the gap larger for efficient decoding. The required gap |
| 4223 | size starts from 2000 which is the magic number used in make_gap. |
| 4224 | But, after one batch of conversion, it will be incremented if we |
| 4225 | find that it is not enough . */ |
| 4226 | require = 2000; |
| 4227 | |
| 4228 | if (GAP_SIZE < require) |
| 4229 | make_gap (require - GAP_SIZE); |
| 4230 | move_gap_both (from, from_byte); |
| 4231 | |
| 4232 | if (GPT - BEG < beg_unchanged) |
| 4233 | beg_unchanged = GPT - BEG; |
| 4234 | if (Z - GPT < end_unchanged) |
| 4235 | end_unchanged = Z - GPT; |
| 4236 | |
| 4237 | inserted = inserted_byte = 0; |
| 4238 | src = GAP_END_ADDR, dst = GPT_ADDR; |
| 4239 | |
| 4240 | GAP_SIZE += len_byte; |
| 4241 | ZV -= len; |
| 4242 | Z -= len; |
| 4243 | ZV_BYTE -= len_byte; |
| 4244 | Z_BYTE -= len_byte; |
| 4245 | |
| 4246 | for (;;) |
| 4247 | { |
| 4248 | int result; |
| 4249 | |
| 4250 | /* The buffer memory is changed from: |
| 4251 | +--------+converted-text+---------+-------original-text------+---+ |
| 4252 | |<-from->|<--inserted-->|---------|<-----------len---------->|---| |
| 4253 | |<------------------- GAP_SIZE -------------------->| */ |
| 4254 | if (encodep) |
| 4255 | result = encode_coding (coding, src, dst, len_byte, 0); |
| 4256 | else |
| 4257 | result = decode_coding (coding, src, dst, len_byte, 0); |
| 4258 | /* to: |
| 4259 | +--------+-------converted-text--------+--+---original-text--+---+ |
| 4260 | |<-from->|<--inserted-->|<--produced-->|--|<-(len-consumed)->|---| |
| 4261 | |<------------------- GAP_SIZE -------------------->| */ |
| 4262 | if (coding->fake_multibyte) |
| 4263 | fake_multibyte = 1; |
| 4264 | |
| 4265 | if (!encodep && !multibyte) |
| 4266 | coding->produced_char = coding->produced; |
| 4267 | inserted += coding->produced_char; |
| 4268 | inserted_byte += coding->produced; |
| 4269 | len_byte -= coding->consumed; |
| 4270 | src += coding->consumed; |
| 4271 | dst += inserted_byte; |
| 4272 | |
| 4273 | if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL) |
| 4274 | { |
| 4275 | unsigned char *pend = dst, *p = pend - inserted_byte; |
| 4276 | |
| 4277 | /* Encode LFs back to the original eol format (CR or CRLF). */ |
| 4278 | if (coding->eol_type == CODING_EOL_CR) |
| 4279 | { |
| 4280 | while (p < pend) if (*p++ == '\n') p[-1] = '\r'; |
| 4281 | } |
| 4282 | else |
| 4283 | { |
| 4284 | int count = 0; |
| 4285 | |
| 4286 | while (p < pend) if (*p++ == '\n') count++; |
| 4287 | if (src - dst < count) |
| 4288 | { |
| 4289 | /* We don't have sufficient room for putting LFs |
| 4290 | back to CRLF. We must record converted and |
| 4291 | not-yet-converted text back to the buffer |
| 4292 | content, enlarge the gap, then record them out of |
| 4293 | the buffer contents again. */ |
| 4294 | int add = len_byte + inserted_byte; |
| 4295 | |
| 4296 | GAP_SIZE -= add; |
| 4297 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; |
| 4298 | GPT += inserted_byte; GPT_BYTE += inserted_byte; |
| 4299 | make_gap (count - GAP_SIZE); |
| 4300 | GAP_SIZE += add; |
| 4301 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; |
| 4302 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; |
| 4303 | /* Don't forget to update SRC, DST, and PEND. */ |
| 4304 | src = GAP_END_ADDR - len_byte; |
| 4305 | dst = GPT_ADDR + inserted_byte; |
| 4306 | pend = dst; |
| 4307 | } |
| 4308 | inserted += count; |
| 4309 | inserted_byte += count; |
| 4310 | coding->produced += count; |
| 4311 | p = dst = pend + count; |
| 4312 | while (count) |
| 4313 | { |
| 4314 | *--p = *--pend; |
| 4315 | if (*p == '\n') count--, *--p = '\r'; |
| 4316 | } |
| 4317 | } |
| 4318 | |
| 4319 | /* Suppress eol-format conversion in the further conversion. */ |
| 4320 | coding->eol_type = CODING_EOL_LF; |
| 4321 | |
| 4322 | /* Restore the original symbol. */ |
| 4323 | coding->symbol = saved_coding_symbol; |
| 4324 | |
| 4325 | continue; |
| 4326 | } |
| 4327 | if (len_byte <= 0) |
| 4328 | break; |
| 4329 | if (result == CODING_FINISH_INSUFFICIENT_SRC) |
| 4330 | { |
| 4331 | /* The source text ends in invalid codes. Let's just |
| 4332 | make them valid buffer contents, and finish conversion. */ |
| 4333 | inserted += len_byte; |
| 4334 | inserted_byte += len_byte; |
| 4335 | while (len_byte--) |
| 4336 | *dst++ = *src++; |
| 4337 | fake_multibyte = 1; |
| 4338 | break; |
| 4339 | } |
| 4340 | if (first) |
| 4341 | { |
| 4342 | /* We have just done the first batch of conversion which was |
| 4343 | stoped because of insufficient gap. Let's reconsider the |
| 4344 | required gap size (i.e. SRT - DST) now. |
| 4345 | |
| 4346 | We have converted ORIG bytes (== coding->consumed) into |
| 4347 | NEW bytes (coding->produced). To convert the remaining |
| 4348 | LEN bytes, we may need REQUIRE bytes of gap, where: |
| 4349 | REQUIRE + LEN_BYTE = LEN_BYTE * (NEW / ORIG) |
| 4350 | REQUIRE = LEN_BYTE * (NEW - ORIG) / ORIG |
| 4351 | Here, we are sure that NEW >= ORIG. */ |
| 4352 | float ratio = coding->produced - coding->consumed; |
| 4353 | ratio /= coding->consumed; |
| 4354 | require = len_byte * ratio; |
| 4355 | first = 0; |
| 4356 | } |
| 4357 | if ((src - dst) < (require + 2000)) |
| 4358 | { |
| 4359 | /* See the comment above the previous call of make_gap. */ |
| 4360 | int add = len_byte + inserted_byte; |
| 4361 | |
| 4362 | GAP_SIZE -= add; |
| 4363 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; |
| 4364 | GPT += inserted_byte; GPT_BYTE += inserted_byte; |
| 4365 | make_gap (require + 2000); |
| 4366 | GAP_SIZE += add; |
| 4367 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; |
| 4368 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; |
| 4369 | /* Don't forget to update SRC, DST. */ |
| 4370 | src = GAP_END_ADDR - len_byte; |
| 4371 | dst = GPT_ADDR + inserted_byte; |
| 4372 | } |
| 4373 | } |
| 4374 | if (src - dst > 0) *dst = 0; /* Put an anchor. */ |
| 4375 | |
| 4376 | if (multibyte |
| 4377 | && (fake_multibyte |
| 4378 | || !encodep && (to - from) != (to_byte - from_byte))) |
| 4379 | inserted = multibyte_chars_in_text (GPT_ADDR, inserted_byte); |
| 4380 | |
| 4381 | /* If we have shrinked the conversion area, adjust it now. */ |
| 4382 | if (total_skip > 0) |
| 4383 | { |
| 4384 | if (tail_skip > 0) |
| 4385 | safe_bcopy (GAP_END_ADDR, GPT_ADDR + inserted_byte, tail_skip); |
| 4386 | inserted += total_skip; inserted_byte += total_skip; |
| 4387 | GAP_SIZE += total_skip; |
| 4388 | GPT -= head_skip; GPT_BYTE -= head_skip; |
| 4389 | ZV -= total_skip; ZV_BYTE -= total_skip; |
| 4390 | Z -= total_skip; Z_BYTE -= total_skip; |
| 4391 | from -= head_skip; from_byte -= head_skip; |
| 4392 | to += tail_skip; to_byte += tail_skip; |
| 4393 | } |
| 4394 | |
| 4395 | adjust_after_replace (from, from_byte, deletion, inserted, inserted_byte); |
| 4396 | |
| 4397 | if (! encodep && ! NILP (coding->post_read_conversion)) |
| 4398 | { |
| 4399 | Lisp_Object val; |
| 4400 | int orig_inserted = inserted, pos = PT; |
| 4401 | |
| 4402 | if (from != pos) |
| 4403 | temp_set_point_both (current_buffer, from, from_byte); |
| 4404 | val = call1 (coding->post_read_conversion, make_number (inserted)); |
| 4405 | if (! NILP (val)) |
| 4406 | { |
| 4407 | CHECK_NUMBER (val, 0); |
| 4408 | inserted = XFASTINT (val); |
| 4409 | } |
| 4410 | if (pos >= from + orig_inserted) |
| 4411 | temp_set_point (current_buffer, pos + (inserted - orig_inserted)); |
| 4412 | } |
| 4413 | |
| 4414 | signal_after_change (from, to - from, inserted); |
| 4415 | |
| 4416 | { |
| 4417 | coding->consumed = to_byte - from_byte; |
| 4418 | coding->consumed_char = to - from; |
| 4419 | coding->produced = inserted_byte; |
| 4420 | coding->produced_char = inserted; |
| 4421 | } |
| 4422 | |
| 4423 | return 0; |
| 4424 | } |
| 4425 | |
| 4426 | Lisp_Object |
| 4427 | code_convert_string (str, coding, encodep, nocopy) |
| 4428 | Lisp_Object str; |
| 4429 | struct coding_system *coding; |
| 4430 | int encodep, nocopy; |
| 4431 | { |
| 4432 | int len; |
| 4433 | char *buf; |
| 4434 | int from = 0, to = XSTRING (str)->size; |
| 4435 | int to_byte = STRING_BYTES (XSTRING (str)); |
| 4436 | struct gcpro gcpro1; |
| 4437 | Lisp_Object saved_coding_symbol; |
| 4438 | int result; |
| 4439 | |
| 4440 | saved_coding_symbol = Qnil; |
| 4441 | if (encodep && !NILP (coding->pre_write_conversion) |
| 4442 | || !encodep && !NILP (coding->post_read_conversion)) |
| 4443 | { |
| 4444 | /* Since we have to call Lisp functions which assume target text |
| 4445 | is in a buffer, after setting a temporary buffer, call |
| 4446 | code_convert_region. */ |
| 4447 | int count = specpdl_ptr - specpdl; |
| 4448 | struct buffer *prev = current_buffer; |
| 4449 | |
| 4450 | record_unwind_protect (Fset_buffer, Fcurrent_buffer ()); |
| 4451 | temp_output_buffer_setup (" *code-converting-work*"); |
| 4452 | set_buffer_internal (XBUFFER (Vstandard_output)); |
| 4453 | if (encodep) |
| 4454 | insert_from_string (str, 0, 0, to, to_byte, 0); |
| 4455 | else |
| 4456 | { |
| 4457 | /* We must insert the contents of STR as is without |
| 4458 | unibyte<->multibyte conversion. */ |
| 4459 | current_buffer->enable_multibyte_characters = Qnil; |
| 4460 | insert_from_string (str, 0, 0, to_byte, to_byte, 0); |
| 4461 | current_buffer->enable_multibyte_characters = Qt; |
| 4462 | } |
| 4463 | code_convert_region (BEGV, BEGV_BYTE, ZV, ZV_BYTE, coding, encodep, 1); |
| 4464 | if (encodep) |
| 4465 | /* We must return the buffer contents as unibyte string. */ |
| 4466 | current_buffer->enable_multibyte_characters = Qnil; |
| 4467 | str = make_buffer_string (BEGV, ZV, 0); |
| 4468 | set_buffer_internal (prev); |
| 4469 | return unbind_to (count, str); |
| 4470 | } |
| 4471 | |
| 4472 | if (! encodep && CODING_REQUIRE_DETECTION (coding)) |
| 4473 | { |
| 4474 | /* See the comments in code_convert_region. */ |
| 4475 | if (coding->type == coding_type_undecided) |
| 4476 | { |
| 4477 | detect_coding (coding, XSTRING (str)->data, to_byte); |
| 4478 | if (coding->type == coding_type_undecided) |
| 4479 | coding->type = coding_type_emacs_mule; |
| 4480 | } |
| 4481 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 4482 | { |
| 4483 | saved_coding_symbol = coding->symbol; |
| 4484 | detect_eol (coding, XSTRING (str)->data, to_byte); |
| 4485 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 4486 | coding->eol_type = CODING_EOL_LF; |
| 4487 | /* We had better recover the original eol format if we |
| 4488 | encounter an inconsitent eol format while decoding. */ |
| 4489 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; |
| 4490 | } |
| 4491 | } |
| 4492 | |
| 4493 | if (encodep |
| 4494 | ? ! CODING_REQUIRE_ENCODING (coding) |
| 4495 | : ! CODING_REQUIRE_DECODING (coding)) |
| 4496 | from = to_byte; |
| 4497 | else |
| 4498 | { |
| 4499 | /* Try to skip the heading and tailing ASCIIs. */ |
| 4500 | if (encodep) |
| 4501 | shrink_encoding_region (&from, &to_byte, coding, XSTRING (str)->data); |
| 4502 | else |
| 4503 | shrink_decoding_region (&from, &to_byte, coding, XSTRING (str)->data); |
| 4504 | } |
| 4505 | if (from == to_byte) |
| 4506 | return (nocopy ? str : Fcopy_sequence (str)); |
| 4507 | |
| 4508 | if (encodep) |
| 4509 | len = encoding_buffer_size (coding, to_byte - from); |
| 4510 | else |
| 4511 | len = decoding_buffer_size (coding, to_byte - from); |
| 4512 | len += from + STRING_BYTES (XSTRING (str)) - to_byte; |
| 4513 | GCPRO1 (str); |
| 4514 | buf = get_conversion_buffer (len); |
| 4515 | UNGCPRO; |
| 4516 | |
| 4517 | if (from > 0) |
| 4518 | bcopy (XSTRING (str)->data, buf, from); |
| 4519 | result = (encodep |
| 4520 | ? encode_coding (coding, XSTRING (str)->data + from, |
| 4521 | buf + from, to_byte - from, len) |
| 4522 | : decode_coding (coding, XSTRING (str)->data + from, |
| 4523 | buf + from, to_byte - from, len)); |
| 4524 | if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL) |
| 4525 | { |
| 4526 | /* We simple try to decode the whole string again but without |
| 4527 | eol-conversion this time. */ |
| 4528 | coding->eol_type = CODING_EOL_LF; |
| 4529 | coding->symbol = saved_coding_symbol; |
| 4530 | return code_convert_string (str, coding, encodep, nocopy); |
| 4531 | } |
| 4532 | |
| 4533 | bcopy (XSTRING (str)->data + to_byte, buf + from + coding->produced, |
| 4534 | STRING_BYTES (XSTRING (str)) - to_byte); |
| 4535 | |
| 4536 | len = from + STRING_BYTES (XSTRING (str)) - to_byte; |
| 4537 | if (encodep) |
| 4538 | str = make_unibyte_string (buf, len + coding->produced); |
| 4539 | else |
| 4540 | { |
| 4541 | int chars= (coding->fake_multibyte |
| 4542 | ? multibyte_chars_in_text (buf + from, coding->produced) |
| 4543 | : coding->produced_char); |
| 4544 | str = make_multibyte_string (buf, len + chars, len + coding->produced); |
| 4545 | } |
| 4546 | |
| 4547 | return str; |
| 4548 | } |
| 4549 | |
| 4550 | \f |
| 4551 | #ifdef emacs |
| 4552 | /*** 8. Emacs Lisp library functions ***/ |
| 4553 | |
| 4554 | DEFUN ("coding-system-p", Fcoding_system_p, Scoding_system_p, 1, 1, 0, |
| 4555 | "Return t if OBJECT is nil or a coding-system.\n\ |
| 4556 | See the documentation of `make-coding-system' for information\n\ |
| 4557 | about coding-system objects.") |
| 4558 | (obj) |
| 4559 | Lisp_Object obj; |
| 4560 | { |
| 4561 | if (NILP (obj)) |
| 4562 | return Qt; |
| 4563 | if (!SYMBOLP (obj)) |
| 4564 | return Qnil; |
| 4565 | /* Get coding-spec vector for OBJ. */ |
| 4566 | obj = Fget (obj, Qcoding_system); |
| 4567 | return ((VECTORP (obj) && XVECTOR (obj)->size == 5) |
| 4568 | ? Qt : Qnil); |
| 4569 | } |
| 4570 | |
| 4571 | DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system, |
| 4572 | Sread_non_nil_coding_system, 1, 1, 0, |
| 4573 | "Read a coding system from the minibuffer, prompting with string PROMPT.") |
| 4574 | (prompt) |
| 4575 | Lisp_Object prompt; |
| 4576 | { |
| 4577 | Lisp_Object val; |
| 4578 | do |
| 4579 | { |
| 4580 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
| 4581 | Qt, Qnil, Qcoding_system_history, Qnil, Qnil); |
| 4582 | } |
| 4583 | while (XSTRING (val)->size == 0); |
| 4584 | return (Fintern (val, Qnil)); |
| 4585 | } |
| 4586 | |
| 4587 | DEFUN ("read-coding-system", Fread_coding_system, Sread_coding_system, 1, 2, 0, |
| 4588 | "Read a coding system from the minibuffer, prompting with string PROMPT.\n\ |
| 4589 | If the user enters null input, return second argument DEFAULT-CODING-SYSTEM.") |
| 4590 | (prompt, default_coding_system) |
| 4591 | Lisp_Object prompt, default_coding_system; |
| 4592 | { |
| 4593 | Lisp_Object val; |
| 4594 | if (SYMBOLP (default_coding_system)) |
| 4595 | XSETSTRING (default_coding_system, XSYMBOL (default_coding_system)->name); |
| 4596 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
| 4597 | Qt, Qnil, Qcoding_system_history, |
| 4598 | default_coding_system, Qnil); |
| 4599 | return (XSTRING (val)->size == 0 ? Qnil : Fintern (val, Qnil)); |
| 4600 | } |
| 4601 | |
| 4602 | DEFUN ("check-coding-system", Fcheck_coding_system, Scheck_coding_system, |
| 4603 | 1, 1, 0, |
| 4604 | "Check validity of CODING-SYSTEM.\n\ |
| 4605 | If valid, return CODING-SYSTEM, else signal a `coding-system-error' error.\n\ |
| 4606 | It is valid if it is a symbol with a non-nil `coding-system' property.\n\ |
| 4607 | The value of property should be a vector of length 5.") |
| 4608 | (coding_system) |
| 4609 | Lisp_Object coding_system; |
| 4610 | { |
| 4611 | CHECK_SYMBOL (coding_system, 0); |
| 4612 | if (!NILP (Fcoding_system_p (coding_system))) |
| 4613 | return coding_system; |
| 4614 | while (1) |
| 4615 | Fsignal (Qcoding_system_error, Fcons (coding_system, Qnil)); |
| 4616 | } |
| 4617 | \f |
| 4618 | Lisp_Object |
| 4619 | detect_coding_system (src, src_bytes, highest) |
| 4620 | unsigned char *src; |
| 4621 | int src_bytes, highest; |
| 4622 | { |
| 4623 | int coding_mask, eol_type; |
| 4624 | Lisp_Object val, tmp; |
| 4625 | int dummy; |
| 4626 | |
| 4627 | coding_mask = detect_coding_mask (src, src_bytes, NULL, &dummy); |
| 4628 | eol_type = detect_eol_type (src, src_bytes, &dummy); |
| 4629 | if (eol_type == CODING_EOL_INCONSISTENT) |
| 4630 | eol_type = CODING_EOL_UNDECIDED; |
| 4631 | |
| 4632 | if (!coding_mask) |
| 4633 | { |
| 4634 | val = Qundecided; |
| 4635 | if (eol_type != CODING_EOL_UNDECIDED) |
| 4636 | { |
| 4637 | Lisp_Object val2; |
| 4638 | val2 = Fget (Qundecided, Qeol_type); |
| 4639 | if (VECTORP (val2)) |
| 4640 | val = XVECTOR (val2)->contents[eol_type]; |
| 4641 | } |
| 4642 | return (highest ? val : Fcons (val, Qnil)); |
| 4643 | } |
| 4644 | |
| 4645 | /* At first, gather possible coding systems in VAL. */ |
| 4646 | val = Qnil; |
| 4647 | for (tmp = Vcoding_category_list; !NILP (tmp); tmp = XCONS (tmp)->cdr) |
| 4648 | { |
| 4649 | int idx |
| 4650 | = XFASTINT (Fget (XCONS (tmp)->car, Qcoding_category_index)); |
| 4651 | if (coding_mask & (1 << idx)) |
| 4652 | { |
| 4653 | val = Fcons (Fsymbol_value (XCONS (tmp)->car), val); |
| 4654 | if (highest) |
| 4655 | break; |
| 4656 | } |
| 4657 | } |
| 4658 | if (!highest) |
| 4659 | val = Fnreverse (val); |
| 4660 | |
| 4661 | /* Then, replace the elements with subsidiary coding systems. */ |
| 4662 | for (tmp = val; !NILP (tmp); tmp = XCONS (tmp)->cdr) |
| 4663 | { |
| 4664 | if (eol_type != CODING_EOL_UNDECIDED |
| 4665 | && eol_type != CODING_EOL_INCONSISTENT) |
| 4666 | { |
| 4667 | Lisp_Object eol; |
| 4668 | eol = Fget (XCONS (tmp)->car, Qeol_type); |
| 4669 | if (VECTORP (eol)) |
| 4670 | XCONS (tmp)->car = XVECTOR (eol)->contents[eol_type]; |
| 4671 | } |
| 4672 | } |
| 4673 | return (highest ? XCONS (val)->car : val); |
| 4674 | } |
| 4675 | |
| 4676 | DEFUN ("detect-coding-region", Fdetect_coding_region, Sdetect_coding_region, |
| 4677 | 2, 3, 0, |
| 4678 | "Detect coding system of the text in the region between START and END.\n\ |
| 4679 | Return a list of possible coding systems ordered by priority.\n\ |
| 4680 | \n\ |
| 4681 | If only ASCII characters are found, it returns a list of single element\n\ |
| 4682 | `undecided' or its subsidiary coding system according to a detected\n\ |
| 4683 | end-of-line format.\n\ |
| 4684 | \n\ |
| 4685 | If optional argument HIGHEST is non-nil, return the coding system of\n\ |
| 4686 | highest priority.") |
| 4687 | (start, end, highest) |
| 4688 | Lisp_Object start, end, highest; |
| 4689 | { |
| 4690 | int from, to; |
| 4691 | int from_byte, to_byte; |
| 4692 | |
| 4693 | CHECK_NUMBER_COERCE_MARKER (start, 0); |
| 4694 | CHECK_NUMBER_COERCE_MARKER (end, 1); |
| 4695 | |
| 4696 | validate_region (&start, &end); |
| 4697 | from = XINT (start), to = XINT (end); |
| 4698 | from_byte = CHAR_TO_BYTE (from); |
| 4699 | to_byte = CHAR_TO_BYTE (to); |
| 4700 | |
| 4701 | if (from < GPT && to >= GPT) |
| 4702 | move_gap_both (to, to_byte); |
| 4703 | |
| 4704 | return detect_coding_system (BYTE_POS_ADDR (from_byte), |
| 4705 | to_byte - from_byte, |
| 4706 | !NILP (highest)); |
| 4707 | } |
| 4708 | |
| 4709 | DEFUN ("detect-coding-string", Fdetect_coding_string, Sdetect_coding_string, |
| 4710 | 1, 2, 0, |
| 4711 | "Detect coding system of the text in STRING.\n\ |
| 4712 | Return a list of possible coding systems ordered by priority.\n\ |
| 4713 | \n\ |
| 4714 | If only ASCII characters are found, it returns a list of single element\n\ |
| 4715 | `undecided' or its subsidiary coding system according to a detected\n\ |
| 4716 | end-of-line format.\n\ |
| 4717 | \n\ |
| 4718 | If optional argument HIGHEST is non-nil, return the coding system of\n\ |
| 4719 | highest priority.") |
| 4720 | (string, highest) |
| 4721 | Lisp_Object string, highest; |
| 4722 | { |
| 4723 | CHECK_STRING (string, 0); |
| 4724 | |
| 4725 | return detect_coding_system (XSTRING (string)->data, |
| 4726 | STRING_BYTES (XSTRING (string)), |
| 4727 | !NILP (highest)); |
| 4728 | } |
| 4729 | |
| 4730 | Lisp_Object |
| 4731 | code_convert_region1 (start, end, coding_system, encodep) |
| 4732 | Lisp_Object start, end, coding_system; |
| 4733 | int encodep; |
| 4734 | { |
| 4735 | struct coding_system coding; |
| 4736 | int from, to, len; |
| 4737 | |
| 4738 | CHECK_NUMBER_COERCE_MARKER (start, 0); |
| 4739 | CHECK_NUMBER_COERCE_MARKER (end, 1); |
| 4740 | CHECK_SYMBOL (coding_system, 2); |
| 4741 | |
| 4742 | validate_region (&start, &end); |
| 4743 | from = XFASTINT (start); |
| 4744 | to = XFASTINT (end); |
| 4745 | |
| 4746 | if (NILP (coding_system)) |
| 4747 | return make_number (to - from); |
| 4748 | |
| 4749 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
| 4750 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); |
| 4751 | |
| 4752 | coding.mode |= CODING_MODE_LAST_BLOCK; |
| 4753 | code_convert_region (from, CHAR_TO_BYTE (from), to, CHAR_TO_BYTE (to), |
| 4754 | &coding, encodep, 1); |
| 4755 | Vlast_coding_system_used = coding.symbol; |
| 4756 | return make_number (coding.produced_char); |
| 4757 | } |
| 4758 | |
| 4759 | DEFUN ("decode-coding-region", Fdecode_coding_region, Sdecode_coding_region, |
| 4760 | 3, 3, "r\nzCoding system: ", |
| 4761 | "Decode the current region by specified coding system.\n\ |
| 4762 | When called from a program, takes three arguments:\n\ |
| 4763 | START, END, and CODING-SYSTEM. START and END are buffer positions.\n\ |
| 4764 | This function sets `last-coding-system-used' to the precise coding system\n\ |
| 4765 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ |
| 4766 | not fully specified.)\n\ |
| 4767 | It returns the length of the decoded text.") |
| 4768 | (start, end, coding_system) |
| 4769 | Lisp_Object start, end, coding_system; |
| 4770 | { |
| 4771 | return code_convert_region1 (start, end, coding_system, 0); |
| 4772 | } |
| 4773 | |
| 4774 | DEFUN ("encode-coding-region", Fencode_coding_region, Sencode_coding_region, |
| 4775 | 3, 3, "r\nzCoding system: ", |
| 4776 | "Encode the current region by specified coding system.\n\ |
| 4777 | When called from a program, takes three arguments:\n\ |
| 4778 | START, END, and CODING-SYSTEM. START and END are buffer positions.\n\ |
| 4779 | This function sets `last-coding-system-used' to the precise coding system\n\ |
| 4780 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ |
| 4781 | not fully specified.)\n\ |
| 4782 | It returns the length of the encoded text.") |
| 4783 | (start, end, coding_system) |
| 4784 | Lisp_Object start, end, coding_system; |
| 4785 | { |
| 4786 | return code_convert_region1 (start, end, coding_system, 1); |
| 4787 | } |
| 4788 | |
| 4789 | Lisp_Object |
| 4790 | code_convert_string1 (string, coding_system, nocopy, encodep) |
| 4791 | Lisp_Object string, coding_system, nocopy; |
| 4792 | int encodep; |
| 4793 | { |
| 4794 | struct coding_system coding; |
| 4795 | |
| 4796 | CHECK_STRING (string, 0); |
| 4797 | CHECK_SYMBOL (coding_system, 1); |
| 4798 | |
| 4799 | if (NILP (coding_system)) |
| 4800 | return (NILP (nocopy) ? Fcopy_sequence (string) : string); |
| 4801 | |
| 4802 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
| 4803 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); |
| 4804 | |
| 4805 | coding.mode |= CODING_MODE_LAST_BLOCK; |
| 4806 | Vlast_coding_system_used = coding.symbol; |
| 4807 | return code_convert_string (string, &coding, encodep, !NILP (nocopy)); |
| 4808 | } |
| 4809 | |
| 4810 | DEFUN ("decode-coding-string", Fdecode_coding_string, Sdecode_coding_string, |
| 4811 | 2, 3, 0, |
| 4812 | "Decode STRING which is encoded in CODING-SYSTEM, and return the result.\n\ |
| 4813 | Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\ |
| 4814 | if the decoding operation is trivial.\n\ |
| 4815 | This function sets `last-coding-system-used' to the precise coding system\n\ |
| 4816 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ |
| 4817 | not fully specified.)") |
| 4818 | (string, coding_system, nocopy) |
| 4819 | Lisp_Object string, coding_system, nocopy; |
| 4820 | { |
| 4821 | return code_convert_string1 (string, coding_system, nocopy, 0); |
| 4822 | } |
| 4823 | |
| 4824 | DEFUN ("encode-coding-string", Fencode_coding_string, Sencode_coding_string, |
| 4825 | 2, 3, 0, |
| 4826 | "Encode STRING to CODING-SYSTEM, and return the result.\n\ |
| 4827 | Optional arg NOCOPY non-nil means it is ok to return STRING itself\n\ |
| 4828 | if the encoding operation is trivial.\n\ |
| 4829 | This function sets `last-coding-system-used' to the precise coding system\n\ |
| 4830 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is\n\ |
| 4831 | not fully specified.)") |
| 4832 | (string, coding_system, nocopy) |
| 4833 | Lisp_Object string, coding_system, nocopy; |
| 4834 | { |
| 4835 | return code_convert_string1 (string, coding_system, nocopy, 1); |
| 4836 | } |
| 4837 | |
| 4838 | /* Encode or decode STRING according to CODING_SYSTEM. |
| 4839 | Do not set Vlast_coding_system_used. */ |
| 4840 | |
| 4841 | Lisp_Object |
| 4842 | code_convert_string_norecord (string, coding_system, encodep) |
| 4843 | Lisp_Object string, coding_system; |
| 4844 | int encodep; |
| 4845 | { |
| 4846 | struct coding_system coding; |
| 4847 | |
| 4848 | CHECK_STRING (string, 0); |
| 4849 | CHECK_SYMBOL (coding_system, 1); |
| 4850 | |
| 4851 | if (NILP (coding_system)) |
| 4852 | return string; |
| 4853 | |
| 4854 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
| 4855 | error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data); |
| 4856 | |
| 4857 | coding.mode |= CODING_MODE_LAST_BLOCK; |
| 4858 | return code_convert_string (string, &coding, encodep, Qt); |
| 4859 | } |
| 4860 | \f |
| 4861 | DEFUN ("decode-sjis-char", Fdecode_sjis_char, Sdecode_sjis_char, 1, 1, 0, |
| 4862 | "Decode a JISX0208 character of shift-jis encoding.\n\ |
| 4863 | CODE is the character code in SJIS.\n\ |
| 4864 | Return the corresponding character.") |
| 4865 | (code) |
| 4866 | Lisp_Object code; |
| 4867 | { |
| 4868 | unsigned char c1, c2, s1, s2; |
| 4869 | Lisp_Object val; |
| 4870 | |
| 4871 | CHECK_NUMBER (code, 0); |
| 4872 | s1 = (XFASTINT (code)) >> 8, s2 = (XFASTINT (code)) & 0xFF; |
| 4873 | DECODE_SJIS (s1, s2, c1, c2); |
| 4874 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset_jisx0208, c1, c2)); |
| 4875 | return val; |
| 4876 | } |
| 4877 | |
| 4878 | DEFUN ("encode-sjis-char", Fencode_sjis_char, Sencode_sjis_char, 1, 1, 0, |
| 4879 | "Encode a JISX0208 character CHAR to SJIS coding system.\n\ |
| 4880 | Return the corresponding character code in SJIS.") |
| 4881 | (ch) |
| 4882 | Lisp_Object ch; |
| 4883 | { |
| 4884 | int charset, c1, c2, s1, s2; |
| 4885 | Lisp_Object val; |
| 4886 | |
| 4887 | CHECK_NUMBER (ch, 0); |
| 4888 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
| 4889 | if (charset == charset_jisx0208) |
| 4890 | { |
| 4891 | ENCODE_SJIS (c1, c2, s1, s2); |
| 4892 | XSETFASTINT (val, (s1 << 8) | s2); |
| 4893 | } |
| 4894 | else |
| 4895 | XSETFASTINT (val, 0); |
| 4896 | return val; |
| 4897 | } |
| 4898 | |
| 4899 | DEFUN ("decode-big5-char", Fdecode_big5_char, Sdecode_big5_char, 1, 1, 0, |
| 4900 | "Decode a Big5 character CODE of BIG5 coding system.\n\ |
| 4901 | CODE is the character code in BIG5.\n\ |
| 4902 | Return the corresponding character.") |
| 4903 | (code) |
| 4904 | Lisp_Object code; |
| 4905 | { |
| 4906 | int charset; |
| 4907 | unsigned char b1, b2, c1, c2; |
| 4908 | Lisp_Object val; |
| 4909 | |
| 4910 | CHECK_NUMBER (code, 0); |
| 4911 | b1 = (XFASTINT (code)) >> 8, b2 = (XFASTINT (code)) & 0xFF; |
| 4912 | DECODE_BIG5 (b1, b2, charset, c1, c2); |
| 4913 | XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset, c1, c2)); |
| 4914 | return val; |
| 4915 | } |
| 4916 | |
| 4917 | DEFUN ("encode-big5-char", Fencode_big5_char, Sencode_big5_char, 1, 1, 0, |
| 4918 | "Encode the Big5 character CHAR to BIG5 coding system.\n\ |
| 4919 | Return the corresponding character code in Big5.") |
| 4920 | (ch) |
| 4921 | Lisp_Object ch; |
| 4922 | { |
| 4923 | int charset, c1, c2, b1, b2; |
| 4924 | Lisp_Object val; |
| 4925 | |
| 4926 | CHECK_NUMBER (ch, 0); |
| 4927 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
| 4928 | if (charset == charset_big5_1 || charset == charset_big5_2) |
| 4929 | { |
| 4930 | ENCODE_BIG5 (charset, c1, c2, b1, b2); |
| 4931 | XSETFASTINT (val, (b1 << 8) | b2); |
| 4932 | } |
| 4933 | else |
| 4934 | XSETFASTINT (val, 0); |
| 4935 | return val; |
| 4936 | } |
| 4937 | \f |
| 4938 | DEFUN ("set-terminal-coding-system-internal", |
| 4939 | Fset_terminal_coding_system_internal, |
| 4940 | Sset_terminal_coding_system_internal, 1, 1, 0, "") |
| 4941 | (coding_system) |
| 4942 | Lisp_Object coding_system; |
| 4943 | { |
| 4944 | CHECK_SYMBOL (coding_system, 0); |
| 4945 | setup_coding_system (Fcheck_coding_system (coding_system), &terminal_coding); |
| 4946 | /* We had better not send unsafe characters to terminal. */ |
| 4947 | terminal_coding.flags |= CODING_FLAG_ISO_SAFE; |
| 4948 | |
| 4949 | return Qnil; |
| 4950 | } |
| 4951 | |
| 4952 | DEFUN ("set-safe-terminal-coding-system-internal", |
| 4953 | Fset_safe_terminal_coding_system_internal, |
| 4954 | Sset_safe_terminal_coding_system_internal, 1, 1, 0, "") |
| 4955 | (coding_system) |
| 4956 | Lisp_Object coding_system; |
| 4957 | { |
| 4958 | CHECK_SYMBOL (coding_system, 0); |
| 4959 | setup_coding_system (Fcheck_coding_system (coding_system), |
| 4960 | &safe_terminal_coding); |
| 4961 | return Qnil; |
| 4962 | } |
| 4963 | |
| 4964 | DEFUN ("terminal-coding-system", |
| 4965 | Fterminal_coding_system, Sterminal_coding_system, 0, 0, 0, |
| 4966 | "Return coding system specified for terminal output.") |
| 4967 | () |
| 4968 | { |
| 4969 | return terminal_coding.symbol; |
| 4970 | } |
| 4971 | |
| 4972 | DEFUN ("set-keyboard-coding-system-internal", |
| 4973 | Fset_keyboard_coding_system_internal, |
| 4974 | Sset_keyboard_coding_system_internal, 1, 1, 0, "") |
| 4975 | (coding_system) |
| 4976 | Lisp_Object coding_system; |
| 4977 | { |
| 4978 | CHECK_SYMBOL (coding_system, 0); |
| 4979 | setup_coding_system (Fcheck_coding_system (coding_system), &keyboard_coding); |
| 4980 | return Qnil; |
| 4981 | } |
| 4982 | |
| 4983 | DEFUN ("keyboard-coding-system", |
| 4984 | Fkeyboard_coding_system, Skeyboard_coding_system, 0, 0, 0, |
| 4985 | "Return coding system specified for decoding keyboard input.") |
| 4986 | () |
| 4987 | { |
| 4988 | return keyboard_coding.symbol; |
| 4989 | } |
| 4990 | |
| 4991 | \f |
| 4992 | DEFUN ("find-operation-coding-system", Ffind_operation_coding_system, |
| 4993 | Sfind_operation_coding_system, 1, MANY, 0, |
| 4994 | "Choose a coding system for an operation based on the target name.\n\ |
| 4995 | The value names a pair of coding systems: (DECODING-SYSTEM ENCODING-SYSTEM).\n\ |
| 4996 | DECODING-SYSTEM is the coding system to use for decoding\n\ |
| 4997 | \(in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system\n\ |
| 4998 | for encoding (in case OPERATION does encoding).\n\ |
| 4999 | \n\ |
| 5000 | The first argument OPERATION specifies an I/O primitive:\n\ |
| 5001 | For file I/O, `insert-file-contents' or `write-region'.\n\ |
| 5002 | For process I/O, `call-process', `call-process-region', or `start-process'.\n\ |
| 5003 | For network I/O, `open-network-stream'.\n\ |
| 5004 | \n\ |
| 5005 | The remaining arguments should be the same arguments that were passed\n\ |
| 5006 | to the primitive. Depending on which primitive, one of those arguments\n\ |
| 5007 | is selected as the TARGET. For example, if OPERATION does file I/O,\n\ |
| 5008 | whichever argument specifies the file name is TARGET.\n\ |
| 5009 | \n\ |
| 5010 | TARGET has a meaning which depends on OPERATION:\n\ |
| 5011 | For file I/O, TARGET is a file name.\n\ |
| 5012 | For process I/O, TARGET is a process name.\n\ |
| 5013 | For network I/O, TARGET is a service name or a port number\n\ |
| 5014 | \n\ |
| 5015 | This function looks up what specified for TARGET in,\n\ |
| 5016 | `file-coding-system-alist', `process-coding-system-alist',\n\ |
| 5017 | or `network-coding-system-alist' depending on OPERATION.\n\ |
| 5018 | They may specify a coding system, a cons of coding systems,\n\ |
| 5019 | or a function symbol to call.\n\ |
| 5020 | In the last case, we call the function with one argument,\n\ |
| 5021 | which is a list of all the arguments given to this function.") |
| 5022 | (nargs, args) |
| 5023 | int nargs; |
| 5024 | Lisp_Object *args; |
| 5025 | { |
| 5026 | Lisp_Object operation, target_idx, target, val; |
| 5027 | register Lisp_Object chain; |
| 5028 | |
| 5029 | if (nargs < 2) |
| 5030 | error ("Too few arguments"); |
| 5031 | operation = args[0]; |
| 5032 | if (!SYMBOLP (operation) |
| 5033 | || !INTEGERP (target_idx = Fget (operation, Qtarget_idx))) |
| 5034 | error ("Invalid first arguement"); |
| 5035 | if (nargs < 1 + XINT (target_idx)) |
| 5036 | error ("Too few arguments for operation: %s", |
| 5037 | XSYMBOL (operation)->name->data); |
| 5038 | target = args[XINT (target_idx) + 1]; |
| 5039 | if (!(STRINGP (target) |
| 5040 | || (EQ (operation, Qopen_network_stream) && INTEGERP (target)))) |
| 5041 | error ("Invalid %dth argument", XINT (target_idx) + 1); |
| 5042 | |
| 5043 | chain = ((EQ (operation, Qinsert_file_contents) |
| 5044 | || EQ (operation, Qwrite_region)) |
| 5045 | ? Vfile_coding_system_alist |
| 5046 | : (EQ (operation, Qopen_network_stream) |
| 5047 | ? Vnetwork_coding_system_alist |
| 5048 | : Vprocess_coding_system_alist)); |
| 5049 | if (NILP (chain)) |
| 5050 | return Qnil; |
| 5051 | |
| 5052 | for (; CONSP (chain); chain = XCONS (chain)->cdr) |
| 5053 | { |
| 5054 | Lisp_Object elt; |
| 5055 | elt = XCONS (chain)->car; |
| 5056 | |
| 5057 | if (CONSP (elt) |
| 5058 | && ((STRINGP (target) |
| 5059 | && STRINGP (XCONS (elt)->car) |
| 5060 | && fast_string_match (XCONS (elt)->car, target) >= 0) |
| 5061 | || (INTEGERP (target) && EQ (target, XCONS (elt)->car)))) |
| 5062 | { |
| 5063 | val = XCONS (elt)->cdr; |
| 5064 | /* Here, if VAL is both a valid coding system and a valid |
| 5065 | function symbol, we return VAL as a coding system. */ |
| 5066 | if (CONSP (val)) |
| 5067 | return val; |
| 5068 | if (! SYMBOLP (val)) |
| 5069 | return Qnil; |
| 5070 | if (! NILP (Fcoding_system_p (val))) |
| 5071 | return Fcons (val, val); |
| 5072 | if (! NILP (Ffboundp (val))) |
| 5073 | { |
| 5074 | val = call1 (val, Flist (nargs, args)); |
| 5075 | if (CONSP (val)) |
| 5076 | return val; |
| 5077 | if (SYMBOLP (val) && ! NILP (Fcoding_system_p (val))) |
| 5078 | return Fcons (val, val); |
| 5079 | } |
| 5080 | return Qnil; |
| 5081 | } |
| 5082 | } |
| 5083 | return Qnil; |
| 5084 | } |
| 5085 | |
| 5086 | DEFUN ("update-coding-systems-internal", Fupdate_coding_systems_internal, |
| 5087 | Supdate_coding_systems_internal, 0, 0, 0, |
| 5088 | "Update internal database for ISO2022 and CCL based coding systems.\n\ |
| 5089 | When values of the following coding categories are changed, you must\n\ |
| 5090 | call this function:\n\ |
| 5091 | coding-category-iso-7, coding-category-iso-7-tight,\n\ |
| 5092 | coding-category-iso-8-1, coding-category-iso-8-2,\n\ |
| 5093 | coding-category-iso-7-else, coding-category-iso-8-else,\n\ |
| 5094 | coding-category-ccl") |
| 5095 | () |
| 5096 | { |
| 5097 | int i; |
| 5098 | |
| 5099 | for (i = CODING_CATEGORY_IDX_ISO_7; i <= CODING_CATEGORY_IDX_CCL; i++) |
| 5100 | { |
| 5101 | Lisp_Object val; |
| 5102 | |
| 5103 | val = XSYMBOL (XVECTOR (Vcoding_category_table)->contents[i])->value; |
| 5104 | if (!NILP (val)) |
| 5105 | { |
| 5106 | if (! coding_system_table[i]) |
| 5107 | coding_system_table[i] = ((struct coding_system *) |
| 5108 | xmalloc (sizeof (struct coding_system))); |
| 5109 | setup_coding_system (val, coding_system_table[i]); |
| 5110 | } |
| 5111 | else if (coding_system_table[i]) |
| 5112 | { |
| 5113 | xfree (coding_system_table[i]); |
| 5114 | coding_system_table[i] = NULL; |
| 5115 | } |
| 5116 | } |
| 5117 | |
| 5118 | return Qnil; |
| 5119 | } |
| 5120 | |
| 5121 | DEFUN ("set-coding-priority-internal", Fset_coding_priority_internal, |
| 5122 | Sset_coding_priority_internal, 0, 0, 0, |
| 5123 | "Update internal database for the current value of `coding-category-list'.\n\ |
| 5124 | This function is internal use only.") |
| 5125 | () |
| 5126 | { |
| 5127 | int i = 0, idx; |
| 5128 | Lisp_Object val; |
| 5129 | |
| 5130 | val = Vcoding_category_list; |
| 5131 | |
| 5132 | while (CONSP (val) && i < CODING_CATEGORY_IDX_MAX) |
| 5133 | { |
| 5134 | if (! SYMBOLP (XCONS (val)->car)) |
| 5135 | break; |
| 5136 | idx = XFASTINT (Fget (XCONS (val)->car, Qcoding_category_index)); |
| 5137 | if (idx >= CODING_CATEGORY_IDX_MAX) |
| 5138 | break; |
| 5139 | coding_priorities[i++] = (1 << idx); |
| 5140 | val = XCONS (val)->cdr; |
| 5141 | } |
| 5142 | /* If coding-category-list is valid and contains all coding |
| 5143 | categories, `i' should be CODING_CATEGORY_IDX_MAX now. If not, |
| 5144 | the following code saves Emacs from craching. */ |
| 5145 | while (i < CODING_CATEGORY_IDX_MAX) |
| 5146 | coding_priorities[i++] = CODING_CATEGORY_MASK_RAW_TEXT; |
| 5147 | |
| 5148 | return Qnil; |
| 5149 | } |
| 5150 | |
| 5151 | #endif /* emacs */ |
| 5152 | |
| 5153 | \f |
| 5154 | /*** 9. Post-amble ***/ |
| 5155 | |
| 5156 | void |
| 5157 | init_coding () |
| 5158 | { |
| 5159 | conversion_buffer = (char *) xmalloc (MINIMUM_CONVERSION_BUFFER_SIZE); |
| 5160 | } |
| 5161 | |
| 5162 | void |
| 5163 | init_coding_once () |
| 5164 | { |
| 5165 | int i; |
| 5166 | |
| 5167 | /* Emacs' internal format specific initialize routine. */ |
| 5168 | for (i = 0; i <= 0x20; i++) |
| 5169 | emacs_code_class[i] = EMACS_control_code; |
| 5170 | emacs_code_class[0x0A] = EMACS_linefeed_code; |
| 5171 | emacs_code_class[0x0D] = EMACS_carriage_return_code; |
| 5172 | for (i = 0x21 ; i < 0x7F; i++) |
| 5173 | emacs_code_class[i] = EMACS_ascii_code; |
| 5174 | emacs_code_class[0x7F] = EMACS_control_code; |
| 5175 | emacs_code_class[0x80] = EMACS_leading_code_composition; |
| 5176 | for (i = 0x81; i < 0xFF; i++) |
| 5177 | emacs_code_class[i] = EMACS_invalid_code; |
| 5178 | emacs_code_class[LEADING_CODE_PRIVATE_11] = EMACS_leading_code_3; |
| 5179 | emacs_code_class[LEADING_CODE_PRIVATE_12] = EMACS_leading_code_3; |
| 5180 | emacs_code_class[LEADING_CODE_PRIVATE_21] = EMACS_leading_code_4; |
| 5181 | emacs_code_class[LEADING_CODE_PRIVATE_22] = EMACS_leading_code_4; |
| 5182 | |
| 5183 | /* ISO2022 specific initialize routine. */ |
| 5184 | for (i = 0; i < 0x20; i++) |
| 5185 | iso_code_class[i] = ISO_control_code; |
| 5186 | for (i = 0x21; i < 0x7F; i++) |
| 5187 | iso_code_class[i] = ISO_graphic_plane_0; |
| 5188 | for (i = 0x80; i < 0xA0; i++) |
| 5189 | iso_code_class[i] = ISO_control_code; |
| 5190 | for (i = 0xA1; i < 0xFF; i++) |
| 5191 | iso_code_class[i] = ISO_graphic_plane_1; |
| 5192 | iso_code_class[0x20] = iso_code_class[0x7F] = ISO_0x20_or_0x7F; |
| 5193 | iso_code_class[0xA0] = iso_code_class[0xFF] = ISO_0xA0_or_0xFF; |
| 5194 | iso_code_class[ISO_CODE_CR] = ISO_carriage_return; |
| 5195 | iso_code_class[ISO_CODE_SO] = ISO_shift_out; |
| 5196 | iso_code_class[ISO_CODE_SI] = ISO_shift_in; |
| 5197 | iso_code_class[ISO_CODE_SS2_7] = ISO_single_shift_2_7; |
| 5198 | iso_code_class[ISO_CODE_ESC] = ISO_escape; |
| 5199 | iso_code_class[ISO_CODE_SS2] = ISO_single_shift_2; |
| 5200 | iso_code_class[ISO_CODE_SS3] = ISO_single_shift_3; |
| 5201 | iso_code_class[ISO_CODE_CSI] = ISO_control_sequence_introducer; |
| 5202 | |
| 5203 | conversion_buffer_size = MINIMUM_CONVERSION_BUFFER_SIZE; |
| 5204 | |
| 5205 | setup_coding_system (Qnil, &keyboard_coding); |
| 5206 | setup_coding_system (Qnil, &terminal_coding); |
| 5207 | setup_coding_system (Qnil, &safe_terminal_coding); |
| 5208 | setup_coding_system (Qnil, &default_buffer_file_coding); |
| 5209 | |
| 5210 | bzero (coding_system_table, sizeof coding_system_table); |
| 5211 | |
| 5212 | bzero (ascii_skip_code, sizeof ascii_skip_code); |
| 5213 | for (i = 0; i < 128; i++) |
| 5214 | ascii_skip_code[i] = 1; |
| 5215 | |
| 5216 | #if defined (MSDOS) || defined (WINDOWSNT) |
| 5217 | system_eol_type = CODING_EOL_CRLF; |
| 5218 | #else |
| 5219 | system_eol_type = CODING_EOL_LF; |
| 5220 | #endif |
| 5221 | } |
| 5222 | |
| 5223 | #ifdef emacs |
| 5224 | |
| 5225 | void |
| 5226 | syms_of_coding () |
| 5227 | { |
| 5228 | Qtarget_idx = intern ("target-idx"); |
| 5229 | staticpro (&Qtarget_idx); |
| 5230 | |
| 5231 | Qcoding_system_history = intern ("coding-system-history"); |
| 5232 | staticpro (&Qcoding_system_history); |
| 5233 | Fset (Qcoding_system_history, Qnil); |
| 5234 | |
| 5235 | /* Target FILENAME is the first argument. */ |
| 5236 | Fput (Qinsert_file_contents, Qtarget_idx, make_number (0)); |
| 5237 | /* Target FILENAME is the third argument. */ |
| 5238 | Fput (Qwrite_region, Qtarget_idx, make_number (2)); |
| 5239 | |
| 5240 | Qcall_process = intern ("call-process"); |
| 5241 | staticpro (&Qcall_process); |
| 5242 | /* Target PROGRAM is the first argument. */ |
| 5243 | Fput (Qcall_process, Qtarget_idx, make_number (0)); |
| 5244 | |
| 5245 | Qcall_process_region = intern ("call-process-region"); |
| 5246 | staticpro (&Qcall_process_region); |
| 5247 | /* Target PROGRAM is the third argument. */ |
| 5248 | Fput (Qcall_process_region, Qtarget_idx, make_number (2)); |
| 5249 | |
| 5250 | Qstart_process = intern ("start-process"); |
| 5251 | staticpro (&Qstart_process); |
| 5252 | /* Target PROGRAM is the third argument. */ |
| 5253 | Fput (Qstart_process, Qtarget_idx, make_number (2)); |
| 5254 | |
| 5255 | Qopen_network_stream = intern ("open-network-stream"); |
| 5256 | staticpro (&Qopen_network_stream); |
| 5257 | /* Target SERVICE is the fourth argument. */ |
| 5258 | Fput (Qopen_network_stream, Qtarget_idx, make_number (3)); |
| 5259 | |
| 5260 | Qcoding_system = intern ("coding-system"); |
| 5261 | staticpro (&Qcoding_system); |
| 5262 | |
| 5263 | Qeol_type = intern ("eol-type"); |
| 5264 | staticpro (&Qeol_type); |
| 5265 | |
| 5266 | Qbuffer_file_coding_system = intern ("buffer-file-coding-system"); |
| 5267 | staticpro (&Qbuffer_file_coding_system); |
| 5268 | |
| 5269 | Qpost_read_conversion = intern ("post-read-conversion"); |
| 5270 | staticpro (&Qpost_read_conversion); |
| 5271 | |
| 5272 | Qpre_write_conversion = intern ("pre-write-conversion"); |
| 5273 | staticpro (&Qpre_write_conversion); |
| 5274 | |
| 5275 | Qno_conversion = intern ("no-conversion"); |
| 5276 | staticpro (&Qno_conversion); |
| 5277 | |
| 5278 | Qundecided = intern ("undecided"); |
| 5279 | staticpro (&Qundecided); |
| 5280 | |
| 5281 | Qcoding_system_p = intern ("coding-system-p"); |
| 5282 | staticpro (&Qcoding_system_p); |
| 5283 | |
| 5284 | Qcoding_system_error = intern ("coding-system-error"); |
| 5285 | staticpro (&Qcoding_system_error); |
| 5286 | |
| 5287 | Fput (Qcoding_system_error, Qerror_conditions, |
| 5288 | Fcons (Qcoding_system_error, Fcons (Qerror, Qnil))); |
| 5289 | Fput (Qcoding_system_error, Qerror_message, |
| 5290 | build_string ("Invalid coding system")); |
| 5291 | |
| 5292 | Qcoding_category = intern ("coding-category"); |
| 5293 | staticpro (&Qcoding_category); |
| 5294 | Qcoding_category_index = intern ("coding-category-index"); |
| 5295 | staticpro (&Qcoding_category_index); |
| 5296 | |
| 5297 | Vcoding_category_table |
| 5298 | = Fmake_vector (make_number (CODING_CATEGORY_IDX_MAX), Qnil); |
| 5299 | staticpro (&Vcoding_category_table); |
| 5300 | { |
| 5301 | int i; |
| 5302 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) |
| 5303 | { |
| 5304 | XVECTOR (Vcoding_category_table)->contents[i] |
| 5305 | = intern (coding_category_name[i]); |
| 5306 | Fput (XVECTOR (Vcoding_category_table)->contents[i], |
| 5307 | Qcoding_category_index, make_number (i)); |
| 5308 | } |
| 5309 | } |
| 5310 | |
| 5311 | Qtranslation_table = intern ("translation-table"); |
| 5312 | staticpro (&Qtranslation_table); |
| 5313 | Fput (Qtranslation_table, Qchar_table_extra_slots, make_number (1)); |
| 5314 | |
| 5315 | Qtranslation_table_id = intern ("translation-table-id"); |
| 5316 | staticpro (&Qtranslation_table_id); |
| 5317 | |
| 5318 | Qtranslation_table_for_decode = intern ("translation-table-for-decode"); |
| 5319 | staticpro (&Qtranslation_table_for_decode); |
| 5320 | |
| 5321 | Qtranslation_table_for_encode = intern ("translation-table-for-encode"); |
| 5322 | staticpro (&Qtranslation_table_for_encode); |
| 5323 | |
| 5324 | Qsafe_charsets = intern ("safe-charsets"); |
| 5325 | staticpro (&Qsafe_charsets); |
| 5326 | |
| 5327 | Qvalid_codes = intern ("valid-codes"); |
| 5328 | staticpro (&Qvalid_codes); |
| 5329 | |
| 5330 | Qemacs_mule = intern ("emacs-mule"); |
| 5331 | staticpro (&Qemacs_mule); |
| 5332 | |
| 5333 | Qraw_text = intern ("raw-text"); |
| 5334 | staticpro (&Qraw_text); |
| 5335 | |
| 5336 | defsubr (&Scoding_system_p); |
| 5337 | defsubr (&Sread_coding_system); |
| 5338 | defsubr (&Sread_non_nil_coding_system); |
| 5339 | defsubr (&Scheck_coding_system); |
| 5340 | defsubr (&Sdetect_coding_region); |
| 5341 | defsubr (&Sdetect_coding_string); |
| 5342 | defsubr (&Sdecode_coding_region); |
| 5343 | defsubr (&Sencode_coding_region); |
| 5344 | defsubr (&Sdecode_coding_string); |
| 5345 | defsubr (&Sencode_coding_string); |
| 5346 | defsubr (&Sdecode_sjis_char); |
| 5347 | defsubr (&Sencode_sjis_char); |
| 5348 | defsubr (&Sdecode_big5_char); |
| 5349 | defsubr (&Sencode_big5_char); |
| 5350 | defsubr (&Sset_terminal_coding_system_internal); |
| 5351 | defsubr (&Sset_safe_terminal_coding_system_internal); |
| 5352 | defsubr (&Sterminal_coding_system); |
| 5353 | defsubr (&Sset_keyboard_coding_system_internal); |
| 5354 | defsubr (&Skeyboard_coding_system); |
| 5355 | defsubr (&Sfind_operation_coding_system); |
| 5356 | defsubr (&Supdate_coding_systems_internal); |
| 5357 | defsubr (&Sset_coding_priority_internal); |
| 5358 | |
| 5359 | DEFVAR_LISP ("coding-system-list", &Vcoding_system_list, |
| 5360 | "List of coding systems.\n\ |
| 5361 | \n\ |
| 5362 | Do not alter the value of this variable manually. This variable should be\n\ |
| 5363 | updated by the functions `make-coding-system' and\n\ |
| 5364 | `define-coding-system-alias'."); |
| 5365 | Vcoding_system_list = Qnil; |
| 5366 | |
| 5367 | DEFVAR_LISP ("coding-system-alist", &Vcoding_system_alist, |
| 5368 | "Alist of coding system names.\n\ |
| 5369 | Each element is one element list of coding system name.\n\ |
| 5370 | This variable is given to `completing-read' as TABLE argument.\n\ |
| 5371 | \n\ |
| 5372 | Do not alter the value of this variable manually. This variable should be\n\ |
| 5373 | updated by the functions `make-coding-system' and\n\ |
| 5374 | `define-coding-system-alias'."); |
| 5375 | Vcoding_system_alist = Qnil; |
| 5376 | |
| 5377 | DEFVAR_LISP ("coding-category-list", &Vcoding_category_list, |
| 5378 | "List of coding-categories (symbols) ordered by priority."); |
| 5379 | { |
| 5380 | int i; |
| 5381 | |
| 5382 | Vcoding_category_list = Qnil; |
| 5383 | for (i = CODING_CATEGORY_IDX_MAX - 1; i >= 0; i--) |
| 5384 | Vcoding_category_list |
| 5385 | = Fcons (XVECTOR (Vcoding_category_table)->contents[i], |
| 5386 | Vcoding_category_list); |
| 5387 | } |
| 5388 | |
| 5389 | DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read, |
| 5390 | "Specify the coding system for read operations.\n\ |
| 5391 | It is useful to bind this variable with `let', but do not set it globally.\n\ |
| 5392 | If the value is a coding system, it is used for decoding on read operation.\n\ |
| 5393 | If not, an appropriate element is used from one of the coding system alists:\n\ |
| 5394 | There are three such tables, `file-coding-system-alist',\n\ |
| 5395 | `process-coding-system-alist', and `network-coding-system-alist'."); |
| 5396 | Vcoding_system_for_read = Qnil; |
| 5397 | |
| 5398 | DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write, |
| 5399 | "Specify the coding system for write operations.\n\ |
| 5400 | It is useful to bind this variable with `let', but do not set it globally.\n\ |
| 5401 | If the value is a coding system, it is used for encoding on write operation.\n\ |
| 5402 | If not, an appropriate element is used from one of the coding system alists:\n\ |
| 5403 | There are three such tables, `file-coding-system-alist',\n\ |
| 5404 | `process-coding-system-alist', and `network-coding-system-alist'."); |
| 5405 | Vcoding_system_for_write = Qnil; |
| 5406 | |
| 5407 | DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used, |
| 5408 | "Coding system used in the latest file or process I/O."); |
| 5409 | Vlast_coding_system_used = Qnil; |
| 5410 | |
| 5411 | DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion, |
| 5412 | "*Non-nil inhibit code conversion of end-of-line format in any cases."); |
| 5413 | inhibit_eol_conversion = 0; |
| 5414 | |
| 5415 | DEFVAR_BOOL ("inherit-process-coding-system", &inherit_process_coding_system, |
| 5416 | "Non-nil means process buffer inherits coding system of process output.\n\ |
| 5417 | Bind it to t if the process output is to be treated as if it were a file\n\ |
| 5418 | read from some filesystem."); |
| 5419 | inherit_process_coding_system = 0; |
| 5420 | |
| 5421 | DEFVAR_LISP ("file-coding-system-alist", &Vfile_coding_system_alist, |
| 5422 | "Alist to decide a coding system to use for a file I/O operation.\n\ |
| 5423 | The format is ((PATTERN . VAL) ...),\n\ |
| 5424 | where PATTERN is a regular expression matching a file name,\n\ |
| 5425 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ |
| 5426 | If VAL is a coding system, it is used for both decoding and encoding\n\ |
| 5427 | the file contents.\n\ |
| 5428 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ |
| 5429 | and the cdr part is used for encoding.\n\ |
| 5430 | If VAL is a function symbol, the function must return a coding system\n\ |
| 5431 | or a cons of coding systems which are used as above.\n\ |
| 5432 | \n\ |
| 5433 | See also the function `find-operation-coding-system'\n\ |
| 5434 | and the variable `auto-coding-alist'."); |
| 5435 | Vfile_coding_system_alist = Qnil; |
| 5436 | |
| 5437 | DEFVAR_LISP ("process-coding-system-alist", &Vprocess_coding_system_alist, |
| 5438 | "Alist to decide a coding system to use for a process I/O operation.\n\ |
| 5439 | The format is ((PATTERN . VAL) ...),\n\ |
| 5440 | where PATTERN is a regular expression matching a program name,\n\ |
| 5441 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ |
| 5442 | If VAL is a coding system, it is used for both decoding what received\n\ |
| 5443 | from the program and encoding what sent to the program.\n\ |
| 5444 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ |
| 5445 | and the cdr part is used for encoding.\n\ |
| 5446 | If VAL is a function symbol, the function must return a coding system\n\ |
| 5447 | or a cons of coding systems which are used as above.\n\ |
| 5448 | \n\ |
| 5449 | See also the function `find-operation-coding-system'."); |
| 5450 | Vprocess_coding_system_alist = Qnil; |
| 5451 | |
| 5452 | DEFVAR_LISP ("network-coding-system-alist", &Vnetwork_coding_system_alist, |
| 5453 | "Alist to decide a coding system to use for a network I/O operation.\n\ |
| 5454 | The format is ((PATTERN . VAL) ...),\n\ |
| 5455 | where PATTERN is a regular expression matching a network service name\n\ |
| 5456 | or is a port number to connect to,\n\ |
| 5457 | VAL is a coding system, a cons of coding systems, or a function symbol.\n\ |
| 5458 | If VAL is a coding system, it is used for both decoding what received\n\ |
| 5459 | from the network stream and encoding what sent to the network stream.\n\ |
| 5460 | If VAL is a cons of coding systems, the car part is used for decoding,\n\ |
| 5461 | and the cdr part is used for encoding.\n\ |
| 5462 | If VAL is a function symbol, the function must return a coding system\n\ |
| 5463 | or a cons of coding systems which are used as above.\n\ |
| 5464 | \n\ |
| 5465 | See also the function `find-operation-coding-system'."); |
| 5466 | Vnetwork_coding_system_alist = Qnil; |
| 5467 | |
| 5468 | DEFVAR_INT ("eol-mnemonic-unix", &eol_mnemonic_unix, |
| 5469 | "Mnemonic character indicating UNIX-like end-of-line format (i.e. LF) ."); |
| 5470 | eol_mnemonic_unix = ':'; |
| 5471 | |
| 5472 | DEFVAR_INT ("eol-mnemonic-dos", &eol_mnemonic_dos, |
| 5473 | "Mnemonic character indicating DOS-like end-of-line format (i.e. CRLF)."); |
| 5474 | eol_mnemonic_dos = '\\'; |
| 5475 | |
| 5476 | DEFVAR_INT ("eol-mnemonic-mac", &eol_mnemonic_mac, |
| 5477 | "Mnemonic character indicating MAC-like end-of-line format (i.e. CR)."); |
| 5478 | eol_mnemonic_mac = '/'; |
| 5479 | |
| 5480 | DEFVAR_INT ("eol-mnemonic-undecided", &eol_mnemonic_undecided, |
| 5481 | "Mnemonic character indicating end-of-line format is not yet decided."); |
| 5482 | eol_mnemonic_undecided = ':'; |
| 5483 | |
| 5484 | DEFVAR_LISP ("enable-character-translation", &Venable_character_translation, |
| 5485 | "*Non-nil enables character translation while encoding and decoding."); |
| 5486 | Venable_character_translation = Qt; |
| 5487 | |
| 5488 | DEFVAR_LISP ("standard-translation-table-for-decode", |
| 5489 | &Vstandard_translation_table_for_decode, |
| 5490 | "Table for translating characters while decoding."); |
| 5491 | Vstandard_translation_table_for_decode = Qnil; |
| 5492 | |
| 5493 | DEFVAR_LISP ("standard-translation-table-for-encode", |
| 5494 | &Vstandard_translation_table_for_encode, |
| 5495 | "Table for translationg characters while encoding."); |
| 5496 | Vstandard_translation_table_for_encode = Qnil; |
| 5497 | |
| 5498 | DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist, |
| 5499 | "Alist of charsets vs revision numbers.\n\ |
| 5500 | While encoding, if a charset (car part of an element) is found,\n\ |
| 5501 | designate it with the escape sequence identifing revision (cdr part of the element)."); |
| 5502 | Vcharset_revision_alist = Qnil; |
| 5503 | |
| 5504 | DEFVAR_LISP ("default-process-coding-system", |
| 5505 | &Vdefault_process_coding_system, |
| 5506 | "Cons of coding systems used for process I/O by default.\n\ |
| 5507 | The car part is used for decoding a process output,\n\ |
| 5508 | the cdr part is used for encoding a text to be sent to a process."); |
| 5509 | Vdefault_process_coding_system = Qnil; |
| 5510 | |
| 5511 | DEFVAR_LISP ("latin-extra-code-table", &Vlatin_extra_code_table, |
| 5512 | "Table of extra Latin codes in the range 128..159 (inclusive).\n\ |
| 5513 | This is a vector of length 256.\n\ |
| 5514 | If Nth element is non-nil, the existence of code N in a file\n\ |
| 5515 | \(or output of subprocess) doesn't prevent it to be detected as\n\ |
| 5516 | a coding system of ISO 2022 variant which has a flag\n\ |
| 5517 | `accept-latin-extra-code' t (e.g. iso-latin-1) on reading a file\n\ |
| 5518 | or reading output of a subprocess.\n\ |
| 5519 | Only 128th through 159th elements has a meaning."); |
| 5520 | Vlatin_extra_code_table = Fmake_vector (make_number (256), Qnil); |
| 5521 | |
| 5522 | DEFVAR_LISP ("select-safe-coding-system-function", |
| 5523 | &Vselect_safe_coding_system_function, |
| 5524 | "Function to call to select safe coding system for encoding a text.\n\ |
| 5525 | \n\ |
| 5526 | If set, this function is called to force a user to select a proper\n\ |
| 5527 | coding system which can encode the text in the case that a default\n\ |
| 5528 | coding system used in each operation can't encode the text.\n\ |
| 5529 | \n\ |
| 5530 | The default value is `select-safe-coding-system' (which see)."); |
| 5531 | Vselect_safe_coding_system_function = Qnil; |
| 5532 | |
| 5533 | } |
| 5534 | |
| 5535 | #endif /* emacs */ |