| 1 | /* Coding system handler (conversion, detection, and etc). |
| 2 | Copyright (C) 1995, 1997, 1998, 2002 Electrotechnical Laboratory, JAPAN. |
| 3 | Licensed to the Free Software Foundation. |
| 4 | Copyright (C) 2001 Free Software Foundation, Inc. |
| 5 | |
| 6 | This file is part of GNU Emacs. |
| 7 | |
| 8 | GNU Emacs is free software; you can redistribute it and/or modify |
| 9 | it under the terms of the GNU General Public License as published by |
| 10 | the Free Software Foundation; either version 2, or (at your option) |
| 11 | any later version. |
| 12 | |
| 13 | GNU Emacs is distributed in the hope that it will be useful, |
| 14 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 15 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 16 | GNU General Public License for more details. |
| 17 | |
| 18 | You should have received a copy of the GNU General Public License |
| 19 | along with GNU Emacs; see the file COPYING. If not, write to |
| 20 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, |
| 21 | Boston, MA 02111-1307, USA. */ |
| 22 | |
| 23 | /*** TABLE OF CONTENTS *** |
| 24 | |
| 25 | 0. General comments |
| 26 | 1. Preamble |
| 27 | 2. Emacs' internal format (emacs-mule) handlers |
| 28 | 3. ISO2022 handlers |
| 29 | 4. Shift-JIS and BIG5 handlers |
| 30 | 5. CCL handlers |
| 31 | 6. End-of-line handlers |
| 32 | 7. C library functions |
| 33 | 8. Emacs Lisp library functions |
| 34 | 9. Post-amble |
| 35 | |
| 36 | */ |
| 37 | |
| 38 | /*** 0. General comments ***/ |
| 39 | |
| 40 | |
| 41 | /*** GENERAL NOTE on CODING SYSTEMS *** |
| 42 | |
| 43 | A coding system is an encoding mechanism for one or more character |
| 44 | sets. Here's a list of coding systems which Emacs can handle. When |
| 45 | we say "decode", it means converting some other coding system to |
| 46 | Emacs' internal format (emacs-mule), and when we say "encode", |
| 47 | it means converting the coding system emacs-mule to some other |
| 48 | coding system. |
| 49 | |
| 50 | 0. Emacs' internal format (emacs-mule) |
| 51 | |
| 52 | Emacs itself holds a multi-lingual character in buffers and strings |
| 53 | in a special format. Details are described in section 2. |
| 54 | |
| 55 | 1. ISO2022 |
| 56 | |
| 57 | The most famous coding system for multiple character sets. X's |
| 58 | Compound Text, various EUCs (Extended Unix Code), and coding |
| 59 | systems used in Internet communication such as ISO-2022-JP are |
| 60 | all variants of ISO2022. Details are described in section 3. |
| 61 | |
| 62 | 2. SJIS (or Shift-JIS or MS-Kanji-Code) |
| 63 | |
| 64 | A coding system to encode character sets: ASCII, JISX0201, and |
| 65 | JISX0208. Widely used for PC's in Japan. Details are described in |
| 66 | section 4. |
| 67 | |
| 68 | 3. BIG5 |
| 69 | |
| 70 | A coding system to encode the character sets ASCII and Big5. Widely |
| 71 | used for Chinese (mainly in Taiwan and Hong Kong). Details are |
| 72 | described in section 4. In this file, when we write "BIG5" |
| 73 | (all uppercase), we mean the coding system, and when we write |
| 74 | "Big5" (capitalized), we mean the character set. |
| 75 | |
| 76 | 4. Raw text |
| 77 | |
| 78 | A coding system for text containing random 8-bit code. Emacs does |
| 79 | no code conversion on such text except for end-of-line format. |
| 80 | |
| 81 | 5. Other |
| 82 | |
| 83 | If a user wants to read/write text encoded in a coding system not |
| 84 | listed above, he can supply a decoder and an encoder for it as CCL |
| 85 | (Code Conversion Language) programs. Emacs executes the CCL program |
| 86 | while reading/writing. |
| 87 | |
| 88 | Emacs represents a coding system by a Lisp symbol that has a property |
| 89 | `coding-system'. But, before actually using the coding system, the |
| 90 | information about it is set in a structure of type `struct |
| 91 | coding_system' for rapid processing. See section 6 for more details. |
| 92 | |
| 93 | */ |
| 94 | |
| 95 | /*** GENERAL NOTES on END-OF-LINE FORMAT *** |
| 96 | |
| 97 | How end-of-line of text is encoded depends on the operating system. |
| 98 | For instance, Unix's format is just one byte of `line-feed' code, |
| 99 | whereas DOS's format is two-byte sequence of `carriage-return' and |
| 100 | `line-feed' codes. MacOS's format is usually one byte of |
| 101 | `carriage-return'. |
| 102 | |
| 103 | Since text character encoding and end-of-line encoding are |
| 104 | independent, any coding system described above can have any |
| 105 | end-of-line format. So Emacs has information about end-of-line |
| 106 | format in each coding-system. See section 6 for more details. |
| 107 | |
| 108 | */ |
| 109 | |
| 110 | /*** GENERAL NOTES on `detect_coding_XXX ()' functions *** |
| 111 | |
| 112 | These functions check if a text between SRC and SRC_END is encoded |
| 113 | in the coding system category XXX. Each returns an integer value in |
| 114 | which appropriate flag bits for the category XXX are set. The flag |
| 115 | bits are defined in macros CODING_CATEGORY_MASK_XXX. Below is the |
| 116 | template for these functions. If MULTIBYTEP is nonzero, 8-bit codes |
| 117 | of the range 0x80..0x9F are in multibyte form. */ |
| 118 | #if 0 |
| 119 | int |
| 120 | detect_coding_emacs_mule (src, src_end, multibytep) |
| 121 | unsigned char *src, *src_end; |
| 122 | int multibytep; |
| 123 | { |
| 124 | ... |
| 125 | } |
| 126 | #endif |
| 127 | |
| 128 | /*** GENERAL NOTES on `decode_coding_XXX ()' functions *** |
| 129 | |
| 130 | These functions decode SRC_BYTES length of unibyte text at SOURCE |
| 131 | encoded in CODING to Emacs' internal format. The resulting |
| 132 | multibyte text goes to a place pointed to by DESTINATION, the length |
| 133 | of which should not exceed DST_BYTES. |
| 134 | |
| 135 | These functions set the information about original and decoded texts |
| 136 | in the members `produced', `produced_char', `consumed', and |
| 137 | `consumed_char' of the structure *CODING. They also set the member |
| 138 | `result' to one of CODING_FINISH_XXX indicating how the decoding |
| 139 | finished. |
| 140 | |
| 141 | DST_BYTES zero means that the source area and destination area are |
| 142 | overlapped, which means that we can produce a decoded text until it |
| 143 | reaches the head of the not-yet-decoded source text. |
| 144 | |
| 145 | Below is a template for these functions. */ |
| 146 | #if 0 |
| 147 | static void |
| 148 | decode_coding_XXX (coding, source, destination, src_bytes, dst_bytes) |
| 149 | struct coding_system *coding; |
| 150 | unsigned char *source, *destination; |
| 151 | int src_bytes, dst_bytes; |
| 152 | { |
| 153 | ... |
| 154 | } |
| 155 | #endif |
| 156 | |
| 157 | /*** GENERAL NOTES on `encode_coding_XXX ()' functions *** |
| 158 | |
| 159 | These functions encode SRC_BYTES length text at SOURCE from Emacs' |
| 160 | internal multibyte format to CODING. The resulting unibyte text |
| 161 | goes to a place pointed to by DESTINATION, the length of which |
| 162 | should not exceed DST_BYTES. |
| 163 | |
| 164 | These functions set the information about original and encoded texts |
| 165 | in the members `produced', `produced_char', `consumed', and |
| 166 | `consumed_char' of the structure *CODING. They also set the member |
| 167 | `result' to one of CODING_FINISH_XXX indicating how the encoding |
| 168 | finished. |
| 169 | |
| 170 | DST_BYTES zero means that the source area and destination area are |
| 171 | overlapped, which means that we can produce encoded text until it |
| 172 | reaches at the head of the not-yet-encoded source text. |
| 173 | |
| 174 | Below is a template for these functions. */ |
| 175 | #if 0 |
| 176 | static void |
| 177 | encode_coding_XXX (coding, source, destination, src_bytes, dst_bytes) |
| 178 | struct coding_system *coding; |
| 179 | unsigned char *source, *destination; |
| 180 | int src_bytes, dst_bytes; |
| 181 | { |
| 182 | ... |
| 183 | } |
| 184 | #endif |
| 185 | |
| 186 | /*** COMMONLY USED MACROS ***/ |
| 187 | |
| 188 | /* The following two macros ONE_MORE_BYTE and TWO_MORE_BYTES safely |
| 189 | get one, two, and three bytes from the source text respectively. |
| 190 | If there are not enough bytes in the source, they jump to |
| 191 | `label_end_of_loop'. The caller should set variables `coding', |
| 192 | `src' and `src_end' to appropriate pointer in advance. These |
| 193 | macros are called from decoding routines `decode_coding_XXX', thus |
| 194 | it is assumed that the source text is unibyte. */ |
| 195 | |
| 196 | #define ONE_MORE_BYTE(c1) \ |
| 197 | do { \ |
| 198 | if (src >= src_end) \ |
| 199 | { \ |
| 200 | coding->result = CODING_FINISH_INSUFFICIENT_SRC; \ |
| 201 | goto label_end_of_loop; \ |
| 202 | } \ |
| 203 | c1 = *src++; \ |
| 204 | } while (0) |
| 205 | |
| 206 | #define TWO_MORE_BYTES(c1, c2) \ |
| 207 | do { \ |
| 208 | if (src + 1 >= src_end) \ |
| 209 | { \ |
| 210 | coding->result = CODING_FINISH_INSUFFICIENT_SRC; \ |
| 211 | goto label_end_of_loop; \ |
| 212 | } \ |
| 213 | c1 = *src++; \ |
| 214 | c2 = *src++; \ |
| 215 | } while (0) |
| 216 | |
| 217 | |
| 218 | /* Like ONE_MORE_BYTE, but 8-bit bytes of data at SRC are in multibyte |
| 219 | form if MULTIBYTEP is nonzero. */ |
| 220 | |
| 221 | #define ONE_MORE_BYTE_CHECK_MULTIBYTE(c1, multibytep) \ |
| 222 | do { \ |
| 223 | if (src >= src_end) \ |
| 224 | { \ |
| 225 | coding->result = CODING_FINISH_INSUFFICIENT_SRC; \ |
| 226 | goto label_end_of_loop; \ |
| 227 | } \ |
| 228 | c1 = *src++; \ |
| 229 | if (multibytep && c1 == LEADING_CODE_8_BIT_CONTROL) \ |
| 230 | c1 = *src++ - 0x20; \ |
| 231 | } while (0) |
| 232 | |
| 233 | /* Set C to the next character at the source text pointed by `src'. |
| 234 | If there are not enough characters in the source, jump to |
| 235 | `label_end_of_loop'. The caller should set variables `coding' |
| 236 | `src', `src_end', and `translation_table' to appropriate pointers |
| 237 | in advance. This macro is used in encoding routines |
| 238 | `encode_coding_XXX', thus it assumes that the source text is in |
| 239 | multibyte form except for 8-bit characters. 8-bit characters are |
| 240 | in multibyte form if coding->src_multibyte is nonzero, else they |
| 241 | are represented by a single byte. */ |
| 242 | |
| 243 | #define ONE_MORE_CHAR(c) \ |
| 244 | do { \ |
| 245 | int len = src_end - src; \ |
| 246 | int bytes; \ |
| 247 | if (len <= 0) \ |
| 248 | { \ |
| 249 | coding->result = CODING_FINISH_INSUFFICIENT_SRC; \ |
| 250 | goto label_end_of_loop; \ |
| 251 | } \ |
| 252 | if (coding->src_multibyte \ |
| 253 | || UNIBYTE_STR_AS_MULTIBYTE_P (src, len, bytes)) \ |
| 254 | c = STRING_CHAR_AND_LENGTH (src, len, bytes); \ |
| 255 | else \ |
| 256 | c = *src, bytes = 1; \ |
| 257 | if (!NILP (translation_table)) \ |
| 258 | c = translate_char (translation_table, c, -1, 0, 0); \ |
| 259 | src += bytes; \ |
| 260 | } while (0) |
| 261 | |
| 262 | |
| 263 | /* Produce a multibyte form of character C to `dst'. Jump to |
| 264 | `label_end_of_loop' if there's not enough space at `dst'. |
| 265 | |
| 266 | If we are now in the middle of a composition sequence, the decoded |
| 267 | character may be ALTCHAR (for the current composition). In that |
| 268 | case, the character goes to coding->cmp_data->data instead of |
| 269 | `dst'. |
| 270 | |
| 271 | This macro is used in decoding routines. */ |
| 272 | |
| 273 | #define EMIT_CHAR(c) \ |
| 274 | do { \ |
| 275 | if (! COMPOSING_P (coding) \ |
| 276 | || coding->composing == COMPOSITION_RELATIVE \ |
| 277 | || coding->composing == COMPOSITION_WITH_RULE) \ |
| 278 | { \ |
| 279 | int bytes = CHAR_BYTES (c); \ |
| 280 | if ((dst + bytes) > (dst_bytes ? dst_end : src)) \ |
| 281 | { \ |
| 282 | coding->result = CODING_FINISH_INSUFFICIENT_DST; \ |
| 283 | goto label_end_of_loop; \ |
| 284 | } \ |
| 285 | dst += CHAR_STRING (c, dst); \ |
| 286 | coding->produced_char++; \ |
| 287 | } \ |
| 288 | \ |
| 289 | if (COMPOSING_P (coding) \ |
| 290 | && coding->composing != COMPOSITION_RELATIVE) \ |
| 291 | { \ |
| 292 | CODING_ADD_COMPOSITION_COMPONENT (coding, c); \ |
| 293 | coding->composition_rule_follows \ |
| 294 | = coding->composing != COMPOSITION_WITH_ALTCHARS; \ |
| 295 | } \ |
| 296 | } while (0) |
| 297 | |
| 298 | |
| 299 | #define EMIT_ONE_BYTE(c) \ |
| 300 | do { \ |
| 301 | if (dst >= (dst_bytes ? dst_end : src)) \ |
| 302 | { \ |
| 303 | coding->result = CODING_FINISH_INSUFFICIENT_DST; \ |
| 304 | goto label_end_of_loop; \ |
| 305 | } \ |
| 306 | *dst++ = c; \ |
| 307 | } while (0) |
| 308 | |
| 309 | #define EMIT_TWO_BYTES(c1, c2) \ |
| 310 | do { \ |
| 311 | if (dst + 2 > (dst_bytes ? dst_end : src)) \ |
| 312 | { \ |
| 313 | coding->result = CODING_FINISH_INSUFFICIENT_DST; \ |
| 314 | goto label_end_of_loop; \ |
| 315 | } \ |
| 316 | *dst++ = c1, *dst++ = c2; \ |
| 317 | } while (0) |
| 318 | |
| 319 | #define EMIT_BYTES(from, to) \ |
| 320 | do { \ |
| 321 | if (dst + (to - from) > (dst_bytes ? dst_end : src)) \ |
| 322 | { \ |
| 323 | coding->result = CODING_FINISH_INSUFFICIENT_DST; \ |
| 324 | goto label_end_of_loop; \ |
| 325 | } \ |
| 326 | while (from < to) \ |
| 327 | *dst++ = *from++; \ |
| 328 | } while (0) |
| 329 | |
| 330 | \f |
| 331 | /*** 1. Preamble ***/ |
| 332 | |
| 333 | #ifdef emacs |
| 334 | #include <config.h> |
| 335 | #endif |
| 336 | |
| 337 | #include <stdio.h> |
| 338 | |
| 339 | #ifdef emacs |
| 340 | |
| 341 | #include "lisp.h" |
| 342 | #include "buffer.h" |
| 343 | #include "charset.h" |
| 344 | #include "composite.h" |
| 345 | #include "ccl.h" |
| 346 | #include "coding.h" |
| 347 | #include "window.h" |
| 348 | |
| 349 | #else /* not emacs */ |
| 350 | |
| 351 | #include "mulelib.h" |
| 352 | |
| 353 | #endif /* not emacs */ |
| 354 | |
| 355 | Lisp_Object Qcoding_system, Qeol_type; |
| 356 | Lisp_Object Qbuffer_file_coding_system; |
| 357 | Lisp_Object Qpost_read_conversion, Qpre_write_conversion; |
| 358 | Lisp_Object Qno_conversion, Qundecided; |
| 359 | Lisp_Object Qcoding_system_history; |
| 360 | Lisp_Object Qsafe_chars; |
| 361 | Lisp_Object Qvalid_codes; |
| 362 | |
| 363 | extern Lisp_Object Qinsert_file_contents, Qwrite_region; |
| 364 | Lisp_Object Qcall_process, Qcall_process_region, Qprocess_argument; |
| 365 | Lisp_Object Qstart_process, Qopen_network_stream; |
| 366 | Lisp_Object Qtarget_idx; |
| 367 | |
| 368 | Lisp_Object Vselect_safe_coding_system_function; |
| 369 | |
| 370 | /* Mnemonic string for each format of end-of-line. */ |
| 371 | Lisp_Object eol_mnemonic_unix, eol_mnemonic_dos, eol_mnemonic_mac; |
| 372 | /* Mnemonic string to indicate format of end-of-line is not yet |
| 373 | decided. */ |
| 374 | Lisp_Object eol_mnemonic_undecided; |
| 375 | |
| 376 | /* Format of end-of-line decided by system. This is CODING_EOL_LF on |
| 377 | Unix, CODING_EOL_CRLF on DOS/Windows, and CODING_EOL_CR on Mac. */ |
| 378 | int system_eol_type; |
| 379 | |
| 380 | #ifdef emacs |
| 381 | |
| 382 | Lisp_Object Vcoding_system_list, Vcoding_system_alist; |
| 383 | |
| 384 | Lisp_Object Qcoding_system_p, Qcoding_system_error; |
| 385 | |
| 386 | /* Coding system emacs-mule and raw-text are for converting only |
| 387 | end-of-line format. */ |
| 388 | Lisp_Object Qemacs_mule, Qraw_text; |
| 389 | |
| 390 | /* Coding-systems are handed between Emacs Lisp programs and C internal |
| 391 | routines by the following three variables. */ |
| 392 | /* Coding-system for reading files and receiving data from process. */ |
| 393 | Lisp_Object Vcoding_system_for_read; |
| 394 | /* Coding-system for writing files and sending data to process. */ |
| 395 | Lisp_Object Vcoding_system_for_write; |
| 396 | /* Coding-system actually used in the latest I/O. */ |
| 397 | Lisp_Object Vlast_coding_system_used; |
| 398 | |
| 399 | /* A vector of length 256 which contains information about special |
| 400 | Latin codes (especially for dealing with Microsoft codes). */ |
| 401 | Lisp_Object Vlatin_extra_code_table; |
| 402 | |
| 403 | /* Flag to inhibit code conversion of end-of-line format. */ |
| 404 | int inhibit_eol_conversion; |
| 405 | |
| 406 | /* Flag to inhibit ISO2022 escape sequence detection. */ |
| 407 | int inhibit_iso_escape_detection; |
| 408 | |
| 409 | /* Flag to make buffer-file-coding-system inherit from process-coding. */ |
| 410 | int inherit_process_coding_system; |
| 411 | |
| 412 | /* Coding system to be used to encode text for terminal display. */ |
| 413 | struct coding_system terminal_coding; |
| 414 | |
| 415 | /* Coding system to be used to encode text for terminal display when |
| 416 | terminal coding system is nil. */ |
| 417 | struct coding_system safe_terminal_coding; |
| 418 | |
| 419 | /* Coding system of what is sent from terminal keyboard. */ |
| 420 | struct coding_system keyboard_coding; |
| 421 | |
| 422 | /* Default coding system to be used to write a file. */ |
| 423 | struct coding_system default_buffer_file_coding; |
| 424 | |
| 425 | Lisp_Object Vfile_coding_system_alist; |
| 426 | Lisp_Object Vprocess_coding_system_alist; |
| 427 | Lisp_Object Vnetwork_coding_system_alist; |
| 428 | |
| 429 | Lisp_Object Vlocale_coding_system; |
| 430 | |
| 431 | #endif /* emacs */ |
| 432 | |
| 433 | Lisp_Object Qcoding_category, Qcoding_category_index; |
| 434 | |
| 435 | /* List of symbols `coding-category-xxx' ordered by priority. */ |
| 436 | Lisp_Object Vcoding_category_list; |
| 437 | |
| 438 | /* Table of coding categories (Lisp symbols). */ |
| 439 | Lisp_Object Vcoding_category_table; |
| 440 | |
| 441 | /* Table of names of symbol for each coding-category. */ |
| 442 | char *coding_category_name[CODING_CATEGORY_IDX_MAX] = { |
| 443 | "coding-category-emacs-mule", |
| 444 | "coding-category-sjis", |
| 445 | "coding-category-iso-7", |
| 446 | "coding-category-iso-7-tight", |
| 447 | "coding-category-iso-8-1", |
| 448 | "coding-category-iso-8-2", |
| 449 | "coding-category-iso-7-else", |
| 450 | "coding-category-iso-8-else", |
| 451 | "coding-category-ccl", |
| 452 | "coding-category-big5", |
| 453 | "coding-category-utf-8", |
| 454 | "coding-category-utf-16-be", |
| 455 | "coding-category-utf-16-le", |
| 456 | "coding-category-raw-text", |
| 457 | "coding-category-binary" |
| 458 | }; |
| 459 | |
| 460 | /* Table of pointers to coding systems corresponding to each coding |
| 461 | categories. */ |
| 462 | struct coding_system *coding_system_table[CODING_CATEGORY_IDX_MAX]; |
| 463 | |
| 464 | /* Table of coding category masks. Nth element is a mask for a coding |
| 465 | category of which priority is Nth. */ |
| 466 | static |
| 467 | int coding_priorities[CODING_CATEGORY_IDX_MAX]; |
| 468 | |
| 469 | /* Flag to tell if we look up translation table on character code |
| 470 | conversion. */ |
| 471 | Lisp_Object Venable_character_translation; |
| 472 | /* Standard translation table to look up on decoding (reading). */ |
| 473 | Lisp_Object Vstandard_translation_table_for_decode; |
| 474 | /* Standard translation table to look up on encoding (writing). */ |
| 475 | Lisp_Object Vstandard_translation_table_for_encode; |
| 476 | |
| 477 | Lisp_Object Qtranslation_table; |
| 478 | Lisp_Object Qtranslation_table_id; |
| 479 | Lisp_Object Qtranslation_table_for_decode; |
| 480 | Lisp_Object Qtranslation_table_for_encode; |
| 481 | |
| 482 | /* Alist of charsets vs revision number. */ |
| 483 | Lisp_Object Vcharset_revision_alist; |
| 484 | |
| 485 | /* Default coding systems used for process I/O. */ |
| 486 | Lisp_Object Vdefault_process_coding_system; |
| 487 | |
| 488 | /* Global flag to tell that we can't call post-read-conversion and |
| 489 | pre-write-conversion functions. Usually the value is zero, but it |
| 490 | is set to 1 temporarily while such functions are running. This is |
| 491 | to avoid infinite recursive call. */ |
| 492 | static int inhibit_pre_post_conversion; |
| 493 | |
| 494 | /* Char-table containing safe coding systems of each character. */ |
| 495 | Lisp_Object Vchar_coding_system_table; |
| 496 | Lisp_Object Qchar_coding_system; |
| 497 | |
| 498 | /* Return `safe-chars' property of coding system CODING. Don't check |
| 499 | validity of CODING. */ |
| 500 | |
| 501 | Lisp_Object |
| 502 | coding_safe_chars (coding) |
| 503 | struct coding_system *coding; |
| 504 | { |
| 505 | Lisp_Object coding_spec, plist, safe_chars; |
| 506 | |
| 507 | coding_spec = Fget (coding->symbol, Qcoding_system); |
| 508 | plist = XVECTOR (coding_spec)->contents[3]; |
| 509 | safe_chars = Fplist_get (XVECTOR (coding_spec)->contents[3], Qsafe_chars); |
| 510 | return (CHAR_TABLE_P (safe_chars) ? safe_chars : Qt); |
| 511 | } |
| 512 | |
| 513 | #define CODING_SAFE_CHAR_P(safe_chars, c) \ |
| 514 | (EQ (safe_chars, Qt) || !NILP (CHAR_TABLE_REF (safe_chars, c))) |
| 515 | |
| 516 | \f |
| 517 | /*** 2. Emacs internal format (emacs-mule) handlers ***/ |
| 518 | |
| 519 | /* Emacs' internal format for representation of multiple character |
| 520 | sets is a kind of multi-byte encoding, i.e. characters are |
| 521 | represented by variable-length sequences of one-byte codes. |
| 522 | |
| 523 | ASCII characters and control characters (e.g. `tab', `newline') are |
| 524 | represented by one-byte sequences which are their ASCII codes, in |
| 525 | the range 0x00 through 0x7F. |
| 526 | |
| 527 | 8-bit characters of the range 0x80..0x9F are represented by |
| 528 | two-byte sequences of LEADING_CODE_8_BIT_CONTROL and (their 8-bit |
| 529 | code + 0x20). |
| 530 | |
| 531 | 8-bit characters of the range 0xA0..0xFF are represented by |
| 532 | one-byte sequences which are their 8-bit code. |
| 533 | |
| 534 | The other characters are represented by a sequence of `base |
| 535 | leading-code', optional `extended leading-code', and one or two |
| 536 | `position-code's. The length of the sequence is determined by the |
| 537 | base leading-code. Leading-code takes the range 0x81 through 0x9D, |
| 538 | whereas extended leading-code and position-code take the range 0xA0 |
| 539 | through 0xFF. See `charset.h' for more details about leading-code |
| 540 | and position-code. |
| 541 | |
| 542 | --- CODE RANGE of Emacs' internal format --- |
| 543 | character set range |
| 544 | ------------- ----- |
| 545 | ascii 0x00..0x7F |
| 546 | eight-bit-control LEADING_CODE_8_BIT_CONTROL + 0xA0..0xBF |
| 547 | eight-bit-graphic 0xA0..0xBF |
| 548 | ELSE 0x81..0x9D + [0xA0..0xFF]+ |
| 549 | --------------------------------------------- |
| 550 | |
| 551 | As this is the internal character representation, the format is |
| 552 | usually not used externally (i.e. in a file or in a data sent to a |
| 553 | process). But, it is possible to have a text externally in this |
| 554 | format (i.e. by encoding by the coding system `emacs-mule'). |
| 555 | |
| 556 | In that case, a sequence of one-byte codes has a slightly different |
| 557 | form. |
| 558 | |
| 559 | Firstly, all characters in eight-bit-control are represented by |
| 560 | one-byte sequences which are their 8-bit code. |
| 561 | |
| 562 | Next, character composition data are represented by the byte |
| 563 | sequence of the form: 0x80 METHOD BYTES CHARS COMPONENT ..., |
| 564 | where, |
| 565 | METHOD is 0xF0 plus one of composition method (enum |
| 566 | composition_method), |
| 567 | |
| 568 | BYTES is 0xA0 plus the byte length of these composition data, |
| 569 | |
| 570 | CHARS is 0xA0 plus the number of characters composed by these |
| 571 | data, |
| 572 | |
| 573 | COMPONENTs are characters of multibyte form or composition |
| 574 | rules encoded by two-byte of ASCII codes. |
| 575 | |
| 576 | In addition, for backward compatibility, the following formats are |
| 577 | also recognized as composition data on decoding. |
| 578 | |
| 579 | 0x80 MSEQ ... |
| 580 | 0x80 0xFF MSEQ RULE MSEQ RULE ... MSEQ |
| 581 | |
| 582 | Here, |
| 583 | MSEQ is a multibyte form but in these special format: |
| 584 | ASCII: 0xA0 ASCII_CODE+0x80, |
| 585 | other: LEADING_CODE+0x20 FOLLOWING-BYTE ..., |
| 586 | RULE is a one byte code of the range 0xA0..0xF0 that |
| 587 | represents a composition rule. |
| 588 | */ |
| 589 | |
| 590 | enum emacs_code_class_type emacs_code_class[256]; |
| 591 | |
| 592 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 593 | Check if a text is encoded in Emacs' internal format. If it is, |
| 594 | return CODING_CATEGORY_MASK_EMACS_MULE, else return 0. */ |
| 595 | |
| 596 | static int |
| 597 | detect_coding_emacs_mule (src, src_end, multibytep) |
| 598 | unsigned char *src, *src_end; |
| 599 | int multibytep; |
| 600 | { |
| 601 | unsigned char c; |
| 602 | int composing = 0; |
| 603 | /* Dummy for ONE_MORE_BYTE. */ |
| 604 | struct coding_system dummy_coding; |
| 605 | struct coding_system *coding = &dummy_coding; |
| 606 | |
| 607 | while (1) |
| 608 | { |
| 609 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 610 | |
| 611 | if (composing) |
| 612 | { |
| 613 | if (c < 0xA0) |
| 614 | composing = 0; |
| 615 | else if (c == 0xA0) |
| 616 | { |
| 617 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 618 | c &= 0x7F; |
| 619 | } |
| 620 | else |
| 621 | c -= 0x20; |
| 622 | } |
| 623 | |
| 624 | if (c < 0x20) |
| 625 | { |
| 626 | if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) |
| 627 | return 0; |
| 628 | } |
| 629 | else if (c >= 0x80 && c < 0xA0) |
| 630 | { |
| 631 | if (c == 0x80) |
| 632 | /* Old leading code for a composite character. */ |
| 633 | composing = 1; |
| 634 | else |
| 635 | { |
| 636 | unsigned char *src_base = src - 1; |
| 637 | int bytes; |
| 638 | |
| 639 | if (!UNIBYTE_STR_AS_MULTIBYTE_P (src_base, src_end - src_base, |
| 640 | bytes)) |
| 641 | return 0; |
| 642 | src = src_base + bytes; |
| 643 | } |
| 644 | } |
| 645 | } |
| 646 | label_end_of_loop: |
| 647 | return CODING_CATEGORY_MASK_EMACS_MULE; |
| 648 | } |
| 649 | |
| 650 | |
| 651 | /* Record the starting position START and METHOD of one composition. */ |
| 652 | |
| 653 | #define CODING_ADD_COMPOSITION_START(coding, start, method) \ |
| 654 | do { \ |
| 655 | struct composition_data *cmp_data = coding->cmp_data; \ |
| 656 | int *data = cmp_data->data + cmp_data->used; \ |
| 657 | coding->cmp_data_start = cmp_data->used; \ |
| 658 | data[0] = -1; \ |
| 659 | data[1] = cmp_data->char_offset + start; \ |
| 660 | data[3] = (int) method; \ |
| 661 | cmp_data->used += 4; \ |
| 662 | } while (0) |
| 663 | |
| 664 | /* Record the ending position END of the current composition. */ |
| 665 | |
| 666 | #define CODING_ADD_COMPOSITION_END(coding, end) \ |
| 667 | do { \ |
| 668 | struct composition_data *cmp_data = coding->cmp_data; \ |
| 669 | int *data = cmp_data->data + coding->cmp_data_start; \ |
| 670 | data[0] = cmp_data->used - coding->cmp_data_start; \ |
| 671 | data[2] = cmp_data->char_offset + end; \ |
| 672 | } while (0) |
| 673 | |
| 674 | /* Record one COMPONENT (alternate character or composition rule). */ |
| 675 | |
| 676 | #define CODING_ADD_COMPOSITION_COMPONENT(coding, component) \ |
| 677 | (coding->cmp_data->data[coding->cmp_data->used++] = component) |
| 678 | |
| 679 | |
| 680 | /* Get one byte from a data pointed by SRC and increment SRC. If SRC |
| 681 | is not less than SRC_END, return -1 without incrementing Src. */ |
| 682 | |
| 683 | #define SAFE_ONE_MORE_BYTE() (src >= src_end ? -1 : *src++) |
| 684 | |
| 685 | |
| 686 | /* Decode a character represented as a component of composition |
| 687 | sequence of Emacs 20 style at SRC. Set C to that character, store |
| 688 | its multibyte form sequence at P, and set P to the end of that |
| 689 | sequence. If no valid character is found, set C to -1. */ |
| 690 | |
| 691 | #define DECODE_EMACS_MULE_COMPOSITION_CHAR(c, p) \ |
| 692 | do { \ |
| 693 | int bytes; \ |
| 694 | \ |
| 695 | c = SAFE_ONE_MORE_BYTE (); \ |
| 696 | if (c < 0) \ |
| 697 | break; \ |
| 698 | if (CHAR_HEAD_P (c)) \ |
| 699 | c = -1; \ |
| 700 | else if (c == 0xA0) \ |
| 701 | { \ |
| 702 | c = SAFE_ONE_MORE_BYTE (); \ |
| 703 | if (c < 0xA0) \ |
| 704 | c = -1; \ |
| 705 | else \ |
| 706 | { \ |
| 707 | c -= 0xA0; \ |
| 708 | *p++ = c; \ |
| 709 | } \ |
| 710 | } \ |
| 711 | else if (BASE_LEADING_CODE_P (c - 0x20)) \ |
| 712 | { \ |
| 713 | unsigned char *p0 = p; \ |
| 714 | \ |
| 715 | c -= 0x20; \ |
| 716 | *p++ = c; \ |
| 717 | bytes = BYTES_BY_CHAR_HEAD (c); \ |
| 718 | while (--bytes) \ |
| 719 | { \ |
| 720 | c = SAFE_ONE_MORE_BYTE (); \ |
| 721 | if (c < 0) \ |
| 722 | break; \ |
| 723 | *p++ = c; \ |
| 724 | } \ |
| 725 | if (UNIBYTE_STR_AS_MULTIBYTE_P (p0, p - p0, bytes)) \ |
| 726 | c = STRING_CHAR (p0, bytes); \ |
| 727 | else \ |
| 728 | c = -1; \ |
| 729 | } \ |
| 730 | else \ |
| 731 | c = -1; \ |
| 732 | } while (0) |
| 733 | |
| 734 | |
| 735 | /* Decode a composition rule represented as a component of composition |
| 736 | sequence of Emacs 20 style at SRC. Set C to the rule. If not |
| 737 | valid rule is found, set C to -1. */ |
| 738 | |
| 739 | #define DECODE_EMACS_MULE_COMPOSITION_RULE(c) \ |
| 740 | do { \ |
| 741 | c = SAFE_ONE_MORE_BYTE (); \ |
| 742 | c -= 0xA0; \ |
| 743 | if (c < 0 || c >= 81) \ |
| 744 | c = -1; \ |
| 745 | else \ |
| 746 | { \ |
| 747 | gref = c / 9, nref = c % 9; \ |
| 748 | c = COMPOSITION_ENCODE_RULE (gref, nref); \ |
| 749 | } \ |
| 750 | } while (0) |
| 751 | |
| 752 | |
| 753 | /* Decode composition sequence encoded by `emacs-mule' at the source |
| 754 | pointed by SRC. SRC_END is the end of source. Store information |
| 755 | of the composition in CODING->cmp_data. |
| 756 | |
| 757 | For backward compatibility, decode also a composition sequence of |
| 758 | Emacs 20 style. In that case, the composition sequence contains |
| 759 | characters that should be extracted into a buffer or string. Store |
| 760 | those characters at *DESTINATION in multibyte form. |
| 761 | |
| 762 | If we encounter an invalid byte sequence, return 0. |
| 763 | If we encounter an insufficient source or destination, or |
| 764 | insufficient space in CODING->cmp_data, return 1. |
| 765 | Otherwise, return consumed bytes in the source. |
| 766 | |
| 767 | */ |
| 768 | static INLINE int |
| 769 | decode_composition_emacs_mule (coding, src, src_end, |
| 770 | destination, dst_end, dst_bytes) |
| 771 | struct coding_system *coding; |
| 772 | unsigned char *src, *src_end, **destination, *dst_end; |
| 773 | int dst_bytes; |
| 774 | { |
| 775 | unsigned char *dst = *destination; |
| 776 | int method, data_len, nchars; |
| 777 | unsigned char *src_base = src++; |
| 778 | /* Store components of composition. */ |
| 779 | int component[COMPOSITION_DATA_MAX_BUNCH_LENGTH]; |
| 780 | int ncomponent; |
| 781 | /* Store multibyte form of characters to be composed. This is for |
| 782 | Emacs 20 style composition sequence. */ |
| 783 | unsigned char buf[MAX_COMPOSITION_COMPONENTS * MAX_MULTIBYTE_LENGTH]; |
| 784 | unsigned char *bufp = buf; |
| 785 | int c, i, gref, nref; |
| 786 | |
| 787 | if (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH |
| 788 | >= COMPOSITION_DATA_SIZE) |
| 789 | { |
| 790 | coding->result = CODING_FINISH_INSUFFICIENT_CMP; |
| 791 | return -1; |
| 792 | } |
| 793 | |
| 794 | ONE_MORE_BYTE (c); |
| 795 | if (c - 0xF0 >= COMPOSITION_RELATIVE |
| 796 | && c - 0xF0 <= COMPOSITION_WITH_RULE_ALTCHARS) |
| 797 | { |
| 798 | int with_rule; |
| 799 | |
| 800 | method = c - 0xF0; |
| 801 | with_rule = (method == COMPOSITION_WITH_RULE |
| 802 | || method == COMPOSITION_WITH_RULE_ALTCHARS); |
| 803 | ONE_MORE_BYTE (c); |
| 804 | data_len = c - 0xA0; |
| 805 | if (data_len < 4 |
| 806 | || src_base + data_len > src_end) |
| 807 | return 0; |
| 808 | ONE_MORE_BYTE (c); |
| 809 | nchars = c - 0xA0; |
| 810 | if (c < 1) |
| 811 | return 0; |
| 812 | for (ncomponent = 0; src < src_base + data_len; ncomponent++) |
| 813 | { |
| 814 | /* If it is longer than this, it can't be valid. */ |
| 815 | if (ncomponent >= COMPOSITION_DATA_MAX_BUNCH_LENGTH) |
| 816 | return 0; |
| 817 | |
| 818 | if (ncomponent % 2 && with_rule) |
| 819 | { |
| 820 | ONE_MORE_BYTE (gref); |
| 821 | gref -= 32; |
| 822 | ONE_MORE_BYTE (nref); |
| 823 | nref -= 32; |
| 824 | c = COMPOSITION_ENCODE_RULE (gref, nref); |
| 825 | } |
| 826 | else |
| 827 | { |
| 828 | int bytes; |
| 829 | if (UNIBYTE_STR_AS_MULTIBYTE_P (src, src_end - src, bytes)) |
| 830 | c = STRING_CHAR (src, bytes); |
| 831 | else |
| 832 | c = *src, bytes = 1; |
| 833 | src += bytes; |
| 834 | } |
| 835 | component[ncomponent] = c; |
| 836 | } |
| 837 | } |
| 838 | else |
| 839 | { |
| 840 | /* This may be an old Emacs 20 style format. See the comment at |
| 841 | the section 2 of this file. */ |
| 842 | while (src < src_end && !CHAR_HEAD_P (*src)) src++; |
| 843 | if (src == src_end |
| 844 | && !(coding->mode & CODING_MODE_LAST_BLOCK)) |
| 845 | goto label_end_of_loop; |
| 846 | |
| 847 | src_end = src; |
| 848 | src = src_base + 1; |
| 849 | if (c < 0xC0) |
| 850 | { |
| 851 | method = COMPOSITION_RELATIVE; |
| 852 | for (ncomponent = 0; ncomponent < MAX_COMPOSITION_COMPONENTS;) |
| 853 | { |
| 854 | DECODE_EMACS_MULE_COMPOSITION_CHAR (c, bufp); |
| 855 | if (c < 0) |
| 856 | break; |
| 857 | component[ncomponent++] = c; |
| 858 | } |
| 859 | if (ncomponent < 2) |
| 860 | return 0; |
| 861 | nchars = ncomponent; |
| 862 | } |
| 863 | else if (c == 0xFF) |
| 864 | { |
| 865 | method = COMPOSITION_WITH_RULE; |
| 866 | src++; |
| 867 | DECODE_EMACS_MULE_COMPOSITION_CHAR (c, bufp); |
| 868 | if (c < 0) |
| 869 | return 0; |
| 870 | component[0] = c; |
| 871 | for (ncomponent = 1; |
| 872 | ncomponent < MAX_COMPOSITION_COMPONENTS * 2 - 1;) |
| 873 | { |
| 874 | DECODE_EMACS_MULE_COMPOSITION_RULE (c); |
| 875 | if (c < 0) |
| 876 | break; |
| 877 | component[ncomponent++] = c; |
| 878 | DECODE_EMACS_MULE_COMPOSITION_CHAR (c, bufp); |
| 879 | if (c < 0) |
| 880 | break; |
| 881 | component[ncomponent++] = c; |
| 882 | } |
| 883 | if (ncomponent < 3) |
| 884 | return 0; |
| 885 | nchars = (ncomponent + 1) / 2; |
| 886 | } |
| 887 | else |
| 888 | return 0; |
| 889 | } |
| 890 | |
| 891 | if (buf == bufp || dst + (bufp - buf) <= (dst_bytes ? dst_end : src)) |
| 892 | { |
| 893 | CODING_ADD_COMPOSITION_START (coding, coding->produced_char, method); |
| 894 | for (i = 0; i < ncomponent; i++) |
| 895 | CODING_ADD_COMPOSITION_COMPONENT (coding, component[i]); |
| 896 | CODING_ADD_COMPOSITION_END (coding, coding->produced_char + nchars); |
| 897 | if (buf < bufp) |
| 898 | { |
| 899 | unsigned char *p = buf; |
| 900 | EMIT_BYTES (p, bufp); |
| 901 | *destination += bufp - buf; |
| 902 | coding->produced_char += nchars; |
| 903 | } |
| 904 | return (src - src_base); |
| 905 | } |
| 906 | label_end_of_loop: |
| 907 | return -1; |
| 908 | } |
| 909 | |
| 910 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */ |
| 911 | |
| 912 | static void |
| 913 | decode_coding_emacs_mule (coding, source, destination, src_bytes, dst_bytes) |
| 914 | struct coding_system *coding; |
| 915 | unsigned char *source, *destination; |
| 916 | int src_bytes, dst_bytes; |
| 917 | { |
| 918 | unsigned char *src = source; |
| 919 | unsigned char *src_end = source + src_bytes; |
| 920 | unsigned char *dst = destination; |
| 921 | unsigned char *dst_end = destination + dst_bytes; |
| 922 | /* SRC_BASE remembers the start position in source in each loop. |
| 923 | The loop will be exited when there's not enough source code, or |
| 924 | when there's not enough destination area to produce a |
| 925 | character. */ |
| 926 | unsigned char *src_base; |
| 927 | |
| 928 | coding->produced_char = 0; |
| 929 | while ((src_base = src) < src_end) |
| 930 | { |
| 931 | unsigned char tmp[MAX_MULTIBYTE_LENGTH], *p; |
| 932 | int bytes; |
| 933 | |
| 934 | if (*src == '\r') |
| 935 | { |
| 936 | int c = *src++; |
| 937 | |
| 938 | if (coding->eol_type == CODING_EOL_CR) |
| 939 | c = '\n'; |
| 940 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 941 | { |
| 942 | ONE_MORE_BYTE (c); |
| 943 | if (c != '\n') |
| 944 | { |
| 945 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 946 | { |
| 947 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 948 | goto label_end_of_loop; |
| 949 | } |
| 950 | src--; |
| 951 | c = '\r'; |
| 952 | } |
| 953 | } |
| 954 | *dst++ = c; |
| 955 | coding->produced_char++; |
| 956 | continue; |
| 957 | } |
| 958 | else if (*src == '\n') |
| 959 | { |
| 960 | if ((coding->eol_type == CODING_EOL_CR |
| 961 | || coding->eol_type == CODING_EOL_CRLF) |
| 962 | && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 963 | { |
| 964 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 965 | goto label_end_of_loop; |
| 966 | } |
| 967 | *dst++ = *src++; |
| 968 | coding->produced_char++; |
| 969 | continue; |
| 970 | } |
| 971 | else if (*src == 0x80) |
| 972 | { |
| 973 | /* Start of composition data. */ |
| 974 | int consumed = decode_composition_emacs_mule (coding, src, src_end, |
| 975 | &dst, dst_end, |
| 976 | dst_bytes); |
| 977 | if (consumed < 0) |
| 978 | goto label_end_of_loop; |
| 979 | else if (consumed > 0) |
| 980 | { |
| 981 | src += consumed; |
| 982 | continue; |
| 983 | } |
| 984 | bytes = CHAR_STRING (*src, tmp); |
| 985 | p = tmp; |
| 986 | src++; |
| 987 | } |
| 988 | else if (UNIBYTE_STR_AS_MULTIBYTE_P (src, src_end - src, bytes)) |
| 989 | { |
| 990 | p = src; |
| 991 | src += bytes; |
| 992 | } |
| 993 | else |
| 994 | { |
| 995 | bytes = CHAR_STRING (*src, tmp); |
| 996 | p = tmp; |
| 997 | src++; |
| 998 | } |
| 999 | if (dst + bytes >= (dst_bytes ? dst_end : src)) |
| 1000 | { |
| 1001 | coding->result = CODING_FINISH_INSUFFICIENT_DST; |
| 1002 | break; |
| 1003 | } |
| 1004 | while (bytes--) *dst++ = *p++; |
| 1005 | coding->produced_char++; |
| 1006 | } |
| 1007 | label_end_of_loop: |
| 1008 | coding->consumed = coding->consumed_char = src_base - source; |
| 1009 | coding->produced = dst - destination; |
| 1010 | } |
| 1011 | |
| 1012 | |
| 1013 | /* Encode composition data stored at DATA into a special byte sequence |
| 1014 | starting by 0x80. Update CODING->cmp_data_start and maybe |
| 1015 | CODING->cmp_data for the next call. */ |
| 1016 | |
| 1017 | #define ENCODE_COMPOSITION_EMACS_MULE(coding, data) \ |
| 1018 | do { \ |
| 1019 | unsigned char buf[1024], *p0 = buf, *p; \ |
| 1020 | int len = data[0]; \ |
| 1021 | int i; \ |
| 1022 | \ |
| 1023 | buf[0] = 0x80; \ |
| 1024 | buf[1] = 0xF0 + data[3]; /* METHOD */ \ |
| 1025 | buf[3] = 0xA0 + (data[2] - data[1]); /* COMPOSED-CHARS */ \ |
| 1026 | p = buf + 4; \ |
| 1027 | if (data[3] == COMPOSITION_WITH_RULE \ |
| 1028 | || data[3] == COMPOSITION_WITH_RULE_ALTCHARS) \ |
| 1029 | { \ |
| 1030 | p += CHAR_STRING (data[4], p); \ |
| 1031 | for (i = 5; i < len; i += 2) \ |
| 1032 | { \ |
| 1033 | int gref, nref; \ |
| 1034 | COMPOSITION_DECODE_RULE (data[i], gref, nref); \ |
| 1035 | *p++ = 0x20 + gref; \ |
| 1036 | *p++ = 0x20 + nref; \ |
| 1037 | p += CHAR_STRING (data[i + 1], p); \ |
| 1038 | } \ |
| 1039 | } \ |
| 1040 | else \ |
| 1041 | { \ |
| 1042 | for (i = 4; i < len; i++) \ |
| 1043 | p += CHAR_STRING (data[i], p); \ |
| 1044 | } \ |
| 1045 | buf[2] = 0xA0 + (p - buf); /* COMPONENTS-BYTES */ \ |
| 1046 | \ |
| 1047 | if (dst + (p - buf) + 4 > (dst_bytes ? dst_end : src)) \ |
| 1048 | { \ |
| 1049 | coding->result = CODING_FINISH_INSUFFICIENT_DST; \ |
| 1050 | goto label_end_of_loop; \ |
| 1051 | } \ |
| 1052 | while (p0 < p) \ |
| 1053 | *dst++ = *p0++; \ |
| 1054 | coding->cmp_data_start += data[0]; \ |
| 1055 | if (coding->cmp_data_start == coding->cmp_data->used \ |
| 1056 | && coding->cmp_data->next) \ |
| 1057 | { \ |
| 1058 | coding->cmp_data = coding->cmp_data->next; \ |
| 1059 | coding->cmp_data_start = 0; \ |
| 1060 | } \ |
| 1061 | } while (0) |
| 1062 | |
| 1063 | |
| 1064 | static void encode_eol P_ ((struct coding_system *, unsigned char *, |
| 1065 | unsigned char *, int, int)); |
| 1066 | |
| 1067 | static void |
| 1068 | encode_coding_emacs_mule (coding, source, destination, src_bytes, dst_bytes) |
| 1069 | struct coding_system *coding; |
| 1070 | unsigned char *source, *destination; |
| 1071 | int src_bytes, dst_bytes; |
| 1072 | { |
| 1073 | unsigned char *src = source; |
| 1074 | unsigned char *src_end = source + src_bytes; |
| 1075 | unsigned char *dst = destination; |
| 1076 | unsigned char *dst_end = destination + dst_bytes; |
| 1077 | unsigned char *src_base; |
| 1078 | int c; |
| 1079 | int char_offset; |
| 1080 | int *data; |
| 1081 | |
| 1082 | Lisp_Object translation_table; |
| 1083 | |
| 1084 | translation_table = Qnil; |
| 1085 | |
| 1086 | /* Optimization for the case that there's no composition. */ |
| 1087 | if (!coding->cmp_data || coding->cmp_data->used == 0) |
| 1088 | { |
| 1089 | encode_eol (coding, source, destination, src_bytes, dst_bytes); |
| 1090 | return; |
| 1091 | } |
| 1092 | |
| 1093 | char_offset = coding->cmp_data->char_offset; |
| 1094 | data = coding->cmp_data->data + coding->cmp_data_start; |
| 1095 | while (1) |
| 1096 | { |
| 1097 | src_base = src; |
| 1098 | |
| 1099 | /* If SRC starts a composition, encode the information about the |
| 1100 | composition in advance. */ |
| 1101 | if (coding->cmp_data_start < coding->cmp_data->used |
| 1102 | && char_offset + coding->consumed_char == data[1]) |
| 1103 | { |
| 1104 | ENCODE_COMPOSITION_EMACS_MULE (coding, data); |
| 1105 | char_offset = coding->cmp_data->char_offset; |
| 1106 | data = coding->cmp_data->data + coding->cmp_data_start; |
| 1107 | } |
| 1108 | |
| 1109 | ONE_MORE_CHAR (c); |
| 1110 | if (c == '\n' && (coding->eol_type == CODING_EOL_CRLF |
| 1111 | || coding->eol_type == CODING_EOL_CR)) |
| 1112 | { |
| 1113 | if (coding->eol_type == CODING_EOL_CRLF) |
| 1114 | EMIT_TWO_BYTES ('\r', c); |
| 1115 | else |
| 1116 | EMIT_ONE_BYTE ('\r'); |
| 1117 | } |
| 1118 | else if (SINGLE_BYTE_CHAR_P (c)) |
| 1119 | EMIT_ONE_BYTE (c); |
| 1120 | else |
| 1121 | EMIT_BYTES (src_base, src); |
| 1122 | coding->consumed_char++; |
| 1123 | } |
| 1124 | label_end_of_loop: |
| 1125 | coding->consumed = src_base - source; |
| 1126 | coding->produced = coding->produced_char = dst - destination; |
| 1127 | return; |
| 1128 | } |
| 1129 | |
| 1130 | \f |
| 1131 | /*** 3. ISO2022 handlers ***/ |
| 1132 | |
| 1133 | /* The following note describes the coding system ISO2022 briefly. |
| 1134 | Since the intention of this note is to help understand the |
| 1135 | functions in this file, some parts are NOT ACCURATE or are OVERLY |
| 1136 | SIMPLIFIED. For thorough understanding, please refer to the |
| 1137 | original document of ISO2022. This is equivalent to the standard |
| 1138 | ECMA-35, obtainable from <URL:http://www.ecma.ch/> (*). |
| 1139 | |
| 1140 | ISO2022 provides many mechanisms to encode several character sets |
| 1141 | in 7-bit and 8-bit environments. For 7-bit environments, all text |
| 1142 | is encoded using bytes less than 128. This may make the encoded |
| 1143 | text a little bit longer, but the text passes more easily through |
| 1144 | several types of gateway, some of which strip off the MSB (Most |
| 1145 | Significant Bit). |
| 1146 | |
| 1147 | There are two kinds of character sets: control character sets and |
| 1148 | graphic character sets. The former contain control characters such |
| 1149 | as `newline' and `escape' to provide control functions (control |
| 1150 | functions are also provided by escape sequences). The latter |
| 1151 | contain graphic characters such as 'A' and '-'. Emacs recognizes |
| 1152 | two control character sets and many graphic character sets. |
| 1153 | |
| 1154 | Graphic character sets are classified into one of the following |
| 1155 | four classes, according to the number of bytes (DIMENSION) and |
| 1156 | number of characters in one dimension (CHARS) of the set: |
| 1157 | - DIMENSION1_CHARS94 |
| 1158 | - DIMENSION1_CHARS96 |
| 1159 | - DIMENSION2_CHARS94 |
| 1160 | - DIMENSION2_CHARS96 |
| 1161 | |
| 1162 | In addition, each character set is assigned an identification tag, |
| 1163 | unique for each set, called the "final character" (denoted as <F> |
| 1164 | hereafter). The <F> of each character set is decided by ECMA(*) |
| 1165 | when it is registered in ISO. The code range of <F> is 0x30..0x7F |
| 1166 | (0x30..0x3F are for private use only). |
| 1167 | |
| 1168 | Note (*): ECMA = European Computer Manufacturers Association |
| 1169 | |
| 1170 | Here are examples of graphic character sets [NAME(<F>)]: |
| 1171 | o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ... |
| 1172 | o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ... |
| 1173 | o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ... |
| 1174 | o DIMENSION2_CHARS96 -- none for the moment |
| 1175 | |
| 1176 | A code area (1 byte=8 bits) is divided into 4 areas, C0, GL, C1, and GR. |
| 1177 | C0 [0x00..0x1F] -- control character plane 0 |
| 1178 | GL [0x20..0x7F] -- graphic character plane 0 |
| 1179 | C1 [0x80..0x9F] -- control character plane 1 |
| 1180 | GR [0xA0..0xFF] -- graphic character plane 1 |
| 1181 | |
| 1182 | A control character set is directly designated and invoked to C0 or |
| 1183 | C1 by an escape sequence. The most common case is that: |
| 1184 | - ISO646's control character set is designated/invoked to C0, and |
| 1185 | - ISO6429's control character set is designated/invoked to C1, |
| 1186 | and usually these designations/invocations are omitted in encoded |
| 1187 | text. In a 7-bit environment, only C0 can be used, and a control |
| 1188 | character for C1 is encoded by an appropriate escape sequence to |
| 1189 | fit into the environment. All control characters for C1 are |
| 1190 | defined to have corresponding escape sequences. |
| 1191 | |
| 1192 | A graphic character set is at first designated to one of four |
| 1193 | graphic registers (G0 through G3), then these graphic registers are |
| 1194 | invoked to GL or GR. These designations and invocations can be |
| 1195 | done independently. The most common case is that G0 is invoked to |
| 1196 | GL, G1 is invoked to GR, and ASCII is designated to G0. Usually |
| 1197 | these invocations and designations are omitted in encoded text. |
| 1198 | In a 7-bit environment, only GL can be used. |
| 1199 | |
| 1200 | When a graphic character set of CHARS94 is invoked to GL, codes |
| 1201 | 0x20 and 0x7F of the GL area work as control characters SPACE and |
| 1202 | DEL respectively, and codes 0xA0 and 0xFF of the GR area should not |
| 1203 | be used. |
| 1204 | |
| 1205 | There are two ways of invocation: locking-shift and single-shift. |
| 1206 | With locking-shift, the invocation lasts until the next different |
| 1207 | invocation, whereas with single-shift, the invocation affects the |
| 1208 | following character only and doesn't affect the locking-shift |
| 1209 | state. Invocations are done by the following control characters or |
| 1210 | escape sequences: |
| 1211 | |
| 1212 | ---------------------------------------------------------------------- |
| 1213 | abbrev function cntrl escape seq description |
| 1214 | ---------------------------------------------------------------------- |
| 1215 | SI/LS0 (shift-in) 0x0F none invoke G0 into GL |
| 1216 | SO/LS1 (shift-out) 0x0E none invoke G1 into GL |
| 1217 | LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL |
| 1218 | LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL |
| 1219 | LS1R (locking-shift-1 right) none ESC '~' invoke G1 into GR (*) |
| 1220 | LS2R (locking-shift-2 right) none ESC '}' invoke G2 into GR (*) |
| 1221 | LS3R (locking-shift 3 right) none ESC '|' invoke G3 into GR (*) |
| 1222 | SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 for one char |
| 1223 | SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 for one char |
| 1224 | ---------------------------------------------------------------------- |
| 1225 | (*) These are not used by any known coding system. |
| 1226 | |
| 1227 | Control characters for these functions are defined by macros |
| 1228 | ISO_CODE_XXX in `coding.h'. |
| 1229 | |
| 1230 | Designations are done by the following escape sequences: |
| 1231 | ---------------------------------------------------------------------- |
| 1232 | escape sequence description |
| 1233 | ---------------------------------------------------------------------- |
| 1234 | ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0 |
| 1235 | ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1 |
| 1236 | ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2 |
| 1237 | ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3 |
| 1238 | ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*) |
| 1239 | ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1 |
| 1240 | ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2 |
| 1241 | ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3 |
| 1242 | ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**) |
| 1243 | ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1 |
| 1244 | ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2 |
| 1245 | ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3 |
| 1246 | ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*) |
| 1247 | ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1 |
| 1248 | ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2 |
| 1249 | ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3 |
| 1250 | ---------------------------------------------------------------------- |
| 1251 | |
| 1252 | In this list, "DIMENSION1_CHARS94<F>" means a graphic character set |
| 1253 | of dimension 1, chars 94, and final character <F>, etc... |
| 1254 | |
| 1255 | Note (*): Although these designations are not allowed in ISO2022, |
| 1256 | Emacs accepts them on decoding, and produces them on encoding |
| 1257 | CHARS96 character sets in a coding system which is characterized as |
| 1258 | 7-bit environment, non-locking-shift, and non-single-shift. |
| 1259 | |
| 1260 | Note (**): If <F> is '@', 'A', or 'B', the intermediate character |
| 1261 | '(' can be omitted. We refer to this as "short-form" hereafter. |
| 1262 | |
| 1263 | Now you may notice that there are a lot of ways of encoding the |
| 1264 | same multilingual text in ISO2022. Actually, there exist many |
| 1265 | coding systems such as Compound Text (used in X11's inter client |
| 1266 | communication, ISO-2022-JP (used in Japanese Internet), ISO-2022-KR |
| 1267 | (used in Korean Internet), EUC (Extended UNIX Code, used in Asian |
| 1268 | localized platforms), and all of these are variants of ISO2022. |
| 1269 | |
| 1270 | In addition to the above, Emacs handles two more kinds of escape |
| 1271 | sequences: ISO6429's direction specification and Emacs' private |
| 1272 | sequence for specifying character composition. |
| 1273 | |
| 1274 | ISO6429's direction specification takes the following form: |
| 1275 | o CSI ']' -- end of the current direction |
| 1276 | o CSI '0' ']' -- end of the current direction |
| 1277 | o CSI '1' ']' -- start of left-to-right text |
| 1278 | o CSI '2' ']' -- start of right-to-left text |
| 1279 | The control character CSI (0x9B: control sequence introducer) is |
| 1280 | abbreviated to the escape sequence ESC '[' in a 7-bit environment. |
| 1281 | |
| 1282 | Character composition specification takes the following form: |
| 1283 | o ESC '0' -- start relative composition |
| 1284 | o ESC '1' -- end composition |
| 1285 | o ESC '2' -- start rule-base composition (*) |
| 1286 | o ESC '3' -- start relative composition with alternate chars (**) |
| 1287 | o ESC '4' -- start rule-base composition with alternate chars (**) |
| 1288 | Since these are not standard escape sequences of any ISO standard, |
| 1289 | the use of them with these meanings is restricted to Emacs only. |
| 1290 | |
| 1291 | (*) This form is used only in Emacs 20.5 and older versions, |
| 1292 | but the newer versions can safely decode it. |
| 1293 | (**) This form is used only in Emacs 21.1 and newer versions, |
| 1294 | and the older versions can't decode it. |
| 1295 | |
| 1296 | Here's a list of example usages of these composition escape |
| 1297 | sequences (categorized by `enum composition_method'). |
| 1298 | |
| 1299 | COMPOSITION_RELATIVE: |
| 1300 | ESC 0 CHAR [ CHAR ] ESC 1 |
| 1301 | COMPOSITION_WITH_RULE: |
| 1302 | ESC 2 CHAR [ RULE CHAR ] ESC 1 |
| 1303 | COMPOSITION_WITH_ALTCHARS: |
| 1304 | ESC 3 ALTCHAR [ ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1 |
| 1305 | COMPOSITION_WITH_RULE_ALTCHARS: |
| 1306 | ESC 4 ALTCHAR [ RULE ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1 */ |
| 1307 | |
| 1308 | enum iso_code_class_type iso_code_class[256]; |
| 1309 | |
| 1310 | #define CHARSET_OK(idx, charset, c) \ |
| 1311 | (coding_system_table[idx] \ |
| 1312 | && (charset == CHARSET_ASCII \ |
| 1313 | || (safe_chars = coding_safe_chars (coding_system_table[idx]), \ |
| 1314 | CODING_SAFE_CHAR_P (safe_chars, c))) \ |
| 1315 | && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding_system_table[idx], \ |
| 1316 | charset) \ |
| 1317 | != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)) |
| 1318 | |
| 1319 | #define SHIFT_OUT_OK(idx) \ |
| 1320 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding_system_table[idx], 1) >= 0) |
| 1321 | |
| 1322 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 1323 | Check if a text is encoded in ISO2022. If it is, return an |
| 1324 | integer in which appropriate flag bits any of: |
| 1325 | CODING_CATEGORY_MASK_ISO_7 |
| 1326 | CODING_CATEGORY_MASK_ISO_7_TIGHT |
| 1327 | CODING_CATEGORY_MASK_ISO_8_1 |
| 1328 | CODING_CATEGORY_MASK_ISO_8_2 |
| 1329 | CODING_CATEGORY_MASK_ISO_7_ELSE |
| 1330 | CODING_CATEGORY_MASK_ISO_8_ELSE |
| 1331 | are set. If a code which should never appear in ISO2022 is found, |
| 1332 | returns 0. */ |
| 1333 | |
| 1334 | static int |
| 1335 | detect_coding_iso2022 (src, src_end, multibytep) |
| 1336 | unsigned char *src, *src_end; |
| 1337 | int multibytep; |
| 1338 | { |
| 1339 | int mask = CODING_CATEGORY_MASK_ISO; |
| 1340 | int mask_found = 0; |
| 1341 | int reg[4], shift_out = 0, single_shifting = 0; |
| 1342 | int c, c1, charset; |
| 1343 | /* Dummy for ONE_MORE_BYTE. */ |
| 1344 | struct coding_system dummy_coding; |
| 1345 | struct coding_system *coding = &dummy_coding; |
| 1346 | Lisp_Object safe_chars; |
| 1347 | |
| 1348 | reg[0] = CHARSET_ASCII, reg[1] = reg[2] = reg[3] = -1; |
| 1349 | while (mask && src < src_end) |
| 1350 | { |
| 1351 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 1352 | switch (c) |
| 1353 | { |
| 1354 | case ISO_CODE_ESC: |
| 1355 | if (inhibit_iso_escape_detection) |
| 1356 | break; |
| 1357 | single_shifting = 0; |
| 1358 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 1359 | if (c >= '(' && c <= '/') |
| 1360 | { |
| 1361 | /* Designation sequence for a charset of dimension 1. */ |
| 1362 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c1, multibytep); |
| 1363 | if (c1 < ' ' || c1 >= 0x80 |
| 1364 | || (charset = iso_charset_table[0][c >= ','][c1]) < 0) |
| 1365 | /* Invalid designation sequence. Just ignore. */ |
| 1366 | break; |
| 1367 | reg[(c - '(') % 4] = charset; |
| 1368 | } |
| 1369 | else if (c == '$') |
| 1370 | { |
| 1371 | /* Designation sequence for a charset of dimension 2. */ |
| 1372 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 1373 | if (c >= '@' && c <= 'B') |
| 1374 | /* Designation for JISX0208.1978, GB2312, or JISX0208. */ |
| 1375 | reg[0] = charset = iso_charset_table[1][0][c]; |
| 1376 | else if (c >= '(' && c <= '/') |
| 1377 | { |
| 1378 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c1, multibytep); |
| 1379 | if (c1 < ' ' || c1 >= 0x80 |
| 1380 | || (charset = iso_charset_table[1][c >= ','][c1]) < 0) |
| 1381 | /* Invalid designation sequence. Just ignore. */ |
| 1382 | break; |
| 1383 | reg[(c - '(') % 4] = charset; |
| 1384 | } |
| 1385 | else |
| 1386 | /* Invalid designation sequence. Just ignore. */ |
| 1387 | break; |
| 1388 | } |
| 1389 | else if (c == 'N' || c == 'O') |
| 1390 | { |
| 1391 | /* ESC <Fe> for SS2 or SS3. */ |
| 1392 | mask &= CODING_CATEGORY_MASK_ISO_7_ELSE; |
| 1393 | break; |
| 1394 | } |
| 1395 | else if (c >= '0' && c <= '4') |
| 1396 | { |
| 1397 | /* ESC <Fp> for start/end composition. */ |
| 1398 | mask_found |= CODING_CATEGORY_MASK_ISO; |
| 1399 | break; |
| 1400 | } |
| 1401 | else |
| 1402 | /* Invalid escape sequence. Just ignore. */ |
| 1403 | break; |
| 1404 | |
| 1405 | /* We found a valid designation sequence for CHARSET. */ |
| 1406 | mask &= ~CODING_CATEGORY_MASK_ISO_8BIT; |
| 1407 | c = MAKE_CHAR (charset, 0, 0); |
| 1408 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7, charset, c)) |
| 1409 | mask_found |= CODING_CATEGORY_MASK_ISO_7; |
| 1410 | else |
| 1411 | mask &= ~CODING_CATEGORY_MASK_ISO_7; |
| 1412 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT, charset, c)) |
| 1413 | mask_found |= CODING_CATEGORY_MASK_ISO_7_TIGHT; |
| 1414 | else |
| 1415 | mask &= ~CODING_CATEGORY_MASK_ISO_7_TIGHT; |
| 1416 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE, charset, c)) |
| 1417 | mask_found |= CODING_CATEGORY_MASK_ISO_7_ELSE; |
| 1418 | else |
| 1419 | mask &= ~CODING_CATEGORY_MASK_ISO_7_ELSE; |
| 1420 | if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE, charset, c)) |
| 1421 | mask_found |= CODING_CATEGORY_MASK_ISO_8_ELSE; |
| 1422 | else |
| 1423 | mask &= ~CODING_CATEGORY_MASK_ISO_8_ELSE; |
| 1424 | break; |
| 1425 | |
| 1426 | case ISO_CODE_SO: |
| 1427 | if (inhibit_iso_escape_detection) |
| 1428 | break; |
| 1429 | single_shifting = 0; |
| 1430 | if (shift_out == 0 |
| 1431 | && (reg[1] >= 0 |
| 1432 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_7_ELSE) |
| 1433 | || SHIFT_OUT_OK (CODING_CATEGORY_IDX_ISO_8_ELSE))) |
| 1434 | { |
| 1435 | /* Locking shift out. */ |
| 1436 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; |
| 1437 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; |
| 1438 | } |
| 1439 | break; |
| 1440 | |
| 1441 | case ISO_CODE_SI: |
| 1442 | if (inhibit_iso_escape_detection) |
| 1443 | break; |
| 1444 | single_shifting = 0; |
| 1445 | if (shift_out == 1) |
| 1446 | { |
| 1447 | /* Locking shift in. */ |
| 1448 | mask &= ~CODING_CATEGORY_MASK_ISO_7BIT; |
| 1449 | mask_found |= CODING_CATEGORY_MASK_ISO_SHIFT; |
| 1450 | } |
| 1451 | break; |
| 1452 | |
| 1453 | case ISO_CODE_CSI: |
| 1454 | single_shifting = 0; |
| 1455 | case ISO_CODE_SS2: |
| 1456 | case ISO_CODE_SS3: |
| 1457 | { |
| 1458 | int newmask = CODING_CATEGORY_MASK_ISO_8_ELSE; |
| 1459 | |
| 1460 | if (inhibit_iso_escape_detection) |
| 1461 | break; |
| 1462 | if (c != ISO_CODE_CSI) |
| 1463 | { |
| 1464 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
| 1465 | & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1466 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
| 1467 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
| 1468 | & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1469 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
| 1470 | single_shifting = 1; |
| 1471 | } |
| 1472 | if (VECTORP (Vlatin_extra_code_table) |
| 1473 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) |
| 1474 | { |
| 1475 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
| 1476 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 1477 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
| 1478 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
| 1479 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 1480 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
| 1481 | } |
| 1482 | mask &= newmask; |
| 1483 | mask_found |= newmask; |
| 1484 | } |
| 1485 | break; |
| 1486 | |
| 1487 | default: |
| 1488 | if (c < 0x80) |
| 1489 | { |
| 1490 | single_shifting = 0; |
| 1491 | break; |
| 1492 | } |
| 1493 | else if (c < 0xA0) |
| 1494 | { |
| 1495 | single_shifting = 0; |
| 1496 | if (VECTORP (Vlatin_extra_code_table) |
| 1497 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) |
| 1498 | { |
| 1499 | int newmask = 0; |
| 1500 | |
| 1501 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags |
| 1502 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 1503 | newmask |= CODING_CATEGORY_MASK_ISO_8_1; |
| 1504 | if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_2]->flags |
| 1505 | & CODING_FLAG_ISO_LATIN_EXTRA) |
| 1506 | newmask |= CODING_CATEGORY_MASK_ISO_8_2; |
| 1507 | mask &= newmask; |
| 1508 | mask_found |= newmask; |
| 1509 | } |
| 1510 | else |
| 1511 | return 0; |
| 1512 | } |
| 1513 | else |
| 1514 | { |
| 1515 | mask &= ~(CODING_CATEGORY_MASK_ISO_7BIT |
| 1516 | | CODING_CATEGORY_MASK_ISO_7_ELSE); |
| 1517 | mask_found |= CODING_CATEGORY_MASK_ISO_8_1; |
| 1518 | /* Check the length of succeeding codes of the range |
| 1519 | 0xA0..0FF. If the byte length is odd, we exclude |
| 1520 | CODING_CATEGORY_MASK_ISO_8_2. We can check this only |
| 1521 | when we are not single shifting. */ |
| 1522 | if (!single_shifting |
| 1523 | && mask & CODING_CATEGORY_MASK_ISO_8_2) |
| 1524 | { |
| 1525 | int i = 1; |
| 1526 | while (src < src_end) |
| 1527 | { |
| 1528 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 1529 | if (c < 0xA0) |
| 1530 | break; |
| 1531 | i++; |
| 1532 | } |
| 1533 | |
| 1534 | if (i & 1 && src < src_end) |
| 1535 | mask &= ~CODING_CATEGORY_MASK_ISO_8_2; |
| 1536 | else |
| 1537 | mask_found |= CODING_CATEGORY_MASK_ISO_8_2; |
| 1538 | } |
| 1539 | } |
| 1540 | break; |
| 1541 | } |
| 1542 | } |
| 1543 | label_end_of_loop: |
| 1544 | return (mask & mask_found); |
| 1545 | } |
| 1546 | |
| 1547 | /* Decode a character of which charset is CHARSET, the 1st position |
| 1548 | code is C1, the 2nd position code is C2, and return the decoded |
| 1549 | character code. If the variable `translation_table' is non-nil, |
| 1550 | returned the translated code. */ |
| 1551 | |
| 1552 | #define DECODE_ISO_CHARACTER(charset, c1, c2) \ |
| 1553 | (NILP (translation_table) \ |
| 1554 | ? MAKE_CHAR (charset, c1, c2) \ |
| 1555 | : translate_char (translation_table, -1, charset, c1, c2)) |
| 1556 | |
| 1557 | /* Set designation state into CODING. */ |
| 1558 | #define DECODE_DESIGNATION(reg, dimension, chars, final_char) \ |
| 1559 | do { \ |
| 1560 | int charset, c; \ |
| 1561 | \ |
| 1562 | if (final_char < '0' || final_char >= 128) \ |
| 1563 | goto label_invalid_code; \ |
| 1564 | charset = ISO_CHARSET_TABLE (make_number (dimension), \ |
| 1565 | make_number (chars), \ |
| 1566 | make_number (final_char)); \ |
| 1567 | c = MAKE_CHAR (charset, 0, 0); \ |
| 1568 | if (charset >= 0 \ |
| 1569 | && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) == reg \ |
| 1570 | || CODING_SAFE_CHAR_P (safe_chars, c))) \ |
| 1571 | { \ |
| 1572 | if (coding->spec.iso2022.last_invalid_designation_register == 0 \ |
| 1573 | && reg == 0 \ |
| 1574 | && charset == CHARSET_ASCII) \ |
| 1575 | { \ |
| 1576 | /* We should insert this designation sequence as is so \ |
| 1577 | that it is surely written back to a file. */ \ |
| 1578 | coding->spec.iso2022.last_invalid_designation_register = -1; \ |
| 1579 | goto label_invalid_code; \ |
| 1580 | } \ |
| 1581 | coding->spec.iso2022.last_invalid_designation_register = -1; \ |
| 1582 | if ((coding->mode & CODING_MODE_DIRECTION) \ |
| 1583 | && CHARSET_REVERSE_CHARSET (charset) >= 0) \ |
| 1584 | charset = CHARSET_REVERSE_CHARSET (charset); \ |
| 1585 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ |
| 1586 | } \ |
| 1587 | else \ |
| 1588 | { \ |
| 1589 | coding->spec.iso2022.last_invalid_designation_register = reg; \ |
| 1590 | goto label_invalid_code; \ |
| 1591 | } \ |
| 1592 | } while (0) |
| 1593 | |
| 1594 | /* Allocate a memory block for storing information about compositions. |
| 1595 | The block is chained to the already allocated blocks. */ |
| 1596 | |
| 1597 | void |
| 1598 | coding_allocate_composition_data (coding, char_offset) |
| 1599 | struct coding_system *coding; |
| 1600 | int char_offset; |
| 1601 | { |
| 1602 | struct composition_data *cmp_data |
| 1603 | = (struct composition_data *) xmalloc (sizeof *cmp_data); |
| 1604 | |
| 1605 | cmp_data->char_offset = char_offset; |
| 1606 | cmp_data->used = 0; |
| 1607 | cmp_data->prev = coding->cmp_data; |
| 1608 | cmp_data->next = NULL; |
| 1609 | if (coding->cmp_data) |
| 1610 | coding->cmp_data->next = cmp_data; |
| 1611 | coding->cmp_data = cmp_data; |
| 1612 | coding->cmp_data_start = 0; |
| 1613 | } |
| 1614 | |
| 1615 | /* Handle composition start sequence ESC 0, ESC 2, ESC 3, or ESC 4. |
| 1616 | ESC 0 : relative composition : ESC 0 CHAR ... ESC 1 |
| 1617 | ESC 2 : rulebase composition : ESC 2 CHAR RULE CHAR RULE ... CHAR ESC 1 |
| 1618 | ESC 3 : altchar composition : ESC 3 ALT ... ESC 0 CHAR ... ESC 1 |
| 1619 | ESC 4 : alt&rule composition : ESC 4 ALT RULE .. ALT ESC 0 CHAR ... ESC 1 |
| 1620 | */ |
| 1621 | |
| 1622 | #define DECODE_COMPOSITION_START(c1) \ |
| 1623 | do { \ |
| 1624 | if (coding->composing == COMPOSITION_DISABLED) \ |
| 1625 | { \ |
| 1626 | *dst++ = ISO_CODE_ESC; \ |
| 1627 | *dst++ = c1 & 0x7f; \ |
| 1628 | coding->produced_char += 2; \ |
| 1629 | } \ |
| 1630 | else if (!COMPOSING_P (coding)) \ |
| 1631 | { \ |
| 1632 | /* This is surely the start of a composition. We must be sure \ |
| 1633 | that coding->cmp_data has enough space to store the \ |
| 1634 | information about the composition. If not, terminate the \ |
| 1635 | current decoding loop, allocate one more memory block for \ |
| 1636 | coding->cmp_data in the caller, then start the decoding \ |
| 1637 | loop again. We can't allocate memory here directly because \ |
| 1638 | it may cause buffer/string relocation. */ \ |
| 1639 | if (!coding->cmp_data \ |
| 1640 | || (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH \ |
| 1641 | >= COMPOSITION_DATA_SIZE)) \ |
| 1642 | { \ |
| 1643 | coding->result = CODING_FINISH_INSUFFICIENT_CMP; \ |
| 1644 | goto label_end_of_loop; \ |
| 1645 | } \ |
| 1646 | coding->composing = (c1 == '0' ? COMPOSITION_RELATIVE \ |
| 1647 | : c1 == '2' ? COMPOSITION_WITH_RULE \ |
| 1648 | : c1 == '3' ? COMPOSITION_WITH_ALTCHARS \ |
| 1649 | : COMPOSITION_WITH_RULE_ALTCHARS); \ |
| 1650 | CODING_ADD_COMPOSITION_START (coding, coding->produced_char, \ |
| 1651 | coding->composing); \ |
| 1652 | coding->composition_rule_follows = 0; \ |
| 1653 | } \ |
| 1654 | else \ |
| 1655 | { \ |
| 1656 | /* We are already handling a composition. If the method is \ |
| 1657 | the following two, the codes following the current escape \ |
| 1658 | sequence are actual characters stored in a buffer. */ \ |
| 1659 | if (coding->composing == COMPOSITION_WITH_ALTCHARS \ |
| 1660 | || coding->composing == COMPOSITION_WITH_RULE_ALTCHARS) \ |
| 1661 | { \ |
| 1662 | coding->composing = COMPOSITION_RELATIVE; \ |
| 1663 | coding->composition_rule_follows = 0; \ |
| 1664 | } \ |
| 1665 | } \ |
| 1666 | } while (0) |
| 1667 | |
| 1668 | /* Handle composition end sequence ESC 1. */ |
| 1669 | |
| 1670 | #define DECODE_COMPOSITION_END(c1) \ |
| 1671 | do { \ |
| 1672 | if (! COMPOSING_P (coding)) \ |
| 1673 | { \ |
| 1674 | *dst++ = ISO_CODE_ESC; \ |
| 1675 | *dst++ = c1; \ |
| 1676 | coding->produced_char += 2; \ |
| 1677 | } \ |
| 1678 | else \ |
| 1679 | { \ |
| 1680 | CODING_ADD_COMPOSITION_END (coding, coding->produced_char); \ |
| 1681 | coding->composing = COMPOSITION_NO; \ |
| 1682 | } \ |
| 1683 | } while (0) |
| 1684 | |
| 1685 | /* Decode a composition rule from the byte C1 (and maybe one more byte |
| 1686 | from SRC) and store one encoded composition rule in |
| 1687 | coding->cmp_data. */ |
| 1688 | |
| 1689 | #define DECODE_COMPOSITION_RULE(c1) \ |
| 1690 | do { \ |
| 1691 | int rule = 0; \ |
| 1692 | (c1) -= 32; \ |
| 1693 | if (c1 < 81) /* old format (before ver.21) */ \ |
| 1694 | { \ |
| 1695 | int gref = (c1) / 9; \ |
| 1696 | int nref = (c1) % 9; \ |
| 1697 | if (gref == 4) gref = 10; \ |
| 1698 | if (nref == 4) nref = 10; \ |
| 1699 | rule = COMPOSITION_ENCODE_RULE (gref, nref); \ |
| 1700 | } \ |
| 1701 | else if (c1 < 93) /* new format (after ver.21) */ \ |
| 1702 | { \ |
| 1703 | ONE_MORE_BYTE (c2); \ |
| 1704 | rule = COMPOSITION_ENCODE_RULE (c1 - 81, c2 - 32); \ |
| 1705 | } \ |
| 1706 | CODING_ADD_COMPOSITION_COMPONENT (coding, rule); \ |
| 1707 | coding->composition_rule_follows = 0; \ |
| 1708 | } while (0) |
| 1709 | |
| 1710 | |
| 1711 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */ |
| 1712 | |
| 1713 | static void |
| 1714 | decode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
| 1715 | struct coding_system *coding; |
| 1716 | unsigned char *source, *destination; |
| 1717 | int src_bytes, dst_bytes; |
| 1718 | { |
| 1719 | unsigned char *src = source; |
| 1720 | unsigned char *src_end = source + src_bytes; |
| 1721 | unsigned char *dst = destination; |
| 1722 | unsigned char *dst_end = destination + dst_bytes; |
| 1723 | /* Charsets invoked to graphic plane 0 and 1 respectively. */ |
| 1724 | int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1725 | int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); |
| 1726 | /* SRC_BASE remembers the start position in source in each loop. |
| 1727 | The loop will be exited when there's not enough source code |
| 1728 | (within macro ONE_MORE_BYTE), or when there's not enough |
| 1729 | destination area to produce a character (within macro |
| 1730 | EMIT_CHAR). */ |
| 1731 | unsigned char *src_base; |
| 1732 | int c, charset; |
| 1733 | Lisp_Object translation_table; |
| 1734 | Lisp_Object safe_chars; |
| 1735 | |
| 1736 | safe_chars = coding_safe_chars (coding); |
| 1737 | |
| 1738 | if (NILP (Venable_character_translation)) |
| 1739 | translation_table = Qnil; |
| 1740 | else |
| 1741 | { |
| 1742 | translation_table = coding->translation_table_for_decode; |
| 1743 | if (NILP (translation_table)) |
| 1744 | translation_table = Vstandard_translation_table_for_decode; |
| 1745 | } |
| 1746 | |
| 1747 | coding->result = CODING_FINISH_NORMAL; |
| 1748 | |
| 1749 | while (1) |
| 1750 | { |
| 1751 | int c1, c2; |
| 1752 | |
| 1753 | src_base = src; |
| 1754 | ONE_MORE_BYTE (c1); |
| 1755 | |
| 1756 | /* We produce no character or one character. */ |
| 1757 | switch (iso_code_class [c1]) |
| 1758 | { |
| 1759 | case ISO_0x20_or_0x7F: |
| 1760 | if (COMPOSING_P (coding) && coding->composition_rule_follows) |
| 1761 | { |
| 1762 | DECODE_COMPOSITION_RULE (c1); |
| 1763 | continue; |
| 1764 | } |
| 1765 | if (charset0 < 0 || CHARSET_CHARS (charset0) == 94) |
| 1766 | { |
| 1767 | /* This is SPACE or DEL. */ |
| 1768 | charset = CHARSET_ASCII; |
| 1769 | break; |
| 1770 | } |
| 1771 | /* This is a graphic character, we fall down ... */ |
| 1772 | |
| 1773 | case ISO_graphic_plane_0: |
| 1774 | if (COMPOSING_P (coding) && coding->composition_rule_follows) |
| 1775 | { |
| 1776 | DECODE_COMPOSITION_RULE (c1); |
| 1777 | continue; |
| 1778 | } |
| 1779 | charset = charset0; |
| 1780 | break; |
| 1781 | |
| 1782 | case ISO_0xA0_or_0xFF: |
| 1783 | if (charset1 < 0 || CHARSET_CHARS (charset1) == 94 |
| 1784 | || coding->flags & CODING_FLAG_ISO_SEVEN_BITS) |
| 1785 | goto label_invalid_code; |
| 1786 | /* This is a graphic character, we fall down ... */ |
| 1787 | |
| 1788 | case ISO_graphic_plane_1: |
| 1789 | if (charset1 < 0) |
| 1790 | goto label_invalid_code; |
| 1791 | charset = charset1; |
| 1792 | break; |
| 1793 | |
| 1794 | case ISO_control_0: |
| 1795 | if (COMPOSING_P (coding)) |
| 1796 | DECODE_COMPOSITION_END ('1'); |
| 1797 | |
| 1798 | /* All ISO2022 control characters in this class have the |
| 1799 | same representation in Emacs internal format. */ |
| 1800 | if (c1 == '\n' |
| 1801 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 1802 | && (coding->eol_type == CODING_EOL_CR |
| 1803 | || coding->eol_type == CODING_EOL_CRLF)) |
| 1804 | { |
| 1805 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 1806 | goto label_end_of_loop; |
| 1807 | } |
| 1808 | charset = CHARSET_ASCII; |
| 1809 | break; |
| 1810 | |
| 1811 | case ISO_control_1: |
| 1812 | if (COMPOSING_P (coding)) |
| 1813 | DECODE_COMPOSITION_END ('1'); |
| 1814 | goto label_invalid_code; |
| 1815 | |
| 1816 | case ISO_carriage_return: |
| 1817 | if (COMPOSING_P (coding)) |
| 1818 | DECODE_COMPOSITION_END ('1'); |
| 1819 | |
| 1820 | if (coding->eol_type == CODING_EOL_CR) |
| 1821 | c1 = '\n'; |
| 1822 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 1823 | { |
| 1824 | ONE_MORE_BYTE (c1); |
| 1825 | if (c1 != ISO_CODE_LF) |
| 1826 | { |
| 1827 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 1828 | { |
| 1829 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 1830 | goto label_end_of_loop; |
| 1831 | } |
| 1832 | src--; |
| 1833 | c1 = '\r'; |
| 1834 | } |
| 1835 | } |
| 1836 | charset = CHARSET_ASCII; |
| 1837 | break; |
| 1838 | |
| 1839 | case ISO_shift_out: |
| 1840 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
| 1841 | || CODING_SPEC_ISO_DESIGNATION (coding, 1) < 0) |
| 1842 | goto label_invalid_code; |
| 1843 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; |
| 1844 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1845 | continue; |
| 1846 | |
| 1847 | case ISO_shift_in: |
| 1848 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
| 1849 | goto label_invalid_code; |
| 1850 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; |
| 1851 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1852 | continue; |
| 1853 | |
| 1854 | case ISO_single_shift_2_7: |
| 1855 | case ISO_single_shift_2: |
| 1856 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
| 1857 | goto label_invalid_code; |
| 1858 | /* SS2 is handled as an escape sequence of ESC 'N' */ |
| 1859 | c1 = 'N'; |
| 1860 | goto label_escape_sequence; |
| 1861 | |
| 1862 | case ISO_single_shift_3: |
| 1863 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
| 1864 | goto label_invalid_code; |
| 1865 | /* SS2 is handled as an escape sequence of ESC 'O' */ |
| 1866 | c1 = 'O'; |
| 1867 | goto label_escape_sequence; |
| 1868 | |
| 1869 | case ISO_control_sequence_introducer: |
| 1870 | /* CSI is handled as an escape sequence of ESC '[' ... */ |
| 1871 | c1 = '['; |
| 1872 | goto label_escape_sequence; |
| 1873 | |
| 1874 | case ISO_escape: |
| 1875 | ONE_MORE_BYTE (c1); |
| 1876 | label_escape_sequence: |
| 1877 | /* Escape sequences handled by Emacs are invocation, |
| 1878 | designation, direction specification, and character |
| 1879 | composition specification. */ |
| 1880 | switch (c1) |
| 1881 | { |
| 1882 | case '&': /* revision of following character set */ |
| 1883 | ONE_MORE_BYTE (c1); |
| 1884 | if (!(c1 >= '@' && c1 <= '~')) |
| 1885 | goto label_invalid_code; |
| 1886 | ONE_MORE_BYTE (c1); |
| 1887 | if (c1 != ISO_CODE_ESC) |
| 1888 | goto label_invalid_code; |
| 1889 | ONE_MORE_BYTE (c1); |
| 1890 | goto label_escape_sequence; |
| 1891 | |
| 1892 | case '$': /* designation of 2-byte character set */ |
| 1893 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
| 1894 | goto label_invalid_code; |
| 1895 | ONE_MORE_BYTE (c1); |
| 1896 | if (c1 >= '@' && c1 <= 'B') |
| 1897 | { /* designation of JISX0208.1978, GB2312.1980, |
| 1898 | or JISX0208.1980 */ |
| 1899 | DECODE_DESIGNATION (0, 2, 94, c1); |
| 1900 | } |
| 1901 | else if (c1 >= 0x28 && c1 <= 0x2B) |
| 1902 | { /* designation of DIMENSION2_CHARS94 character set */ |
| 1903 | ONE_MORE_BYTE (c2); |
| 1904 | DECODE_DESIGNATION (c1 - 0x28, 2, 94, c2); |
| 1905 | } |
| 1906 | else if (c1 >= 0x2C && c1 <= 0x2F) |
| 1907 | { /* designation of DIMENSION2_CHARS96 character set */ |
| 1908 | ONE_MORE_BYTE (c2); |
| 1909 | DECODE_DESIGNATION (c1 - 0x2C, 2, 96, c2); |
| 1910 | } |
| 1911 | else |
| 1912 | goto label_invalid_code; |
| 1913 | /* We must update these variables now. */ |
| 1914 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1915 | charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); |
| 1916 | continue; |
| 1917 | |
| 1918 | case 'n': /* invocation of locking-shift-2 */ |
| 1919 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
| 1920 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) |
| 1921 | goto label_invalid_code; |
| 1922 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; |
| 1923 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1924 | continue; |
| 1925 | |
| 1926 | case 'o': /* invocation of locking-shift-3 */ |
| 1927 | if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT) |
| 1928 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) |
| 1929 | goto label_invalid_code; |
| 1930 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; |
| 1931 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 1932 | continue; |
| 1933 | |
| 1934 | case 'N': /* invocation of single-shift-2 */ |
| 1935 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1936 | || CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0) |
| 1937 | goto label_invalid_code; |
| 1938 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 2); |
| 1939 | ONE_MORE_BYTE (c1); |
| 1940 | if (c1 < 0x20 || (c1 >= 0x80 && c1 < 0xA0)) |
| 1941 | goto label_invalid_code; |
| 1942 | break; |
| 1943 | |
| 1944 | case 'O': /* invocation of single-shift-3 */ |
| 1945 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 1946 | || CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0) |
| 1947 | goto label_invalid_code; |
| 1948 | charset = CODING_SPEC_ISO_DESIGNATION (coding, 3); |
| 1949 | ONE_MORE_BYTE (c1); |
| 1950 | if (c1 < 0x20 || (c1 >= 0x80 && c1 < 0xA0)) |
| 1951 | goto label_invalid_code; |
| 1952 | break; |
| 1953 | |
| 1954 | case '0': case '2': case '3': case '4': /* start composition */ |
| 1955 | DECODE_COMPOSITION_START (c1); |
| 1956 | continue; |
| 1957 | |
| 1958 | case '1': /* end composition */ |
| 1959 | DECODE_COMPOSITION_END (c1); |
| 1960 | continue; |
| 1961 | |
| 1962 | case '[': /* specification of direction */ |
| 1963 | if (coding->flags & CODING_FLAG_ISO_NO_DIRECTION) |
| 1964 | goto label_invalid_code; |
| 1965 | /* For the moment, nested direction is not supported. |
| 1966 | So, `coding->mode & CODING_MODE_DIRECTION' zero means |
| 1967 | left-to-right, and nonzero means right-to-left. */ |
| 1968 | ONE_MORE_BYTE (c1); |
| 1969 | switch (c1) |
| 1970 | { |
| 1971 | case ']': /* end of the current direction */ |
| 1972 | coding->mode &= ~CODING_MODE_DIRECTION; |
| 1973 | |
| 1974 | case '0': /* end of the current direction */ |
| 1975 | case '1': /* start of left-to-right direction */ |
| 1976 | ONE_MORE_BYTE (c1); |
| 1977 | if (c1 == ']') |
| 1978 | coding->mode &= ~CODING_MODE_DIRECTION; |
| 1979 | else |
| 1980 | goto label_invalid_code; |
| 1981 | break; |
| 1982 | |
| 1983 | case '2': /* start of right-to-left direction */ |
| 1984 | ONE_MORE_BYTE (c1); |
| 1985 | if (c1 == ']') |
| 1986 | coding->mode |= CODING_MODE_DIRECTION; |
| 1987 | else |
| 1988 | goto label_invalid_code; |
| 1989 | break; |
| 1990 | |
| 1991 | default: |
| 1992 | goto label_invalid_code; |
| 1993 | } |
| 1994 | continue; |
| 1995 | |
| 1996 | default: |
| 1997 | if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION)) |
| 1998 | goto label_invalid_code; |
| 1999 | if (c1 >= 0x28 && c1 <= 0x2B) |
| 2000 | { /* designation of DIMENSION1_CHARS94 character set */ |
| 2001 | ONE_MORE_BYTE (c2); |
| 2002 | DECODE_DESIGNATION (c1 - 0x28, 1, 94, c2); |
| 2003 | } |
| 2004 | else if (c1 >= 0x2C && c1 <= 0x2F) |
| 2005 | { /* designation of DIMENSION1_CHARS96 character set */ |
| 2006 | ONE_MORE_BYTE (c2); |
| 2007 | DECODE_DESIGNATION (c1 - 0x2C, 1, 96, c2); |
| 2008 | } |
| 2009 | else |
| 2010 | goto label_invalid_code; |
| 2011 | /* We must update these variables now. */ |
| 2012 | charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0); |
| 2013 | charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1); |
| 2014 | continue; |
| 2015 | } |
| 2016 | } |
| 2017 | |
| 2018 | /* Now we know CHARSET and 1st position code C1 of a character. |
| 2019 | Produce a multibyte sequence for that character while getting |
| 2020 | 2nd position code C2 if necessary. */ |
| 2021 | if (CHARSET_DIMENSION (charset) == 2) |
| 2022 | { |
| 2023 | ONE_MORE_BYTE (c2); |
| 2024 | if (c1 < 0x80 ? c2 < 0x20 || c2 >= 0x80 : c2 < 0xA0) |
| 2025 | /* C2 is not in a valid range. */ |
| 2026 | goto label_invalid_code; |
| 2027 | } |
| 2028 | c = DECODE_ISO_CHARACTER (charset, c1, c2); |
| 2029 | EMIT_CHAR (c); |
| 2030 | continue; |
| 2031 | |
| 2032 | label_invalid_code: |
| 2033 | coding->errors++; |
| 2034 | if (COMPOSING_P (coding)) |
| 2035 | DECODE_COMPOSITION_END ('1'); |
| 2036 | src = src_base; |
| 2037 | c = *src++; |
| 2038 | EMIT_CHAR (c); |
| 2039 | } |
| 2040 | |
| 2041 | label_end_of_loop: |
| 2042 | coding->consumed = coding->consumed_char = src_base - source; |
| 2043 | coding->produced = dst - destination; |
| 2044 | return; |
| 2045 | } |
| 2046 | |
| 2047 | |
| 2048 | /* ISO2022 encoding stuff. */ |
| 2049 | |
| 2050 | /* |
| 2051 | It is not enough to say just "ISO2022" on encoding, we have to |
| 2052 | specify more details. In Emacs, each ISO2022 coding system |
| 2053 | variant has the following specifications: |
| 2054 | 1. Initial designation to G0 through G3. |
| 2055 | 2. Allows short-form designation? |
| 2056 | 3. ASCII should be designated to G0 before control characters? |
| 2057 | 4. ASCII should be designated to G0 at end of line? |
| 2058 | 5. 7-bit environment or 8-bit environment? |
| 2059 | 6. Use locking-shift? |
| 2060 | 7. Use Single-shift? |
| 2061 | And the following two are only for Japanese: |
| 2062 | 8. Use ASCII in place of JIS0201-1976-Roman? |
| 2063 | 9. Use JISX0208-1983 in place of JISX0208-1978? |
| 2064 | These specifications are encoded in `coding->flags' as flag bits |
| 2065 | defined by macros CODING_FLAG_ISO_XXX. See `coding.h' for more |
| 2066 | details. |
| 2067 | */ |
| 2068 | |
| 2069 | /* Produce codes (escape sequence) for designating CHARSET to graphic |
| 2070 | register REG at DST, and increment DST. If <final-char> of CHARSET is |
| 2071 | '@', 'A', or 'B' and the coding system CODING allows, produce |
| 2072 | designation sequence of short-form. */ |
| 2073 | |
| 2074 | #define ENCODE_DESIGNATION(charset, reg, coding) \ |
| 2075 | do { \ |
| 2076 | unsigned char final_char = CHARSET_ISO_FINAL_CHAR (charset); \ |
| 2077 | char *intermediate_char_94 = "()*+"; \ |
| 2078 | char *intermediate_char_96 = ",-./"; \ |
| 2079 | int revision = CODING_SPEC_ISO_REVISION_NUMBER(coding, charset); \ |
| 2080 | \ |
| 2081 | if (revision < 255) \ |
| 2082 | { \ |
| 2083 | *dst++ = ISO_CODE_ESC; \ |
| 2084 | *dst++ = '&'; \ |
| 2085 | *dst++ = '@' + revision; \ |
| 2086 | } \ |
| 2087 | *dst++ = ISO_CODE_ESC; \ |
| 2088 | if (CHARSET_DIMENSION (charset) == 1) \ |
| 2089 | { \ |
| 2090 | if (CHARSET_CHARS (charset) == 94) \ |
| 2091 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ |
| 2092 | else \ |
| 2093 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ |
| 2094 | } \ |
| 2095 | else \ |
| 2096 | { \ |
| 2097 | *dst++ = '$'; \ |
| 2098 | if (CHARSET_CHARS (charset) == 94) \ |
| 2099 | { \ |
| 2100 | if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \ |
| 2101 | || reg != 0 \ |
| 2102 | || final_char < '@' || final_char > 'B') \ |
| 2103 | *dst++ = (unsigned char) (intermediate_char_94[reg]); \ |
| 2104 | } \ |
| 2105 | else \ |
| 2106 | *dst++ = (unsigned char) (intermediate_char_96[reg]); \ |
| 2107 | } \ |
| 2108 | *dst++ = final_char; \ |
| 2109 | CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \ |
| 2110 | } while (0) |
| 2111 | |
| 2112 | /* The following two macros produce codes (control character or escape |
| 2113 | sequence) for ISO2022 single-shift functions (single-shift-2 and |
| 2114 | single-shift-3). */ |
| 2115 | |
| 2116 | #define ENCODE_SINGLE_SHIFT_2 \ |
| 2117 | do { \ |
| 2118 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 2119 | *dst++ = ISO_CODE_ESC, *dst++ = 'N'; \ |
| 2120 | else \ |
| 2121 | *dst++ = ISO_CODE_SS2; \ |
| 2122 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
| 2123 | } while (0) |
| 2124 | |
| 2125 | #define ENCODE_SINGLE_SHIFT_3 \ |
| 2126 | do { \ |
| 2127 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 2128 | *dst++ = ISO_CODE_ESC, *dst++ = 'O'; \ |
| 2129 | else \ |
| 2130 | *dst++ = ISO_CODE_SS3; \ |
| 2131 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \ |
| 2132 | } while (0) |
| 2133 | |
| 2134 | /* The following four macros produce codes (control character or |
| 2135 | escape sequence) for ISO2022 locking-shift functions (shift-in, |
| 2136 | shift-out, locking-shift-2, and locking-shift-3). */ |
| 2137 | |
| 2138 | #define ENCODE_SHIFT_IN \ |
| 2139 | do { \ |
| 2140 | *dst++ = ISO_CODE_SI; \ |
| 2141 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; \ |
| 2142 | } while (0) |
| 2143 | |
| 2144 | #define ENCODE_SHIFT_OUT \ |
| 2145 | do { \ |
| 2146 | *dst++ = ISO_CODE_SO; \ |
| 2147 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; \ |
| 2148 | } while (0) |
| 2149 | |
| 2150 | #define ENCODE_LOCKING_SHIFT_2 \ |
| 2151 | do { \ |
| 2152 | *dst++ = ISO_CODE_ESC, *dst++ = 'n'; \ |
| 2153 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; \ |
| 2154 | } while (0) |
| 2155 | |
| 2156 | #define ENCODE_LOCKING_SHIFT_3 \ |
| 2157 | do { \ |
| 2158 | *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \ |
| 2159 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; \ |
| 2160 | } while (0) |
| 2161 | |
| 2162 | /* Produce codes for a DIMENSION1 character whose character set is |
| 2163 | CHARSET and whose position-code is C1. Designation and invocation |
| 2164 | sequences are also produced in advance if necessary. */ |
| 2165 | |
| 2166 | #define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \ |
| 2167 | do { \ |
| 2168 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ |
| 2169 | { \ |
| 2170 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 2171 | *dst++ = c1 & 0x7F; \ |
| 2172 | else \ |
| 2173 | *dst++ = c1 | 0x80; \ |
| 2174 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ |
| 2175 | break; \ |
| 2176 | } \ |
| 2177 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ |
| 2178 | { \ |
| 2179 | *dst++ = c1 & 0x7F; \ |
| 2180 | break; \ |
| 2181 | } \ |
| 2182 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ |
| 2183 | { \ |
| 2184 | *dst++ = c1 | 0x80; \ |
| 2185 | break; \ |
| 2186 | } \ |
| 2187 | else \ |
| 2188 | /* Since CHARSET is not yet invoked to any graphic planes, we \ |
| 2189 | must invoke it, or, at first, designate it to some graphic \ |
| 2190 | register. Then repeat the loop to actually produce the \ |
| 2191 | character. */ \ |
| 2192 | dst = encode_invocation_designation (charset, coding, dst); \ |
| 2193 | } while (1) |
| 2194 | |
| 2195 | /* Produce codes for a DIMENSION2 character whose character set is |
| 2196 | CHARSET and whose position-codes are C1 and C2. Designation and |
| 2197 | invocation codes are also produced in advance if necessary. */ |
| 2198 | |
| 2199 | #define ENCODE_ISO_CHARACTER_DIMENSION2(charset, c1, c2) \ |
| 2200 | do { \ |
| 2201 | if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \ |
| 2202 | { \ |
| 2203 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \ |
| 2204 | *dst++ = c1 & 0x7F, *dst++ = c2 & 0x7F; \ |
| 2205 | else \ |
| 2206 | *dst++ = c1 | 0x80, *dst++ = c2 | 0x80; \ |
| 2207 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; \ |
| 2208 | break; \ |
| 2209 | } \ |
| 2210 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 0)) \ |
| 2211 | { \ |
| 2212 | *dst++ = c1 & 0x7F, *dst++= c2 & 0x7F; \ |
| 2213 | break; \ |
| 2214 | } \ |
| 2215 | else if (charset == CODING_SPEC_ISO_PLANE_CHARSET (coding, 1)) \ |
| 2216 | { \ |
| 2217 | *dst++ = c1 | 0x80, *dst++= c2 | 0x80; \ |
| 2218 | break; \ |
| 2219 | } \ |
| 2220 | else \ |
| 2221 | /* Since CHARSET is not yet invoked to any graphic planes, we \ |
| 2222 | must invoke it, or, at first, designate it to some graphic \ |
| 2223 | register. Then repeat the loop to actually produce the \ |
| 2224 | character. */ \ |
| 2225 | dst = encode_invocation_designation (charset, coding, dst); \ |
| 2226 | } while (1) |
| 2227 | |
| 2228 | #define ENCODE_ISO_CHARACTER(c) \ |
| 2229 | do { \ |
| 2230 | int charset, c1, c2; \ |
| 2231 | \ |
| 2232 | SPLIT_CHAR (c, charset, c1, c2); \ |
| 2233 | if (CHARSET_DEFINED_P (charset)) \ |
| 2234 | { \ |
| 2235 | if (CHARSET_DIMENSION (charset) == 1) \ |
| 2236 | { \ |
| 2237 | if (charset == CHARSET_ASCII \ |
| 2238 | && coding->flags & CODING_FLAG_ISO_USE_ROMAN) \ |
| 2239 | charset = charset_latin_jisx0201; \ |
| 2240 | ENCODE_ISO_CHARACTER_DIMENSION1 (charset, c1); \ |
| 2241 | } \ |
| 2242 | else \ |
| 2243 | { \ |
| 2244 | if (charset == charset_jisx0208 \ |
| 2245 | && coding->flags & CODING_FLAG_ISO_USE_OLDJIS) \ |
| 2246 | charset = charset_jisx0208_1978; \ |
| 2247 | ENCODE_ISO_CHARACTER_DIMENSION2 (charset, c1, c2); \ |
| 2248 | } \ |
| 2249 | } \ |
| 2250 | else \ |
| 2251 | { \ |
| 2252 | *dst++ = c1; \ |
| 2253 | if (c2 >= 0) \ |
| 2254 | *dst++ = c2; \ |
| 2255 | } \ |
| 2256 | } while (0) |
| 2257 | |
| 2258 | |
| 2259 | /* Instead of encoding character C, produce one or two `?'s. */ |
| 2260 | |
| 2261 | #define ENCODE_UNSAFE_CHARACTER(c) \ |
| 2262 | do { \ |
| 2263 | ENCODE_ISO_CHARACTER (CODING_INHIBIT_CHARACTER_SUBSTITUTION); \ |
| 2264 | if (CHARSET_WIDTH (CHAR_CHARSET (c)) > 1) \ |
| 2265 | ENCODE_ISO_CHARACTER (CODING_INHIBIT_CHARACTER_SUBSTITUTION); \ |
| 2266 | } while (0) |
| 2267 | |
| 2268 | |
| 2269 | /* Produce designation and invocation codes at a place pointed by DST |
| 2270 | to use CHARSET. The element `spec.iso2022' of *CODING is updated. |
| 2271 | Return new DST. */ |
| 2272 | |
| 2273 | unsigned char * |
| 2274 | encode_invocation_designation (charset, coding, dst) |
| 2275 | int charset; |
| 2276 | struct coding_system *coding; |
| 2277 | unsigned char *dst; |
| 2278 | { |
| 2279 | int reg; /* graphic register number */ |
| 2280 | |
| 2281 | /* At first, check designations. */ |
| 2282 | for (reg = 0; reg < 4; reg++) |
| 2283 | if (charset == CODING_SPEC_ISO_DESIGNATION (coding, reg)) |
| 2284 | break; |
| 2285 | |
| 2286 | if (reg >= 4) |
| 2287 | { |
| 2288 | /* CHARSET is not yet designated to any graphic registers. */ |
| 2289 | /* At first check the requested designation. */ |
| 2290 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); |
| 2291 | if (reg == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION) |
| 2292 | /* Since CHARSET requests no special designation, designate it |
| 2293 | to graphic register 0. */ |
| 2294 | reg = 0; |
| 2295 | |
| 2296 | ENCODE_DESIGNATION (charset, reg, coding); |
| 2297 | } |
| 2298 | |
| 2299 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != reg |
| 2300 | && CODING_SPEC_ISO_INVOCATION (coding, 1) != reg) |
| 2301 | { |
| 2302 | /* Since the graphic register REG is not invoked to any graphic |
| 2303 | planes, invoke it to graphic plane 0. */ |
| 2304 | switch (reg) |
| 2305 | { |
| 2306 | case 0: /* graphic register 0 */ |
| 2307 | ENCODE_SHIFT_IN; |
| 2308 | break; |
| 2309 | |
| 2310 | case 1: /* graphic register 1 */ |
| 2311 | ENCODE_SHIFT_OUT; |
| 2312 | break; |
| 2313 | |
| 2314 | case 2: /* graphic register 2 */ |
| 2315 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 2316 | ENCODE_SINGLE_SHIFT_2; |
| 2317 | else |
| 2318 | ENCODE_LOCKING_SHIFT_2; |
| 2319 | break; |
| 2320 | |
| 2321 | case 3: /* graphic register 3 */ |
| 2322 | if (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT) |
| 2323 | ENCODE_SINGLE_SHIFT_3; |
| 2324 | else |
| 2325 | ENCODE_LOCKING_SHIFT_3; |
| 2326 | break; |
| 2327 | } |
| 2328 | } |
| 2329 | |
| 2330 | return dst; |
| 2331 | } |
| 2332 | |
| 2333 | /* Produce 2-byte codes for encoded composition rule RULE. */ |
| 2334 | |
| 2335 | #define ENCODE_COMPOSITION_RULE(rule) \ |
| 2336 | do { \ |
| 2337 | int gref, nref; \ |
| 2338 | COMPOSITION_DECODE_RULE (rule, gref, nref); \ |
| 2339 | *dst++ = 32 + 81 + gref; \ |
| 2340 | *dst++ = 32 + nref; \ |
| 2341 | } while (0) |
| 2342 | |
| 2343 | /* Produce codes for indicating the start of a composition sequence |
| 2344 | (ESC 0, ESC 3, or ESC 4). DATA points to an array of integers |
| 2345 | which specify information about the composition. See the comment |
| 2346 | in coding.h for the format of DATA. */ |
| 2347 | |
| 2348 | #define ENCODE_COMPOSITION_START(coding, data) \ |
| 2349 | do { \ |
| 2350 | coding->composing = data[3]; \ |
| 2351 | *dst++ = ISO_CODE_ESC; \ |
| 2352 | if (coding->composing == COMPOSITION_RELATIVE) \ |
| 2353 | *dst++ = '0'; \ |
| 2354 | else \ |
| 2355 | { \ |
| 2356 | *dst++ = (coding->composing == COMPOSITION_WITH_ALTCHARS \ |
| 2357 | ? '3' : '4'); \ |
| 2358 | coding->cmp_data_index = coding->cmp_data_start + 4; \ |
| 2359 | coding->composition_rule_follows = 0; \ |
| 2360 | } \ |
| 2361 | } while (0) |
| 2362 | |
| 2363 | /* Produce codes for indicating the end of the current composition. */ |
| 2364 | |
| 2365 | #define ENCODE_COMPOSITION_END(coding, data) \ |
| 2366 | do { \ |
| 2367 | *dst++ = ISO_CODE_ESC; \ |
| 2368 | *dst++ = '1'; \ |
| 2369 | coding->cmp_data_start += data[0]; \ |
| 2370 | coding->composing = COMPOSITION_NO; \ |
| 2371 | if (coding->cmp_data_start == coding->cmp_data->used \ |
| 2372 | && coding->cmp_data->next) \ |
| 2373 | { \ |
| 2374 | coding->cmp_data = coding->cmp_data->next; \ |
| 2375 | coding->cmp_data_start = 0; \ |
| 2376 | } \ |
| 2377 | } while (0) |
| 2378 | |
| 2379 | /* Produce composition start sequence ESC 0. Here, this sequence |
| 2380 | doesn't mean the start of a new composition but means that we have |
| 2381 | just produced components (alternate chars and composition rules) of |
| 2382 | the composition and the actual text follows in SRC. */ |
| 2383 | |
| 2384 | #define ENCODE_COMPOSITION_FAKE_START(coding) \ |
| 2385 | do { \ |
| 2386 | *dst++ = ISO_CODE_ESC; \ |
| 2387 | *dst++ = '0'; \ |
| 2388 | coding->composing = COMPOSITION_RELATIVE; \ |
| 2389 | } while (0) |
| 2390 | |
| 2391 | /* The following three macros produce codes for indicating direction |
| 2392 | of text. */ |
| 2393 | #define ENCODE_CONTROL_SEQUENCE_INTRODUCER \ |
| 2394 | do { \ |
| 2395 | if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS) \ |
| 2396 | *dst++ = ISO_CODE_ESC, *dst++ = '['; \ |
| 2397 | else \ |
| 2398 | *dst++ = ISO_CODE_CSI; \ |
| 2399 | } while (0) |
| 2400 | |
| 2401 | #define ENCODE_DIRECTION_R2L \ |
| 2402 | ENCODE_CONTROL_SEQUENCE_INTRODUCER (dst), *dst++ = '2', *dst++ = ']' |
| 2403 | |
| 2404 | #define ENCODE_DIRECTION_L2R \ |
| 2405 | ENCODE_CONTROL_SEQUENCE_INTRODUCER (dst), *dst++ = '0', *dst++ = ']' |
| 2406 | |
| 2407 | /* Produce codes for designation and invocation to reset the graphic |
| 2408 | planes and registers to initial state. */ |
| 2409 | #define ENCODE_RESET_PLANE_AND_REGISTER \ |
| 2410 | do { \ |
| 2411 | int reg; \ |
| 2412 | if (CODING_SPEC_ISO_INVOCATION (coding, 0) != 0) \ |
| 2413 | ENCODE_SHIFT_IN; \ |
| 2414 | for (reg = 0; reg < 4; reg++) \ |
| 2415 | if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg) >= 0 \ |
| 2416 | && (CODING_SPEC_ISO_DESIGNATION (coding, reg) \ |
| 2417 | != CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg))) \ |
| 2418 | ENCODE_DESIGNATION \ |
| 2419 | (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, reg), reg, coding); \ |
| 2420 | } while (0) |
| 2421 | |
| 2422 | /* Produce designation sequences of charsets in the line started from |
| 2423 | SRC to a place pointed by DST, and return updated DST. |
| 2424 | |
| 2425 | If the current block ends before any end-of-line, we may fail to |
| 2426 | find all the necessary designations. */ |
| 2427 | |
| 2428 | static unsigned char * |
| 2429 | encode_designation_at_bol (coding, translation_table, src, src_end, dst) |
| 2430 | struct coding_system *coding; |
| 2431 | Lisp_Object translation_table; |
| 2432 | unsigned char *src, *src_end, *dst; |
| 2433 | { |
| 2434 | int charset, c, found = 0, reg; |
| 2435 | /* Table of charsets to be designated to each graphic register. */ |
| 2436 | int r[4]; |
| 2437 | |
| 2438 | for (reg = 0; reg < 4; reg++) |
| 2439 | r[reg] = -1; |
| 2440 | |
| 2441 | while (found < 4) |
| 2442 | { |
| 2443 | ONE_MORE_CHAR (c); |
| 2444 | if (c == '\n') |
| 2445 | break; |
| 2446 | |
| 2447 | charset = CHAR_CHARSET (c); |
| 2448 | reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset); |
| 2449 | if (reg != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION && r[reg] < 0) |
| 2450 | { |
| 2451 | found++; |
| 2452 | r[reg] = charset; |
| 2453 | } |
| 2454 | } |
| 2455 | |
| 2456 | label_end_of_loop: |
| 2457 | if (found) |
| 2458 | { |
| 2459 | for (reg = 0; reg < 4; reg++) |
| 2460 | if (r[reg] >= 0 |
| 2461 | && CODING_SPEC_ISO_DESIGNATION (coding, reg) != r[reg]) |
| 2462 | ENCODE_DESIGNATION (r[reg], reg, coding); |
| 2463 | } |
| 2464 | |
| 2465 | return dst; |
| 2466 | } |
| 2467 | |
| 2468 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */ |
| 2469 | |
| 2470 | static void |
| 2471 | encode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes) |
| 2472 | struct coding_system *coding; |
| 2473 | unsigned char *source, *destination; |
| 2474 | int src_bytes, dst_bytes; |
| 2475 | { |
| 2476 | unsigned char *src = source; |
| 2477 | unsigned char *src_end = source + src_bytes; |
| 2478 | unsigned char *dst = destination; |
| 2479 | unsigned char *dst_end = destination + dst_bytes; |
| 2480 | /* Since the maximum bytes produced by each loop is 20, we subtract 19 |
| 2481 | from DST_END to assure overflow checking is necessary only at the |
| 2482 | head of loop. */ |
| 2483 | unsigned char *adjusted_dst_end = dst_end - 19; |
| 2484 | /* SRC_BASE remembers the start position in source in each loop. |
| 2485 | The loop will be exited when there's not enough source text to |
| 2486 | analyze multi-byte codes (within macro ONE_MORE_CHAR), or when |
| 2487 | there's not enough destination area to produce encoded codes |
| 2488 | (within macro EMIT_BYTES). */ |
| 2489 | unsigned char *src_base; |
| 2490 | int c; |
| 2491 | Lisp_Object translation_table; |
| 2492 | Lisp_Object safe_chars; |
| 2493 | |
| 2494 | safe_chars = coding_safe_chars (coding); |
| 2495 | |
| 2496 | if (NILP (Venable_character_translation)) |
| 2497 | translation_table = Qnil; |
| 2498 | else |
| 2499 | { |
| 2500 | translation_table = coding->translation_table_for_encode; |
| 2501 | if (NILP (translation_table)) |
| 2502 | translation_table = Vstandard_translation_table_for_encode; |
| 2503 | } |
| 2504 | |
| 2505 | coding->consumed_char = 0; |
| 2506 | coding->errors = 0; |
| 2507 | while (1) |
| 2508 | { |
| 2509 | src_base = src; |
| 2510 | |
| 2511 | if (dst >= (dst_bytes ? adjusted_dst_end : (src - 19))) |
| 2512 | { |
| 2513 | coding->result = CODING_FINISH_INSUFFICIENT_DST; |
| 2514 | break; |
| 2515 | } |
| 2516 | |
| 2517 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL |
| 2518 | && CODING_SPEC_ISO_BOL (coding)) |
| 2519 | { |
| 2520 | /* We have to produce designation sequences if any now. */ |
| 2521 | dst = encode_designation_at_bol (coding, translation_table, |
| 2522 | src, src_end, dst); |
| 2523 | CODING_SPEC_ISO_BOL (coding) = 0; |
| 2524 | } |
| 2525 | |
| 2526 | /* Check composition start and end. */ |
| 2527 | if (coding->composing != COMPOSITION_DISABLED |
| 2528 | && coding->cmp_data_start < coding->cmp_data->used) |
| 2529 | { |
| 2530 | struct composition_data *cmp_data = coding->cmp_data; |
| 2531 | int *data = cmp_data->data + coding->cmp_data_start; |
| 2532 | int this_pos = cmp_data->char_offset + coding->consumed_char; |
| 2533 | |
| 2534 | if (coding->composing == COMPOSITION_RELATIVE) |
| 2535 | { |
| 2536 | if (this_pos == data[2]) |
| 2537 | { |
| 2538 | ENCODE_COMPOSITION_END (coding, data); |
| 2539 | cmp_data = coding->cmp_data; |
| 2540 | data = cmp_data->data + coding->cmp_data_start; |
| 2541 | } |
| 2542 | } |
| 2543 | else if (COMPOSING_P (coding)) |
| 2544 | { |
| 2545 | /* COMPOSITION_WITH_ALTCHARS or COMPOSITION_WITH_RULE_ALTCHAR */ |
| 2546 | if (coding->cmp_data_index == coding->cmp_data_start + data[0]) |
| 2547 | /* We have consumed components of the composition. |
| 2548 | What follows in SRC is the composition's base |
| 2549 | text. */ |
| 2550 | ENCODE_COMPOSITION_FAKE_START (coding); |
| 2551 | else |
| 2552 | { |
| 2553 | int c = cmp_data->data[coding->cmp_data_index++]; |
| 2554 | if (coding->composition_rule_follows) |
| 2555 | { |
| 2556 | ENCODE_COMPOSITION_RULE (c); |
| 2557 | coding->composition_rule_follows = 0; |
| 2558 | } |
| 2559 | else |
| 2560 | { |
| 2561 | if (coding->flags & CODING_FLAG_ISO_SAFE |
| 2562 | && ! CODING_SAFE_CHAR_P (safe_chars, c)) |
| 2563 | ENCODE_UNSAFE_CHARACTER (c); |
| 2564 | else |
| 2565 | ENCODE_ISO_CHARACTER (c); |
| 2566 | if (coding->composing == COMPOSITION_WITH_RULE_ALTCHARS) |
| 2567 | coding->composition_rule_follows = 1; |
| 2568 | } |
| 2569 | continue; |
| 2570 | } |
| 2571 | } |
| 2572 | if (!COMPOSING_P (coding)) |
| 2573 | { |
| 2574 | if (this_pos == data[1]) |
| 2575 | { |
| 2576 | ENCODE_COMPOSITION_START (coding, data); |
| 2577 | continue; |
| 2578 | } |
| 2579 | } |
| 2580 | } |
| 2581 | |
| 2582 | ONE_MORE_CHAR (c); |
| 2583 | |
| 2584 | /* Now encode the character C. */ |
| 2585 | if (c < 0x20 || c == 0x7F) |
| 2586 | { |
| 2587 | if (c == '\r') |
| 2588 | { |
| 2589 | if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
| 2590 | { |
| 2591 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) |
| 2592 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 2593 | *dst++ = c; |
| 2594 | continue; |
| 2595 | } |
| 2596 | /* fall down to treat '\r' as '\n' ... */ |
| 2597 | c = '\n'; |
| 2598 | } |
| 2599 | if (c == '\n') |
| 2600 | { |
| 2601 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL) |
| 2602 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 2603 | if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL) |
| 2604 | bcopy (coding->spec.iso2022.initial_designation, |
| 2605 | coding->spec.iso2022.current_designation, |
| 2606 | sizeof coding->spec.iso2022.initial_designation); |
| 2607 | if (coding->eol_type == CODING_EOL_LF |
| 2608 | || coding->eol_type == CODING_EOL_UNDECIDED) |
| 2609 | *dst++ = ISO_CODE_LF; |
| 2610 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 2611 | *dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF; |
| 2612 | else |
| 2613 | *dst++ = ISO_CODE_CR; |
| 2614 | CODING_SPEC_ISO_BOL (coding) = 1; |
| 2615 | } |
| 2616 | else |
| 2617 | { |
| 2618 | if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL) |
| 2619 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 2620 | *dst++ = c; |
| 2621 | } |
| 2622 | } |
| 2623 | else if (ASCII_BYTE_P (c)) |
| 2624 | ENCODE_ISO_CHARACTER (c); |
| 2625 | else if (SINGLE_BYTE_CHAR_P (c)) |
| 2626 | { |
| 2627 | *dst++ = c; |
| 2628 | coding->errors++; |
| 2629 | } |
| 2630 | else if (coding->flags & CODING_FLAG_ISO_SAFE |
| 2631 | && ! CODING_SAFE_CHAR_P (safe_chars, c)) |
| 2632 | ENCODE_UNSAFE_CHARACTER (c); |
| 2633 | else |
| 2634 | ENCODE_ISO_CHARACTER (c); |
| 2635 | |
| 2636 | coding->consumed_char++; |
| 2637 | } |
| 2638 | |
| 2639 | label_end_of_loop: |
| 2640 | coding->consumed = src_base - source; |
| 2641 | coding->produced = coding->produced_char = dst - destination; |
| 2642 | } |
| 2643 | |
| 2644 | \f |
| 2645 | /*** 4. SJIS and BIG5 handlers ***/ |
| 2646 | |
| 2647 | /* Although SJIS and BIG5 are not ISO coding systems, they are used |
| 2648 | quite widely. So, for the moment, Emacs supports them in the bare |
| 2649 | C code. But, in the future, they may be supported only by CCL. */ |
| 2650 | |
| 2651 | /* SJIS is a coding system encoding three character sets: ASCII, right |
| 2652 | half of JISX0201-Kana, and JISX0208. An ASCII character is encoded |
| 2653 | as is. A character of charset katakana-jisx0201 is encoded by |
| 2654 | "position-code + 0x80". A character of charset japanese-jisx0208 |
| 2655 | is encoded in 2-byte but two position-codes are divided and shifted |
| 2656 | so that it fits in the range below. |
| 2657 | |
| 2658 | --- CODE RANGE of SJIS --- |
| 2659 | (character set) (range) |
| 2660 | ASCII 0x00 .. 0x7F |
| 2661 | KATAKANA-JISX0201 0xA1 .. 0xDF |
| 2662 | JISX0208 (1st byte) 0x81 .. 0x9F and 0xE0 .. 0xEF |
| 2663 | (2nd byte) 0x40 .. 0x7E and 0x80 .. 0xFC |
| 2664 | ------------------------------- |
| 2665 | |
| 2666 | */ |
| 2667 | |
| 2668 | /* BIG5 is a coding system encoding two character sets: ASCII and |
| 2669 | Big5. An ASCII character is encoded as is. Big5 is a two-byte |
| 2670 | character set and is encoded in two bytes. |
| 2671 | |
| 2672 | --- CODE RANGE of BIG5 --- |
| 2673 | (character set) (range) |
| 2674 | ASCII 0x00 .. 0x7F |
| 2675 | Big5 (1st byte) 0xA1 .. 0xFE |
| 2676 | (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE |
| 2677 | -------------------------- |
| 2678 | |
| 2679 | Since the number of characters in Big5 is larger than maximum |
| 2680 | characters in Emacs' charset (96x96), it can't be handled as one |
| 2681 | charset. So, in Emacs, Big5 is divided into two: `charset-big5-1' |
| 2682 | and `charset-big5-2'. Both are DIMENSION2 and CHARS94. The former |
| 2683 | contains frequently used characters and the latter contains less |
| 2684 | frequently used characters. */ |
| 2685 | |
| 2686 | /* Macros to decode or encode a character of Big5 in BIG5. B1 and B2 |
| 2687 | are the 1st and 2nd position-codes of Big5 in BIG5 coding system. |
| 2688 | C1 and C2 are the 1st and 2nd position-codes of Emacs' internal |
| 2689 | format. CHARSET is `charset_big5_1' or `charset_big5_2'. */ |
| 2690 | |
| 2691 | /* Number of Big5 characters which have the same code in 1st byte. */ |
| 2692 | #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) |
| 2693 | |
| 2694 | #define DECODE_BIG5(b1, b2, charset, c1, c2) \ |
| 2695 | do { \ |
| 2696 | unsigned int temp \ |
| 2697 | = (b1 - 0xA1) * BIG5_SAME_ROW + b2 - (b2 < 0x7F ? 0x40 : 0x62); \ |
| 2698 | if (b1 < 0xC9) \ |
| 2699 | charset = charset_big5_1; \ |
| 2700 | else \ |
| 2701 | { \ |
| 2702 | charset = charset_big5_2; \ |
| 2703 | temp -= (0xC9 - 0xA1) * BIG5_SAME_ROW; \ |
| 2704 | } \ |
| 2705 | c1 = temp / (0xFF - 0xA1) + 0x21; \ |
| 2706 | c2 = temp % (0xFF - 0xA1) + 0x21; \ |
| 2707 | } while (0) |
| 2708 | |
| 2709 | #define ENCODE_BIG5(charset, c1, c2, b1, b2) \ |
| 2710 | do { \ |
| 2711 | unsigned int temp = (c1 - 0x21) * (0xFF - 0xA1) + (c2 - 0x21); \ |
| 2712 | if (charset == charset_big5_2) \ |
| 2713 | temp += BIG5_SAME_ROW * (0xC9 - 0xA1); \ |
| 2714 | b1 = temp / BIG5_SAME_ROW + 0xA1; \ |
| 2715 | b2 = temp % BIG5_SAME_ROW; \ |
| 2716 | b2 += b2 < 0x3F ? 0x40 : 0x62; \ |
| 2717 | } while (0) |
| 2718 | |
| 2719 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 2720 | Check if a text is encoded in SJIS. If it is, return |
| 2721 | CODING_CATEGORY_MASK_SJIS, else return 0. */ |
| 2722 | |
| 2723 | static int |
| 2724 | detect_coding_sjis (src, src_end, multibytep) |
| 2725 | unsigned char *src, *src_end; |
| 2726 | int multibytep; |
| 2727 | { |
| 2728 | int c; |
| 2729 | /* Dummy for ONE_MORE_BYTE. */ |
| 2730 | struct coding_system dummy_coding; |
| 2731 | struct coding_system *coding = &dummy_coding; |
| 2732 | |
| 2733 | while (1) |
| 2734 | { |
| 2735 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 2736 | if (c < 0x80) |
| 2737 | continue; |
| 2738 | if (c == 0x80 || c == 0xA0 || c > 0xEF) |
| 2739 | return 0; |
| 2740 | if (c <= 0x9F || c >= 0xE0) |
| 2741 | { |
| 2742 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 2743 | if (c < 0x40 || c == 0x7F || c > 0xFC) |
| 2744 | return 0; |
| 2745 | } |
| 2746 | } |
| 2747 | label_end_of_loop: |
| 2748 | return CODING_CATEGORY_MASK_SJIS; |
| 2749 | } |
| 2750 | |
| 2751 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 2752 | Check if a text is encoded in BIG5. If it is, return |
| 2753 | CODING_CATEGORY_MASK_BIG5, else return 0. */ |
| 2754 | |
| 2755 | static int |
| 2756 | detect_coding_big5 (src, src_end, multibytep) |
| 2757 | unsigned char *src, *src_end; |
| 2758 | int multibytep; |
| 2759 | { |
| 2760 | int c; |
| 2761 | /* Dummy for ONE_MORE_BYTE. */ |
| 2762 | struct coding_system dummy_coding; |
| 2763 | struct coding_system *coding = &dummy_coding; |
| 2764 | |
| 2765 | while (1) |
| 2766 | { |
| 2767 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 2768 | if (c < 0x80) |
| 2769 | continue; |
| 2770 | if (c < 0xA1 || c > 0xFE) |
| 2771 | return 0; |
| 2772 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 2773 | if (c < 0x40 || (c > 0x7F && c < 0xA1) || c > 0xFE) |
| 2774 | return 0; |
| 2775 | } |
| 2776 | label_end_of_loop: |
| 2777 | return CODING_CATEGORY_MASK_BIG5; |
| 2778 | } |
| 2779 | |
| 2780 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 2781 | Check if a text is encoded in UTF-8. If it is, return |
| 2782 | CODING_CATEGORY_MASK_UTF_8, else return 0. */ |
| 2783 | |
| 2784 | #define UTF_8_1_OCTET_P(c) ((c) < 0x80) |
| 2785 | #define UTF_8_EXTRA_OCTET_P(c) (((c) & 0xC0) == 0x80) |
| 2786 | #define UTF_8_2_OCTET_LEADING_P(c) (((c) & 0xE0) == 0xC0) |
| 2787 | #define UTF_8_3_OCTET_LEADING_P(c) (((c) & 0xF0) == 0xE0) |
| 2788 | #define UTF_8_4_OCTET_LEADING_P(c) (((c) & 0xF8) == 0xF0) |
| 2789 | #define UTF_8_5_OCTET_LEADING_P(c) (((c) & 0xFC) == 0xF8) |
| 2790 | #define UTF_8_6_OCTET_LEADING_P(c) (((c) & 0xFE) == 0xFC) |
| 2791 | |
| 2792 | static int |
| 2793 | detect_coding_utf_8 (src, src_end, multibytep) |
| 2794 | unsigned char *src, *src_end; |
| 2795 | int multibytep; |
| 2796 | { |
| 2797 | unsigned char c; |
| 2798 | int seq_maybe_bytes; |
| 2799 | /* Dummy for ONE_MORE_BYTE. */ |
| 2800 | struct coding_system dummy_coding; |
| 2801 | struct coding_system *coding = &dummy_coding; |
| 2802 | |
| 2803 | while (1) |
| 2804 | { |
| 2805 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 2806 | if (UTF_8_1_OCTET_P (c)) |
| 2807 | continue; |
| 2808 | else if (UTF_8_2_OCTET_LEADING_P (c)) |
| 2809 | seq_maybe_bytes = 1; |
| 2810 | else if (UTF_8_3_OCTET_LEADING_P (c)) |
| 2811 | seq_maybe_bytes = 2; |
| 2812 | else if (UTF_8_4_OCTET_LEADING_P (c)) |
| 2813 | seq_maybe_bytes = 3; |
| 2814 | else if (UTF_8_5_OCTET_LEADING_P (c)) |
| 2815 | seq_maybe_bytes = 4; |
| 2816 | else if (UTF_8_6_OCTET_LEADING_P (c)) |
| 2817 | seq_maybe_bytes = 5; |
| 2818 | else |
| 2819 | return 0; |
| 2820 | |
| 2821 | do |
| 2822 | { |
| 2823 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 2824 | if (!UTF_8_EXTRA_OCTET_P (c)) |
| 2825 | return 0; |
| 2826 | seq_maybe_bytes--; |
| 2827 | } |
| 2828 | while (seq_maybe_bytes > 0); |
| 2829 | } |
| 2830 | |
| 2831 | label_end_of_loop: |
| 2832 | return CODING_CATEGORY_MASK_UTF_8; |
| 2833 | } |
| 2834 | |
| 2835 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 2836 | Check if a text is encoded in UTF-16 Big Endian (endian == 1) or |
| 2837 | Little Endian (otherwise). If it is, return |
| 2838 | CODING_CATEGORY_MASK_UTF_16_BE or CODING_CATEGORY_MASK_UTF_16_LE, |
| 2839 | else return 0. */ |
| 2840 | |
| 2841 | #define UTF_16_INVALID_P(val) \ |
| 2842 | (((val) == 0xFFFE) \ |
| 2843 | || ((val) == 0xFFFF)) |
| 2844 | |
| 2845 | #define UTF_16_HIGH_SURROGATE_P(val) \ |
| 2846 | (((val) & 0xD800) == 0xD800) |
| 2847 | |
| 2848 | #define UTF_16_LOW_SURROGATE_P(val) \ |
| 2849 | (((val) & 0xDC00) == 0xDC00) |
| 2850 | |
| 2851 | static int |
| 2852 | detect_coding_utf_16 (src, src_end, multibytep) |
| 2853 | unsigned char *src, *src_end; |
| 2854 | int multibytep; |
| 2855 | { |
| 2856 | unsigned char c1, c2; |
| 2857 | /* Dummy for TWO_MORE_BYTES. */ |
| 2858 | struct coding_system dummy_coding; |
| 2859 | struct coding_system *coding = &dummy_coding; |
| 2860 | |
| 2861 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c1, multibytep); |
| 2862 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c2, multibytep); |
| 2863 | |
| 2864 | if ((c1 == 0xFF) && (c2 == 0xFE)) |
| 2865 | return CODING_CATEGORY_MASK_UTF_16_LE; |
| 2866 | else if ((c1 == 0xFE) && (c2 == 0xFF)) |
| 2867 | return CODING_CATEGORY_MASK_UTF_16_BE; |
| 2868 | |
| 2869 | label_end_of_loop: |
| 2870 | return 0; |
| 2871 | } |
| 2872 | |
| 2873 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". |
| 2874 | If SJIS_P is 1, decode SJIS text, else decode BIG5 test. */ |
| 2875 | |
| 2876 | static void |
| 2877 | decode_coding_sjis_big5 (coding, source, destination, |
| 2878 | src_bytes, dst_bytes, sjis_p) |
| 2879 | struct coding_system *coding; |
| 2880 | unsigned char *source, *destination; |
| 2881 | int src_bytes, dst_bytes; |
| 2882 | int sjis_p; |
| 2883 | { |
| 2884 | unsigned char *src = source; |
| 2885 | unsigned char *src_end = source + src_bytes; |
| 2886 | unsigned char *dst = destination; |
| 2887 | unsigned char *dst_end = destination + dst_bytes; |
| 2888 | /* SRC_BASE remembers the start position in source in each loop. |
| 2889 | The loop will be exited when there's not enough source code |
| 2890 | (within macro ONE_MORE_BYTE), or when there's not enough |
| 2891 | destination area to produce a character (within macro |
| 2892 | EMIT_CHAR). */ |
| 2893 | unsigned char *src_base; |
| 2894 | Lisp_Object translation_table; |
| 2895 | |
| 2896 | if (NILP (Venable_character_translation)) |
| 2897 | translation_table = Qnil; |
| 2898 | else |
| 2899 | { |
| 2900 | translation_table = coding->translation_table_for_decode; |
| 2901 | if (NILP (translation_table)) |
| 2902 | translation_table = Vstandard_translation_table_for_decode; |
| 2903 | } |
| 2904 | |
| 2905 | coding->produced_char = 0; |
| 2906 | while (1) |
| 2907 | { |
| 2908 | int c, charset, c1, c2; |
| 2909 | |
| 2910 | src_base = src; |
| 2911 | ONE_MORE_BYTE (c1); |
| 2912 | |
| 2913 | if (c1 < 0x80) |
| 2914 | { |
| 2915 | charset = CHARSET_ASCII; |
| 2916 | if (c1 < 0x20) |
| 2917 | { |
| 2918 | if (c1 == '\r') |
| 2919 | { |
| 2920 | if (coding->eol_type == CODING_EOL_CRLF) |
| 2921 | { |
| 2922 | ONE_MORE_BYTE (c2); |
| 2923 | if (c2 == '\n') |
| 2924 | c1 = c2; |
| 2925 | else if (coding->mode |
| 2926 | & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 2927 | { |
| 2928 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 2929 | goto label_end_of_loop; |
| 2930 | } |
| 2931 | else |
| 2932 | /* To process C2 again, SRC is subtracted by 1. */ |
| 2933 | src--; |
| 2934 | } |
| 2935 | else if (coding->eol_type == CODING_EOL_CR) |
| 2936 | c1 = '\n'; |
| 2937 | } |
| 2938 | else if (c1 == '\n' |
| 2939 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 2940 | && (coding->eol_type == CODING_EOL_CR |
| 2941 | || coding->eol_type == CODING_EOL_CRLF)) |
| 2942 | { |
| 2943 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 2944 | goto label_end_of_loop; |
| 2945 | } |
| 2946 | } |
| 2947 | } |
| 2948 | else |
| 2949 | { |
| 2950 | if (sjis_p) |
| 2951 | { |
| 2952 | if (c1 == 0x80 || c1 == 0xA0 || c1 > 0xEF) |
| 2953 | goto label_invalid_code; |
| 2954 | if (c1 <= 0x9F || c1 >= 0xE0) |
| 2955 | { |
| 2956 | /* SJIS -> JISX0208 */ |
| 2957 | ONE_MORE_BYTE (c2); |
| 2958 | if (c2 < 0x40 || c2 == 0x7F || c2 > 0xFC) |
| 2959 | goto label_invalid_code; |
| 2960 | DECODE_SJIS (c1, c2, c1, c2); |
| 2961 | charset = charset_jisx0208; |
| 2962 | } |
| 2963 | else |
| 2964 | /* SJIS -> JISX0201-Kana */ |
| 2965 | charset = charset_katakana_jisx0201; |
| 2966 | } |
| 2967 | else |
| 2968 | { |
| 2969 | /* BIG5 -> Big5 */ |
| 2970 | if (c1 < 0xA0 || c1 > 0xFE) |
| 2971 | goto label_invalid_code; |
| 2972 | ONE_MORE_BYTE (c2); |
| 2973 | if (c2 < 0x40 || (c2 > 0x7E && c2 < 0xA1) || c2 > 0xFE) |
| 2974 | goto label_invalid_code; |
| 2975 | DECODE_BIG5 (c1, c2, charset, c1, c2); |
| 2976 | } |
| 2977 | } |
| 2978 | |
| 2979 | c = DECODE_ISO_CHARACTER (charset, c1, c2); |
| 2980 | EMIT_CHAR (c); |
| 2981 | continue; |
| 2982 | |
| 2983 | label_invalid_code: |
| 2984 | coding->errors++; |
| 2985 | src = src_base; |
| 2986 | c = *src++; |
| 2987 | EMIT_CHAR (c); |
| 2988 | } |
| 2989 | |
| 2990 | label_end_of_loop: |
| 2991 | coding->consumed = coding->consumed_char = src_base - source; |
| 2992 | coding->produced = dst - destination; |
| 2993 | return; |
| 2994 | } |
| 2995 | |
| 2996 | /* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". |
| 2997 | This function can encode charsets `ascii', `katakana-jisx0201', |
| 2998 | `japanese-jisx0208', `chinese-big5-1', and `chinese-big5-2'. We |
| 2999 | are sure that all these charsets are registered as official charset |
| 3000 | (i.e. do not have extended leading-codes). Characters of other |
| 3001 | charsets are produced without any encoding. If SJIS_P is 1, encode |
| 3002 | SJIS text, else encode BIG5 text. */ |
| 3003 | |
| 3004 | static void |
| 3005 | encode_coding_sjis_big5 (coding, source, destination, |
| 3006 | src_bytes, dst_bytes, sjis_p) |
| 3007 | struct coding_system *coding; |
| 3008 | unsigned char *source, *destination; |
| 3009 | int src_bytes, dst_bytes; |
| 3010 | int sjis_p; |
| 3011 | { |
| 3012 | unsigned char *src = source; |
| 3013 | unsigned char *src_end = source + src_bytes; |
| 3014 | unsigned char *dst = destination; |
| 3015 | unsigned char *dst_end = destination + dst_bytes; |
| 3016 | /* SRC_BASE remembers the start position in source in each loop. |
| 3017 | The loop will be exited when there's not enough source text to |
| 3018 | analyze multi-byte codes (within macro ONE_MORE_CHAR), or when |
| 3019 | there's not enough destination area to produce encoded codes |
| 3020 | (within macro EMIT_BYTES). */ |
| 3021 | unsigned char *src_base; |
| 3022 | Lisp_Object translation_table; |
| 3023 | |
| 3024 | if (NILP (Venable_character_translation)) |
| 3025 | translation_table = Qnil; |
| 3026 | else |
| 3027 | { |
| 3028 | translation_table = coding->translation_table_for_encode; |
| 3029 | if (NILP (translation_table)) |
| 3030 | translation_table = Vstandard_translation_table_for_encode; |
| 3031 | } |
| 3032 | |
| 3033 | while (1) |
| 3034 | { |
| 3035 | int c, charset, c1, c2; |
| 3036 | |
| 3037 | src_base = src; |
| 3038 | ONE_MORE_CHAR (c); |
| 3039 | |
| 3040 | /* Now encode the character C. */ |
| 3041 | if (SINGLE_BYTE_CHAR_P (c)) |
| 3042 | { |
| 3043 | switch (c) |
| 3044 | { |
| 3045 | case '\r': |
| 3046 | if (!(coding->mode & CODING_MODE_SELECTIVE_DISPLAY)) |
| 3047 | { |
| 3048 | EMIT_ONE_BYTE (c); |
| 3049 | break; |
| 3050 | } |
| 3051 | c = '\n'; |
| 3052 | case '\n': |
| 3053 | if (coding->eol_type == CODING_EOL_CRLF) |
| 3054 | { |
| 3055 | EMIT_TWO_BYTES ('\r', c); |
| 3056 | break; |
| 3057 | } |
| 3058 | else if (coding->eol_type == CODING_EOL_CR) |
| 3059 | c = '\r'; |
| 3060 | default: |
| 3061 | EMIT_ONE_BYTE (c); |
| 3062 | } |
| 3063 | } |
| 3064 | else |
| 3065 | { |
| 3066 | SPLIT_CHAR (c, charset, c1, c2); |
| 3067 | if (sjis_p) |
| 3068 | { |
| 3069 | if (charset == charset_jisx0208 |
| 3070 | || charset == charset_jisx0208_1978) |
| 3071 | { |
| 3072 | ENCODE_SJIS (c1, c2, c1, c2); |
| 3073 | EMIT_TWO_BYTES (c1, c2); |
| 3074 | } |
| 3075 | else if (charset == charset_katakana_jisx0201) |
| 3076 | EMIT_ONE_BYTE (c1 | 0x80); |
| 3077 | else if (charset == charset_latin_jisx0201) |
| 3078 | EMIT_ONE_BYTE (c1); |
| 3079 | else |
| 3080 | /* There's no way other than producing the internal |
| 3081 | codes as is. */ |
| 3082 | EMIT_BYTES (src_base, src); |
| 3083 | } |
| 3084 | else |
| 3085 | { |
| 3086 | if (charset == charset_big5_1 || charset == charset_big5_2) |
| 3087 | { |
| 3088 | ENCODE_BIG5 (charset, c1, c2, c1, c2); |
| 3089 | EMIT_TWO_BYTES (c1, c2); |
| 3090 | } |
| 3091 | else |
| 3092 | /* There's no way other than producing the internal |
| 3093 | codes as is. */ |
| 3094 | EMIT_BYTES (src_base, src); |
| 3095 | } |
| 3096 | } |
| 3097 | coding->consumed_char++; |
| 3098 | } |
| 3099 | |
| 3100 | label_end_of_loop: |
| 3101 | coding->consumed = src_base - source; |
| 3102 | coding->produced = coding->produced_char = dst - destination; |
| 3103 | } |
| 3104 | |
| 3105 | \f |
| 3106 | /*** 5. CCL handlers ***/ |
| 3107 | |
| 3108 | /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions". |
| 3109 | Check if a text is encoded in a coding system of which |
| 3110 | encoder/decoder are written in CCL program. If it is, return |
| 3111 | CODING_CATEGORY_MASK_CCL, else return 0. */ |
| 3112 | |
| 3113 | static int |
| 3114 | detect_coding_ccl (src, src_end, multibytep) |
| 3115 | unsigned char *src, *src_end; |
| 3116 | int multibytep; |
| 3117 | { |
| 3118 | unsigned char *valid; |
| 3119 | int c; |
| 3120 | /* Dummy for ONE_MORE_BYTE. */ |
| 3121 | struct coding_system dummy_coding; |
| 3122 | struct coding_system *coding = &dummy_coding; |
| 3123 | |
| 3124 | /* No coding system is assigned to coding-category-ccl. */ |
| 3125 | if (!coding_system_table[CODING_CATEGORY_IDX_CCL]) |
| 3126 | return 0; |
| 3127 | |
| 3128 | valid = coding_system_table[CODING_CATEGORY_IDX_CCL]->spec.ccl.valid_codes; |
| 3129 | while (1) |
| 3130 | { |
| 3131 | ONE_MORE_BYTE_CHECK_MULTIBYTE (c, multibytep); |
| 3132 | if (! valid[c]) |
| 3133 | return 0; |
| 3134 | } |
| 3135 | label_end_of_loop: |
| 3136 | return CODING_CATEGORY_MASK_CCL; |
| 3137 | } |
| 3138 | |
| 3139 | \f |
| 3140 | /*** 6. End-of-line handlers ***/ |
| 3141 | |
| 3142 | /* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */ |
| 3143 | |
| 3144 | static void |
| 3145 | decode_eol (coding, source, destination, src_bytes, dst_bytes) |
| 3146 | struct coding_system *coding; |
| 3147 | unsigned char *source, *destination; |
| 3148 | int src_bytes, dst_bytes; |
| 3149 | { |
| 3150 | unsigned char *src = source; |
| 3151 | unsigned char *dst = destination; |
| 3152 | unsigned char *src_end = src + src_bytes; |
| 3153 | unsigned char *dst_end = dst + dst_bytes; |
| 3154 | Lisp_Object translation_table; |
| 3155 | /* SRC_BASE remembers the start position in source in each loop. |
| 3156 | The loop will be exited when there's not enough source code |
| 3157 | (within macro ONE_MORE_BYTE), or when there's not enough |
| 3158 | destination area to produce a character (within macro |
| 3159 | EMIT_CHAR). */ |
| 3160 | unsigned char *src_base; |
| 3161 | int c; |
| 3162 | |
| 3163 | translation_table = Qnil; |
| 3164 | switch (coding->eol_type) |
| 3165 | { |
| 3166 | case CODING_EOL_CRLF: |
| 3167 | while (1) |
| 3168 | { |
| 3169 | src_base = src; |
| 3170 | ONE_MORE_BYTE (c); |
| 3171 | if (c == '\r') |
| 3172 | { |
| 3173 | ONE_MORE_BYTE (c); |
| 3174 | if (c != '\n') |
| 3175 | { |
| 3176 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 3177 | { |
| 3178 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 3179 | goto label_end_of_loop; |
| 3180 | } |
| 3181 | src--; |
| 3182 | c = '\r'; |
| 3183 | } |
| 3184 | } |
| 3185 | else if (c == '\n' |
| 3186 | && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)) |
| 3187 | { |
| 3188 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 3189 | goto label_end_of_loop; |
| 3190 | } |
| 3191 | EMIT_CHAR (c); |
| 3192 | } |
| 3193 | break; |
| 3194 | |
| 3195 | case CODING_EOL_CR: |
| 3196 | while (1) |
| 3197 | { |
| 3198 | src_base = src; |
| 3199 | ONE_MORE_BYTE (c); |
| 3200 | if (c == '\n') |
| 3201 | { |
| 3202 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 3203 | { |
| 3204 | coding->result = CODING_FINISH_INCONSISTENT_EOL; |
| 3205 | goto label_end_of_loop; |
| 3206 | } |
| 3207 | } |
| 3208 | else if (c == '\r') |
| 3209 | c = '\n'; |
| 3210 | EMIT_CHAR (c); |
| 3211 | } |
| 3212 | break; |
| 3213 | |
| 3214 | default: /* no need for EOL handling */ |
| 3215 | while (1) |
| 3216 | { |
| 3217 | src_base = src; |
| 3218 | ONE_MORE_BYTE (c); |
| 3219 | EMIT_CHAR (c); |
| 3220 | } |
| 3221 | } |
| 3222 | |
| 3223 | label_end_of_loop: |
| 3224 | coding->consumed = coding->consumed_char = src_base - source; |
| 3225 | coding->produced = dst - destination; |
| 3226 | return; |
| 3227 | } |
| 3228 | |
| 3229 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". Encode |
| 3230 | format of end-of-line according to `coding->eol_type'. It also |
| 3231 | convert multibyte form 8-bit characters to unibyte if |
| 3232 | CODING->src_multibyte is nonzero. If `coding->mode & |
| 3233 | CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code '\r' in source text |
| 3234 | also means end-of-line. */ |
| 3235 | |
| 3236 | static void |
| 3237 | encode_eol (coding, source, destination, src_bytes, dst_bytes) |
| 3238 | struct coding_system *coding; |
| 3239 | unsigned char *source, *destination; |
| 3240 | int src_bytes, dst_bytes; |
| 3241 | { |
| 3242 | unsigned char *src = source; |
| 3243 | unsigned char *dst = destination; |
| 3244 | unsigned char *src_end = src + src_bytes; |
| 3245 | unsigned char *dst_end = dst + dst_bytes; |
| 3246 | Lisp_Object translation_table; |
| 3247 | /* SRC_BASE remembers the start position in source in each loop. |
| 3248 | The loop will be exited when there's not enough source text to |
| 3249 | analyze multi-byte codes (within macro ONE_MORE_CHAR), or when |
| 3250 | there's not enough destination area to produce encoded codes |
| 3251 | (within macro EMIT_BYTES). */ |
| 3252 | unsigned char *src_base; |
| 3253 | int c; |
| 3254 | int selective_display = coding->mode & CODING_MODE_SELECTIVE_DISPLAY; |
| 3255 | |
| 3256 | translation_table = Qnil; |
| 3257 | if (coding->src_multibyte |
| 3258 | && *(src_end - 1) == LEADING_CODE_8_BIT_CONTROL) |
| 3259 | { |
| 3260 | src_end--; |
| 3261 | src_bytes--; |
| 3262 | coding->result = CODING_FINISH_INSUFFICIENT_SRC; |
| 3263 | } |
| 3264 | |
| 3265 | if (coding->eol_type == CODING_EOL_CRLF) |
| 3266 | { |
| 3267 | while (src < src_end) |
| 3268 | { |
| 3269 | src_base = src; |
| 3270 | c = *src++; |
| 3271 | if (c >= 0x20) |
| 3272 | EMIT_ONE_BYTE (c); |
| 3273 | else if (c == '\n' || (c == '\r' && selective_display)) |
| 3274 | EMIT_TWO_BYTES ('\r', '\n'); |
| 3275 | else |
| 3276 | EMIT_ONE_BYTE (c); |
| 3277 | } |
| 3278 | src_base = src; |
| 3279 | label_end_of_loop: |
| 3280 | ; |
| 3281 | } |
| 3282 | else |
| 3283 | { |
| 3284 | if (!dst_bytes || src_bytes <= dst_bytes) |
| 3285 | { |
| 3286 | safe_bcopy (src, dst, src_bytes); |
| 3287 | src_base = src_end; |
| 3288 | dst += src_bytes; |
| 3289 | } |
| 3290 | else |
| 3291 | { |
| 3292 | if (coding->src_multibyte |
| 3293 | && *(src + dst_bytes - 1) == LEADING_CODE_8_BIT_CONTROL) |
| 3294 | dst_bytes--; |
| 3295 | safe_bcopy (src, dst, dst_bytes); |
| 3296 | src_base = src + dst_bytes; |
| 3297 | dst = destination + dst_bytes; |
| 3298 | coding->result = CODING_FINISH_INSUFFICIENT_DST; |
| 3299 | } |
| 3300 | if (coding->eol_type == CODING_EOL_CR) |
| 3301 | { |
| 3302 | for (src = destination; src < dst; src++) |
| 3303 | if (*src == '\n') *src = '\r'; |
| 3304 | } |
| 3305 | else if (selective_display) |
| 3306 | { |
| 3307 | for (src = destination; src < dst; src++) |
| 3308 | if (*src == '\r') *src = '\n'; |
| 3309 | } |
| 3310 | } |
| 3311 | if (coding->src_multibyte) |
| 3312 | dst = destination + str_as_unibyte (destination, dst - destination); |
| 3313 | |
| 3314 | coding->consumed = src_base - source; |
| 3315 | coding->produced = dst - destination; |
| 3316 | coding->produced_char = coding->produced; |
| 3317 | } |
| 3318 | |
| 3319 | \f |
| 3320 | /*** 7. C library functions ***/ |
| 3321 | |
| 3322 | /* In Emacs Lisp, a coding system is represented by a Lisp symbol which |
| 3323 | has a property `coding-system'. The value of this property is a |
| 3324 | vector of length 5 (called the coding-vector). Among elements of |
| 3325 | this vector, the first (element[0]) and the fifth (element[4]) |
| 3326 | carry important information for decoding/encoding. Before |
| 3327 | decoding/encoding, this information should be set in fields of a |
| 3328 | structure of type `coding_system'. |
| 3329 | |
| 3330 | The value of the property `coding-system' can be a symbol of another |
| 3331 | subsidiary coding-system. In that case, Emacs gets coding-vector |
| 3332 | from that symbol. |
| 3333 | |
| 3334 | `element[0]' contains information to be set in `coding->type'. The |
| 3335 | value and its meaning is as follows: |
| 3336 | |
| 3337 | 0 -- coding_type_emacs_mule |
| 3338 | 1 -- coding_type_sjis |
| 3339 | 2 -- coding_type_iso2022 |
| 3340 | 3 -- coding_type_big5 |
| 3341 | 4 -- coding_type_ccl encoder/decoder written in CCL |
| 3342 | nil -- coding_type_no_conversion |
| 3343 | t -- coding_type_undecided (automatic conversion on decoding, |
| 3344 | no-conversion on encoding) |
| 3345 | |
| 3346 | `element[4]' contains information to be set in `coding->flags' and |
| 3347 | `coding->spec'. The meaning varies by `coding->type'. |
| 3348 | |
| 3349 | If `coding->type' is `coding_type_iso2022', element[4] is a vector |
| 3350 | of length 32 (of which the first 13 sub-elements are used now). |
| 3351 | Meanings of these sub-elements are: |
| 3352 | |
| 3353 | sub-element[N] where N is 0 through 3: to be set in `coding->spec.iso2022' |
| 3354 | If the value is an integer of valid charset, the charset is |
| 3355 | assumed to be designated to graphic register N initially. |
| 3356 | |
| 3357 | If the value is minus, it is a minus value of charset which |
| 3358 | reserves graphic register N, which means that the charset is |
| 3359 | not designated initially but should be designated to graphic |
| 3360 | register N just before encoding a character in that charset. |
| 3361 | |
| 3362 | If the value is nil, graphic register N is never used on |
| 3363 | encoding. |
| 3364 | |
| 3365 | sub-element[N] where N is 4 through 11: to be set in `coding->flags' |
| 3366 | Each value takes t or nil. See the section ISO2022 of |
| 3367 | `coding.h' for more information. |
| 3368 | |
| 3369 | If `coding->type' is `coding_type_big5', element[4] is t to denote |
| 3370 | BIG5-ETen or nil to denote BIG5-HKU. |
| 3371 | |
| 3372 | If `coding->type' takes the other value, element[4] is ignored. |
| 3373 | |
| 3374 | Emacs Lisp's coding systems also carry information about format of |
| 3375 | end-of-line in a value of property `eol-type'. If the value is |
| 3376 | integer, 0 means CODING_EOL_LF, 1 means CODING_EOL_CRLF, and 2 |
| 3377 | means CODING_EOL_CR. If it is not integer, it should be a vector |
| 3378 | of subsidiary coding systems of which property `eol-type' has one |
| 3379 | of the above values. |
| 3380 | |
| 3381 | */ |
| 3382 | |
| 3383 | /* Extract information for decoding/encoding from CODING_SYSTEM_SYMBOL |
| 3384 | and set it in CODING. If CODING_SYSTEM_SYMBOL is invalid, CODING |
| 3385 | is setup so that no conversion is necessary and return -1, else |
| 3386 | return 0. */ |
| 3387 | |
| 3388 | int |
| 3389 | setup_coding_system (coding_system, coding) |
| 3390 | Lisp_Object coding_system; |
| 3391 | struct coding_system *coding; |
| 3392 | { |
| 3393 | Lisp_Object coding_spec, coding_type, eol_type, plist; |
| 3394 | Lisp_Object val; |
| 3395 | |
| 3396 | /* At first, zero clear all members. */ |
| 3397 | bzero (coding, sizeof (struct coding_system)); |
| 3398 | |
| 3399 | /* Initialize some fields required for all kinds of coding systems. */ |
| 3400 | coding->symbol = coding_system; |
| 3401 | coding->heading_ascii = -1; |
| 3402 | coding->post_read_conversion = coding->pre_write_conversion = Qnil; |
| 3403 | coding->composing = COMPOSITION_DISABLED; |
| 3404 | coding->cmp_data = NULL; |
| 3405 | |
| 3406 | if (NILP (coding_system)) |
| 3407 | goto label_invalid_coding_system; |
| 3408 | |
| 3409 | coding_spec = Fget (coding_system, Qcoding_system); |
| 3410 | |
| 3411 | if (!VECTORP (coding_spec) |
| 3412 | || XVECTOR (coding_spec)->size != 5 |
| 3413 | || !CONSP (XVECTOR (coding_spec)->contents[3])) |
| 3414 | goto label_invalid_coding_system; |
| 3415 | |
| 3416 | eol_type = inhibit_eol_conversion ? Qnil : Fget (coding_system, Qeol_type); |
| 3417 | if (VECTORP (eol_type)) |
| 3418 | { |
| 3419 | coding->eol_type = CODING_EOL_UNDECIDED; |
| 3420 | coding->common_flags = CODING_REQUIRE_DETECTION_MASK; |
| 3421 | } |
| 3422 | else if (XFASTINT (eol_type) == 1) |
| 3423 | { |
| 3424 | coding->eol_type = CODING_EOL_CRLF; |
| 3425 | coding->common_flags |
| 3426 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3427 | } |
| 3428 | else if (XFASTINT (eol_type) == 2) |
| 3429 | { |
| 3430 | coding->eol_type = CODING_EOL_CR; |
| 3431 | coding->common_flags |
| 3432 | = CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3433 | } |
| 3434 | else |
| 3435 | coding->eol_type = CODING_EOL_LF; |
| 3436 | |
| 3437 | coding_type = XVECTOR (coding_spec)->contents[0]; |
| 3438 | /* Try short cut. */ |
| 3439 | if (SYMBOLP (coding_type)) |
| 3440 | { |
| 3441 | if (EQ (coding_type, Qt)) |
| 3442 | { |
| 3443 | coding->type = coding_type_undecided; |
| 3444 | coding->common_flags |= CODING_REQUIRE_DETECTION_MASK; |
| 3445 | } |
| 3446 | else |
| 3447 | coding->type = coding_type_no_conversion; |
| 3448 | /* Initialize this member. Any thing other than |
| 3449 | CODING_CATEGORY_IDX_UTF_16_BE and |
| 3450 | CODING_CATEGORY_IDX_UTF_16_LE are ok because they have |
| 3451 | special treatment in detect_eol. */ |
| 3452 | coding->category_idx = CODING_CATEGORY_IDX_EMACS_MULE; |
| 3453 | |
| 3454 | return 0; |
| 3455 | } |
| 3456 | |
| 3457 | /* Get values of coding system properties: |
| 3458 | `post-read-conversion', `pre-write-conversion', |
| 3459 | `translation-table-for-decode', `translation-table-for-encode'. */ |
| 3460 | plist = XVECTOR (coding_spec)->contents[3]; |
| 3461 | /* Pre & post conversion functions should be disabled if |
| 3462 | inhibit_eol_conversion is nonzero. This is the case that a code |
| 3463 | conversion function is called while those functions are running. */ |
| 3464 | if (! inhibit_pre_post_conversion) |
| 3465 | { |
| 3466 | coding->post_read_conversion = Fplist_get (plist, Qpost_read_conversion); |
| 3467 | coding->pre_write_conversion = Fplist_get (plist, Qpre_write_conversion); |
| 3468 | } |
| 3469 | val = Fplist_get (plist, Qtranslation_table_for_decode); |
| 3470 | if (SYMBOLP (val)) |
| 3471 | val = Fget (val, Qtranslation_table_for_decode); |
| 3472 | coding->translation_table_for_decode = CHAR_TABLE_P (val) ? val : Qnil; |
| 3473 | val = Fplist_get (plist, Qtranslation_table_for_encode); |
| 3474 | if (SYMBOLP (val)) |
| 3475 | val = Fget (val, Qtranslation_table_for_encode); |
| 3476 | coding->translation_table_for_encode = CHAR_TABLE_P (val) ? val : Qnil; |
| 3477 | val = Fplist_get (plist, Qcoding_category); |
| 3478 | if (!NILP (val)) |
| 3479 | { |
| 3480 | val = Fget (val, Qcoding_category_index); |
| 3481 | if (INTEGERP (val)) |
| 3482 | coding->category_idx = XINT (val); |
| 3483 | else |
| 3484 | goto label_invalid_coding_system; |
| 3485 | } |
| 3486 | else |
| 3487 | goto label_invalid_coding_system; |
| 3488 | |
| 3489 | /* If the coding system has non-nil `composition' property, enable |
| 3490 | composition handling. */ |
| 3491 | val = Fplist_get (plist, Qcomposition); |
| 3492 | if (!NILP (val)) |
| 3493 | coding->composing = COMPOSITION_NO; |
| 3494 | |
| 3495 | switch (XFASTINT (coding_type)) |
| 3496 | { |
| 3497 | case 0: |
| 3498 | coding->type = coding_type_emacs_mule; |
| 3499 | coding->common_flags |
| 3500 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3501 | coding->composing = COMPOSITION_NO; |
| 3502 | if (!NILP (coding->post_read_conversion)) |
| 3503 | coding->common_flags |= CODING_REQUIRE_DECODING_MASK; |
| 3504 | if (!NILP (coding->pre_write_conversion)) |
| 3505 | coding->common_flags |= CODING_REQUIRE_ENCODING_MASK; |
| 3506 | break; |
| 3507 | |
| 3508 | case 1: |
| 3509 | coding->type = coding_type_sjis; |
| 3510 | coding->common_flags |
| 3511 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3512 | break; |
| 3513 | |
| 3514 | case 2: |
| 3515 | coding->type = coding_type_iso2022; |
| 3516 | coding->common_flags |
| 3517 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3518 | { |
| 3519 | Lisp_Object val, temp; |
| 3520 | Lisp_Object *flags; |
| 3521 | int i, charset, reg_bits = 0; |
| 3522 | |
| 3523 | val = XVECTOR (coding_spec)->contents[4]; |
| 3524 | |
| 3525 | if (!VECTORP (val) || XVECTOR (val)->size != 32) |
| 3526 | goto label_invalid_coding_system; |
| 3527 | |
| 3528 | flags = XVECTOR (val)->contents; |
| 3529 | coding->flags |
| 3530 | = ((NILP (flags[4]) ? 0 : CODING_FLAG_ISO_SHORT_FORM) |
| 3531 | | (NILP (flags[5]) ? 0 : CODING_FLAG_ISO_RESET_AT_EOL) |
| 3532 | | (NILP (flags[6]) ? 0 : CODING_FLAG_ISO_RESET_AT_CNTL) |
| 3533 | | (NILP (flags[7]) ? 0 : CODING_FLAG_ISO_SEVEN_BITS) |
| 3534 | | (NILP (flags[8]) ? 0 : CODING_FLAG_ISO_LOCKING_SHIFT) |
| 3535 | | (NILP (flags[9]) ? 0 : CODING_FLAG_ISO_SINGLE_SHIFT) |
| 3536 | | (NILP (flags[10]) ? 0 : CODING_FLAG_ISO_USE_ROMAN) |
| 3537 | | (NILP (flags[11]) ? 0 : CODING_FLAG_ISO_USE_OLDJIS) |
| 3538 | | (NILP (flags[12]) ? 0 : CODING_FLAG_ISO_NO_DIRECTION) |
| 3539 | | (NILP (flags[13]) ? 0 : CODING_FLAG_ISO_INIT_AT_BOL) |
| 3540 | | (NILP (flags[14]) ? 0 : CODING_FLAG_ISO_DESIGNATE_AT_BOL) |
| 3541 | | (NILP (flags[15]) ? 0 : CODING_FLAG_ISO_SAFE) |
| 3542 | | (NILP (flags[16]) ? 0 : CODING_FLAG_ISO_LATIN_EXTRA) |
| 3543 | ); |
| 3544 | |
| 3545 | /* Invoke graphic register 0 to plane 0. */ |
| 3546 | CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; |
| 3547 | /* Invoke graphic register 1 to plane 1 if we can use full 8-bit. */ |
| 3548 | CODING_SPEC_ISO_INVOCATION (coding, 1) |
| 3549 | = (coding->flags & CODING_FLAG_ISO_SEVEN_BITS ? -1 : 1); |
| 3550 | /* Not single shifting at first. */ |
| 3551 | CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 0; |
| 3552 | /* Beginning of buffer should also be regarded as bol. */ |
| 3553 | CODING_SPEC_ISO_BOL (coding) = 1; |
| 3554 | |
| 3555 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
| 3556 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = 255; |
| 3557 | val = Vcharset_revision_alist; |
| 3558 | while (CONSP (val)) |
| 3559 | { |
| 3560 | charset = get_charset_id (Fcar_safe (XCAR (val))); |
| 3561 | if (charset >= 0 |
| 3562 | && (temp = Fcdr_safe (XCAR (val)), INTEGERP (temp)) |
| 3563 | && (i = XINT (temp), (i >= 0 && (i + '@') < 128))) |
| 3564 | CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = i; |
| 3565 | val = XCDR (val); |
| 3566 | } |
| 3567 | |
| 3568 | /* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations. |
| 3569 | FLAGS[REG] can be one of below: |
| 3570 | integer CHARSET: CHARSET occupies register I, |
| 3571 | t: designate nothing to REG initially, but can be used |
| 3572 | by any charsets, |
| 3573 | list of integer, nil, or t: designate the first |
| 3574 | element (if integer) to REG initially, the remaining |
| 3575 | elements (if integer) is designated to REG on request, |
| 3576 | if an element is t, REG can be used by any charsets, |
| 3577 | nil: REG is never used. */ |
| 3578 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
| 3579 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3580 | = CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION; |
| 3581 | for (i = 0; i < 4; i++) |
| 3582 | { |
| 3583 | if ((INTEGERP (flags[i]) |
| 3584 | && (charset = XINT (flags[i]), CHARSET_VALID_P (charset))) |
| 3585 | || (charset = get_charset_id (flags[i])) >= 0) |
| 3586 | { |
| 3587 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; |
| 3588 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) = i; |
| 3589 | } |
| 3590 | else if (EQ (flags[i], Qt)) |
| 3591 | { |
| 3592 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; |
| 3593 | reg_bits |= 1 << i; |
| 3594 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; |
| 3595 | } |
| 3596 | else if (CONSP (flags[i])) |
| 3597 | { |
| 3598 | Lisp_Object tail; |
| 3599 | tail = flags[i]; |
| 3600 | |
| 3601 | coding->flags |= CODING_FLAG_ISO_DESIGNATION; |
| 3602 | if ((INTEGERP (XCAR (tail)) |
| 3603 | && (charset = XINT (XCAR (tail)), |
| 3604 | CHARSET_VALID_P (charset))) |
| 3605 | || (charset = get_charset_id (XCAR (tail))) >= 0) |
| 3606 | { |
| 3607 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset; |
| 3608 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) =i; |
| 3609 | } |
| 3610 | else |
| 3611 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; |
| 3612 | tail = XCDR (tail); |
| 3613 | while (CONSP (tail)) |
| 3614 | { |
| 3615 | if ((INTEGERP (XCAR (tail)) |
| 3616 | && (charset = XINT (XCAR (tail)), |
| 3617 | CHARSET_VALID_P (charset))) |
| 3618 | || (charset = get_charset_id (XCAR (tail))) >= 0) |
| 3619 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3620 | = i; |
| 3621 | else if (EQ (XCAR (tail), Qt)) |
| 3622 | reg_bits |= 1 << i; |
| 3623 | tail = XCDR (tail); |
| 3624 | } |
| 3625 | } |
| 3626 | else |
| 3627 | CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1; |
| 3628 | |
| 3629 | CODING_SPEC_ISO_DESIGNATION (coding, i) |
| 3630 | = CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i); |
| 3631 | } |
| 3632 | |
| 3633 | if (reg_bits && ! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)) |
| 3634 | { |
| 3635 | /* REG 1 can be used only by locking shift in 7-bit env. */ |
| 3636 | if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) |
| 3637 | reg_bits &= ~2; |
| 3638 | if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)) |
| 3639 | /* Without any shifting, only REG 0 and 1 can be used. */ |
| 3640 | reg_bits &= 3; |
| 3641 | } |
| 3642 | |
| 3643 | if (reg_bits) |
| 3644 | for (charset = 0; charset <= MAX_CHARSET; charset++) |
| 3645 | { |
| 3646 | if (CHARSET_DEFINED_P (charset) |
| 3647 | && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3648 | == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)) |
| 3649 | { |
| 3650 | /* There exist some default graphic registers to be |
| 3651 | used by CHARSET. */ |
| 3652 | |
| 3653 | /* We had better avoid designating a charset of |
| 3654 | CHARS96 to REG 0 as far as possible. */ |
| 3655 | if (CHARSET_CHARS (charset) == 96) |
| 3656 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3657 | = (reg_bits & 2 |
| 3658 | ? 1 : (reg_bits & 4 ? 2 : (reg_bits & 8 ? 3 : 0))); |
| 3659 | else |
| 3660 | CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) |
| 3661 | = (reg_bits & 1 |
| 3662 | ? 0 : (reg_bits & 2 ? 1 : (reg_bits & 4 ? 2 : 3))); |
| 3663 | } |
| 3664 | } |
| 3665 | } |
| 3666 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
| 3667 | coding->spec.iso2022.last_invalid_designation_register = -1; |
| 3668 | break; |
| 3669 | |
| 3670 | case 3: |
| 3671 | coding->type = coding_type_big5; |
| 3672 | coding->common_flags |
| 3673 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3674 | coding->flags |
| 3675 | = (NILP (XVECTOR (coding_spec)->contents[4]) |
| 3676 | ? CODING_FLAG_BIG5_HKU |
| 3677 | : CODING_FLAG_BIG5_ETEN); |
| 3678 | break; |
| 3679 | |
| 3680 | case 4: |
| 3681 | coding->type = coding_type_ccl; |
| 3682 | coding->common_flags |
| 3683 | |= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK; |
| 3684 | { |
| 3685 | val = XVECTOR (coding_spec)->contents[4]; |
| 3686 | if (! CONSP (val) |
| 3687 | || setup_ccl_program (&(coding->spec.ccl.decoder), |
| 3688 | XCAR (val)) < 0 |
| 3689 | || setup_ccl_program (&(coding->spec.ccl.encoder), |
| 3690 | XCDR (val)) < 0) |
| 3691 | goto label_invalid_coding_system; |
| 3692 | |
| 3693 | bzero (coding->spec.ccl.valid_codes, 256); |
| 3694 | val = Fplist_get (plist, Qvalid_codes); |
| 3695 | if (CONSP (val)) |
| 3696 | { |
| 3697 | Lisp_Object this; |
| 3698 | |
| 3699 | for (; CONSP (val); val = XCDR (val)) |
| 3700 | { |
| 3701 | this = XCAR (val); |
| 3702 | if (INTEGERP (this) |
| 3703 | && XINT (this) >= 0 && XINT (this) < 256) |
| 3704 | coding->spec.ccl.valid_codes[XINT (this)] = 1; |
| 3705 | else if (CONSP (this) |
| 3706 | && INTEGERP (XCAR (this)) |
| 3707 | && INTEGERP (XCDR (this))) |
| 3708 | { |
| 3709 | int start = XINT (XCAR (this)); |
| 3710 | int end = XINT (XCDR (this)); |
| 3711 | |
| 3712 | if (start >= 0 && start <= end && end < 256) |
| 3713 | while (start <= end) |
| 3714 | coding->spec.ccl.valid_codes[start++] = 1; |
| 3715 | } |
| 3716 | } |
| 3717 | } |
| 3718 | } |
| 3719 | coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK; |
| 3720 | coding->spec.ccl.cr_carryover = 0; |
| 3721 | coding->spec.ccl.eight_bit_carryover[0] = 0; |
| 3722 | break; |
| 3723 | |
| 3724 | case 5: |
| 3725 | coding->type = coding_type_raw_text; |
| 3726 | break; |
| 3727 | |
| 3728 | default: |
| 3729 | goto label_invalid_coding_system; |
| 3730 | } |
| 3731 | return 0; |
| 3732 | |
| 3733 | label_invalid_coding_system: |
| 3734 | coding->type = coding_type_no_conversion; |
| 3735 | coding->category_idx = CODING_CATEGORY_IDX_BINARY; |
| 3736 | coding->common_flags = 0; |
| 3737 | coding->eol_type = CODING_EOL_LF; |
| 3738 | coding->pre_write_conversion = coding->post_read_conversion = Qnil; |
| 3739 | return -1; |
| 3740 | } |
| 3741 | |
| 3742 | /* Free memory blocks allocated for storing composition information. */ |
| 3743 | |
| 3744 | void |
| 3745 | coding_free_composition_data (coding) |
| 3746 | struct coding_system *coding; |
| 3747 | { |
| 3748 | struct composition_data *cmp_data = coding->cmp_data, *next; |
| 3749 | |
| 3750 | if (!cmp_data) |
| 3751 | return; |
| 3752 | /* Memory blocks are chained. At first, rewind to the first, then, |
| 3753 | free blocks one by one. */ |
| 3754 | while (cmp_data->prev) |
| 3755 | cmp_data = cmp_data->prev; |
| 3756 | while (cmp_data) |
| 3757 | { |
| 3758 | next = cmp_data->next; |
| 3759 | xfree (cmp_data); |
| 3760 | cmp_data = next; |
| 3761 | } |
| 3762 | coding->cmp_data = NULL; |
| 3763 | } |
| 3764 | |
| 3765 | /* Set `char_offset' member of all memory blocks pointed by |
| 3766 | coding->cmp_data to POS. */ |
| 3767 | |
| 3768 | void |
| 3769 | coding_adjust_composition_offset (coding, pos) |
| 3770 | struct coding_system *coding; |
| 3771 | int pos; |
| 3772 | { |
| 3773 | struct composition_data *cmp_data; |
| 3774 | |
| 3775 | for (cmp_data = coding->cmp_data; cmp_data; cmp_data = cmp_data->next) |
| 3776 | cmp_data->char_offset = pos; |
| 3777 | } |
| 3778 | |
| 3779 | /* Setup raw-text or one of its subsidiaries in the structure |
| 3780 | coding_system CODING according to the already setup value eol_type |
| 3781 | in CODING. CODING should be setup for some coding system in |
| 3782 | advance. */ |
| 3783 | |
| 3784 | void |
| 3785 | setup_raw_text_coding_system (coding) |
| 3786 | struct coding_system *coding; |
| 3787 | { |
| 3788 | if (coding->type != coding_type_raw_text) |
| 3789 | { |
| 3790 | coding->symbol = Qraw_text; |
| 3791 | coding->type = coding_type_raw_text; |
| 3792 | if (coding->eol_type != CODING_EOL_UNDECIDED) |
| 3793 | { |
| 3794 | Lisp_Object subsidiaries; |
| 3795 | subsidiaries = Fget (Qraw_text, Qeol_type); |
| 3796 | |
| 3797 | if (VECTORP (subsidiaries) |
| 3798 | && XVECTOR (subsidiaries)->size == 3) |
| 3799 | coding->symbol |
| 3800 | = XVECTOR (subsidiaries)->contents[coding->eol_type]; |
| 3801 | } |
| 3802 | setup_coding_system (coding->symbol, coding); |
| 3803 | } |
| 3804 | return; |
| 3805 | } |
| 3806 | |
| 3807 | /* Emacs has a mechanism to automatically detect a coding system if it |
| 3808 | is one of Emacs' internal format, ISO2022, SJIS, and BIG5. But, |
| 3809 | it's impossible to distinguish some coding systems accurately |
| 3810 | because they use the same range of codes. So, at first, coding |
| 3811 | systems are categorized into 7, those are: |
| 3812 | |
| 3813 | o coding-category-emacs-mule |
| 3814 | |
| 3815 | The category for a coding system which has the same code range |
| 3816 | as Emacs' internal format. Assigned the coding-system (Lisp |
| 3817 | symbol) `emacs-mule' by default. |
| 3818 | |
| 3819 | o coding-category-sjis |
| 3820 | |
| 3821 | The category for a coding system which has the same code range |
| 3822 | as SJIS. Assigned the coding-system (Lisp |
| 3823 | symbol) `japanese-shift-jis' by default. |
| 3824 | |
| 3825 | o coding-category-iso-7 |
| 3826 | |
| 3827 | The category for a coding system which has the same code range |
| 3828 | as ISO2022 of 7-bit environment. This doesn't use any locking |
| 3829 | shift and single shift functions. This can encode/decode all |
| 3830 | charsets. Assigned the coding-system (Lisp symbol) |
| 3831 | `iso-2022-7bit' by default. |
| 3832 | |
| 3833 | o coding-category-iso-7-tight |
| 3834 | |
| 3835 | Same as coding-category-iso-7 except that this can |
| 3836 | encode/decode only the specified charsets. |
| 3837 | |
| 3838 | o coding-category-iso-8-1 |
| 3839 | |
| 3840 | The category for a coding system which has the same code range |
| 3841 | as ISO2022 of 8-bit environment and graphic plane 1 used only |
| 3842 | for DIMENSION1 charset. This doesn't use any locking shift |
| 3843 | and single shift functions. Assigned the coding-system (Lisp |
| 3844 | symbol) `iso-latin-1' by default. |
| 3845 | |
| 3846 | o coding-category-iso-8-2 |
| 3847 | |
| 3848 | The category for a coding system which has the same code range |
| 3849 | as ISO2022 of 8-bit environment and graphic plane 1 used only |
| 3850 | for DIMENSION2 charset. This doesn't use any locking shift |
| 3851 | and single shift functions. Assigned the coding-system (Lisp |
| 3852 | symbol) `japanese-iso-8bit' by default. |
| 3853 | |
| 3854 | o coding-category-iso-7-else |
| 3855 | |
| 3856 | The category for a coding system which has the same code range |
| 3857 | as ISO2022 of 7-bit environment but uses locking shift or |
| 3858 | single shift functions. Assigned the coding-system (Lisp |
| 3859 | symbol) `iso-2022-7bit-lock' by default. |
| 3860 | |
| 3861 | o coding-category-iso-8-else |
| 3862 | |
| 3863 | The category for a coding system which has the same code range |
| 3864 | as ISO2022 of 8-bit environment but uses locking shift or |
| 3865 | single shift functions. Assigned the coding-system (Lisp |
| 3866 | symbol) `iso-2022-8bit-ss2' by default. |
| 3867 | |
| 3868 | o coding-category-big5 |
| 3869 | |
| 3870 | The category for a coding system which has the same code range |
| 3871 | as BIG5. Assigned the coding-system (Lisp symbol) |
| 3872 | `cn-big5' by default. |
| 3873 | |
| 3874 | o coding-category-utf-8 |
| 3875 | |
| 3876 | The category for a coding system which has the same code range |
| 3877 | as UTF-8 (cf. RFC2279). Assigned the coding-system (Lisp |
| 3878 | symbol) `utf-8' by default. |
| 3879 | |
| 3880 | o coding-category-utf-16-be |
| 3881 | |
| 3882 | The category for a coding system in which a text has an |
| 3883 | Unicode signature (cf. Unicode Standard) in the order of BIG |
| 3884 | endian at the head. Assigned the coding-system (Lisp symbol) |
| 3885 | `utf-16-be' by default. |
| 3886 | |
| 3887 | o coding-category-utf-16-le |
| 3888 | |
| 3889 | The category for a coding system in which a text has an |
| 3890 | Unicode signature (cf. Unicode Standard) in the order of |
| 3891 | LITTLE endian at the head. Assigned the coding-system (Lisp |
| 3892 | symbol) `utf-16-le' by default. |
| 3893 | |
| 3894 | o coding-category-ccl |
| 3895 | |
| 3896 | The category for a coding system of which encoder/decoder is |
| 3897 | written in CCL programs. The default value is nil, i.e., no |
| 3898 | coding system is assigned. |
| 3899 | |
| 3900 | o coding-category-binary |
| 3901 | |
| 3902 | The category for a coding system not categorized in any of the |
| 3903 | above. Assigned the coding-system (Lisp symbol) |
| 3904 | `no-conversion' by default. |
| 3905 | |
| 3906 | Each of them is a Lisp symbol and the value is an actual |
| 3907 | `coding-system' (this is also a Lisp symbol) assigned by a user. |
| 3908 | What Emacs does actually is to detect a category of coding system. |
| 3909 | Then, it uses a `coding-system' assigned to it. If Emacs can't |
| 3910 | decide a single possible category, it selects a category of the |
| 3911 | highest priority. Priorities of categories are also specified by a |
| 3912 | user in a Lisp variable `coding-category-list'. |
| 3913 | |
| 3914 | */ |
| 3915 | |
| 3916 | static |
| 3917 | int ascii_skip_code[256]; |
| 3918 | |
| 3919 | /* Detect how a text of length SRC_BYTES pointed by SOURCE is encoded. |
| 3920 | If it detects possible coding systems, return an integer in which |
| 3921 | appropriate flag bits are set. Flag bits are defined by macros |
| 3922 | CODING_CATEGORY_MASK_XXX in `coding.h'. If PRIORITIES is non-NULL, |
| 3923 | it should point the table `coding_priorities'. In that case, only |
| 3924 | the flag bit for a coding system of the highest priority is set in |
| 3925 | the returned value. If MULTIBYTEP is nonzero, 8-bit codes of the |
| 3926 | range 0x80..0x9F are in multibyte form. |
| 3927 | |
| 3928 | How many ASCII characters are at the head is returned as *SKIP. */ |
| 3929 | |
| 3930 | static int |
| 3931 | detect_coding_mask (source, src_bytes, priorities, skip, multibytep) |
| 3932 | unsigned char *source; |
| 3933 | int src_bytes, *priorities, *skip; |
| 3934 | int multibytep; |
| 3935 | { |
| 3936 | register unsigned char c; |
| 3937 | unsigned char *src = source, *src_end = source + src_bytes; |
| 3938 | unsigned int mask, utf16_examined_p, iso2022_examined_p; |
| 3939 | int i; |
| 3940 | |
| 3941 | /* At first, skip all ASCII characters and control characters except |
| 3942 | for three ISO2022 specific control characters. */ |
| 3943 | ascii_skip_code[ISO_CODE_SO] = 0; |
| 3944 | ascii_skip_code[ISO_CODE_SI] = 0; |
| 3945 | ascii_skip_code[ISO_CODE_ESC] = 0; |
| 3946 | |
| 3947 | label_loop_detect_coding: |
| 3948 | while (src < src_end && ascii_skip_code[*src]) src++; |
| 3949 | *skip = src - source; |
| 3950 | |
| 3951 | if (src >= src_end) |
| 3952 | /* We found nothing other than ASCII. There's nothing to do. */ |
| 3953 | return 0; |
| 3954 | |
| 3955 | c = *src; |
| 3956 | /* The text seems to be encoded in some multilingual coding system. |
| 3957 | Now, try to find in which coding system the text is encoded. */ |
| 3958 | if (c < 0x80) |
| 3959 | { |
| 3960 | /* i.e. (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO) */ |
| 3961 | /* C is an ISO2022 specific control code of C0. */ |
| 3962 | mask = detect_coding_iso2022 (src, src_end, multibytep); |
| 3963 | if (mask == 0) |
| 3964 | { |
| 3965 | /* No valid ISO2022 code follows C. Try again. */ |
| 3966 | src++; |
| 3967 | if (c == ISO_CODE_ESC) |
| 3968 | ascii_skip_code[ISO_CODE_ESC] = 1; |
| 3969 | else |
| 3970 | ascii_skip_code[ISO_CODE_SO] = ascii_skip_code[ISO_CODE_SI] = 1; |
| 3971 | goto label_loop_detect_coding; |
| 3972 | } |
| 3973 | if (priorities) |
| 3974 | { |
| 3975 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) |
| 3976 | { |
| 3977 | if (mask & priorities[i]) |
| 3978 | return priorities[i]; |
| 3979 | } |
| 3980 | return CODING_CATEGORY_MASK_RAW_TEXT; |
| 3981 | } |
| 3982 | } |
| 3983 | else |
| 3984 | { |
| 3985 | int try; |
| 3986 | |
| 3987 | if (multibytep && c == LEADING_CODE_8_BIT_CONTROL) |
| 3988 | c = src[1] - 0x20; |
| 3989 | |
| 3990 | if (c < 0xA0) |
| 3991 | { |
| 3992 | /* C is the first byte of SJIS character code, |
| 3993 | or a leading-code of Emacs' internal format (emacs-mule), |
| 3994 | or the first byte of UTF-16. */ |
| 3995 | try = (CODING_CATEGORY_MASK_SJIS |
| 3996 | | CODING_CATEGORY_MASK_EMACS_MULE |
| 3997 | | CODING_CATEGORY_MASK_UTF_16_BE |
| 3998 | | CODING_CATEGORY_MASK_UTF_16_LE); |
| 3999 | |
| 4000 | /* Or, if C is a special latin extra code, |
| 4001 | or is an ISO2022 specific control code of C1 (SS2 or SS3), |
| 4002 | or is an ISO2022 control-sequence-introducer (CSI), |
| 4003 | we should also consider the possibility of ISO2022 codings. */ |
| 4004 | if ((VECTORP (Vlatin_extra_code_table) |
| 4005 | && !NILP (XVECTOR (Vlatin_extra_code_table)->contents[c])) |
| 4006 | || (c == ISO_CODE_SS2 || c == ISO_CODE_SS3) |
| 4007 | || (c == ISO_CODE_CSI |
| 4008 | && (src < src_end |
| 4009 | && (*src == ']' |
| 4010 | || ((*src == '0' || *src == '1' || *src == '2') |
| 4011 | && src + 1 < src_end |
| 4012 | && src[1] == ']'))))) |
| 4013 | try |= (CODING_CATEGORY_MASK_ISO_8_ELSE |
| 4014 | | CODING_CATEGORY_MASK_ISO_8BIT); |
| 4015 | } |
| 4016 | else |
| 4017 | /* C is a character of ISO2022 in graphic plane right, |
| 4018 | or a SJIS's 1-byte character code (i.e. JISX0201), |
| 4019 | or the first byte of BIG5's 2-byte code, |
| 4020 | or the first byte of UTF-8/16. */ |
| 4021 | try = (CODING_CATEGORY_MASK_ISO_8_ELSE |
| 4022 | | CODING_CATEGORY_MASK_ISO_8BIT |
| 4023 | | CODING_CATEGORY_MASK_SJIS |
| 4024 | | CODING_CATEGORY_MASK_BIG5 |
| 4025 | | CODING_CATEGORY_MASK_UTF_8 |
| 4026 | | CODING_CATEGORY_MASK_UTF_16_BE |
| 4027 | | CODING_CATEGORY_MASK_UTF_16_LE); |
| 4028 | |
| 4029 | /* Or, we may have to consider the possibility of CCL. */ |
| 4030 | if (coding_system_table[CODING_CATEGORY_IDX_CCL] |
| 4031 | && (coding_system_table[CODING_CATEGORY_IDX_CCL] |
| 4032 | ->spec.ccl.valid_codes)[c]) |
| 4033 | try |= CODING_CATEGORY_MASK_CCL; |
| 4034 | |
| 4035 | mask = 0; |
| 4036 | utf16_examined_p = iso2022_examined_p = 0; |
| 4037 | if (priorities) |
| 4038 | { |
| 4039 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) |
| 4040 | { |
| 4041 | if (!iso2022_examined_p |
| 4042 | && (priorities[i] & try & CODING_CATEGORY_MASK_ISO)) |
| 4043 | { |
| 4044 | mask |= detect_coding_iso2022 (src, src_end, multibytep); |
| 4045 | iso2022_examined_p = 1; |
| 4046 | } |
| 4047 | else if (priorities[i] & try & CODING_CATEGORY_MASK_SJIS) |
| 4048 | mask |= detect_coding_sjis (src, src_end, multibytep); |
| 4049 | else if (priorities[i] & try & CODING_CATEGORY_MASK_UTF_8) |
| 4050 | mask |= detect_coding_utf_8 (src, src_end, multibytep); |
| 4051 | else if (!utf16_examined_p |
| 4052 | && (priorities[i] & try & |
| 4053 | CODING_CATEGORY_MASK_UTF_16_BE_LE)) |
| 4054 | { |
| 4055 | mask |= detect_coding_utf_16 (src, src_end, multibytep); |
| 4056 | utf16_examined_p = 1; |
| 4057 | } |
| 4058 | else if (priorities[i] & try & CODING_CATEGORY_MASK_BIG5) |
| 4059 | mask |= detect_coding_big5 (src, src_end, multibytep); |
| 4060 | else if (priorities[i] & try & CODING_CATEGORY_MASK_EMACS_MULE) |
| 4061 | mask |= detect_coding_emacs_mule (src, src_end, multibytep); |
| 4062 | else if (priorities[i] & try & CODING_CATEGORY_MASK_CCL) |
| 4063 | mask |= detect_coding_ccl (src, src_end, multibytep); |
| 4064 | else if (priorities[i] & CODING_CATEGORY_MASK_RAW_TEXT) |
| 4065 | mask |= CODING_CATEGORY_MASK_RAW_TEXT; |
| 4066 | else if (priorities[i] & CODING_CATEGORY_MASK_BINARY) |
| 4067 | mask |= CODING_CATEGORY_MASK_BINARY; |
| 4068 | if (mask & priorities[i]) |
| 4069 | return priorities[i]; |
| 4070 | } |
| 4071 | return CODING_CATEGORY_MASK_RAW_TEXT; |
| 4072 | } |
| 4073 | if (try & CODING_CATEGORY_MASK_ISO) |
| 4074 | mask |= detect_coding_iso2022 (src, src_end, multibytep); |
| 4075 | if (try & CODING_CATEGORY_MASK_SJIS) |
| 4076 | mask |= detect_coding_sjis (src, src_end, multibytep); |
| 4077 | if (try & CODING_CATEGORY_MASK_BIG5) |
| 4078 | mask |= detect_coding_big5 (src, src_end, multibytep); |
| 4079 | if (try & CODING_CATEGORY_MASK_UTF_8) |
| 4080 | mask |= detect_coding_utf_8 (src, src_end, multibytep); |
| 4081 | if (try & CODING_CATEGORY_MASK_UTF_16_BE_LE) |
| 4082 | mask |= detect_coding_utf_16 (src, src_end, multibytep); |
| 4083 | if (try & CODING_CATEGORY_MASK_EMACS_MULE) |
| 4084 | mask |= detect_coding_emacs_mule (src, src_end, multibytep); |
| 4085 | if (try & CODING_CATEGORY_MASK_CCL) |
| 4086 | mask |= detect_coding_ccl (src, src_end, multibytep); |
| 4087 | } |
| 4088 | return (mask | CODING_CATEGORY_MASK_RAW_TEXT | CODING_CATEGORY_MASK_BINARY); |
| 4089 | } |
| 4090 | |
| 4091 | /* Detect how a text of length SRC_BYTES pointed by SRC is encoded. |
| 4092 | The information of the detected coding system is set in CODING. */ |
| 4093 | |
| 4094 | void |
| 4095 | detect_coding (coding, src, src_bytes) |
| 4096 | struct coding_system *coding; |
| 4097 | unsigned char *src; |
| 4098 | int src_bytes; |
| 4099 | { |
| 4100 | unsigned int idx; |
| 4101 | int skip, mask; |
| 4102 | Lisp_Object val; |
| 4103 | |
| 4104 | val = Vcoding_category_list; |
| 4105 | mask = detect_coding_mask (src, src_bytes, coding_priorities, &skip, |
| 4106 | coding->src_multibyte); |
| 4107 | coding->heading_ascii = skip; |
| 4108 | |
| 4109 | if (!mask) return; |
| 4110 | |
| 4111 | /* We found a single coding system of the highest priority in MASK. */ |
| 4112 | idx = 0; |
| 4113 | while (mask && ! (mask & 1)) mask >>= 1, idx++; |
| 4114 | if (! mask) |
| 4115 | idx = CODING_CATEGORY_IDX_RAW_TEXT; |
| 4116 | |
| 4117 | val = SYMBOL_VALUE (XVECTOR (Vcoding_category_table)->contents[idx]); |
| 4118 | |
| 4119 | if (coding->eol_type != CODING_EOL_UNDECIDED) |
| 4120 | { |
| 4121 | Lisp_Object tmp; |
| 4122 | |
| 4123 | tmp = Fget (val, Qeol_type); |
| 4124 | if (VECTORP (tmp)) |
| 4125 | val = XVECTOR (tmp)->contents[coding->eol_type]; |
| 4126 | } |
| 4127 | |
| 4128 | /* Setup this new coding system while preserving some slots. */ |
| 4129 | { |
| 4130 | int src_multibyte = coding->src_multibyte; |
| 4131 | int dst_multibyte = coding->dst_multibyte; |
| 4132 | |
| 4133 | setup_coding_system (val, coding); |
| 4134 | coding->src_multibyte = src_multibyte; |
| 4135 | coding->dst_multibyte = dst_multibyte; |
| 4136 | coding->heading_ascii = skip; |
| 4137 | } |
| 4138 | } |
| 4139 | |
| 4140 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by |
| 4141 | SOURCE is encoded. Return one of CODING_EOL_LF, CODING_EOL_CRLF, |
| 4142 | CODING_EOL_CR, and CODING_EOL_UNDECIDED. |
| 4143 | |
| 4144 | How many non-eol characters are at the head is returned as *SKIP. */ |
| 4145 | |
| 4146 | #define MAX_EOL_CHECK_COUNT 3 |
| 4147 | |
| 4148 | static int |
| 4149 | detect_eol_type (source, src_bytes, skip) |
| 4150 | unsigned char *source; |
| 4151 | int src_bytes, *skip; |
| 4152 | { |
| 4153 | unsigned char *src = source, *src_end = src + src_bytes; |
| 4154 | unsigned char c; |
| 4155 | int total = 0; /* How many end-of-lines are found so far. */ |
| 4156 | int eol_type = CODING_EOL_UNDECIDED; |
| 4157 | int this_eol_type; |
| 4158 | |
| 4159 | *skip = 0; |
| 4160 | |
| 4161 | while (src < src_end && total < MAX_EOL_CHECK_COUNT) |
| 4162 | { |
| 4163 | c = *src++; |
| 4164 | if (c == '\n' || c == '\r') |
| 4165 | { |
| 4166 | if (*skip == 0) |
| 4167 | *skip = src - 1 - source; |
| 4168 | total++; |
| 4169 | if (c == '\n') |
| 4170 | this_eol_type = CODING_EOL_LF; |
| 4171 | else if (src >= src_end || *src != '\n') |
| 4172 | this_eol_type = CODING_EOL_CR; |
| 4173 | else |
| 4174 | this_eol_type = CODING_EOL_CRLF, src++; |
| 4175 | |
| 4176 | if (eol_type == CODING_EOL_UNDECIDED) |
| 4177 | /* This is the first end-of-line. */ |
| 4178 | eol_type = this_eol_type; |
| 4179 | else if (eol_type != this_eol_type) |
| 4180 | { |
| 4181 | /* The found type is different from what found before. */ |
| 4182 | eol_type = CODING_EOL_INCONSISTENT; |
| 4183 | break; |
| 4184 | } |
| 4185 | } |
| 4186 | } |
| 4187 | |
| 4188 | if (*skip == 0) |
| 4189 | *skip = src_end - source; |
| 4190 | return eol_type; |
| 4191 | } |
| 4192 | |
| 4193 | /* Like detect_eol_type, but detect EOL type in 2-octet |
| 4194 | big-endian/little-endian format for coding systems utf-16-be and |
| 4195 | utf-16-le. */ |
| 4196 | |
| 4197 | static int |
| 4198 | detect_eol_type_in_2_octet_form (source, src_bytes, skip, big_endian_p) |
| 4199 | unsigned char *source; |
| 4200 | int src_bytes, *skip, big_endian_p; |
| 4201 | { |
| 4202 | unsigned char *src = source, *src_end = src + src_bytes; |
| 4203 | unsigned int c1, c2; |
| 4204 | int total = 0; /* How many end-of-lines are found so far. */ |
| 4205 | int eol_type = CODING_EOL_UNDECIDED; |
| 4206 | int this_eol_type; |
| 4207 | int msb, lsb; |
| 4208 | |
| 4209 | if (big_endian_p) |
| 4210 | msb = 0, lsb = 1; |
| 4211 | else |
| 4212 | msb = 1, lsb = 0; |
| 4213 | |
| 4214 | *skip = 0; |
| 4215 | |
| 4216 | while ((src + 1) < src_end && total < MAX_EOL_CHECK_COUNT) |
| 4217 | { |
| 4218 | c1 = (src[msb] << 8) | (src[lsb]); |
| 4219 | src += 2; |
| 4220 | |
| 4221 | if (c1 == '\n' || c1 == '\r') |
| 4222 | { |
| 4223 | if (*skip == 0) |
| 4224 | *skip = src - 2 - source; |
| 4225 | total++; |
| 4226 | if (c1 == '\n') |
| 4227 | { |
| 4228 | this_eol_type = CODING_EOL_LF; |
| 4229 | } |
| 4230 | else |
| 4231 | { |
| 4232 | if ((src + 1) >= src_end) |
| 4233 | { |
| 4234 | this_eol_type = CODING_EOL_CR; |
| 4235 | } |
| 4236 | else |
| 4237 | { |
| 4238 | c2 = (src[msb] << 8) | (src[lsb]); |
| 4239 | if (c2 == '\n') |
| 4240 | this_eol_type = CODING_EOL_CRLF, src += 2; |
| 4241 | else |
| 4242 | this_eol_type = CODING_EOL_CR; |
| 4243 | } |
| 4244 | } |
| 4245 | |
| 4246 | if (eol_type == CODING_EOL_UNDECIDED) |
| 4247 | /* This is the first end-of-line. */ |
| 4248 | eol_type = this_eol_type; |
| 4249 | else if (eol_type != this_eol_type) |
| 4250 | { |
| 4251 | /* The found type is different from what found before. */ |
| 4252 | eol_type = CODING_EOL_INCONSISTENT; |
| 4253 | break; |
| 4254 | } |
| 4255 | } |
| 4256 | } |
| 4257 | |
| 4258 | if (*skip == 0) |
| 4259 | *skip = src_end - source; |
| 4260 | return eol_type; |
| 4261 | } |
| 4262 | |
| 4263 | /* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC |
| 4264 | is encoded. If it detects an appropriate format of end-of-line, it |
| 4265 | sets the information in *CODING. */ |
| 4266 | |
| 4267 | void |
| 4268 | detect_eol (coding, src, src_bytes) |
| 4269 | struct coding_system *coding; |
| 4270 | unsigned char *src; |
| 4271 | int src_bytes; |
| 4272 | { |
| 4273 | Lisp_Object val; |
| 4274 | int skip; |
| 4275 | int eol_type; |
| 4276 | |
| 4277 | switch (coding->category_idx) |
| 4278 | { |
| 4279 | case CODING_CATEGORY_IDX_UTF_16_BE: |
| 4280 | eol_type = detect_eol_type_in_2_octet_form (src, src_bytes, &skip, 1); |
| 4281 | break; |
| 4282 | case CODING_CATEGORY_IDX_UTF_16_LE: |
| 4283 | eol_type = detect_eol_type_in_2_octet_form (src, src_bytes, &skip, 0); |
| 4284 | break; |
| 4285 | default: |
| 4286 | eol_type = detect_eol_type (src, src_bytes, &skip); |
| 4287 | break; |
| 4288 | } |
| 4289 | |
| 4290 | if (coding->heading_ascii > skip) |
| 4291 | coding->heading_ascii = skip; |
| 4292 | else |
| 4293 | skip = coding->heading_ascii; |
| 4294 | |
| 4295 | if (eol_type == CODING_EOL_UNDECIDED) |
| 4296 | return; |
| 4297 | if (eol_type == CODING_EOL_INCONSISTENT) |
| 4298 | { |
| 4299 | #if 0 |
| 4300 | /* This code is suppressed until we find a better way to |
| 4301 | distinguish raw text file and binary file. */ |
| 4302 | |
| 4303 | /* If we have already detected that the coding is raw-text, the |
| 4304 | coding should actually be no-conversion. */ |
| 4305 | if (coding->type == coding_type_raw_text) |
| 4306 | { |
| 4307 | setup_coding_system (Qno_conversion, coding); |
| 4308 | return; |
| 4309 | } |
| 4310 | /* Else, let's decode only text code anyway. */ |
| 4311 | #endif /* 0 */ |
| 4312 | eol_type = CODING_EOL_LF; |
| 4313 | } |
| 4314 | |
| 4315 | val = Fget (coding->symbol, Qeol_type); |
| 4316 | if (VECTORP (val) && XVECTOR (val)->size == 3) |
| 4317 | { |
| 4318 | int src_multibyte = coding->src_multibyte; |
| 4319 | int dst_multibyte = coding->dst_multibyte; |
| 4320 | struct composition_data *cmp_data = coding->cmp_data; |
| 4321 | |
| 4322 | setup_coding_system (XVECTOR (val)->contents[eol_type], coding); |
| 4323 | coding->src_multibyte = src_multibyte; |
| 4324 | coding->dst_multibyte = dst_multibyte; |
| 4325 | coding->heading_ascii = skip; |
| 4326 | coding->cmp_data = cmp_data; |
| 4327 | } |
| 4328 | } |
| 4329 | |
| 4330 | #define CONVERSION_BUFFER_EXTRA_ROOM 256 |
| 4331 | |
| 4332 | #define DECODING_BUFFER_MAG(coding) \ |
| 4333 | (coding->type == coding_type_iso2022 \ |
| 4334 | ? 3 \ |
| 4335 | : (coding->type == coding_type_ccl \ |
| 4336 | ? coding->spec.ccl.decoder.buf_magnification \ |
| 4337 | : 2)) |
| 4338 | |
| 4339 | /* Return maximum size (bytes) of a buffer enough for decoding |
| 4340 | SRC_BYTES of text encoded in CODING. */ |
| 4341 | |
| 4342 | int |
| 4343 | decoding_buffer_size (coding, src_bytes) |
| 4344 | struct coding_system *coding; |
| 4345 | int src_bytes; |
| 4346 | { |
| 4347 | return (src_bytes * DECODING_BUFFER_MAG (coding) |
| 4348 | + CONVERSION_BUFFER_EXTRA_ROOM); |
| 4349 | } |
| 4350 | |
| 4351 | /* Return maximum size (bytes) of a buffer enough for encoding |
| 4352 | SRC_BYTES of text to CODING. */ |
| 4353 | |
| 4354 | int |
| 4355 | encoding_buffer_size (coding, src_bytes) |
| 4356 | struct coding_system *coding; |
| 4357 | int src_bytes; |
| 4358 | { |
| 4359 | int magnification; |
| 4360 | |
| 4361 | if (coding->type == coding_type_ccl) |
| 4362 | magnification = coding->spec.ccl.encoder.buf_magnification; |
| 4363 | else if (CODING_REQUIRE_ENCODING (coding)) |
| 4364 | magnification = 3; |
| 4365 | else |
| 4366 | magnification = 1; |
| 4367 | |
| 4368 | return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM); |
| 4369 | } |
| 4370 | |
| 4371 | /* Working buffer for code conversion. */ |
| 4372 | struct conversion_buffer |
| 4373 | { |
| 4374 | int size; /* size of data. */ |
| 4375 | int on_stack; /* 1 if allocated by alloca. */ |
| 4376 | unsigned char *data; |
| 4377 | }; |
| 4378 | |
| 4379 | /* Don't use alloca for allocating memory space larger than this, lest |
| 4380 | we overflow their stack. */ |
| 4381 | #define MAX_ALLOCA 16*1024 |
| 4382 | |
| 4383 | /* Allocate LEN bytes of memory for BUF (struct conversion_buffer). */ |
| 4384 | #define allocate_conversion_buffer(buf, len) \ |
| 4385 | do { \ |
| 4386 | if (len < MAX_ALLOCA) \ |
| 4387 | { \ |
| 4388 | buf.data = (unsigned char *) alloca (len); \ |
| 4389 | buf.on_stack = 1; \ |
| 4390 | } \ |
| 4391 | else \ |
| 4392 | { \ |
| 4393 | buf.data = (unsigned char *) xmalloc (len); \ |
| 4394 | buf.on_stack = 0; \ |
| 4395 | } \ |
| 4396 | buf.size = len; \ |
| 4397 | } while (0) |
| 4398 | |
| 4399 | /* Double the allocated memory for *BUF. */ |
| 4400 | static void |
| 4401 | extend_conversion_buffer (buf) |
| 4402 | struct conversion_buffer *buf; |
| 4403 | { |
| 4404 | if (buf->on_stack) |
| 4405 | { |
| 4406 | unsigned char *save = buf->data; |
| 4407 | buf->data = (unsigned char *) xmalloc (buf->size * 2); |
| 4408 | bcopy (save, buf->data, buf->size); |
| 4409 | buf->on_stack = 0; |
| 4410 | } |
| 4411 | else |
| 4412 | { |
| 4413 | buf->data = (unsigned char *) xrealloc (buf->data, buf->size * 2); |
| 4414 | } |
| 4415 | buf->size *= 2; |
| 4416 | } |
| 4417 | |
| 4418 | /* Free the allocated memory for BUF if it is not on stack. */ |
| 4419 | static void |
| 4420 | free_conversion_buffer (buf) |
| 4421 | struct conversion_buffer *buf; |
| 4422 | { |
| 4423 | if (!buf->on_stack) |
| 4424 | xfree (buf->data); |
| 4425 | } |
| 4426 | |
| 4427 | int |
| 4428 | ccl_coding_driver (coding, source, destination, src_bytes, dst_bytes, encodep) |
| 4429 | struct coding_system *coding; |
| 4430 | unsigned char *source, *destination; |
| 4431 | int src_bytes, dst_bytes, encodep; |
| 4432 | { |
| 4433 | struct ccl_program *ccl |
| 4434 | = encodep ? &coding->spec.ccl.encoder : &coding->spec.ccl.decoder; |
| 4435 | unsigned char *dst = destination; |
| 4436 | |
| 4437 | ccl->suppress_error = coding->suppress_error; |
| 4438 | ccl->last_block = coding->mode & CODING_MODE_LAST_BLOCK; |
| 4439 | if (encodep) |
| 4440 | { |
| 4441 | /* On encoding, EOL format is converted within ccl_driver. For |
| 4442 | that, setup proper information in the structure CCL. */ |
| 4443 | ccl->eol_type = coding->eol_type; |
| 4444 | if (ccl->eol_type ==CODING_EOL_UNDECIDED) |
| 4445 | ccl->eol_type = CODING_EOL_LF; |
| 4446 | ccl->cr_consumed = coding->spec.ccl.cr_carryover; |
| 4447 | } |
| 4448 | ccl->multibyte = coding->src_multibyte; |
| 4449 | if (coding->spec.ccl.eight_bit_carryover[0] != 0) |
| 4450 | { |
| 4451 | /* Move carryover bytes to DESTINATION. */ |
| 4452 | unsigned char *p = coding->spec.ccl.eight_bit_carryover; |
| 4453 | while (*p) |
| 4454 | *dst++ = *p++; |
| 4455 | coding->spec.ccl.eight_bit_carryover[0] = 0; |
| 4456 | if (dst_bytes) |
| 4457 | dst_bytes -= dst - destination; |
| 4458 | } |
| 4459 | |
| 4460 | coding->produced = (ccl_driver (ccl, source, dst, src_bytes, dst_bytes, |
| 4461 | &(coding->consumed)) |
| 4462 | + dst - destination); |
| 4463 | |
| 4464 | if (encodep) |
| 4465 | { |
| 4466 | coding->produced_char = coding->produced; |
| 4467 | coding->spec.ccl.cr_carryover = ccl->cr_consumed; |
| 4468 | } |
| 4469 | else if (!ccl->eight_bit_control) |
| 4470 | { |
| 4471 | /* The produced bytes forms a valid multibyte sequence. */ |
| 4472 | coding->produced_char |
| 4473 | = multibyte_chars_in_text (destination, coding->produced); |
| 4474 | coding->spec.ccl.eight_bit_carryover[0] = 0; |
| 4475 | } |
| 4476 | else |
| 4477 | { |
| 4478 | /* On decoding, the destination should always multibyte. But, |
| 4479 | CCL program might have been generated an invalid multibyte |
| 4480 | sequence. Here we make such a sequence valid as |
| 4481 | multibyte. */ |
| 4482 | int bytes |
| 4483 | = dst_bytes ? dst_bytes : source + coding->consumed - destination; |
| 4484 | |
| 4485 | if ((coding->consumed < src_bytes |
| 4486 | || !ccl->last_block) |
| 4487 | && coding->produced >= 1 |
| 4488 | && destination[coding->produced - 1] >= 0x80) |
| 4489 | { |
| 4490 | /* We should not convert the tailing 8-bit codes to |
| 4491 | multibyte form even if they doesn't form a valid |
| 4492 | multibyte sequence. They may form a valid sequence in |
| 4493 | the next call. */ |
| 4494 | int carryover = 0; |
| 4495 | |
| 4496 | if (destination[coding->produced - 1] < 0xA0) |
| 4497 | carryover = 1; |
| 4498 | else if (coding->produced >= 2) |
| 4499 | { |
| 4500 | if (destination[coding->produced - 2] >= 0x80) |
| 4501 | { |
| 4502 | if (destination[coding->produced - 2] < 0xA0) |
| 4503 | carryover = 2; |
| 4504 | else if (coding->produced >= 3 |
| 4505 | && destination[coding->produced - 3] >= 0x80 |
| 4506 | && destination[coding->produced - 3] < 0xA0) |
| 4507 | carryover = 3; |
| 4508 | } |
| 4509 | } |
| 4510 | if (carryover > 0) |
| 4511 | { |
| 4512 | BCOPY_SHORT (destination + coding->produced - carryover, |
| 4513 | coding->spec.ccl.eight_bit_carryover, |
| 4514 | carryover); |
| 4515 | coding->spec.ccl.eight_bit_carryover[carryover] = 0; |
| 4516 | coding->produced -= carryover; |
| 4517 | } |
| 4518 | } |
| 4519 | coding->produced = str_as_multibyte (destination, bytes, |
| 4520 | coding->produced, |
| 4521 | &(coding->produced_char)); |
| 4522 | } |
| 4523 | |
| 4524 | switch (ccl->status) |
| 4525 | { |
| 4526 | case CCL_STAT_SUSPEND_BY_SRC: |
| 4527 | coding->result = CODING_FINISH_INSUFFICIENT_SRC; |
| 4528 | break; |
| 4529 | case CCL_STAT_SUSPEND_BY_DST: |
| 4530 | coding->result = CODING_FINISH_INSUFFICIENT_DST; |
| 4531 | break; |
| 4532 | case CCL_STAT_QUIT: |
| 4533 | case CCL_STAT_INVALID_CMD: |
| 4534 | coding->result = CODING_FINISH_INTERRUPT; |
| 4535 | break; |
| 4536 | default: |
| 4537 | coding->result = CODING_FINISH_NORMAL; |
| 4538 | break; |
| 4539 | } |
| 4540 | return coding->result; |
| 4541 | } |
| 4542 | |
| 4543 | /* Decode EOL format of the text at PTR of BYTES length destructively |
| 4544 | according to CODING->eol_type. This is called after the CCL |
| 4545 | program produced a decoded text at PTR. If we do CRLF->LF |
| 4546 | conversion, update CODING->produced and CODING->produced_char. */ |
| 4547 | |
| 4548 | static void |
| 4549 | decode_eol_post_ccl (coding, ptr, bytes) |
| 4550 | struct coding_system *coding; |
| 4551 | unsigned char *ptr; |
| 4552 | int bytes; |
| 4553 | { |
| 4554 | Lisp_Object val, saved_coding_symbol; |
| 4555 | unsigned char *pend = ptr + bytes; |
| 4556 | int dummy; |
| 4557 | |
| 4558 | /* Remember the current coding system symbol. We set it back when |
| 4559 | an inconsistent EOL is found so that `last-coding-system-used' is |
| 4560 | set to the coding system that doesn't specify EOL conversion. */ |
| 4561 | saved_coding_symbol = coding->symbol; |
| 4562 | |
| 4563 | coding->spec.ccl.cr_carryover = 0; |
| 4564 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 4565 | { |
| 4566 | /* Here, to avoid the call of setup_coding_system, we directly |
| 4567 | call detect_eol_type. */ |
| 4568 | coding->eol_type = detect_eol_type (ptr, bytes, &dummy); |
| 4569 | if (coding->eol_type == CODING_EOL_INCONSISTENT) |
| 4570 | coding->eol_type = CODING_EOL_LF; |
| 4571 | if (coding->eol_type != CODING_EOL_UNDECIDED) |
| 4572 | { |
| 4573 | val = Fget (coding->symbol, Qeol_type); |
| 4574 | if (VECTORP (val) && XVECTOR (val)->size == 3) |
| 4575 | coding->symbol = XVECTOR (val)->contents[coding->eol_type]; |
| 4576 | } |
| 4577 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; |
| 4578 | } |
| 4579 | |
| 4580 | if (coding->eol_type == CODING_EOL_LF |
| 4581 | || coding->eol_type == CODING_EOL_UNDECIDED) |
| 4582 | { |
| 4583 | /* We have nothing to do. */ |
| 4584 | ptr = pend; |
| 4585 | } |
| 4586 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 4587 | { |
| 4588 | unsigned char *pstart = ptr, *p = ptr; |
| 4589 | |
| 4590 | if (! (coding->mode & CODING_MODE_LAST_BLOCK) |
| 4591 | && *(pend - 1) == '\r') |
| 4592 | { |
| 4593 | /* If the last character is CR, we can't handle it here |
| 4594 | because LF will be in the not-yet-decoded source text. |
| 4595 | Record that the CR is not yet processed. */ |
| 4596 | coding->spec.ccl.cr_carryover = 1; |
| 4597 | coding->produced--; |
| 4598 | coding->produced_char--; |
| 4599 | pend--; |
| 4600 | } |
| 4601 | while (ptr < pend) |
| 4602 | { |
| 4603 | if (*ptr == '\r') |
| 4604 | { |
| 4605 | if (ptr + 1 < pend && *(ptr + 1) == '\n') |
| 4606 | { |
| 4607 | *p++ = '\n'; |
| 4608 | ptr += 2; |
| 4609 | } |
| 4610 | else |
| 4611 | { |
| 4612 | if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 4613 | goto undo_eol_conversion; |
| 4614 | *p++ = *ptr++; |
| 4615 | } |
| 4616 | } |
| 4617 | else if (*ptr == '\n' |
| 4618 | && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 4619 | goto undo_eol_conversion; |
| 4620 | else |
| 4621 | *p++ = *ptr++; |
| 4622 | continue; |
| 4623 | |
| 4624 | undo_eol_conversion: |
| 4625 | /* We have faced with inconsistent EOL format at PTR. |
| 4626 | Convert all LFs before PTR back to CRLFs. */ |
| 4627 | for (p--, ptr--; p >= pstart; p--) |
| 4628 | { |
| 4629 | if (*p == '\n') |
| 4630 | *ptr-- = '\n', *ptr-- = '\r'; |
| 4631 | else |
| 4632 | *ptr-- = *p; |
| 4633 | } |
| 4634 | /* If carryover is recorded, cancel it because we don't |
| 4635 | convert CRLF anymore. */ |
| 4636 | if (coding->spec.ccl.cr_carryover) |
| 4637 | { |
| 4638 | coding->spec.ccl.cr_carryover = 0; |
| 4639 | coding->produced++; |
| 4640 | coding->produced_char++; |
| 4641 | pend++; |
| 4642 | } |
| 4643 | p = ptr = pend; |
| 4644 | coding->eol_type = CODING_EOL_LF; |
| 4645 | coding->symbol = saved_coding_symbol; |
| 4646 | } |
| 4647 | if (p < pend) |
| 4648 | { |
| 4649 | /* As each two-byte sequence CRLF was converted to LF, (PEND |
| 4650 | - P) is the number of deleted characters. */ |
| 4651 | coding->produced -= pend - p; |
| 4652 | coding->produced_char -= pend - p; |
| 4653 | } |
| 4654 | } |
| 4655 | else /* i.e. coding->eol_type == CODING_EOL_CR */ |
| 4656 | { |
| 4657 | unsigned char *p = ptr; |
| 4658 | |
| 4659 | for (; ptr < pend; ptr++) |
| 4660 | { |
| 4661 | if (*ptr == '\r') |
| 4662 | *ptr = '\n'; |
| 4663 | else if (*ptr == '\n' |
| 4664 | && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL) |
| 4665 | { |
| 4666 | for (; p < ptr; p++) |
| 4667 | { |
| 4668 | if (*p == '\n') |
| 4669 | *p = '\r'; |
| 4670 | } |
| 4671 | ptr = pend; |
| 4672 | coding->eol_type = CODING_EOL_LF; |
| 4673 | coding->symbol = saved_coding_symbol; |
| 4674 | } |
| 4675 | } |
| 4676 | } |
| 4677 | } |
| 4678 | |
| 4679 | /* See "GENERAL NOTES about `decode_coding_XXX ()' functions". Before |
| 4680 | decoding, it may detect coding system and format of end-of-line if |
| 4681 | those are not yet decided. The source should be unibyte, the |
| 4682 | result is multibyte if CODING->dst_multibyte is nonzero, else |
| 4683 | unibyte. */ |
| 4684 | |
| 4685 | int |
| 4686 | decode_coding (coding, source, destination, src_bytes, dst_bytes) |
| 4687 | struct coding_system *coding; |
| 4688 | unsigned char *source, *destination; |
| 4689 | int src_bytes, dst_bytes; |
| 4690 | { |
| 4691 | int extra = 0; |
| 4692 | |
| 4693 | if (coding->type == coding_type_undecided) |
| 4694 | detect_coding (coding, source, src_bytes); |
| 4695 | |
| 4696 | if (coding->eol_type == CODING_EOL_UNDECIDED |
| 4697 | && coding->type != coding_type_ccl) |
| 4698 | { |
| 4699 | detect_eol (coding, source, src_bytes); |
| 4700 | /* We had better recover the original eol format if we |
| 4701 | encounter an inconsistent eol format while decoding. */ |
| 4702 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; |
| 4703 | } |
| 4704 | |
| 4705 | coding->produced = coding->produced_char = 0; |
| 4706 | coding->consumed = coding->consumed_char = 0; |
| 4707 | coding->errors = 0; |
| 4708 | coding->result = CODING_FINISH_NORMAL; |
| 4709 | |
| 4710 | switch (coding->type) |
| 4711 | { |
| 4712 | case coding_type_sjis: |
| 4713 | decode_coding_sjis_big5 (coding, source, destination, |
| 4714 | src_bytes, dst_bytes, 1); |
| 4715 | break; |
| 4716 | |
| 4717 | case coding_type_iso2022: |
| 4718 | decode_coding_iso2022 (coding, source, destination, |
| 4719 | src_bytes, dst_bytes); |
| 4720 | break; |
| 4721 | |
| 4722 | case coding_type_big5: |
| 4723 | decode_coding_sjis_big5 (coding, source, destination, |
| 4724 | src_bytes, dst_bytes, 0); |
| 4725 | break; |
| 4726 | |
| 4727 | case coding_type_emacs_mule: |
| 4728 | decode_coding_emacs_mule (coding, source, destination, |
| 4729 | src_bytes, dst_bytes); |
| 4730 | break; |
| 4731 | |
| 4732 | case coding_type_ccl: |
| 4733 | if (coding->spec.ccl.cr_carryover) |
| 4734 | { |
| 4735 | /* Put the CR which was not processed by the previous call |
| 4736 | of decode_eol_post_ccl in DESTINATION. It will be |
| 4737 | decoded together with the following LF by the call to |
| 4738 | decode_eol_post_ccl below. */ |
| 4739 | *destination = '\r'; |
| 4740 | coding->produced++; |
| 4741 | coding->produced_char++; |
| 4742 | dst_bytes--; |
| 4743 | extra = coding->spec.ccl.cr_carryover; |
| 4744 | } |
| 4745 | ccl_coding_driver (coding, source, destination + extra, |
| 4746 | src_bytes, dst_bytes, 0); |
| 4747 | if (coding->eol_type != CODING_EOL_LF) |
| 4748 | { |
| 4749 | coding->produced += extra; |
| 4750 | coding->produced_char += extra; |
| 4751 | decode_eol_post_ccl (coding, destination, coding->produced); |
| 4752 | } |
| 4753 | break; |
| 4754 | |
| 4755 | default: |
| 4756 | decode_eol (coding, source, destination, src_bytes, dst_bytes); |
| 4757 | } |
| 4758 | |
| 4759 | if (coding->result == CODING_FINISH_INSUFFICIENT_SRC |
| 4760 | && coding->mode & CODING_MODE_LAST_BLOCK |
| 4761 | && coding->consumed == src_bytes) |
| 4762 | coding->result = CODING_FINISH_NORMAL; |
| 4763 | |
| 4764 | if (coding->mode & CODING_MODE_LAST_BLOCK |
| 4765 | && coding->result == CODING_FINISH_INSUFFICIENT_SRC) |
| 4766 | { |
| 4767 | unsigned char *src = source + coding->consumed; |
| 4768 | unsigned char *dst = destination + coding->produced; |
| 4769 | |
| 4770 | src_bytes -= coding->consumed; |
| 4771 | coding->errors++; |
| 4772 | if (COMPOSING_P (coding)) |
| 4773 | DECODE_COMPOSITION_END ('1'); |
| 4774 | while (src_bytes--) |
| 4775 | { |
| 4776 | int c = *src++; |
| 4777 | dst += CHAR_STRING (c, dst); |
| 4778 | coding->produced_char++; |
| 4779 | } |
| 4780 | coding->consumed = coding->consumed_char = src - source; |
| 4781 | coding->produced = dst - destination; |
| 4782 | coding->result = CODING_FINISH_NORMAL; |
| 4783 | } |
| 4784 | |
| 4785 | if (!coding->dst_multibyte) |
| 4786 | { |
| 4787 | coding->produced = str_as_unibyte (destination, coding->produced); |
| 4788 | coding->produced_char = coding->produced; |
| 4789 | } |
| 4790 | |
| 4791 | return coding->result; |
| 4792 | } |
| 4793 | |
| 4794 | /* See "GENERAL NOTES about `encode_coding_XXX ()' functions". The |
| 4795 | multibyteness of the source is CODING->src_multibyte, the |
| 4796 | multibyteness of the result is always unibyte. */ |
| 4797 | |
| 4798 | int |
| 4799 | encode_coding (coding, source, destination, src_bytes, dst_bytes) |
| 4800 | struct coding_system *coding; |
| 4801 | unsigned char *source, *destination; |
| 4802 | int src_bytes, dst_bytes; |
| 4803 | { |
| 4804 | coding->produced = coding->produced_char = 0; |
| 4805 | coding->consumed = coding->consumed_char = 0; |
| 4806 | coding->errors = 0; |
| 4807 | coding->result = CODING_FINISH_NORMAL; |
| 4808 | |
| 4809 | switch (coding->type) |
| 4810 | { |
| 4811 | case coding_type_sjis: |
| 4812 | encode_coding_sjis_big5 (coding, source, destination, |
| 4813 | src_bytes, dst_bytes, 1); |
| 4814 | break; |
| 4815 | |
| 4816 | case coding_type_iso2022: |
| 4817 | encode_coding_iso2022 (coding, source, destination, |
| 4818 | src_bytes, dst_bytes); |
| 4819 | break; |
| 4820 | |
| 4821 | case coding_type_big5: |
| 4822 | encode_coding_sjis_big5 (coding, source, destination, |
| 4823 | src_bytes, dst_bytes, 0); |
| 4824 | break; |
| 4825 | |
| 4826 | case coding_type_emacs_mule: |
| 4827 | encode_coding_emacs_mule (coding, source, destination, |
| 4828 | src_bytes, dst_bytes); |
| 4829 | break; |
| 4830 | |
| 4831 | case coding_type_ccl: |
| 4832 | ccl_coding_driver (coding, source, destination, |
| 4833 | src_bytes, dst_bytes, 1); |
| 4834 | break; |
| 4835 | |
| 4836 | default: |
| 4837 | encode_eol (coding, source, destination, src_bytes, dst_bytes); |
| 4838 | } |
| 4839 | |
| 4840 | if (coding->mode & CODING_MODE_LAST_BLOCK |
| 4841 | && coding->result == CODING_FINISH_INSUFFICIENT_SRC) |
| 4842 | { |
| 4843 | unsigned char *src = source + coding->consumed; |
| 4844 | unsigned char *dst = destination + coding->produced; |
| 4845 | |
| 4846 | if (coding->type == coding_type_iso2022) |
| 4847 | ENCODE_RESET_PLANE_AND_REGISTER; |
| 4848 | if (COMPOSING_P (coding)) |
| 4849 | *dst++ = ISO_CODE_ESC, *dst++ = '1'; |
| 4850 | if (coding->consumed < src_bytes) |
| 4851 | { |
| 4852 | int len = src_bytes - coding->consumed; |
| 4853 | |
| 4854 | BCOPY_SHORT (src, dst, len); |
| 4855 | if (coding->src_multibyte) |
| 4856 | len = str_as_unibyte (dst, len); |
| 4857 | dst += len; |
| 4858 | coding->consumed = src_bytes; |
| 4859 | } |
| 4860 | coding->produced = coding->produced_char = dst - destination; |
| 4861 | coding->result = CODING_FINISH_NORMAL; |
| 4862 | } |
| 4863 | |
| 4864 | if (coding->result == CODING_FINISH_INSUFFICIENT_SRC |
| 4865 | && coding->consumed == src_bytes) |
| 4866 | coding->result = CODING_FINISH_NORMAL; |
| 4867 | |
| 4868 | return coding->result; |
| 4869 | } |
| 4870 | |
| 4871 | /* Scan text in the region between *BEG and *END (byte positions), |
| 4872 | skip characters which we don't have to decode by coding system |
| 4873 | CODING at the head and tail, then set *BEG and *END to the region |
| 4874 | of the text we actually have to convert. The caller should move |
| 4875 | the gap out of the region in advance if the region is from a |
| 4876 | buffer. |
| 4877 | |
| 4878 | If STR is not NULL, *BEG and *END are indices into STR. */ |
| 4879 | |
| 4880 | static void |
| 4881 | shrink_decoding_region (beg, end, coding, str) |
| 4882 | int *beg, *end; |
| 4883 | struct coding_system *coding; |
| 4884 | unsigned char *str; |
| 4885 | { |
| 4886 | unsigned char *begp_orig, *begp, *endp_orig, *endp, c; |
| 4887 | int eol_conversion; |
| 4888 | Lisp_Object translation_table; |
| 4889 | |
| 4890 | if (coding->type == coding_type_ccl |
| 4891 | || coding->type == coding_type_undecided |
| 4892 | || coding->eol_type != CODING_EOL_LF |
| 4893 | || !NILP (coding->post_read_conversion) |
| 4894 | || coding->composing != COMPOSITION_DISABLED) |
| 4895 | { |
| 4896 | /* We can't skip any data. */ |
| 4897 | return; |
| 4898 | } |
| 4899 | if (coding->type == coding_type_no_conversion |
| 4900 | || coding->type == coding_type_raw_text |
| 4901 | || coding->type == coding_type_emacs_mule) |
| 4902 | { |
| 4903 | /* We need no conversion, but don't have to skip any data here. |
| 4904 | Decoding routine handles them effectively anyway. */ |
| 4905 | return; |
| 4906 | } |
| 4907 | |
| 4908 | translation_table = coding->translation_table_for_decode; |
| 4909 | if (NILP (translation_table) && !NILP (Venable_character_translation)) |
| 4910 | translation_table = Vstandard_translation_table_for_decode; |
| 4911 | if (CHAR_TABLE_P (translation_table)) |
| 4912 | { |
| 4913 | int i; |
| 4914 | for (i = 0; i < 128; i++) |
| 4915 | if (!NILP (CHAR_TABLE_REF (translation_table, i))) |
| 4916 | break; |
| 4917 | if (i < 128) |
| 4918 | /* Some ASCII character should be translated. We give up |
| 4919 | shrinking. */ |
| 4920 | return; |
| 4921 | } |
| 4922 | |
| 4923 | if (coding->heading_ascii >= 0) |
| 4924 | /* Detection routine has already found how much we can skip at the |
| 4925 | head. */ |
| 4926 | *beg += coding->heading_ascii; |
| 4927 | |
| 4928 | if (str) |
| 4929 | { |
| 4930 | begp_orig = begp = str + *beg; |
| 4931 | endp_orig = endp = str + *end; |
| 4932 | } |
| 4933 | else |
| 4934 | { |
| 4935 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
| 4936 | endp_orig = endp = begp + *end - *beg; |
| 4937 | } |
| 4938 | |
| 4939 | eol_conversion = (coding->eol_type == CODING_EOL_CR |
| 4940 | || coding->eol_type == CODING_EOL_CRLF); |
| 4941 | |
| 4942 | switch (coding->type) |
| 4943 | { |
| 4944 | case coding_type_sjis: |
| 4945 | case coding_type_big5: |
| 4946 | /* We can skip all ASCII characters at the head. */ |
| 4947 | if (coding->heading_ascii < 0) |
| 4948 | { |
| 4949 | if (eol_conversion) |
| 4950 | while (begp < endp && *begp < 0x80 && *begp != '\r') begp++; |
| 4951 | else |
| 4952 | while (begp < endp && *begp < 0x80) begp++; |
| 4953 | } |
| 4954 | /* We can skip all ASCII characters at the tail except for the |
| 4955 | second byte of SJIS or BIG5 code. */ |
| 4956 | if (eol_conversion) |
| 4957 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\r') endp--; |
| 4958 | else |
| 4959 | while (begp < endp && endp[-1] < 0x80) endp--; |
| 4960 | /* Do not consider LF as ascii if preceded by CR, since that |
| 4961 | confuses eol decoding. */ |
| 4962 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') |
| 4963 | endp++; |
| 4964 | if (begp < endp && endp < endp_orig && endp[-1] >= 0x80) |
| 4965 | endp++; |
| 4966 | break; |
| 4967 | |
| 4968 | case coding_type_iso2022: |
| 4969 | if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, 0) != CHARSET_ASCII) |
| 4970 | /* We can't skip any data. */ |
| 4971 | break; |
| 4972 | if (coding->heading_ascii < 0) |
| 4973 | { |
| 4974 | /* We can skip all ASCII characters at the head except for a |
| 4975 | few control codes. */ |
| 4976 | while (begp < endp && (c = *begp) < 0x80 |
| 4977 | && c != ISO_CODE_CR && c != ISO_CODE_SO |
| 4978 | && c != ISO_CODE_SI && c != ISO_CODE_ESC |
| 4979 | && (!eol_conversion || c != ISO_CODE_LF)) |
| 4980 | begp++; |
| 4981 | } |
| 4982 | switch (coding->category_idx) |
| 4983 | { |
| 4984 | case CODING_CATEGORY_IDX_ISO_8_1: |
| 4985 | case CODING_CATEGORY_IDX_ISO_8_2: |
| 4986 | /* We can skip all ASCII characters at the tail. */ |
| 4987 | if (eol_conversion) |
| 4988 | while (begp < endp && (c = endp[-1]) < 0x80 && c != '\r') endp--; |
| 4989 | else |
| 4990 | while (begp < endp && endp[-1] < 0x80) endp--; |
| 4991 | /* Do not consider LF as ascii if preceded by CR, since that |
| 4992 | confuses eol decoding. */ |
| 4993 | if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') |
| 4994 | endp++; |
| 4995 | break; |
| 4996 | |
| 4997 | case CODING_CATEGORY_IDX_ISO_7: |
| 4998 | case CODING_CATEGORY_IDX_ISO_7_TIGHT: |
| 4999 | { |
| 5000 | /* We can skip all characters at the tail except for 8-bit |
| 5001 | codes and ESC and the following 2-byte at the tail. */ |
| 5002 | unsigned char *eight_bit = NULL; |
| 5003 | |
| 5004 | if (eol_conversion) |
| 5005 | while (begp < endp |
| 5006 | && (c = endp[-1]) != ISO_CODE_ESC && c != '\r') |
| 5007 | { |
| 5008 | if (!eight_bit && c & 0x80) eight_bit = endp; |
| 5009 | endp--; |
| 5010 | } |
| 5011 | else |
| 5012 | while (begp < endp |
| 5013 | && (c = endp[-1]) != ISO_CODE_ESC) |
| 5014 | { |
| 5015 | if (!eight_bit && c & 0x80) eight_bit = endp; |
| 5016 | endp--; |
| 5017 | } |
| 5018 | /* Do not consider LF as ascii if preceded by CR, since that |
| 5019 | confuses eol decoding. */ |
| 5020 | if (begp < endp && endp < endp_orig |
| 5021 | && endp[-1] == '\r' && endp[0] == '\n') |
| 5022 | endp++; |
| 5023 | if (begp < endp && endp[-1] == ISO_CODE_ESC) |
| 5024 | { |
| 5025 | if (endp + 1 < endp_orig && end[0] == '(' && end[1] == 'B') |
| 5026 | /* This is an ASCII designation sequence. We can |
| 5027 | surely skip the tail. But, if we have |
| 5028 | encountered an 8-bit code, skip only the codes |
| 5029 | after that. */ |
| 5030 | endp = eight_bit ? eight_bit : endp + 2; |
| 5031 | else |
| 5032 | /* Hmmm, we can't skip the tail. */ |
| 5033 | endp = endp_orig; |
| 5034 | } |
| 5035 | else if (eight_bit) |
| 5036 | endp = eight_bit; |
| 5037 | } |
| 5038 | } |
| 5039 | break; |
| 5040 | |
| 5041 | default: |
| 5042 | abort (); |
| 5043 | } |
| 5044 | *beg += begp - begp_orig; |
| 5045 | *end += endp - endp_orig; |
| 5046 | return; |
| 5047 | } |
| 5048 | |
| 5049 | /* Like shrink_decoding_region but for encoding. */ |
| 5050 | |
| 5051 | static void |
| 5052 | shrink_encoding_region (beg, end, coding, str) |
| 5053 | int *beg, *end; |
| 5054 | struct coding_system *coding; |
| 5055 | unsigned char *str; |
| 5056 | { |
| 5057 | unsigned char *begp_orig, *begp, *endp_orig, *endp; |
| 5058 | int eol_conversion; |
| 5059 | Lisp_Object translation_table; |
| 5060 | |
| 5061 | if (coding->type == coding_type_ccl |
| 5062 | || coding->eol_type == CODING_EOL_CRLF |
| 5063 | || coding->eol_type == CODING_EOL_CR |
| 5064 | || (coding->cmp_data && coding->cmp_data->used > 0)) |
| 5065 | { |
| 5066 | /* We can't skip any data. */ |
| 5067 | return; |
| 5068 | } |
| 5069 | if (coding->type == coding_type_no_conversion |
| 5070 | || coding->type == coding_type_raw_text |
| 5071 | || coding->type == coding_type_emacs_mule |
| 5072 | || coding->type == coding_type_undecided) |
| 5073 | { |
| 5074 | /* We need no conversion, but don't have to skip any data here. |
| 5075 | Encoding routine handles them effectively anyway. */ |
| 5076 | return; |
| 5077 | } |
| 5078 | |
| 5079 | translation_table = coding->translation_table_for_encode; |
| 5080 | if (NILP (translation_table) && !NILP (Venable_character_translation)) |
| 5081 | translation_table = Vstandard_translation_table_for_encode; |
| 5082 | if (CHAR_TABLE_P (translation_table)) |
| 5083 | { |
| 5084 | int i; |
| 5085 | for (i = 0; i < 128; i++) |
| 5086 | if (!NILP (CHAR_TABLE_REF (translation_table, i))) |
| 5087 | break; |
| 5088 | if (i < 128) |
| 5089 | /* Some ASCII character should be translated. We give up |
| 5090 | shrinking. */ |
| 5091 | return; |
| 5092 | } |
| 5093 | |
| 5094 | if (str) |
| 5095 | { |
| 5096 | begp_orig = begp = str + *beg; |
| 5097 | endp_orig = endp = str + *end; |
| 5098 | } |
| 5099 | else |
| 5100 | { |
| 5101 | begp_orig = begp = BYTE_POS_ADDR (*beg); |
| 5102 | endp_orig = endp = begp + *end - *beg; |
| 5103 | } |
| 5104 | |
| 5105 | eol_conversion = (coding->eol_type == CODING_EOL_CR |
| 5106 | || coding->eol_type == CODING_EOL_CRLF); |
| 5107 | |
| 5108 | /* Here, we don't have to check coding->pre_write_conversion because |
| 5109 | the caller is expected to have handled it already. */ |
| 5110 | switch (coding->type) |
| 5111 | { |
| 5112 | case coding_type_iso2022: |
| 5113 | if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, 0) != CHARSET_ASCII) |
| 5114 | /* We can't skip any data. */ |
| 5115 | break; |
| 5116 | if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL) |
| 5117 | { |
| 5118 | unsigned char *bol = begp; |
| 5119 | while (begp < endp && *begp < 0x80) |
| 5120 | { |
| 5121 | begp++; |
| 5122 | if (begp[-1] == '\n') |
| 5123 | bol = begp; |
| 5124 | } |
| 5125 | begp = bol; |
| 5126 | goto label_skip_tail; |
| 5127 | } |
| 5128 | /* fall down ... */ |
| 5129 | |
| 5130 | case coding_type_sjis: |
| 5131 | case coding_type_big5: |
| 5132 | /* We can skip all ASCII characters at the head and tail. */ |
| 5133 | if (eol_conversion) |
| 5134 | while (begp < endp && *begp < 0x80 && *begp != '\n') begp++; |
| 5135 | else |
| 5136 | while (begp < endp && *begp < 0x80) begp++; |
| 5137 | label_skip_tail: |
| 5138 | if (eol_conversion) |
| 5139 | while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\n') endp--; |
| 5140 | else |
| 5141 | while (begp < endp && *(endp - 1) < 0x80) endp--; |
| 5142 | break; |
| 5143 | |
| 5144 | default: |
| 5145 | abort (); |
| 5146 | } |
| 5147 | |
| 5148 | *beg += begp - begp_orig; |
| 5149 | *end += endp - endp_orig; |
| 5150 | return; |
| 5151 | } |
| 5152 | |
| 5153 | /* As shrinking conversion region requires some overhead, we don't try |
| 5154 | shrinking if the length of conversion region is less than this |
| 5155 | value. */ |
| 5156 | static int shrink_conversion_region_threshhold = 1024; |
| 5157 | |
| 5158 | #define SHRINK_CONVERSION_REGION(beg, end, coding, str, encodep) \ |
| 5159 | do { \ |
| 5160 | if (*(end) - *(beg) > shrink_conversion_region_threshhold) \ |
| 5161 | { \ |
| 5162 | if (encodep) shrink_encoding_region (beg, end, coding, str); \ |
| 5163 | else shrink_decoding_region (beg, end, coding, str); \ |
| 5164 | } \ |
| 5165 | } while (0) |
| 5166 | |
| 5167 | static Lisp_Object |
| 5168 | code_convert_region_unwind (dummy) |
| 5169 | Lisp_Object dummy; |
| 5170 | { |
| 5171 | inhibit_pre_post_conversion = 0; |
| 5172 | return Qnil; |
| 5173 | } |
| 5174 | |
| 5175 | /* Store information about all compositions in the range FROM and TO |
| 5176 | of OBJ in memory blocks pointed by CODING->cmp_data. OBJ is a |
| 5177 | buffer or a string, defaults to the current buffer. */ |
| 5178 | |
| 5179 | void |
| 5180 | coding_save_composition (coding, from, to, obj) |
| 5181 | struct coding_system *coding; |
| 5182 | int from, to; |
| 5183 | Lisp_Object obj; |
| 5184 | { |
| 5185 | Lisp_Object prop; |
| 5186 | int start, end; |
| 5187 | |
| 5188 | if (coding->composing == COMPOSITION_DISABLED) |
| 5189 | return; |
| 5190 | if (!coding->cmp_data) |
| 5191 | coding_allocate_composition_data (coding, from); |
| 5192 | if (!find_composition (from, to, &start, &end, &prop, obj) |
| 5193 | || end > to) |
| 5194 | return; |
| 5195 | if (start < from |
| 5196 | && (!find_composition (end, to, &start, &end, &prop, obj) |
| 5197 | || end > to)) |
| 5198 | return; |
| 5199 | coding->composing = COMPOSITION_NO; |
| 5200 | do |
| 5201 | { |
| 5202 | if (COMPOSITION_VALID_P (start, end, prop)) |
| 5203 | { |
| 5204 | enum composition_method method = COMPOSITION_METHOD (prop); |
| 5205 | if (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH |
| 5206 | >= COMPOSITION_DATA_SIZE) |
| 5207 | coding_allocate_composition_data (coding, from); |
| 5208 | /* For relative composition, we remember start and end |
| 5209 | positions, for the other compositions, we also remember |
| 5210 | components. */ |
| 5211 | CODING_ADD_COMPOSITION_START (coding, start - from, method); |
| 5212 | if (method != COMPOSITION_RELATIVE) |
| 5213 | { |
| 5214 | /* We must store a*/ |
| 5215 | Lisp_Object val, ch; |
| 5216 | |
| 5217 | val = COMPOSITION_COMPONENTS (prop); |
| 5218 | if (CONSP (val)) |
| 5219 | while (CONSP (val)) |
| 5220 | { |
| 5221 | ch = XCAR (val), val = XCDR (val); |
| 5222 | CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (ch)); |
| 5223 | } |
| 5224 | else if (VECTORP (val) || STRINGP (val)) |
| 5225 | { |
| 5226 | int len = (VECTORP (val) |
| 5227 | ? XVECTOR (val)->size : XSTRING (val)->size); |
| 5228 | int i; |
| 5229 | for (i = 0; i < len; i++) |
| 5230 | { |
| 5231 | ch = (STRINGP (val) |
| 5232 | ? Faref (val, make_number (i)) |
| 5233 | : XVECTOR (val)->contents[i]); |
| 5234 | CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (ch)); |
| 5235 | } |
| 5236 | } |
| 5237 | else /* INTEGERP (val) */ |
| 5238 | CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (val)); |
| 5239 | } |
| 5240 | CODING_ADD_COMPOSITION_END (coding, end - from); |
| 5241 | } |
| 5242 | start = end; |
| 5243 | } |
| 5244 | while (start < to |
| 5245 | && find_composition (start, to, &start, &end, &prop, obj) |
| 5246 | && end <= to); |
| 5247 | |
| 5248 | /* Make coding->cmp_data point to the first memory block. */ |
| 5249 | while (coding->cmp_data->prev) |
| 5250 | coding->cmp_data = coding->cmp_data->prev; |
| 5251 | coding->cmp_data_start = 0; |
| 5252 | } |
| 5253 | |
| 5254 | /* Reflect the saved information about compositions to OBJ. |
| 5255 | CODING->cmp_data points to a memory block for the information. OBJ |
| 5256 | is a buffer or a string, defaults to the current buffer. */ |
| 5257 | |
| 5258 | void |
| 5259 | coding_restore_composition (coding, obj) |
| 5260 | struct coding_system *coding; |
| 5261 | Lisp_Object obj; |
| 5262 | { |
| 5263 | struct composition_data *cmp_data = coding->cmp_data; |
| 5264 | |
| 5265 | if (!cmp_data) |
| 5266 | return; |
| 5267 | |
| 5268 | while (cmp_data->prev) |
| 5269 | cmp_data = cmp_data->prev; |
| 5270 | |
| 5271 | while (cmp_data) |
| 5272 | { |
| 5273 | int i; |
| 5274 | |
| 5275 | for (i = 0; i < cmp_data->used && cmp_data->data[i] > 0; |
| 5276 | i += cmp_data->data[i]) |
| 5277 | { |
| 5278 | int *data = cmp_data->data + i; |
| 5279 | enum composition_method method = (enum composition_method) data[3]; |
| 5280 | Lisp_Object components; |
| 5281 | |
| 5282 | if (method == COMPOSITION_RELATIVE) |
| 5283 | components = Qnil; |
| 5284 | else |
| 5285 | { |
| 5286 | int len = data[0] - 4, j; |
| 5287 | Lisp_Object args[MAX_COMPOSITION_COMPONENTS * 2 - 1]; |
| 5288 | |
| 5289 | for (j = 0; j < len; j++) |
| 5290 | args[j] = make_number (data[4 + j]); |
| 5291 | components = (method == COMPOSITION_WITH_ALTCHARS |
| 5292 | ? Fstring (len, args) : Fvector (len, args)); |
| 5293 | } |
| 5294 | compose_text (data[1], data[2], components, Qnil, obj); |
| 5295 | } |
| 5296 | cmp_data = cmp_data->next; |
| 5297 | } |
| 5298 | } |
| 5299 | |
| 5300 | /* Decode (if ENCODEP is zero) or encode (if ENCODEP is nonzero) the |
| 5301 | text from FROM to TO (byte positions are FROM_BYTE and TO_BYTE) by |
| 5302 | coding system CODING, and return the status code of code conversion |
| 5303 | (currently, this value has no meaning). |
| 5304 | |
| 5305 | How many characters (and bytes) are converted to how many |
| 5306 | characters (and bytes) are recorded in members of the structure |
| 5307 | CODING. |
| 5308 | |
| 5309 | If REPLACE is nonzero, we do various things as if the original text |
| 5310 | is deleted and a new text is inserted. See the comments in |
| 5311 | replace_range (insdel.c) to know what we are doing. |
| 5312 | |
| 5313 | If REPLACE is zero, it is assumed that the source text is unibyte. |
| 5314 | Otherwise, it is assumed that the source text is multibyte. */ |
| 5315 | |
| 5316 | int |
| 5317 | code_convert_region (from, from_byte, to, to_byte, coding, encodep, replace) |
| 5318 | int from, from_byte, to, to_byte, encodep, replace; |
| 5319 | struct coding_system *coding; |
| 5320 | { |
| 5321 | int len = to - from, len_byte = to_byte - from_byte; |
| 5322 | int nchars_del = 0, nbytes_del = 0; |
| 5323 | int require, inserted, inserted_byte; |
| 5324 | int head_skip, tail_skip, total_skip = 0; |
| 5325 | Lisp_Object saved_coding_symbol; |
| 5326 | int first = 1; |
| 5327 | unsigned char *src, *dst; |
| 5328 | Lisp_Object deletion; |
| 5329 | int orig_point = PT, orig_len = len; |
| 5330 | int prev_Z; |
| 5331 | int multibyte_p = !NILP (current_buffer->enable_multibyte_characters); |
| 5332 | |
| 5333 | deletion = Qnil; |
| 5334 | saved_coding_symbol = coding->symbol; |
| 5335 | |
| 5336 | if (from < PT && PT < to) |
| 5337 | { |
| 5338 | TEMP_SET_PT_BOTH (from, from_byte); |
| 5339 | orig_point = from; |
| 5340 | } |
| 5341 | |
| 5342 | if (replace) |
| 5343 | { |
| 5344 | int saved_from = from; |
| 5345 | int saved_inhibit_modification_hooks; |
| 5346 | |
| 5347 | prepare_to_modify_buffer (from, to, &from); |
| 5348 | if (saved_from != from) |
| 5349 | { |
| 5350 | to = from + len; |
| 5351 | from_byte = CHAR_TO_BYTE (from), to_byte = CHAR_TO_BYTE (to); |
| 5352 | len_byte = to_byte - from_byte; |
| 5353 | } |
| 5354 | |
| 5355 | /* The code conversion routine can not preserve text properties |
| 5356 | for now. So, we must remove all text properties in the |
| 5357 | region. Here, we must suppress all modification hooks. */ |
| 5358 | saved_inhibit_modification_hooks = inhibit_modification_hooks; |
| 5359 | inhibit_modification_hooks = 1; |
| 5360 | Fset_text_properties (make_number (from), make_number (to), Qnil, Qnil); |
| 5361 | inhibit_modification_hooks = saved_inhibit_modification_hooks; |
| 5362 | } |
| 5363 | |
| 5364 | if (! encodep && CODING_REQUIRE_DETECTION (coding)) |
| 5365 | { |
| 5366 | /* We must detect encoding of text and eol format. */ |
| 5367 | |
| 5368 | if (from < GPT && to > GPT) |
| 5369 | move_gap_both (from, from_byte); |
| 5370 | if (coding->type == coding_type_undecided) |
| 5371 | { |
| 5372 | detect_coding (coding, BYTE_POS_ADDR (from_byte), len_byte); |
| 5373 | if (coding->type == coding_type_undecided) |
| 5374 | { |
| 5375 | /* It seems that the text contains only ASCII, but we |
| 5376 | should not leave it undecided because the deeper |
| 5377 | decoding routine (decode_coding) tries to detect the |
| 5378 | encodings again in vain. */ |
| 5379 | coding->type = coding_type_emacs_mule; |
| 5380 | coding->category_idx = CODING_CATEGORY_IDX_EMACS_MULE; |
| 5381 | /* As emacs-mule decoder will handle composition, we |
| 5382 | need this setting to allocate coding->cmp_data |
| 5383 | later. */ |
| 5384 | coding->composing = COMPOSITION_NO; |
| 5385 | } |
| 5386 | } |
| 5387 | if (coding->eol_type == CODING_EOL_UNDECIDED |
| 5388 | && coding->type != coding_type_ccl) |
| 5389 | { |
| 5390 | detect_eol (coding, BYTE_POS_ADDR (from_byte), len_byte); |
| 5391 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 5392 | coding->eol_type = CODING_EOL_LF; |
| 5393 | /* We had better recover the original eol format if we |
| 5394 | encounter an inconsistent eol format while decoding. */ |
| 5395 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; |
| 5396 | } |
| 5397 | } |
| 5398 | |
| 5399 | /* Now we convert the text. */ |
| 5400 | |
| 5401 | /* For encoding, we must process pre-write-conversion in advance. */ |
| 5402 | if (! inhibit_pre_post_conversion |
| 5403 | && encodep |
| 5404 | && SYMBOLP (coding->pre_write_conversion) |
| 5405 | && ! NILP (Ffboundp (coding->pre_write_conversion))) |
| 5406 | { |
| 5407 | /* The function in pre-write-conversion may put a new text in a |
| 5408 | new buffer. */ |
| 5409 | struct buffer *prev = current_buffer; |
| 5410 | Lisp_Object new; |
| 5411 | |
| 5412 | record_unwind_protect (code_convert_region_unwind, Qnil); |
| 5413 | /* We should not call any more pre-write/post-read-conversion |
| 5414 | functions while this pre-write-conversion is running. */ |
| 5415 | inhibit_pre_post_conversion = 1; |
| 5416 | call2 (coding->pre_write_conversion, |
| 5417 | make_number (from), make_number (to)); |
| 5418 | inhibit_pre_post_conversion = 0; |
| 5419 | /* Discard the unwind protect. */ |
| 5420 | specpdl_ptr--; |
| 5421 | |
| 5422 | if (current_buffer != prev) |
| 5423 | { |
| 5424 | len = ZV - BEGV; |
| 5425 | new = Fcurrent_buffer (); |
| 5426 | set_buffer_internal_1 (prev); |
| 5427 | del_range_2 (from, from_byte, to, to_byte, 0); |
| 5428 | TEMP_SET_PT_BOTH (from, from_byte); |
| 5429 | insert_from_buffer (XBUFFER (new), 1, len, 0); |
| 5430 | Fkill_buffer (new); |
| 5431 | if (orig_point >= to) |
| 5432 | orig_point += len - orig_len; |
| 5433 | else if (orig_point > from) |
| 5434 | orig_point = from; |
| 5435 | orig_len = len; |
| 5436 | to = from + len; |
| 5437 | from_byte = CHAR_TO_BYTE (from); |
| 5438 | to_byte = CHAR_TO_BYTE (to); |
| 5439 | len_byte = to_byte - from_byte; |
| 5440 | TEMP_SET_PT_BOTH (from, from_byte); |
| 5441 | } |
| 5442 | } |
| 5443 | |
| 5444 | if (replace) |
| 5445 | { |
| 5446 | if (! EQ (current_buffer->undo_list, Qt)) |
| 5447 | deletion = make_buffer_string_both (from, from_byte, to, to_byte, 1); |
| 5448 | else |
| 5449 | { |
| 5450 | nchars_del = to - from; |
| 5451 | nbytes_del = to_byte - from_byte; |
| 5452 | } |
| 5453 | } |
| 5454 | |
| 5455 | if (coding->composing != COMPOSITION_DISABLED) |
| 5456 | { |
| 5457 | if (encodep) |
| 5458 | coding_save_composition (coding, from, to, Fcurrent_buffer ()); |
| 5459 | else |
| 5460 | coding_allocate_composition_data (coding, from); |
| 5461 | } |
| 5462 | |
| 5463 | /* Try to skip the heading and tailing ASCIIs. */ |
| 5464 | if (coding->type != coding_type_ccl) |
| 5465 | { |
| 5466 | int from_byte_orig = from_byte, to_byte_orig = to_byte; |
| 5467 | |
| 5468 | if (from < GPT && GPT < to) |
| 5469 | move_gap_both (from, from_byte); |
| 5470 | SHRINK_CONVERSION_REGION (&from_byte, &to_byte, coding, NULL, encodep); |
| 5471 | if (from_byte == to_byte |
| 5472 | && (encodep || NILP (coding->post_read_conversion)) |
| 5473 | && ! CODING_REQUIRE_FLUSHING (coding)) |
| 5474 | { |
| 5475 | coding->produced = len_byte; |
| 5476 | coding->produced_char = len; |
| 5477 | if (!replace) |
| 5478 | /* We must record and adjust for this new text now. */ |
| 5479 | adjust_after_insert (from, from_byte_orig, to, to_byte_orig, len); |
| 5480 | return 0; |
| 5481 | } |
| 5482 | |
| 5483 | head_skip = from_byte - from_byte_orig; |
| 5484 | tail_skip = to_byte_orig - to_byte; |
| 5485 | total_skip = head_skip + tail_skip; |
| 5486 | from += head_skip; |
| 5487 | to -= tail_skip; |
| 5488 | len -= total_skip; len_byte -= total_skip; |
| 5489 | } |
| 5490 | |
| 5491 | /* For conversion, we must put the gap before the text in addition to |
| 5492 | making the gap larger for efficient decoding. The required gap |
| 5493 | size starts from 2000 which is the magic number used in make_gap. |
| 5494 | But, after one batch of conversion, it will be incremented if we |
| 5495 | find that it is not enough . */ |
| 5496 | require = 2000; |
| 5497 | |
| 5498 | if (GAP_SIZE < require) |
| 5499 | make_gap (require - GAP_SIZE); |
| 5500 | move_gap_both (from, from_byte); |
| 5501 | |
| 5502 | inserted = inserted_byte = 0; |
| 5503 | |
| 5504 | GAP_SIZE += len_byte; |
| 5505 | ZV -= len; |
| 5506 | Z -= len; |
| 5507 | ZV_BYTE -= len_byte; |
| 5508 | Z_BYTE -= len_byte; |
| 5509 | |
| 5510 | if (GPT - BEG < BEG_UNCHANGED) |
| 5511 | BEG_UNCHANGED = GPT - BEG; |
| 5512 | if (Z - GPT < END_UNCHANGED) |
| 5513 | END_UNCHANGED = Z - GPT; |
| 5514 | |
| 5515 | if (!encodep && coding->src_multibyte) |
| 5516 | { |
| 5517 | /* Decoding routines expects that the source text is unibyte. |
| 5518 | We must convert 8-bit characters of multibyte form to |
| 5519 | unibyte. */ |
| 5520 | int len_byte_orig = len_byte; |
| 5521 | len_byte = str_as_unibyte (GAP_END_ADDR - len_byte, len_byte); |
| 5522 | if (len_byte < len_byte_orig) |
| 5523 | safe_bcopy (GAP_END_ADDR - len_byte_orig, GAP_END_ADDR - len_byte, |
| 5524 | len_byte); |
| 5525 | coding->src_multibyte = 0; |
| 5526 | } |
| 5527 | |
| 5528 | for (;;) |
| 5529 | { |
| 5530 | int result; |
| 5531 | |
| 5532 | /* The buffer memory is now: |
| 5533 | +--------+converted-text+---------+-------original-text-------+---+ |
| 5534 | |<-from->|<--inserted-->|---------|<--------len_byte--------->|---| |
| 5535 | |<---------------------- GAP ----------------------->| */ |
| 5536 | src = GAP_END_ADDR - len_byte; |
| 5537 | dst = GPT_ADDR + inserted_byte; |
| 5538 | |
| 5539 | if (encodep) |
| 5540 | result = encode_coding (coding, src, dst, len_byte, 0); |
| 5541 | else |
| 5542 | { |
| 5543 | if (coding->composing != COMPOSITION_DISABLED) |
| 5544 | coding->cmp_data->char_offset = from + inserted; |
| 5545 | result = decode_coding (coding, src, dst, len_byte, 0); |
| 5546 | } |
| 5547 | |
| 5548 | /* The buffer memory is now: |
| 5549 | +--------+-------converted-text----+--+------original-text----+---+ |
| 5550 | |<-from->|<-inserted->|<-produced->|--|<-(len_byte-consumed)->|---| |
| 5551 | |<---------------------- GAP ----------------------->| */ |
| 5552 | |
| 5553 | inserted += coding->produced_char; |
| 5554 | inserted_byte += coding->produced; |
| 5555 | len_byte -= coding->consumed; |
| 5556 | |
| 5557 | if (result == CODING_FINISH_INSUFFICIENT_CMP) |
| 5558 | { |
| 5559 | coding_allocate_composition_data (coding, from + inserted); |
| 5560 | continue; |
| 5561 | } |
| 5562 | |
| 5563 | src += coding->consumed; |
| 5564 | dst += coding->produced; |
| 5565 | |
| 5566 | if (result == CODING_FINISH_NORMAL) |
| 5567 | { |
| 5568 | src += len_byte; |
| 5569 | break; |
| 5570 | } |
| 5571 | if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL) |
| 5572 | { |
| 5573 | unsigned char *pend = dst, *p = pend - inserted_byte; |
| 5574 | Lisp_Object eol_type; |
| 5575 | |
| 5576 | /* Encode LFs back to the original eol format (CR or CRLF). */ |
| 5577 | if (coding->eol_type == CODING_EOL_CR) |
| 5578 | { |
| 5579 | while (p < pend) if (*p++ == '\n') p[-1] = '\r'; |
| 5580 | } |
| 5581 | else |
| 5582 | { |
| 5583 | int count = 0; |
| 5584 | |
| 5585 | while (p < pend) if (*p++ == '\n') count++; |
| 5586 | if (src - dst < count) |
| 5587 | { |
| 5588 | /* We don't have sufficient room for encoding LFs |
| 5589 | back to CRLF. We must record converted and |
| 5590 | not-yet-converted text back to the buffer |
| 5591 | content, enlarge the gap, then record them out of |
| 5592 | the buffer contents again. */ |
| 5593 | int add = len_byte + inserted_byte; |
| 5594 | |
| 5595 | GAP_SIZE -= add; |
| 5596 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; |
| 5597 | GPT += inserted_byte; GPT_BYTE += inserted_byte; |
| 5598 | make_gap (count - GAP_SIZE); |
| 5599 | GAP_SIZE += add; |
| 5600 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; |
| 5601 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; |
| 5602 | /* Don't forget to update SRC, DST, and PEND. */ |
| 5603 | src = GAP_END_ADDR - len_byte; |
| 5604 | dst = GPT_ADDR + inserted_byte; |
| 5605 | pend = dst; |
| 5606 | } |
| 5607 | inserted += count; |
| 5608 | inserted_byte += count; |
| 5609 | coding->produced += count; |
| 5610 | p = dst = pend + count; |
| 5611 | while (count) |
| 5612 | { |
| 5613 | *--p = *--pend; |
| 5614 | if (*p == '\n') count--, *--p = '\r'; |
| 5615 | } |
| 5616 | } |
| 5617 | |
| 5618 | /* Suppress eol-format conversion in the further conversion. */ |
| 5619 | coding->eol_type = CODING_EOL_LF; |
| 5620 | |
| 5621 | /* Set the coding system symbol to that for Unix-like EOL. */ |
| 5622 | eol_type = Fget (saved_coding_symbol, Qeol_type); |
| 5623 | if (VECTORP (eol_type) |
| 5624 | && XVECTOR (eol_type)->size == 3 |
| 5625 | && SYMBOLP (XVECTOR (eol_type)->contents[CODING_EOL_LF])) |
| 5626 | coding->symbol = XVECTOR (eol_type)->contents[CODING_EOL_LF]; |
| 5627 | else |
| 5628 | coding->symbol = saved_coding_symbol; |
| 5629 | |
| 5630 | continue; |
| 5631 | } |
| 5632 | if (len_byte <= 0) |
| 5633 | { |
| 5634 | if (coding->type != coding_type_ccl |
| 5635 | || coding->mode & CODING_MODE_LAST_BLOCK) |
| 5636 | break; |
| 5637 | coding->mode |= CODING_MODE_LAST_BLOCK; |
| 5638 | continue; |
| 5639 | } |
| 5640 | if (result == CODING_FINISH_INSUFFICIENT_SRC) |
| 5641 | { |
| 5642 | /* The source text ends in invalid codes. Let's just |
| 5643 | make them valid buffer contents, and finish conversion. */ |
| 5644 | if (multibyte_p) |
| 5645 | { |
| 5646 | unsigned char *start = dst; |
| 5647 | |
| 5648 | inserted += len_byte; |
| 5649 | while (len_byte--) |
| 5650 | { |
| 5651 | int c = *src++; |
| 5652 | dst += CHAR_STRING (c, dst); |
| 5653 | } |
| 5654 | |
| 5655 | inserted_byte += dst - start; |
| 5656 | } |
| 5657 | else |
| 5658 | { |
| 5659 | inserted += len_byte; |
| 5660 | inserted_byte += len_byte; |
| 5661 | while (len_byte--) |
| 5662 | *dst++ = *src++; |
| 5663 | } |
| 5664 | break; |
| 5665 | } |
| 5666 | if (result == CODING_FINISH_INTERRUPT) |
| 5667 | { |
| 5668 | /* The conversion procedure was interrupted by a user. */ |
| 5669 | break; |
| 5670 | } |
| 5671 | /* Now RESULT == CODING_FINISH_INSUFFICIENT_DST */ |
| 5672 | if (coding->consumed < 1) |
| 5673 | { |
| 5674 | /* It's quite strange to require more memory without |
| 5675 | consuming any bytes. Perhaps CCL program bug. */ |
| 5676 | break; |
| 5677 | } |
| 5678 | if (first) |
| 5679 | { |
| 5680 | /* We have just done the first batch of conversion which was |
| 5681 | stopped because of insufficient gap. Let's reconsider the |
| 5682 | required gap size (i.e. SRT - DST) now. |
| 5683 | |
| 5684 | We have converted ORIG bytes (== coding->consumed) into |
| 5685 | NEW bytes (coding->produced). To convert the remaining |
| 5686 | LEN bytes, we may need REQUIRE bytes of gap, where: |
| 5687 | REQUIRE + LEN_BYTE = LEN_BYTE * (NEW / ORIG) |
| 5688 | REQUIRE = LEN_BYTE * (NEW - ORIG) / ORIG |
| 5689 | Here, we are sure that NEW >= ORIG. */ |
| 5690 | float ratio = coding->produced - coding->consumed; |
| 5691 | ratio /= coding->consumed; |
| 5692 | require = len_byte * ratio; |
| 5693 | first = 0; |
| 5694 | } |
| 5695 | if ((src - dst) < (require + 2000)) |
| 5696 | { |
| 5697 | /* See the comment above the previous call of make_gap. */ |
| 5698 | int add = len_byte + inserted_byte; |
| 5699 | |
| 5700 | GAP_SIZE -= add; |
| 5701 | ZV += add; Z += add; ZV_BYTE += add; Z_BYTE += add; |
| 5702 | GPT += inserted_byte; GPT_BYTE += inserted_byte; |
| 5703 | make_gap (require + 2000); |
| 5704 | GAP_SIZE += add; |
| 5705 | ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add; |
| 5706 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; |
| 5707 | } |
| 5708 | } |
| 5709 | if (src - dst > 0) *dst = 0; /* Put an anchor. */ |
| 5710 | |
| 5711 | if (encodep && coding->dst_multibyte) |
| 5712 | { |
| 5713 | /* The output is unibyte. We must convert 8-bit characters to |
| 5714 | multibyte form. */ |
| 5715 | if (inserted_byte * 2 > GAP_SIZE) |
| 5716 | { |
| 5717 | GAP_SIZE -= inserted_byte; |
| 5718 | ZV += inserted_byte; Z += inserted_byte; |
| 5719 | ZV_BYTE += inserted_byte; Z_BYTE += inserted_byte; |
| 5720 | GPT += inserted_byte; GPT_BYTE += inserted_byte; |
| 5721 | make_gap (inserted_byte - GAP_SIZE); |
| 5722 | GAP_SIZE += inserted_byte; |
| 5723 | ZV -= inserted_byte; Z -= inserted_byte; |
| 5724 | ZV_BYTE -= inserted_byte; Z_BYTE -= inserted_byte; |
| 5725 | GPT -= inserted_byte; GPT_BYTE -= inserted_byte; |
| 5726 | } |
| 5727 | inserted_byte = str_to_multibyte (GPT_ADDR, GAP_SIZE, inserted_byte); |
| 5728 | } |
| 5729 | |
| 5730 | /* If we shrank the conversion area, adjust it now. */ |
| 5731 | if (total_skip > 0) |
| 5732 | { |
| 5733 | if (tail_skip > 0) |
| 5734 | safe_bcopy (GAP_END_ADDR, GPT_ADDR + inserted_byte, tail_skip); |
| 5735 | inserted += total_skip; inserted_byte += total_skip; |
| 5736 | GAP_SIZE += total_skip; |
| 5737 | GPT -= head_skip; GPT_BYTE -= head_skip; |
| 5738 | ZV -= total_skip; ZV_BYTE -= total_skip; |
| 5739 | Z -= total_skip; Z_BYTE -= total_skip; |
| 5740 | from -= head_skip; from_byte -= head_skip; |
| 5741 | to += tail_skip; to_byte += tail_skip; |
| 5742 | } |
| 5743 | |
| 5744 | prev_Z = Z; |
| 5745 | if (! EQ (current_buffer->undo_list, Qt)) |
| 5746 | adjust_after_replace (from, from_byte, deletion, inserted, inserted_byte); |
| 5747 | else |
| 5748 | adjust_after_replace_noundo (from, from_byte, nchars_del, nbytes_del, |
| 5749 | inserted, inserted_byte); |
| 5750 | inserted = Z - prev_Z; |
| 5751 | |
| 5752 | if (!encodep && coding->cmp_data && coding->cmp_data->used) |
| 5753 | coding_restore_composition (coding, Fcurrent_buffer ()); |
| 5754 | coding_free_composition_data (coding); |
| 5755 | |
| 5756 | if (! inhibit_pre_post_conversion |
| 5757 | && ! encodep && ! NILP (coding->post_read_conversion)) |
| 5758 | { |
| 5759 | Lisp_Object val; |
| 5760 | |
| 5761 | if (from != PT) |
| 5762 | TEMP_SET_PT_BOTH (from, from_byte); |
| 5763 | prev_Z = Z; |
| 5764 | record_unwind_protect (code_convert_region_unwind, Qnil); |
| 5765 | /* We should not call any more pre-write/post-read-conversion |
| 5766 | functions while this post-read-conversion is running. */ |
| 5767 | inhibit_pre_post_conversion = 1; |
| 5768 | val = call1 (coding->post_read_conversion, make_number (inserted)); |
| 5769 | inhibit_pre_post_conversion = 0; |
| 5770 | /* Discard the unwind protect. */ |
| 5771 | specpdl_ptr--; |
| 5772 | CHECK_NUMBER (val); |
| 5773 | inserted += Z - prev_Z; |
| 5774 | } |
| 5775 | |
| 5776 | if (orig_point >= from) |
| 5777 | { |
| 5778 | if (orig_point >= from + orig_len) |
| 5779 | orig_point += inserted - orig_len; |
| 5780 | else |
| 5781 | orig_point = from; |
| 5782 | TEMP_SET_PT (orig_point); |
| 5783 | } |
| 5784 | |
| 5785 | if (replace) |
| 5786 | { |
| 5787 | signal_after_change (from, to - from, inserted); |
| 5788 | update_compositions (from, from + inserted, CHECK_BORDER); |
| 5789 | } |
| 5790 | |
| 5791 | { |
| 5792 | coding->consumed = to_byte - from_byte; |
| 5793 | coding->consumed_char = to - from; |
| 5794 | coding->produced = inserted_byte; |
| 5795 | coding->produced_char = inserted; |
| 5796 | } |
| 5797 | |
| 5798 | return 0; |
| 5799 | } |
| 5800 | |
| 5801 | Lisp_Object |
| 5802 | run_pre_post_conversion_on_str (str, coding, encodep) |
| 5803 | Lisp_Object str; |
| 5804 | struct coding_system *coding; |
| 5805 | int encodep; |
| 5806 | { |
| 5807 | int count = SPECPDL_INDEX (); |
| 5808 | struct gcpro gcpro1; |
| 5809 | int multibyte = STRING_MULTIBYTE (str); |
| 5810 | Lisp_Object buffer; |
| 5811 | struct buffer *buf; |
| 5812 | |
| 5813 | record_unwind_protect (Fset_buffer, Fcurrent_buffer ()); |
| 5814 | record_unwind_protect (code_convert_region_unwind, Qnil); |
| 5815 | GCPRO1 (str); |
| 5816 | |
| 5817 | buffer = Fget_buffer_create (build_string (" *code-converting-work*")); |
| 5818 | buf = XBUFFER (buffer); |
| 5819 | |
| 5820 | buf->directory = current_buffer->directory; |
| 5821 | buf->read_only = Qnil; |
| 5822 | buf->filename = Qnil; |
| 5823 | buf->undo_list = Qt; |
| 5824 | buf->overlays_before = Qnil; |
| 5825 | buf->overlays_after = Qnil; |
| 5826 | |
| 5827 | set_buffer_internal (buf); |
| 5828 | /* We must insert the contents of STR as is without |
| 5829 | unibyte<->multibyte conversion. For that, we adjust the |
| 5830 | multibyteness of the working buffer to that of STR. */ |
| 5831 | Ferase_buffer (); |
| 5832 | buf->enable_multibyte_characters = multibyte ? Qt : Qnil; |
| 5833 | |
| 5834 | insert_from_string (str, 0, 0, |
| 5835 | XSTRING (str)->size, STRING_BYTES (XSTRING (str)), 0); |
| 5836 | UNGCPRO; |
| 5837 | inhibit_pre_post_conversion = 1; |
| 5838 | if (encodep) |
| 5839 | call2 (coding->pre_write_conversion, make_number (BEG), make_number (Z)); |
| 5840 | else |
| 5841 | { |
| 5842 | TEMP_SET_PT_BOTH (BEG, BEG_BYTE); |
| 5843 | call1 (coding->post_read_conversion, make_number (Z - BEG)); |
| 5844 | } |
| 5845 | inhibit_pre_post_conversion = 0; |
| 5846 | str = make_buffer_string (BEG, Z, 1); |
| 5847 | return unbind_to (count, str); |
| 5848 | } |
| 5849 | |
| 5850 | Lisp_Object |
| 5851 | decode_coding_string (str, coding, nocopy) |
| 5852 | Lisp_Object str; |
| 5853 | struct coding_system *coding; |
| 5854 | int nocopy; |
| 5855 | { |
| 5856 | int len; |
| 5857 | struct conversion_buffer buf; |
| 5858 | int from, to_byte; |
| 5859 | Lisp_Object saved_coding_symbol; |
| 5860 | int result; |
| 5861 | int require_decoding; |
| 5862 | int shrinked_bytes = 0; |
| 5863 | Lisp_Object newstr; |
| 5864 | int consumed, consumed_char, produced, produced_char; |
| 5865 | |
| 5866 | from = 0; |
| 5867 | to_byte = STRING_BYTES (XSTRING (str)); |
| 5868 | |
| 5869 | saved_coding_symbol = coding->symbol; |
| 5870 | coding->src_multibyte = STRING_MULTIBYTE (str); |
| 5871 | coding->dst_multibyte = 1; |
| 5872 | if (CODING_REQUIRE_DETECTION (coding)) |
| 5873 | { |
| 5874 | /* See the comments in code_convert_region. */ |
| 5875 | if (coding->type == coding_type_undecided) |
| 5876 | { |
| 5877 | detect_coding (coding, XSTRING (str)->data, to_byte); |
| 5878 | if (coding->type == coding_type_undecided) |
| 5879 | { |
| 5880 | coding->type = coding_type_emacs_mule; |
| 5881 | coding->category_idx = CODING_CATEGORY_IDX_EMACS_MULE; |
| 5882 | /* As emacs-mule decoder will handle composition, we |
| 5883 | need this setting to allocate coding->cmp_data |
| 5884 | later. */ |
| 5885 | coding->composing = COMPOSITION_NO; |
| 5886 | } |
| 5887 | } |
| 5888 | if (coding->eol_type == CODING_EOL_UNDECIDED |
| 5889 | && coding->type != coding_type_ccl) |
| 5890 | { |
| 5891 | saved_coding_symbol = coding->symbol; |
| 5892 | detect_eol (coding, XSTRING (str)->data, to_byte); |
| 5893 | if (coding->eol_type == CODING_EOL_UNDECIDED) |
| 5894 | coding->eol_type = CODING_EOL_LF; |
| 5895 | /* We had better recover the original eol format if we |
| 5896 | encounter an inconsistent eol format while decoding. */ |
| 5897 | coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL; |
| 5898 | } |
| 5899 | } |
| 5900 | |
| 5901 | if (coding->type == coding_type_no_conversion |
| 5902 | || coding->type == coding_type_raw_text) |
| 5903 | coding->dst_multibyte = 0; |
| 5904 | |
| 5905 | require_decoding = CODING_REQUIRE_DECODING (coding); |
| 5906 | |
| 5907 | if (STRING_MULTIBYTE (str)) |
| 5908 | { |
| 5909 | /* Decoding routines expect the source text to be unibyte. */ |
| 5910 | str = Fstring_as_unibyte (str); |
| 5911 | to_byte = STRING_BYTES (XSTRING (str)); |
| 5912 | nocopy = 1; |
| 5913 | coding->src_multibyte = 0; |
| 5914 | } |
| 5915 | |
| 5916 | /* Try to skip the heading and tailing ASCIIs. */ |
| 5917 | if (require_decoding && coding->type != coding_type_ccl) |
| 5918 | { |
| 5919 | SHRINK_CONVERSION_REGION (&from, &to_byte, coding, XSTRING (str)->data, |
| 5920 | 0); |
| 5921 | if (from == to_byte) |
| 5922 | require_decoding = 0; |
| 5923 | shrinked_bytes = from + (STRING_BYTES (XSTRING (str)) - to_byte); |
| 5924 | } |
| 5925 | |
| 5926 | if (!require_decoding) |
| 5927 | { |
| 5928 | coding->consumed = STRING_BYTES (XSTRING (str)); |
| 5929 | coding->consumed_char = XSTRING (str)->size; |
| 5930 | if (coding->dst_multibyte) |
| 5931 | { |
| 5932 | str = Fstring_as_multibyte (str); |
| 5933 | nocopy = 1; |
| 5934 | } |
| 5935 | coding->produced = STRING_BYTES (XSTRING (str)); |
| 5936 | coding->produced_char = XSTRING (str)->size; |
| 5937 | return (nocopy ? str : Fcopy_sequence (str)); |
| 5938 | } |
| 5939 | |
| 5940 | if (coding->composing != COMPOSITION_DISABLED) |
| 5941 | coding_allocate_composition_data (coding, from); |
| 5942 | len = decoding_buffer_size (coding, to_byte - from); |
| 5943 | allocate_conversion_buffer (buf, len); |
| 5944 | |
| 5945 | consumed = consumed_char = produced = produced_char = 0; |
| 5946 | while (1) |
| 5947 | { |
| 5948 | result = decode_coding (coding, XSTRING (str)->data + from + consumed, |
| 5949 | buf.data + produced, to_byte - from - consumed, |
| 5950 | buf.size - produced); |
| 5951 | consumed += coding->consumed; |
| 5952 | consumed_char += coding->consumed_char; |
| 5953 | produced += coding->produced; |
| 5954 | produced_char += coding->produced_char; |
| 5955 | if (result == CODING_FINISH_NORMAL |
| 5956 | || (result == CODING_FINISH_INSUFFICIENT_SRC |
| 5957 | && coding->consumed == 0)) |
| 5958 | break; |
| 5959 | if (result == CODING_FINISH_INSUFFICIENT_CMP) |
| 5960 | coding_allocate_composition_data (coding, from + produced_char); |
| 5961 | else if (result == CODING_FINISH_INSUFFICIENT_DST) |
| 5962 | extend_conversion_buffer (&buf); |
| 5963 | else if (result == CODING_FINISH_INCONSISTENT_EOL) |
| 5964 | { |
| 5965 | Lisp_Object eol_type; |
| 5966 | |
| 5967 | /* Recover the original EOL format. */ |
| 5968 | if (coding->eol_type == CODING_EOL_CR) |
| 5969 | { |
| 5970 | unsigned char *p; |
| 5971 | for (p = buf.data; p < buf.data + produced; p++) |
| 5972 | if (*p == '\n') *p = '\r'; |
| 5973 | } |
| 5974 | else if (coding->eol_type == CODING_EOL_CRLF) |
| 5975 | { |
| 5976 | int num_eol = 0; |
| 5977 | unsigned char *p0, *p1; |
| 5978 | for (p0 = buf.data, p1 = p0 + produced; p0 < p1; p0++) |
| 5979 | if (*p0 == '\n') num_eol++; |
| 5980 | if (produced + num_eol >= buf.size) |
| 5981 | extend_conversion_buffer (&buf); |
| 5982 | for (p0 = buf.data + produced, p1 = p0 + num_eol; p0 > buf.data;) |
| 5983 | { |
| 5984 | *--p1 = *--p0; |
| 5985 | if (*p0 == '\n') *--p1 = '\r'; |
| 5986 | } |
| 5987 | produced += num_eol; |
| 5988 | produced_char += num_eol; |
| 5989 | } |
| 5990 | /* Suppress eol-format conversion in the further conversion. */ |
| 5991 | coding->eol_type = CODING_EOL_LF; |
| 5992 | |
| 5993 | /* Set the coding system symbol to that for Unix-like EOL. */ |
| 5994 | eol_type = Fget (saved_coding_symbol, Qeol_type); |
| 5995 | if (VECTORP (eol_type) |
| 5996 | && XVECTOR (eol_type)->size == 3 |
| 5997 | && SYMBOLP (XVECTOR (eol_type)->contents[CODING_EOL_LF])) |
| 5998 | coding->symbol = XVECTOR (eol_type)->contents[CODING_EOL_LF]; |
| 5999 | else |
| 6000 | coding->symbol = saved_coding_symbol; |
| 6001 | |
| 6002 | |
| 6003 | } |
| 6004 | } |
| 6005 | |
| 6006 | coding->consumed = consumed; |
| 6007 | coding->consumed_char = consumed_char; |
| 6008 | coding->produced = produced; |
| 6009 | coding->produced_char = produced_char; |
| 6010 | |
| 6011 | if (coding->dst_multibyte) |
| 6012 | newstr = make_uninit_multibyte_string (produced_char + shrinked_bytes, |
| 6013 | produced + shrinked_bytes); |
| 6014 | else |
| 6015 | newstr = make_uninit_string (produced + shrinked_bytes); |
| 6016 | if (from > 0) |
| 6017 | bcopy (XSTRING (str)->data, XSTRING (newstr)->data, from); |
| 6018 | bcopy (buf.data, XSTRING (newstr)->data + from, produced); |
| 6019 | if (shrinked_bytes > from) |
| 6020 | bcopy (XSTRING (str)->data + to_byte, |
| 6021 | XSTRING (newstr)->data + from + produced, |
| 6022 | shrinked_bytes - from); |
| 6023 | free_conversion_buffer (&buf); |
| 6024 | |
| 6025 | if (coding->cmp_data && coding->cmp_data->used) |
| 6026 | coding_restore_composition (coding, newstr); |
| 6027 | coding_free_composition_data (coding); |
| 6028 | |
| 6029 | if (SYMBOLP (coding->post_read_conversion) |
| 6030 | && !NILP (Ffboundp (coding->post_read_conversion))) |
| 6031 | newstr = run_pre_post_conversion_on_str (newstr, coding, 0); |
| 6032 | |
| 6033 | return newstr; |
| 6034 | } |
| 6035 | |
| 6036 | Lisp_Object |
| 6037 | encode_coding_string (str, coding, nocopy) |
| 6038 | Lisp_Object str; |
| 6039 | struct coding_system *coding; |
| 6040 | int nocopy; |
| 6041 | { |
| 6042 | int len; |
| 6043 | struct conversion_buffer buf; |
| 6044 | int from, to, to_byte; |
| 6045 | int result; |
| 6046 | int shrinked_bytes = 0; |
| 6047 | Lisp_Object newstr; |
| 6048 | int consumed, consumed_char, produced, produced_char; |
| 6049 | |
| 6050 | if (SYMBOLP (coding->pre_write_conversion) |
| 6051 | && !NILP (Ffboundp (coding->pre_write_conversion))) |
| 6052 | str = run_pre_post_conversion_on_str (str, coding, 1); |
| 6053 | |
| 6054 | from = 0; |
| 6055 | to = XSTRING (str)->size; |
| 6056 | to_byte = STRING_BYTES (XSTRING (str)); |
| 6057 | |
| 6058 | /* Encoding routines determine the multibyteness of the source text |
| 6059 | by coding->src_multibyte. */ |
| 6060 | coding->src_multibyte = STRING_MULTIBYTE (str); |
| 6061 | coding->dst_multibyte = 0; |
| 6062 | if (! CODING_REQUIRE_ENCODING (coding)) |
| 6063 | { |
| 6064 | coding->consumed = STRING_BYTES (XSTRING (str)); |
| 6065 | coding->consumed_char = XSTRING (str)->size; |
| 6066 | if (STRING_MULTIBYTE (str)) |
| 6067 | { |
| 6068 | str = Fstring_as_unibyte (str); |
| 6069 | nocopy = 1; |
| 6070 | } |
| 6071 | coding->produced = STRING_BYTES (XSTRING (str)); |
| 6072 | coding->produced_char = XSTRING (str)->size; |
| 6073 | return (nocopy ? str : Fcopy_sequence (str)); |
| 6074 | } |
| 6075 | |
| 6076 | if (coding->composing != COMPOSITION_DISABLED) |
| 6077 | coding_save_composition (coding, from, to, str); |
| 6078 | |
| 6079 | /* Try to skip the heading and tailing ASCIIs. */ |
| 6080 | if (coding->type != coding_type_ccl) |
| 6081 | { |
| 6082 | SHRINK_CONVERSION_REGION (&from, &to_byte, coding, XSTRING (str)->data, |
| 6083 | 1); |
| 6084 | if (from == to_byte) |
| 6085 | return (nocopy ? str : Fcopy_sequence (str)); |
| 6086 | shrinked_bytes = from + (STRING_BYTES (XSTRING (str)) - to_byte); |
| 6087 | } |
| 6088 | |
| 6089 | len = encoding_buffer_size (coding, to_byte - from); |
| 6090 | allocate_conversion_buffer (buf, len); |
| 6091 | |
| 6092 | consumed = consumed_char = produced = produced_char = 0; |
| 6093 | while (1) |
| 6094 | { |
| 6095 | result = encode_coding (coding, XSTRING (str)->data + from + consumed, |
| 6096 | buf.data + produced, to_byte - from - consumed, |
| 6097 | buf.size - produced); |
| 6098 | consumed += coding->consumed; |
| 6099 | consumed_char += coding->consumed_char; |
| 6100 | produced += coding->produced; |
| 6101 | produced_char += coding->produced_char; |
| 6102 | if (result == CODING_FINISH_NORMAL |
| 6103 | || (result == CODING_FINISH_INSUFFICIENT_SRC |
| 6104 | && coding->consumed == 0)) |
| 6105 | break; |
| 6106 | /* Now result should be CODING_FINISH_INSUFFICIENT_DST. */ |
| 6107 | extend_conversion_buffer (&buf); |
| 6108 | } |
| 6109 | |
| 6110 | coding->consumed = consumed; |
| 6111 | coding->consumed_char = consumed_char; |
| 6112 | coding->produced = produced; |
| 6113 | coding->produced_char = produced_char; |
| 6114 | |
| 6115 | newstr = make_uninit_string (produced + shrinked_bytes); |
| 6116 | if (from > 0) |
| 6117 | bcopy (XSTRING (str)->data, XSTRING (newstr)->data, from); |
| 6118 | bcopy (buf.data, XSTRING (newstr)->data + from, produced); |
| 6119 | if (shrinked_bytes > from) |
| 6120 | bcopy (XSTRING (str)->data + to_byte, |
| 6121 | XSTRING (newstr)->data + from + produced, |
| 6122 | shrinked_bytes - from); |
| 6123 | |
| 6124 | free_conversion_buffer (&buf); |
| 6125 | coding_free_composition_data (coding); |
| 6126 | |
| 6127 | return newstr; |
| 6128 | } |
| 6129 | |
| 6130 | \f |
| 6131 | #ifdef emacs |
| 6132 | /*** 8. Emacs Lisp library functions ***/ |
| 6133 | |
| 6134 | DEFUN ("coding-system-p", Fcoding_system_p, Scoding_system_p, 1, 1, 0, |
| 6135 | doc: /* Return t if OBJECT is nil or a coding-system. |
| 6136 | See the documentation of `make-coding-system' for information |
| 6137 | about coding-system objects. */) |
| 6138 | (obj) |
| 6139 | Lisp_Object obj; |
| 6140 | { |
| 6141 | if (NILP (obj)) |
| 6142 | return Qt; |
| 6143 | if (!SYMBOLP (obj)) |
| 6144 | return Qnil; |
| 6145 | /* Get coding-spec vector for OBJ. */ |
| 6146 | obj = Fget (obj, Qcoding_system); |
| 6147 | return ((VECTORP (obj) && XVECTOR (obj)->size == 5) |
| 6148 | ? Qt : Qnil); |
| 6149 | } |
| 6150 | |
| 6151 | DEFUN ("read-non-nil-coding-system", Fread_non_nil_coding_system, |
| 6152 | Sread_non_nil_coding_system, 1, 1, 0, |
| 6153 | doc: /* Read a coding system from the minibuffer, prompting with string PROMPT. */) |
| 6154 | (prompt) |
| 6155 | Lisp_Object prompt; |
| 6156 | { |
| 6157 | Lisp_Object val; |
| 6158 | do |
| 6159 | { |
| 6160 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
| 6161 | Qt, Qnil, Qcoding_system_history, Qnil, Qnil); |
| 6162 | } |
| 6163 | while (XSTRING (val)->size == 0); |
| 6164 | return (Fintern (val, Qnil)); |
| 6165 | } |
| 6166 | |
| 6167 | DEFUN ("read-coding-system", Fread_coding_system, Sread_coding_system, 1, 2, 0, |
| 6168 | doc: /* Read a coding system from the minibuffer, prompting with string PROMPT. |
| 6169 | If the user enters null input, return second argument DEFAULT-CODING-SYSTEM. */) |
| 6170 | (prompt, default_coding_system) |
| 6171 | Lisp_Object prompt, default_coding_system; |
| 6172 | { |
| 6173 | Lisp_Object val; |
| 6174 | if (SYMBOLP (default_coding_system)) |
| 6175 | default_coding_system = SYMBOL_NAME (default_coding_system); |
| 6176 | val = Fcompleting_read (prompt, Vcoding_system_alist, Qnil, |
| 6177 | Qt, Qnil, Qcoding_system_history, |
| 6178 | default_coding_system, Qnil); |
| 6179 | return (XSTRING (val)->size == 0 ? Qnil : Fintern (val, Qnil)); |
| 6180 | } |
| 6181 | |
| 6182 | DEFUN ("check-coding-system", Fcheck_coding_system, Scheck_coding_system, |
| 6183 | 1, 1, 0, |
| 6184 | doc: /* Check validity of CODING-SYSTEM. |
| 6185 | If valid, return CODING-SYSTEM, else signal a `coding-system-error' error. |
| 6186 | It is valid if it is a symbol with a non-nil `coding-system' property. |
| 6187 | The value of property should be a vector of length 5. */) |
| 6188 | (coding_system) |
| 6189 | Lisp_Object coding_system; |
| 6190 | { |
| 6191 | CHECK_SYMBOL (coding_system); |
| 6192 | if (!NILP (Fcoding_system_p (coding_system))) |
| 6193 | return coding_system; |
| 6194 | while (1) |
| 6195 | Fsignal (Qcoding_system_error, Fcons (coding_system, Qnil)); |
| 6196 | } |
| 6197 | \f |
| 6198 | Lisp_Object |
| 6199 | detect_coding_system (src, src_bytes, highest, multibytep) |
| 6200 | unsigned char *src; |
| 6201 | int src_bytes, highest; |
| 6202 | int multibytep; |
| 6203 | { |
| 6204 | int coding_mask, eol_type; |
| 6205 | Lisp_Object val, tmp; |
| 6206 | int dummy; |
| 6207 | |
| 6208 | coding_mask = detect_coding_mask (src, src_bytes, NULL, &dummy, multibytep); |
| 6209 | eol_type = detect_eol_type (src, src_bytes, &dummy); |
| 6210 | if (eol_type == CODING_EOL_INCONSISTENT) |
| 6211 | eol_type = CODING_EOL_UNDECIDED; |
| 6212 | |
| 6213 | if (!coding_mask) |
| 6214 | { |
| 6215 | val = Qundecided; |
| 6216 | if (eol_type != CODING_EOL_UNDECIDED) |
| 6217 | { |
| 6218 | Lisp_Object val2; |
| 6219 | val2 = Fget (Qundecided, Qeol_type); |
| 6220 | if (VECTORP (val2)) |
| 6221 | val = XVECTOR (val2)->contents[eol_type]; |
| 6222 | } |
| 6223 | return (highest ? val : Fcons (val, Qnil)); |
| 6224 | } |
| 6225 | |
| 6226 | /* At first, gather possible coding systems in VAL. */ |
| 6227 | val = Qnil; |
| 6228 | for (tmp = Vcoding_category_list; CONSP (tmp); tmp = XCDR (tmp)) |
| 6229 | { |
| 6230 | Lisp_Object category_val, category_index; |
| 6231 | |
| 6232 | category_index = Fget (XCAR (tmp), Qcoding_category_index); |
| 6233 | category_val = Fsymbol_value (XCAR (tmp)); |
| 6234 | if (!NILP (category_val) |
| 6235 | && NATNUMP (category_index) |
| 6236 | && (coding_mask & (1 << XFASTINT (category_index)))) |
| 6237 | { |
| 6238 | val = Fcons (category_val, val); |
| 6239 | if (highest) |
| 6240 | break; |
| 6241 | } |
| 6242 | } |
| 6243 | if (!highest) |
| 6244 | val = Fnreverse (val); |
| 6245 | |
| 6246 | /* Then, replace the elements with subsidiary coding systems. */ |
| 6247 | for (tmp = val; CONSP (tmp); tmp = XCDR (tmp)) |
| 6248 | { |
| 6249 | if (eol_type != CODING_EOL_UNDECIDED |
| 6250 | && eol_type != CODING_EOL_INCONSISTENT) |
| 6251 | { |
| 6252 | Lisp_Object eol; |
| 6253 | eol = Fget (XCAR (tmp), Qeol_type); |
| 6254 | if (VECTORP (eol)) |
| 6255 | XSETCAR (tmp, XVECTOR (eol)->contents[eol_type]); |
| 6256 | } |
| 6257 | } |
| 6258 | return (highest ? XCAR (val) : val); |
| 6259 | } |
| 6260 | |
| 6261 | DEFUN ("detect-coding-region", Fdetect_coding_region, Sdetect_coding_region, |
| 6262 | 2, 3, 0, |
| 6263 | doc: /* Detect coding system of the text in the region between START and END. |
| 6264 | Return a list of possible coding systems ordered by priority. |
| 6265 | |
| 6266 | If only ASCII characters are found, it returns a list of single element |
| 6267 | `undecided' or its subsidiary coding system according to a detected |
| 6268 | end-of-line format. |
| 6269 | |
| 6270 | If optional argument HIGHEST is non-nil, return the coding system of |
| 6271 | highest priority. */) |
| 6272 | (start, end, highest) |
| 6273 | Lisp_Object start, end, highest; |
| 6274 | { |
| 6275 | int from, to; |
| 6276 | int from_byte, to_byte; |
| 6277 | int include_anchor_byte = 0; |
| 6278 | |
| 6279 | CHECK_NUMBER_COERCE_MARKER (start); |
| 6280 | CHECK_NUMBER_COERCE_MARKER (end); |
| 6281 | |
| 6282 | validate_region (&start, &end); |
| 6283 | from = XINT (start), to = XINT (end); |
| 6284 | from_byte = CHAR_TO_BYTE (from); |
| 6285 | to_byte = CHAR_TO_BYTE (to); |
| 6286 | |
| 6287 | if (from < GPT && to >= GPT) |
| 6288 | move_gap_both (to, to_byte); |
| 6289 | /* If we an anchor byte `\0' follows the region, we include it in |
| 6290 | the detecting source. Then code detectors can handle the tailing |
| 6291 | byte sequence more accurately. |
| 6292 | |
| 6293 | Fix me: This is not an perfect solution. It is better that we |
| 6294 | add one more argument, say LAST_BLOCK, to all detect_coding_XXX. |
| 6295 | */ |
| 6296 | if (to == Z || (to == GPT && GAP_SIZE > 0)) |
| 6297 | include_anchor_byte = 1; |
| 6298 | return detect_coding_system (BYTE_POS_ADDR (from_byte), |
| 6299 | to_byte - from_byte + include_anchor_byte, |
| 6300 | !NILP (highest), |
| 6301 | !NILP (current_buffer |
| 6302 | ->enable_multibyte_characters)); |
| 6303 | } |
| 6304 | |
| 6305 | DEFUN ("detect-coding-string", Fdetect_coding_string, Sdetect_coding_string, |
| 6306 | 1, 2, 0, |
| 6307 | doc: /* Detect coding system of the text in STRING. |
| 6308 | Return a list of possible coding systems ordered by priority. |
| 6309 | |
| 6310 | If only ASCII characters are found, it returns a list of single element |
| 6311 | `undecided' or its subsidiary coding system according to a detected |
| 6312 | end-of-line format. |
| 6313 | |
| 6314 | If optional argument HIGHEST is non-nil, return the coding system of |
| 6315 | highest priority. */) |
| 6316 | (string, highest) |
| 6317 | Lisp_Object string, highest; |
| 6318 | { |
| 6319 | CHECK_STRING (string); |
| 6320 | |
| 6321 | return detect_coding_system (XSTRING (string)->data, |
| 6322 | /* "+ 1" is to include the anchor byte |
| 6323 | `\0'. With this, code detectors can |
| 6324 | handle the tailing bytes more |
| 6325 | accurately. */ |
| 6326 | STRING_BYTES (XSTRING (string)) + 1, |
| 6327 | !NILP (highest), |
| 6328 | STRING_MULTIBYTE (string)); |
| 6329 | } |
| 6330 | |
| 6331 | /* Return an intersection of lists L1 and L2. */ |
| 6332 | |
| 6333 | static Lisp_Object |
| 6334 | intersection (l1, l2) |
| 6335 | Lisp_Object l1, l2; |
| 6336 | { |
| 6337 | Lisp_Object val = Fcons (Qnil, Qnil), tail; |
| 6338 | |
| 6339 | for (tail = val; CONSP (l1); l1 = XCDR (l1)) |
| 6340 | { |
| 6341 | if (!NILP (Fmemq (XCAR (l1), l2))) |
| 6342 | { |
| 6343 | XSETCDR (tail, Fcons (XCAR (l1), Qnil)); |
| 6344 | tail = XCDR (tail); |
| 6345 | } |
| 6346 | } |
| 6347 | return XCDR (val); |
| 6348 | } |
| 6349 | |
| 6350 | |
| 6351 | /* Subroutine for Fsafe_coding_systems_region_internal. |
| 6352 | |
| 6353 | Return a list of coding systems that safely encode the multibyte |
| 6354 | text between P and PEND. SAFE_CODINGS, if non-nil, is a list of |
| 6355 | possible coding systems. If it is nil, it means that we have not |
| 6356 | yet found any coding systems. |
| 6357 | |
| 6358 | WORK_TABLE is a copy of the char-table Vchar_coding_system_table. An |
| 6359 | element of WORK_TABLE is set to t once the element is looked up. |
| 6360 | |
| 6361 | If a non-ASCII single byte char is found, set |
| 6362 | *single_byte_char_found to 1. */ |
| 6363 | |
| 6364 | static Lisp_Object |
| 6365 | find_safe_codings (p, pend, safe_codings, work_table, single_byte_char_found) |
| 6366 | unsigned char *p, *pend; |
| 6367 | Lisp_Object safe_codings, work_table; |
| 6368 | int *single_byte_char_found; |
| 6369 | { |
| 6370 | int c, len, idx; |
| 6371 | Lisp_Object val; |
| 6372 | |
| 6373 | while (p < pend) |
| 6374 | { |
| 6375 | c = STRING_CHAR_AND_LENGTH (p, pend - p, len); |
| 6376 | p += len; |
| 6377 | if (ASCII_BYTE_P (c)) |
| 6378 | /* We can ignore ASCII characters here. */ |
| 6379 | continue; |
| 6380 | if (SINGLE_BYTE_CHAR_P (c)) |
| 6381 | *single_byte_char_found = 1; |
| 6382 | if (NILP (safe_codings)) |
| 6383 | continue; |
| 6384 | /* Check the safe coding systems for C. */ |
| 6385 | val = char_table_ref_and_index (work_table, c, &idx); |
| 6386 | if (EQ (val, Qt)) |
| 6387 | /* This element was already checked. Ignore it. */ |
| 6388 | continue; |
| 6389 | /* Remember that we checked this element. */ |
| 6390 | CHAR_TABLE_SET (work_table, make_number (idx), Qt); |
| 6391 | |
| 6392 | /* If there are some safe coding systems for C and we have |
| 6393 | already found the other set of coding systems for the |
| 6394 | different characters, get the intersection of them. */ |
| 6395 | if (!EQ (safe_codings, Qt) && !NILP (val)) |
| 6396 | val = intersection (safe_codings, val); |
| 6397 | safe_codings = val; |
| 6398 | } |
| 6399 | return safe_codings; |
| 6400 | } |
| 6401 | |
| 6402 | |
| 6403 | /* Return a list of coding systems that safely encode the text between |
| 6404 | START and END. If the text contains only ASCII or is unibyte, |
| 6405 | return t. */ |
| 6406 | |
| 6407 | DEFUN ("find-coding-systems-region-internal", |
| 6408 | Ffind_coding_systems_region_internal, |
| 6409 | Sfind_coding_systems_region_internal, 2, 2, 0, |
| 6410 | doc: /* Internal use only. */) |
| 6411 | (start, end) |
| 6412 | Lisp_Object start, end; |
| 6413 | { |
| 6414 | Lisp_Object work_table, safe_codings; |
| 6415 | int non_ascii_p = 0; |
| 6416 | int single_byte_char_found = 0; |
| 6417 | unsigned char *p1, *p1end, *p2, *p2end, *p; |
| 6418 | |
| 6419 | if (STRINGP (start)) |
| 6420 | { |
| 6421 | if (!STRING_MULTIBYTE (start)) |
| 6422 | return Qt; |
| 6423 | p1 = XSTRING (start)->data, p1end = p1 + STRING_BYTES (XSTRING (start)); |
| 6424 | p2 = p2end = p1end; |
| 6425 | if (XSTRING (start)->size != STRING_BYTES (XSTRING (start))) |
| 6426 | non_ascii_p = 1; |
| 6427 | } |
| 6428 | else |
| 6429 | { |
| 6430 | int from, to, stop; |
| 6431 | |
| 6432 | CHECK_NUMBER_COERCE_MARKER (start); |
| 6433 | CHECK_NUMBER_COERCE_MARKER (end); |
| 6434 | if (XINT (start) < BEG || XINT (end) > Z || XINT (start) > XINT (end)) |
| 6435 | args_out_of_range (start, end); |
| 6436 | if (NILP (current_buffer->enable_multibyte_characters)) |
| 6437 | return Qt; |
| 6438 | from = CHAR_TO_BYTE (XINT (start)); |
| 6439 | to = CHAR_TO_BYTE (XINT (end)); |
| 6440 | stop = from < GPT_BYTE && GPT_BYTE < to ? GPT_BYTE : to; |
| 6441 | p1 = BYTE_POS_ADDR (from), p1end = p1 + (stop - from); |
| 6442 | if (stop == to) |
| 6443 | p2 = p2end = p1end; |
| 6444 | else |
| 6445 | p2 = BYTE_POS_ADDR (stop), p2end = p2 + (to - stop); |
| 6446 | if (XINT (end) - XINT (start) != to - from) |
| 6447 | non_ascii_p = 1; |
| 6448 | } |
| 6449 | |
| 6450 | if (!non_ascii_p) |
| 6451 | { |
| 6452 | /* We are sure that the text contains no multibyte character. |
| 6453 | Check if it contains eight-bit-graphic. */ |
| 6454 | p = p1; |
| 6455 | for (p = p1; p < p1end && ASCII_BYTE_P (*p); p++); |
| 6456 | if (p == p1end) |
| 6457 | { |
| 6458 | for (p = p2; p < p2end && ASCII_BYTE_P (*p); p++); |
| 6459 | if (p == p2end) |
| 6460 | return Qt; |
| 6461 | } |
| 6462 | } |
| 6463 | |
| 6464 | /* The text contains non-ASCII characters. */ |
| 6465 | work_table = Fcopy_sequence (Vchar_coding_system_table); |
| 6466 | safe_codings = find_safe_codings (p1, p1end, Qt, work_table, |
| 6467 | &single_byte_char_found); |
| 6468 | if (p2 < p2end) |
| 6469 | safe_codings = find_safe_codings (p2, p2end, safe_codings, work_table, |
| 6470 | &single_byte_char_found); |
| 6471 | |
| 6472 | if (EQ (safe_codings, Qt)) |
| 6473 | ; /* Nothing to be done. */ |
| 6474 | else if (!single_byte_char_found) |
| 6475 | { |
| 6476 | /* Append generic coding systems. */ |
| 6477 | Lisp_Object args[2]; |
| 6478 | args[0] = safe_codings; |
| 6479 | args[1] = Fchar_table_extra_slot (Vchar_coding_system_table, |
| 6480 | make_number (0)); |
| 6481 | safe_codings = Fappend (2, args); |
| 6482 | } |
| 6483 | else |
| 6484 | safe_codings = Fcons (Qraw_text, |
| 6485 | Fcons (Qemacs_mule, |
| 6486 | Fcons (Qno_conversion, safe_codings))); |
| 6487 | return safe_codings; |
| 6488 | } |
| 6489 | |
| 6490 | |
| 6491 | Lisp_Object |
| 6492 | code_convert_region1 (start, end, coding_system, encodep) |
| 6493 | Lisp_Object start, end, coding_system; |
| 6494 | int encodep; |
| 6495 | { |
| 6496 | struct coding_system coding; |
| 6497 | int from, to; |
| 6498 | |
| 6499 | CHECK_NUMBER_COERCE_MARKER (start); |
| 6500 | CHECK_NUMBER_COERCE_MARKER (end); |
| 6501 | CHECK_SYMBOL (coding_system); |
| 6502 | |
| 6503 | validate_region (&start, &end); |
| 6504 | from = XFASTINT (start); |
| 6505 | to = XFASTINT (end); |
| 6506 | |
| 6507 | if (NILP (coding_system)) |
| 6508 | return make_number (to - from); |
| 6509 | |
| 6510 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
| 6511 | error ("Invalid coding system: %s", XSTRING (SYMBOL_NAME (coding_system))->data); |
| 6512 | |
| 6513 | coding.mode |= CODING_MODE_LAST_BLOCK; |
| 6514 | coding.src_multibyte = coding.dst_multibyte |
| 6515 | = !NILP (current_buffer->enable_multibyte_characters); |
| 6516 | code_convert_region (from, CHAR_TO_BYTE (from), to, CHAR_TO_BYTE (to), |
| 6517 | &coding, encodep, 1); |
| 6518 | Vlast_coding_system_used = coding.symbol; |
| 6519 | return make_number (coding.produced_char); |
| 6520 | } |
| 6521 | |
| 6522 | DEFUN ("decode-coding-region", Fdecode_coding_region, Sdecode_coding_region, |
| 6523 | 3, 3, "r\nzCoding system: ", |
| 6524 | doc: /* Decode the current region from the specified coding system. |
| 6525 | When called from a program, takes three arguments: |
| 6526 | START, END, and CODING-SYSTEM. START and END are buffer positions. |
| 6527 | This function sets `last-coding-system-used' to the precise coding system |
| 6528 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is |
| 6529 | not fully specified.) |
| 6530 | It returns the length of the decoded text. */) |
| 6531 | (start, end, coding_system) |
| 6532 | Lisp_Object start, end, coding_system; |
| 6533 | { |
| 6534 | return code_convert_region1 (start, end, coding_system, 0); |
| 6535 | } |
| 6536 | |
| 6537 | DEFUN ("encode-coding-region", Fencode_coding_region, Sencode_coding_region, |
| 6538 | 3, 3, "r\nzCoding system: ", |
| 6539 | doc: /* Encode the current region into the specified coding system. |
| 6540 | When called from a program, takes three arguments: |
| 6541 | START, END, and CODING-SYSTEM. START and END are buffer positions. |
| 6542 | This function sets `last-coding-system-used' to the precise coding system |
| 6543 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is |
| 6544 | not fully specified.) |
| 6545 | It returns the length of the encoded text. */) |
| 6546 | (start, end, coding_system) |
| 6547 | Lisp_Object start, end, coding_system; |
| 6548 | { |
| 6549 | return code_convert_region1 (start, end, coding_system, 1); |
| 6550 | } |
| 6551 | |
| 6552 | Lisp_Object |
| 6553 | code_convert_string1 (string, coding_system, nocopy, encodep) |
| 6554 | Lisp_Object string, coding_system, nocopy; |
| 6555 | int encodep; |
| 6556 | { |
| 6557 | struct coding_system coding; |
| 6558 | |
| 6559 | CHECK_STRING (string); |
| 6560 | CHECK_SYMBOL (coding_system); |
| 6561 | |
| 6562 | if (NILP (coding_system)) |
| 6563 | return (NILP (nocopy) ? Fcopy_sequence (string) : string); |
| 6564 | |
| 6565 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
| 6566 | error ("Invalid coding system: %s", XSTRING (SYMBOL_NAME (coding_system))->data); |
| 6567 | |
| 6568 | coding.mode |= CODING_MODE_LAST_BLOCK; |
| 6569 | string = (encodep |
| 6570 | ? encode_coding_string (string, &coding, !NILP (nocopy)) |
| 6571 | : decode_coding_string (string, &coding, !NILP (nocopy))); |
| 6572 | Vlast_coding_system_used = coding.symbol; |
| 6573 | |
| 6574 | return string; |
| 6575 | } |
| 6576 | |
| 6577 | DEFUN ("decode-coding-string", Fdecode_coding_string, Sdecode_coding_string, |
| 6578 | 2, 3, 0, |
| 6579 | doc: /* Decode STRING which is encoded in CODING-SYSTEM, and return the result. |
| 6580 | Optional arg NOCOPY non-nil means it is OK to return STRING itself |
| 6581 | if the decoding operation is trivial. |
| 6582 | This function sets `last-coding-system-used' to the precise coding system |
| 6583 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is |
| 6584 | not fully specified.) */) |
| 6585 | (string, coding_system, nocopy) |
| 6586 | Lisp_Object string, coding_system, nocopy; |
| 6587 | { |
| 6588 | return code_convert_string1 (string, coding_system, nocopy, 0); |
| 6589 | } |
| 6590 | |
| 6591 | DEFUN ("encode-coding-string", Fencode_coding_string, Sencode_coding_string, |
| 6592 | 2, 3, 0, |
| 6593 | doc: /* Encode STRING to CODING-SYSTEM, and return the result. |
| 6594 | Optional arg NOCOPY non-nil means it is OK to return STRING itself |
| 6595 | if the encoding operation is trivial. |
| 6596 | This function sets `last-coding-system-used' to the precise coding system |
| 6597 | used (which may be different from CODING-SYSTEM if CODING-SYSTEM is |
| 6598 | not fully specified.) */) |
| 6599 | (string, coding_system, nocopy) |
| 6600 | Lisp_Object string, coding_system, nocopy; |
| 6601 | { |
| 6602 | return code_convert_string1 (string, coding_system, nocopy, 1); |
| 6603 | } |
| 6604 | |
| 6605 | /* Encode or decode STRING according to CODING_SYSTEM. |
| 6606 | Do not set Vlast_coding_system_used. |
| 6607 | |
| 6608 | This function is called only from macros DECODE_FILE and |
| 6609 | ENCODE_FILE, thus we ignore character composition. */ |
| 6610 | |
| 6611 | Lisp_Object |
| 6612 | code_convert_string_norecord (string, coding_system, encodep) |
| 6613 | Lisp_Object string, coding_system; |
| 6614 | int encodep; |
| 6615 | { |
| 6616 | struct coding_system coding; |
| 6617 | |
| 6618 | CHECK_STRING (string); |
| 6619 | CHECK_SYMBOL (coding_system); |
| 6620 | |
| 6621 | if (NILP (coding_system)) |
| 6622 | return string; |
| 6623 | |
| 6624 | if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0) |
| 6625 | error ("Invalid coding system: %s", XSTRING (SYMBOL_NAME (coding_system))->data); |
| 6626 | |
| 6627 | coding.composing = COMPOSITION_DISABLED; |
| 6628 | coding.mode |= CODING_MODE_LAST_BLOCK; |
| 6629 | return (encodep |
| 6630 | ? encode_coding_string (string, &coding, 1) |
| 6631 | : decode_coding_string (string, &coding, 1)); |
| 6632 | } |
| 6633 | \f |
| 6634 | DEFUN ("decode-sjis-char", Fdecode_sjis_char, Sdecode_sjis_char, 1, 1, 0, |
| 6635 | doc: /* Decode a Japanese character which has CODE in shift_jis encoding. |
| 6636 | Return the corresponding character. */) |
| 6637 | (code) |
| 6638 | Lisp_Object code; |
| 6639 | { |
| 6640 | unsigned char c1, c2, s1, s2; |
| 6641 | Lisp_Object val; |
| 6642 | |
| 6643 | CHECK_NUMBER (code); |
| 6644 | s1 = (XFASTINT (code)) >> 8, s2 = (XFASTINT (code)) & 0xFF; |
| 6645 | if (s1 == 0) |
| 6646 | { |
| 6647 | if (s2 < 0x80) |
| 6648 | XSETFASTINT (val, s2); |
| 6649 | else if (s2 >= 0xA0 || s2 <= 0xDF) |
| 6650 | XSETFASTINT (val, MAKE_CHAR (charset_katakana_jisx0201, s2, 0)); |
| 6651 | else |
| 6652 | error ("Invalid Shift JIS code: %x", XFASTINT (code)); |
| 6653 | } |
| 6654 | else |
| 6655 | { |
| 6656 | if ((s1 < 0x80 || (s1 > 0x9F && s1 < 0xE0) || s1 > 0xEF) |
| 6657 | || (s2 < 0x40 || s2 == 0x7F || s2 > 0xFC)) |
| 6658 | error ("Invalid Shift JIS code: %x", XFASTINT (code)); |
| 6659 | DECODE_SJIS (s1, s2, c1, c2); |
| 6660 | XSETFASTINT (val, MAKE_CHAR (charset_jisx0208, c1, c2)); |
| 6661 | } |
| 6662 | return val; |
| 6663 | } |
| 6664 | |
| 6665 | DEFUN ("encode-sjis-char", Fencode_sjis_char, Sencode_sjis_char, 1, 1, 0, |
| 6666 | doc: /* Encode a Japanese character CHAR to shift_jis encoding. |
| 6667 | Return the corresponding code in SJIS. */) |
| 6668 | (ch) |
| 6669 | Lisp_Object ch; |
| 6670 | { |
| 6671 | int charset, c1, c2, s1, s2; |
| 6672 | Lisp_Object val; |
| 6673 | |
| 6674 | CHECK_NUMBER (ch); |
| 6675 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
| 6676 | if (charset == CHARSET_ASCII) |
| 6677 | { |
| 6678 | val = ch; |
| 6679 | } |
| 6680 | else if (charset == charset_jisx0208 |
| 6681 | && c1 > 0x20 && c1 < 0x7F && c2 > 0x20 && c2 < 0x7F) |
| 6682 | { |
| 6683 | ENCODE_SJIS (c1, c2, s1, s2); |
| 6684 | XSETFASTINT (val, (s1 << 8) | s2); |
| 6685 | } |
| 6686 | else if (charset == charset_katakana_jisx0201 |
| 6687 | && c1 > 0x20 && c2 < 0xE0) |
| 6688 | { |
| 6689 | XSETFASTINT (val, c1 | 0x80); |
| 6690 | } |
| 6691 | else |
| 6692 | error ("Can't encode to shift_jis: %d", XFASTINT (ch)); |
| 6693 | return val; |
| 6694 | } |
| 6695 | |
| 6696 | DEFUN ("decode-big5-char", Fdecode_big5_char, Sdecode_big5_char, 1, 1, 0, |
| 6697 | doc: /* Decode a Big5 character which has CODE in BIG5 coding system. |
| 6698 | Return the corresponding character. */) |
| 6699 | (code) |
| 6700 | Lisp_Object code; |
| 6701 | { |
| 6702 | int charset; |
| 6703 | unsigned char b1, b2, c1, c2; |
| 6704 | Lisp_Object val; |
| 6705 | |
| 6706 | CHECK_NUMBER (code); |
| 6707 | b1 = (XFASTINT (code)) >> 8, b2 = (XFASTINT (code)) & 0xFF; |
| 6708 | if (b1 == 0) |
| 6709 | { |
| 6710 | if (b2 >= 0x80) |
| 6711 | error ("Invalid BIG5 code: %x", XFASTINT (code)); |
| 6712 | val = code; |
| 6713 | } |
| 6714 | else |
| 6715 | { |
| 6716 | if ((b1 < 0xA1 || b1 > 0xFE) |
| 6717 | || (b2 < 0x40 || (b2 > 0x7E && b2 < 0xA1) || b2 > 0xFE)) |
| 6718 | error ("Invalid BIG5 code: %x", XFASTINT (code)); |
| 6719 | DECODE_BIG5 (b1, b2, charset, c1, c2); |
| 6720 | XSETFASTINT (val, MAKE_CHAR (charset, c1, c2)); |
| 6721 | } |
| 6722 | return val; |
| 6723 | } |
| 6724 | |
| 6725 | DEFUN ("encode-big5-char", Fencode_big5_char, Sencode_big5_char, 1, 1, 0, |
| 6726 | doc: /* Encode the Big5 character CHAR to BIG5 coding system. |
| 6727 | Return the corresponding character code in Big5. */) |
| 6728 | (ch) |
| 6729 | Lisp_Object ch; |
| 6730 | { |
| 6731 | int charset, c1, c2, b1, b2; |
| 6732 | Lisp_Object val; |
| 6733 | |
| 6734 | CHECK_NUMBER (ch); |
| 6735 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
| 6736 | if (charset == CHARSET_ASCII) |
| 6737 | { |
| 6738 | val = ch; |
| 6739 | } |
| 6740 | else if ((charset == charset_big5_1 |
| 6741 | && (XFASTINT (ch) >= 0x250a1 && XFASTINT (ch) <= 0x271ec)) |
| 6742 | || (charset == charset_big5_2 |
| 6743 | && XFASTINT (ch) >= 0x290a1 && XFASTINT (ch) <= 0x2bdb2)) |
| 6744 | { |
| 6745 | ENCODE_BIG5 (charset, c1, c2, b1, b2); |
| 6746 | XSETFASTINT (val, (b1 << 8) | b2); |
| 6747 | } |
| 6748 | else |
| 6749 | error ("Can't encode to Big5: %d", XFASTINT (ch)); |
| 6750 | return val; |
| 6751 | } |
| 6752 | \f |
| 6753 | DEFUN ("set-terminal-coding-system-internal", |
| 6754 | Fset_terminal_coding_system_internal, |
| 6755 | Sset_terminal_coding_system_internal, 1, 1, 0, |
| 6756 | doc: /* Internal use only. */) |
| 6757 | (coding_system) |
| 6758 | Lisp_Object coding_system; |
| 6759 | { |
| 6760 | CHECK_SYMBOL (coding_system); |
| 6761 | setup_coding_system (Fcheck_coding_system (coding_system), &terminal_coding); |
| 6762 | /* We had better not send unsafe characters to terminal. */ |
| 6763 | terminal_coding.flags |= CODING_FLAG_ISO_SAFE; |
| 6764 | /* Character composition should be disabled. */ |
| 6765 | terminal_coding.composing = COMPOSITION_DISABLED; |
| 6766 | /* Error notification should be suppressed. */ |
| 6767 | terminal_coding.suppress_error = 1; |
| 6768 | terminal_coding.src_multibyte = 1; |
| 6769 | terminal_coding.dst_multibyte = 0; |
| 6770 | return Qnil; |
| 6771 | } |
| 6772 | |
| 6773 | DEFUN ("set-safe-terminal-coding-system-internal", |
| 6774 | Fset_safe_terminal_coding_system_internal, |
| 6775 | Sset_safe_terminal_coding_system_internal, 1, 1, 0, |
| 6776 | doc: /* Internal use only. */) |
| 6777 | (coding_system) |
| 6778 | Lisp_Object coding_system; |
| 6779 | { |
| 6780 | CHECK_SYMBOL (coding_system); |
| 6781 | setup_coding_system (Fcheck_coding_system (coding_system), |
| 6782 | &safe_terminal_coding); |
| 6783 | /* Character composition should be disabled. */ |
| 6784 | safe_terminal_coding.composing = COMPOSITION_DISABLED; |
| 6785 | /* Error notification should be suppressed. */ |
| 6786 | terminal_coding.suppress_error = 1; |
| 6787 | safe_terminal_coding.src_multibyte = 1; |
| 6788 | safe_terminal_coding.dst_multibyte = 0; |
| 6789 | return Qnil; |
| 6790 | } |
| 6791 | |
| 6792 | DEFUN ("terminal-coding-system", |
| 6793 | Fterminal_coding_system, Sterminal_coding_system, 0, 0, 0, |
| 6794 | doc: /* Return coding system specified for terminal output. */) |
| 6795 | () |
| 6796 | { |
| 6797 | return terminal_coding.symbol; |
| 6798 | } |
| 6799 | |
| 6800 | DEFUN ("set-keyboard-coding-system-internal", |
| 6801 | Fset_keyboard_coding_system_internal, |
| 6802 | Sset_keyboard_coding_system_internal, 1, 1, 0, |
| 6803 | doc: /* Internal use only. */) |
| 6804 | (coding_system) |
| 6805 | Lisp_Object coding_system; |
| 6806 | { |
| 6807 | CHECK_SYMBOL (coding_system); |
| 6808 | setup_coding_system (Fcheck_coding_system (coding_system), &keyboard_coding); |
| 6809 | /* Character composition should be disabled. */ |
| 6810 | keyboard_coding.composing = COMPOSITION_DISABLED; |
| 6811 | return Qnil; |
| 6812 | } |
| 6813 | |
| 6814 | DEFUN ("keyboard-coding-system", |
| 6815 | Fkeyboard_coding_system, Skeyboard_coding_system, 0, 0, 0, |
| 6816 | doc: /* Return coding system specified for decoding keyboard input. */) |
| 6817 | () |
| 6818 | { |
| 6819 | return keyboard_coding.symbol; |
| 6820 | } |
| 6821 | |
| 6822 | \f |
| 6823 | DEFUN ("find-operation-coding-system", Ffind_operation_coding_system, |
| 6824 | Sfind_operation_coding_system, 1, MANY, 0, |
| 6825 | doc: /* Choose a coding system for an operation based on the target name. |
| 6826 | The value names a pair of coding systems: (DECODING-SYSTEM . ENCODING-SYSTEM). |
| 6827 | DECODING-SYSTEM is the coding system to use for decoding |
| 6828 | \(in case OPERATION does decoding), and ENCODING-SYSTEM is the coding system |
| 6829 | for encoding (in case OPERATION does encoding). |
| 6830 | |
| 6831 | The first argument OPERATION specifies an I/O primitive: |
| 6832 | For file I/O, `insert-file-contents' or `write-region'. |
| 6833 | For process I/O, `call-process', `call-process-region', or `start-process'. |
| 6834 | For network I/O, `open-network-stream'. |
| 6835 | |
| 6836 | The remaining arguments should be the same arguments that were passed |
| 6837 | to the primitive. Depending on which primitive, one of those arguments |
| 6838 | is selected as the TARGET. For example, if OPERATION does file I/O, |
| 6839 | whichever argument specifies the file name is TARGET. |
| 6840 | |
| 6841 | TARGET has a meaning which depends on OPERATION: |
| 6842 | For file I/O, TARGET is a file name. |
| 6843 | For process I/O, TARGET is a process name. |
| 6844 | For network I/O, TARGET is a service name or a port number |
| 6845 | |
| 6846 | This function looks up what specified for TARGET in, |
| 6847 | `file-coding-system-alist', `process-coding-system-alist', |
| 6848 | or `network-coding-system-alist' depending on OPERATION. |
| 6849 | They may specify a coding system, a cons of coding systems, |
| 6850 | or a function symbol to call. |
| 6851 | In the last case, we call the function with one argument, |
| 6852 | which is a list of all the arguments given to this function. |
| 6853 | |
| 6854 | usage: (find-operation-coding-system OPERATION ARGUMENTS ...) */) |
| 6855 | (nargs, args) |
| 6856 | int nargs; |
| 6857 | Lisp_Object *args; |
| 6858 | { |
| 6859 | Lisp_Object operation, target_idx, target, val; |
| 6860 | register Lisp_Object chain; |
| 6861 | |
| 6862 | if (nargs < 2) |
| 6863 | error ("Too few arguments"); |
| 6864 | operation = args[0]; |
| 6865 | if (!SYMBOLP (operation) |
| 6866 | || !INTEGERP (target_idx = Fget (operation, Qtarget_idx))) |
| 6867 | error ("Invalid first argument"); |
| 6868 | if (nargs < 1 + XINT (target_idx)) |
| 6869 | error ("Too few arguments for operation: %s", |
| 6870 | XSTRING (SYMBOL_NAME (operation))->data); |
| 6871 | target = args[XINT (target_idx) + 1]; |
| 6872 | if (!(STRINGP (target) |
| 6873 | || (EQ (operation, Qopen_network_stream) && INTEGERP (target)))) |
| 6874 | error ("Invalid argument %d", XINT (target_idx) + 1); |
| 6875 | |
| 6876 | chain = ((EQ (operation, Qinsert_file_contents) |
| 6877 | || EQ (operation, Qwrite_region)) |
| 6878 | ? Vfile_coding_system_alist |
| 6879 | : (EQ (operation, Qopen_network_stream) |
| 6880 | ? Vnetwork_coding_system_alist |
| 6881 | : Vprocess_coding_system_alist)); |
| 6882 | if (NILP (chain)) |
| 6883 | return Qnil; |
| 6884 | |
| 6885 | for (; CONSP (chain); chain = XCDR (chain)) |
| 6886 | { |
| 6887 | Lisp_Object elt; |
| 6888 | elt = XCAR (chain); |
| 6889 | |
| 6890 | if (CONSP (elt) |
| 6891 | && ((STRINGP (target) |
| 6892 | && STRINGP (XCAR (elt)) |
| 6893 | && fast_string_match (XCAR (elt), target) >= 0) |
| 6894 | || (INTEGERP (target) && EQ (target, XCAR (elt))))) |
| 6895 | { |
| 6896 | val = XCDR (elt); |
| 6897 | /* Here, if VAL is both a valid coding system and a valid |
| 6898 | function symbol, we return VAL as a coding system. */ |
| 6899 | if (CONSP (val)) |
| 6900 | return val; |
| 6901 | if (! SYMBOLP (val)) |
| 6902 | return Qnil; |
| 6903 | if (! NILP (Fcoding_system_p (val))) |
| 6904 | return Fcons (val, val); |
| 6905 | if (! NILP (Ffboundp (val))) |
| 6906 | { |
| 6907 | val = call1 (val, Flist (nargs, args)); |
| 6908 | if (CONSP (val)) |
| 6909 | return val; |
| 6910 | if (SYMBOLP (val) && ! NILP (Fcoding_system_p (val))) |
| 6911 | return Fcons (val, val); |
| 6912 | } |
| 6913 | return Qnil; |
| 6914 | } |
| 6915 | } |
| 6916 | return Qnil; |
| 6917 | } |
| 6918 | |
| 6919 | DEFUN ("update-coding-systems-internal", Fupdate_coding_systems_internal, |
| 6920 | Supdate_coding_systems_internal, 0, 0, 0, |
| 6921 | doc: /* Update internal database for ISO2022 and CCL based coding systems. |
| 6922 | When values of any coding categories are changed, you must |
| 6923 | call this function. */) |
| 6924 | () |
| 6925 | { |
| 6926 | int i; |
| 6927 | |
| 6928 | for (i = CODING_CATEGORY_IDX_EMACS_MULE; i < CODING_CATEGORY_IDX_MAX; i++) |
| 6929 | { |
| 6930 | Lisp_Object val; |
| 6931 | |
| 6932 | val = SYMBOL_VALUE (XVECTOR (Vcoding_category_table)->contents[i]); |
| 6933 | if (!NILP (val)) |
| 6934 | { |
| 6935 | if (! coding_system_table[i]) |
| 6936 | coding_system_table[i] = ((struct coding_system *) |
| 6937 | xmalloc (sizeof (struct coding_system))); |
| 6938 | setup_coding_system (val, coding_system_table[i]); |
| 6939 | } |
| 6940 | else if (coding_system_table[i]) |
| 6941 | { |
| 6942 | xfree (coding_system_table[i]); |
| 6943 | coding_system_table[i] = NULL; |
| 6944 | } |
| 6945 | } |
| 6946 | |
| 6947 | return Qnil; |
| 6948 | } |
| 6949 | |
| 6950 | DEFUN ("set-coding-priority-internal", Fset_coding_priority_internal, |
| 6951 | Sset_coding_priority_internal, 0, 0, 0, |
| 6952 | doc: /* Update internal database for the current value of `coding-category-list'. |
| 6953 | This function is internal use only. */) |
| 6954 | () |
| 6955 | { |
| 6956 | int i = 0, idx; |
| 6957 | Lisp_Object val; |
| 6958 | |
| 6959 | val = Vcoding_category_list; |
| 6960 | |
| 6961 | while (CONSP (val) && i < CODING_CATEGORY_IDX_MAX) |
| 6962 | { |
| 6963 | if (! SYMBOLP (XCAR (val))) |
| 6964 | break; |
| 6965 | idx = XFASTINT (Fget (XCAR (val), Qcoding_category_index)); |
| 6966 | if (idx >= CODING_CATEGORY_IDX_MAX) |
| 6967 | break; |
| 6968 | coding_priorities[i++] = (1 << idx); |
| 6969 | val = XCDR (val); |
| 6970 | } |
| 6971 | /* If coding-category-list is valid and contains all coding |
| 6972 | categories, `i' should be CODING_CATEGORY_IDX_MAX now. If not, |
| 6973 | the following code saves Emacs from crashing. */ |
| 6974 | while (i < CODING_CATEGORY_IDX_MAX) |
| 6975 | coding_priorities[i++] = CODING_CATEGORY_MASK_RAW_TEXT; |
| 6976 | |
| 6977 | return Qnil; |
| 6978 | } |
| 6979 | |
| 6980 | #endif /* emacs */ |
| 6981 | |
| 6982 | \f |
| 6983 | /*** 9. Post-amble ***/ |
| 6984 | |
| 6985 | void |
| 6986 | init_coding_once () |
| 6987 | { |
| 6988 | int i; |
| 6989 | |
| 6990 | /* Emacs' internal format specific initialize routine. */ |
| 6991 | for (i = 0; i <= 0x20; i++) |
| 6992 | emacs_code_class[i] = EMACS_control_code; |
| 6993 | emacs_code_class[0x0A] = EMACS_linefeed_code; |
| 6994 | emacs_code_class[0x0D] = EMACS_carriage_return_code; |
| 6995 | for (i = 0x21 ; i < 0x7F; i++) |
| 6996 | emacs_code_class[i] = EMACS_ascii_code; |
| 6997 | emacs_code_class[0x7F] = EMACS_control_code; |
| 6998 | for (i = 0x80; i < 0xFF; i++) |
| 6999 | emacs_code_class[i] = EMACS_invalid_code; |
| 7000 | emacs_code_class[LEADING_CODE_PRIVATE_11] = EMACS_leading_code_3; |
| 7001 | emacs_code_class[LEADING_CODE_PRIVATE_12] = EMACS_leading_code_3; |
| 7002 | emacs_code_class[LEADING_CODE_PRIVATE_21] = EMACS_leading_code_4; |
| 7003 | emacs_code_class[LEADING_CODE_PRIVATE_22] = EMACS_leading_code_4; |
| 7004 | |
| 7005 | /* ISO2022 specific initialize routine. */ |
| 7006 | for (i = 0; i < 0x20; i++) |
| 7007 | iso_code_class[i] = ISO_control_0; |
| 7008 | for (i = 0x21; i < 0x7F; i++) |
| 7009 | iso_code_class[i] = ISO_graphic_plane_0; |
| 7010 | for (i = 0x80; i < 0xA0; i++) |
| 7011 | iso_code_class[i] = ISO_control_1; |
| 7012 | for (i = 0xA1; i < 0xFF; i++) |
| 7013 | iso_code_class[i] = ISO_graphic_plane_1; |
| 7014 | iso_code_class[0x20] = iso_code_class[0x7F] = ISO_0x20_or_0x7F; |
| 7015 | iso_code_class[0xA0] = iso_code_class[0xFF] = ISO_0xA0_or_0xFF; |
| 7016 | iso_code_class[ISO_CODE_CR] = ISO_carriage_return; |
| 7017 | iso_code_class[ISO_CODE_SO] = ISO_shift_out; |
| 7018 | iso_code_class[ISO_CODE_SI] = ISO_shift_in; |
| 7019 | iso_code_class[ISO_CODE_SS2_7] = ISO_single_shift_2_7; |
| 7020 | iso_code_class[ISO_CODE_ESC] = ISO_escape; |
| 7021 | iso_code_class[ISO_CODE_SS2] = ISO_single_shift_2; |
| 7022 | iso_code_class[ISO_CODE_SS3] = ISO_single_shift_3; |
| 7023 | iso_code_class[ISO_CODE_CSI] = ISO_control_sequence_introducer; |
| 7024 | |
| 7025 | setup_coding_system (Qnil, &keyboard_coding); |
| 7026 | setup_coding_system (Qnil, &terminal_coding); |
| 7027 | setup_coding_system (Qnil, &safe_terminal_coding); |
| 7028 | setup_coding_system (Qnil, &default_buffer_file_coding); |
| 7029 | |
| 7030 | bzero (coding_system_table, sizeof coding_system_table); |
| 7031 | |
| 7032 | bzero (ascii_skip_code, sizeof ascii_skip_code); |
| 7033 | for (i = 0; i < 128; i++) |
| 7034 | ascii_skip_code[i] = 1; |
| 7035 | |
| 7036 | #if defined (MSDOS) || defined (WINDOWSNT) |
| 7037 | system_eol_type = CODING_EOL_CRLF; |
| 7038 | #else |
| 7039 | system_eol_type = CODING_EOL_LF; |
| 7040 | #endif |
| 7041 | |
| 7042 | inhibit_pre_post_conversion = 0; |
| 7043 | } |
| 7044 | |
| 7045 | #ifdef emacs |
| 7046 | |
| 7047 | void |
| 7048 | syms_of_coding () |
| 7049 | { |
| 7050 | Qtarget_idx = intern ("target-idx"); |
| 7051 | staticpro (&Qtarget_idx); |
| 7052 | |
| 7053 | Qcoding_system_history = intern ("coding-system-history"); |
| 7054 | staticpro (&Qcoding_system_history); |
| 7055 | Fset (Qcoding_system_history, Qnil); |
| 7056 | |
| 7057 | /* Target FILENAME is the first argument. */ |
| 7058 | Fput (Qinsert_file_contents, Qtarget_idx, make_number (0)); |
| 7059 | /* Target FILENAME is the third argument. */ |
| 7060 | Fput (Qwrite_region, Qtarget_idx, make_number (2)); |
| 7061 | |
| 7062 | Qcall_process = intern ("call-process"); |
| 7063 | staticpro (&Qcall_process); |
| 7064 | /* Target PROGRAM is the first argument. */ |
| 7065 | Fput (Qcall_process, Qtarget_idx, make_number (0)); |
| 7066 | |
| 7067 | Qcall_process_region = intern ("call-process-region"); |
| 7068 | staticpro (&Qcall_process_region); |
| 7069 | /* Target PROGRAM is the third argument. */ |
| 7070 | Fput (Qcall_process_region, Qtarget_idx, make_number (2)); |
| 7071 | |
| 7072 | Qstart_process = intern ("start-process"); |
| 7073 | staticpro (&Qstart_process); |
| 7074 | /* Target PROGRAM is the third argument. */ |
| 7075 | Fput (Qstart_process, Qtarget_idx, make_number (2)); |
| 7076 | |
| 7077 | Qopen_network_stream = intern ("open-network-stream"); |
| 7078 | staticpro (&Qopen_network_stream); |
| 7079 | /* Target SERVICE is the fourth argument. */ |
| 7080 | Fput (Qopen_network_stream, Qtarget_idx, make_number (3)); |
| 7081 | |
| 7082 | Qcoding_system = intern ("coding-system"); |
| 7083 | staticpro (&Qcoding_system); |
| 7084 | |
| 7085 | Qeol_type = intern ("eol-type"); |
| 7086 | staticpro (&Qeol_type); |
| 7087 | |
| 7088 | Qbuffer_file_coding_system = intern ("buffer-file-coding-system"); |
| 7089 | staticpro (&Qbuffer_file_coding_system); |
| 7090 | |
| 7091 | Qpost_read_conversion = intern ("post-read-conversion"); |
| 7092 | staticpro (&Qpost_read_conversion); |
| 7093 | |
| 7094 | Qpre_write_conversion = intern ("pre-write-conversion"); |
| 7095 | staticpro (&Qpre_write_conversion); |
| 7096 | |
| 7097 | Qno_conversion = intern ("no-conversion"); |
| 7098 | staticpro (&Qno_conversion); |
| 7099 | |
| 7100 | Qundecided = intern ("undecided"); |
| 7101 | staticpro (&Qundecided); |
| 7102 | |
| 7103 | Qcoding_system_p = intern ("coding-system-p"); |
| 7104 | staticpro (&Qcoding_system_p); |
| 7105 | |
| 7106 | Qcoding_system_error = intern ("coding-system-error"); |
| 7107 | staticpro (&Qcoding_system_error); |
| 7108 | |
| 7109 | Fput (Qcoding_system_error, Qerror_conditions, |
| 7110 | Fcons (Qcoding_system_error, Fcons (Qerror, Qnil))); |
| 7111 | Fput (Qcoding_system_error, Qerror_message, |
| 7112 | build_string ("Invalid coding system")); |
| 7113 | |
| 7114 | Qcoding_category = intern ("coding-category"); |
| 7115 | staticpro (&Qcoding_category); |
| 7116 | Qcoding_category_index = intern ("coding-category-index"); |
| 7117 | staticpro (&Qcoding_category_index); |
| 7118 | |
| 7119 | Vcoding_category_table |
| 7120 | = Fmake_vector (make_number (CODING_CATEGORY_IDX_MAX), Qnil); |
| 7121 | staticpro (&Vcoding_category_table); |
| 7122 | { |
| 7123 | int i; |
| 7124 | for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++) |
| 7125 | { |
| 7126 | XVECTOR (Vcoding_category_table)->contents[i] |
| 7127 | = intern (coding_category_name[i]); |
| 7128 | Fput (XVECTOR (Vcoding_category_table)->contents[i], |
| 7129 | Qcoding_category_index, make_number (i)); |
| 7130 | } |
| 7131 | } |
| 7132 | |
| 7133 | Qtranslation_table = intern ("translation-table"); |
| 7134 | staticpro (&Qtranslation_table); |
| 7135 | Fput (Qtranslation_table, Qchar_table_extra_slots, make_number (1)); |
| 7136 | |
| 7137 | Qtranslation_table_id = intern ("translation-table-id"); |
| 7138 | staticpro (&Qtranslation_table_id); |
| 7139 | |
| 7140 | Qtranslation_table_for_decode = intern ("translation-table-for-decode"); |
| 7141 | staticpro (&Qtranslation_table_for_decode); |
| 7142 | |
| 7143 | Qtranslation_table_for_encode = intern ("translation-table-for-encode"); |
| 7144 | staticpro (&Qtranslation_table_for_encode); |
| 7145 | |
| 7146 | Qsafe_chars = intern ("safe-chars"); |
| 7147 | staticpro (&Qsafe_chars); |
| 7148 | |
| 7149 | Qchar_coding_system = intern ("char-coding-system"); |
| 7150 | staticpro (&Qchar_coding_system); |
| 7151 | |
| 7152 | /* Intern this now in case it isn't already done. |
| 7153 | Setting this variable twice is harmless. |
| 7154 | But don't staticpro it here--that is done in alloc.c. */ |
| 7155 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); |
| 7156 | Fput (Qsafe_chars, Qchar_table_extra_slots, make_number (0)); |
| 7157 | Fput (Qchar_coding_system, Qchar_table_extra_slots, make_number (2)); |
| 7158 | |
| 7159 | Qvalid_codes = intern ("valid-codes"); |
| 7160 | staticpro (&Qvalid_codes); |
| 7161 | |
| 7162 | Qemacs_mule = intern ("emacs-mule"); |
| 7163 | staticpro (&Qemacs_mule); |
| 7164 | |
| 7165 | Qraw_text = intern ("raw-text"); |
| 7166 | staticpro (&Qraw_text); |
| 7167 | |
| 7168 | defsubr (&Scoding_system_p); |
| 7169 | defsubr (&Sread_coding_system); |
| 7170 | defsubr (&Sread_non_nil_coding_system); |
| 7171 | defsubr (&Scheck_coding_system); |
| 7172 | defsubr (&Sdetect_coding_region); |
| 7173 | defsubr (&Sdetect_coding_string); |
| 7174 | defsubr (&Sfind_coding_systems_region_internal); |
| 7175 | defsubr (&Sdecode_coding_region); |
| 7176 | defsubr (&Sencode_coding_region); |
| 7177 | defsubr (&Sdecode_coding_string); |
| 7178 | defsubr (&Sencode_coding_string); |
| 7179 | defsubr (&Sdecode_sjis_char); |
| 7180 | defsubr (&Sencode_sjis_char); |
| 7181 | defsubr (&Sdecode_big5_char); |
| 7182 | defsubr (&Sencode_big5_char); |
| 7183 | defsubr (&Sset_terminal_coding_system_internal); |
| 7184 | defsubr (&Sset_safe_terminal_coding_system_internal); |
| 7185 | defsubr (&Sterminal_coding_system); |
| 7186 | defsubr (&Sset_keyboard_coding_system_internal); |
| 7187 | defsubr (&Skeyboard_coding_system); |
| 7188 | defsubr (&Sfind_operation_coding_system); |
| 7189 | defsubr (&Supdate_coding_systems_internal); |
| 7190 | defsubr (&Sset_coding_priority_internal); |
| 7191 | |
| 7192 | DEFVAR_LISP ("coding-system-list", &Vcoding_system_list, |
| 7193 | doc: /* List of coding systems. |
| 7194 | |
| 7195 | Do not alter the value of this variable manually. This variable should be |
| 7196 | updated by the functions `make-coding-system' and |
| 7197 | `define-coding-system-alias'. */); |
| 7198 | Vcoding_system_list = Qnil; |
| 7199 | |
| 7200 | DEFVAR_LISP ("coding-system-alist", &Vcoding_system_alist, |
| 7201 | doc: /* Alist of coding system names. |
| 7202 | Each element is one element list of coding system name. |
| 7203 | This variable is given to `completing-read' as TABLE argument. |
| 7204 | |
| 7205 | Do not alter the value of this variable manually. This variable should be |
| 7206 | updated by the functions `make-coding-system' and |
| 7207 | `define-coding-system-alias'. */); |
| 7208 | Vcoding_system_alist = Qnil; |
| 7209 | |
| 7210 | DEFVAR_LISP ("coding-category-list", &Vcoding_category_list, |
| 7211 | doc: /* List of coding-categories (symbols) ordered by priority. |
| 7212 | |
| 7213 | On detecting a coding system, Emacs tries code detection algorithms |
| 7214 | associated with each coding-category one by one in this order. When |
| 7215 | one algorithm agrees with a byte sequence of source text, the coding |
| 7216 | system bound to the corresponding coding-category is selected. */); |
| 7217 | { |
| 7218 | int i; |
| 7219 | |
| 7220 | Vcoding_category_list = Qnil; |
| 7221 | for (i = CODING_CATEGORY_IDX_MAX - 1; i >= 0; i--) |
| 7222 | Vcoding_category_list |
| 7223 | = Fcons (XVECTOR (Vcoding_category_table)->contents[i], |
| 7224 | Vcoding_category_list); |
| 7225 | } |
| 7226 | |
| 7227 | DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read, |
| 7228 | doc: /* Specify the coding system for read operations. |
| 7229 | It is useful to bind this variable with `let', but do not set it globally. |
| 7230 | If the value is a coding system, it is used for decoding on read operation. |
| 7231 | If not, an appropriate element is used from one of the coding system alists: |
| 7232 | There are three such tables, `file-coding-system-alist', |
| 7233 | `process-coding-system-alist', and `network-coding-system-alist'. */); |
| 7234 | Vcoding_system_for_read = Qnil; |
| 7235 | |
| 7236 | DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write, |
| 7237 | doc: /* Specify the coding system for write operations. |
| 7238 | Programs bind this variable with `let', but you should not set it globally. |
| 7239 | If the value is a coding system, it is used for encoding of output, |
| 7240 | when writing it to a file and when sending it to a file or subprocess. |
| 7241 | |
| 7242 | If this does not specify a coding system, an appropriate element |
| 7243 | is used from one of the coding system alists: |
| 7244 | There are three such tables, `file-coding-system-alist', |
| 7245 | `process-coding-system-alist', and `network-coding-system-alist'. |
| 7246 | For output to files, if the above procedure does not specify a coding system, |
| 7247 | the value of `buffer-file-coding-system' is used. */); |
| 7248 | Vcoding_system_for_write = Qnil; |
| 7249 | |
| 7250 | DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used, |
| 7251 | doc: /* Coding system used in the latest file or process I/O. */); |
| 7252 | Vlast_coding_system_used = Qnil; |
| 7253 | |
| 7254 | DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion, |
| 7255 | doc: /* *Non-nil means always inhibit code conversion of end-of-line format. |
| 7256 | See info node `Coding Systems' and info node `Text and Binary' concerning |
| 7257 | such conversion. */); |
| 7258 | inhibit_eol_conversion = 0; |
| 7259 | |
| 7260 | DEFVAR_BOOL ("inherit-process-coding-system", &inherit_process_coding_system, |
| 7261 | doc: /* Non-nil means process buffer inherits coding system of process output. |
| 7262 | Bind it to t if the process output is to be treated as if it were a file |
| 7263 | read from some filesystem. */); |
| 7264 | inherit_process_coding_system = 0; |
| 7265 | |
| 7266 | DEFVAR_LISP ("file-coding-system-alist", &Vfile_coding_system_alist, |
| 7267 | doc: /* Alist to decide a coding system to use for a file I/O operation. |
| 7268 | The format is ((PATTERN . VAL) ...), |
| 7269 | where PATTERN is a regular expression matching a file name, |
| 7270 | VAL is a coding system, a cons of coding systems, or a function symbol. |
| 7271 | If VAL is a coding system, it is used for both decoding and encoding |
| 7272 | the file contents. |
| 7273 | If VAL is a cons of coding systems, the car part is used for decoding, |
| 7274 | and the cdr part is used for encoding. |
| 7275 | If VAL is a function symbol, the function must return a coding system |
| 7276 | or a cons of coding systems which are used as above. The function gets |
| 7277 | the arguments with which `find-operation-coding-system' was called. |
| 7278 | |
| 7279 | See also the function `find-operation-coding-system' |
| 7280 | and the variable `auto-coding-alist'. */); |
| 7281 | Vfile_coding_system_alist = Qnil; |
| 7282 | |
| 7283 | DEFVAR_LISP ("process-coding-system-alist", &Vprocess_coding_system_alist, |
| 7284 | doc: /* Alist to decide a coding system to use for a process I/O operation. |
| 7285 | The format is ((PATTERN . VAL) ...), |
| 7286 | where PATTERN is a regular expression matching a program name, |
| 7287 | VAL is a coding system, a cons of coding systems, or a function symbol. |
| 7288 | If VAL is a coding system, it is used for both decoding what received |
| 7289 | from the program and encoding what sent to the program. |
| 7290 | If VAL is a cons of coding systems, the car part is used for decoding, |
| 7291 | and the cdr part is used for encoding. |
| 7292 | If VAL is a function symbol, the function must return a coding system |
| 7293 | or a cons of coding systems which are used as above. |
| 7294 | |
| 7295 | See also the function `find-operation-coding-system'. */); |
| 7296 | Vprocess_coding_system_alist = Qnil; |
| 7297 | |
| 7298 | DEFVAR_LISP ("network-coding-system-alist", &Vnetwork_coding_system_alist, |
| 7299 | doc: /* Alist to decide a coding system to use for a network I/O operation. |
| 7300 | The format is ((PATTERN . VAL) ...), |
| 7301 | where PATTERN is a regular expression matching a network service name |
| 7302 | or is a port number to connect to, |
| 7303 | VAL is a coding system, a cons of coding systems, or a function symbol. |
| 7304 | If VAL is a coding system, it is used for both decoding what received |
| 7305 | from the network stream and encoding what sent to the network stream. |
| 7306 | If VAL is a cons of coding systems, the car part is used for decoding, |
| 7307 | and the cdr part is used for encoding. |
| 7308 | If VAL is a function symbol, the function must return a coding system |
| 7309 | or a cons of coding systems which are used as above. |
| 7310 | |
| 7311 | See also the function `find-operation-coding-system'. */); |
| 7312 | Vnetwork_coding_system_alist = Qnil; |
| 7313 | |
| 7314 | DEFVAR_LISP ("locale-coding-system", &Vlocale_coding_system, |
| 7315 | doc: /* Coding system to use with system messages. |
| 7316 | Also used for decoding keyboard input on X Window system. */); |
| 7317 | Vlocale_coding_system = Qnil; |
| 7318 | |
| 7319 | /* The eol mnemonics are reset in startup.el system-dependently. */ |
| 7320 | DEFVAR_LISP ("eol-mnemonic-unix", &eol_mnemonic_unix, |
| 7321 | doc: /* *String displayed in mode line for UNIX-like (LF) end-of-line format. */); |
| 7322 | eol_mnemonic_unix = build_string (":"); |
| 7323 | |
| 7324 | DEFVAR_LISP ("eol-mnemonic-dos", &eol_mnemonic_dos, |
| 7325 | doc: /* *String displayed in mode line for DOS-like (CRLF) end-of-line format. */); |
| 7326 | eol_mnemonic_dos = build_string ("\\"); |
| 7327 | |
| 7328 | DEFVAR_LISP ("eol-mnemonic-mac", &eol_mnemonic_mac, |
| 7329 | doc: /* *String displayed in mode line for MAC-like (CR) end-of-line format. */); |
| 7330 | eol_mnemonic_mac = build_string ("/"); |
| 7331 | |
| 7332 | DEFVAR_LISP ("eol-mnemonic-undecided", &eol_mnemonic_undecided, |
| 7333 | doc: /* *String displayed in mode line when end-of-line format is not yet determined. */); |
| 7334 | eol_mnemonic_undecided = build_string (":"); |
| 7335 | |
| 7336 | DEFVAR_LISP ("enable-character-translation", &Venable_character_translation, |
| 7337 | doc: /* *Non-nil enables character translation while encoding and decoding. */); |
| 7338 | Venable_character_translation = Qt; |
| 7339 | |
| 7340 | DEFVAR_LISP ("standard-translation-table-for-decode", |
| 7341 | &Vstandard_translation_table_for_decode, |
| 7342 | doc: /* Table for translating characters while decoding. */); |
| 7343 | Vstandard_translation_table_for_decode = Qnil; |
| 7344 | |
| 7345 | DEFVAR_LISP ("standard-translation-table-for-encode", |
| 7346 | &Vstandard_translation_table_for_encode, |
| 7347 | doc: /* Table for translating characters while encoding. */); |
| 7348 | Vstandard_translation_table_for_encode = Qnil; |
| 7349 | |
| 7350 | DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist, |
| 7351 | doc: /* Alist of charsets vs revision numbers. |
| 7352 | While encoding, if a charset (car part of an element) is found, |
| 7353 | designate it with the escape sequence identifying revision (cdr part of the element). */); |
| 7354 | Vcharset_revision_alist = Qnil; |
| 7355 | |
| 7356 | DEFVAR_LISP ("default-process-coding-system", |
| 7357 | &Vdefault_process_coding_system, |
| 7358 | doc: /* Cons of coding systems used for process I/O by default. |
| 7359 | The car part is used for decoding a process output, |
| 7360 | the cdr part is used for encoding a text to be sent to a process. */); |
| 7361 | Vdefault_process_coding_system = Qnil; |
| 7362 | |
| 7363 | DEFVAR_LISP ("latin-extra-code-table", &Vlatin_extra_code_table, |
| 7364 | doc: /* Table of extra Latin codes in the range 128..159 (inclusive). |
| 7365 | This is a vector of length 256. |
| 7366 | If Nth element is non-nil, the existence of code N in a file |
| 7367 | \(or output of subprocess) doesn't prevent it to be detected as |
| 7368 | a coding system of ISO 2022 variant which has a flag |
| 7369 | `accept-latin-extra-code' t (e.g. iso-latin-1) on reading a file |
| 7370 | or reading output of a subprocess. |
| 7371 | Only 128th through 159th elements has a meaning. */); |
| 7372 | Vlatin_extra_code_table = Fmake_vector (make_number (256), Qnil); |
| 7373 | |
| 7374 | DEFVAR_LISP ("select-safe-coding-system-function", |
| 7375 | &Vselect_safe_coding_system_function, |
| 7376 | doc: /* Function to call to select safe coding system for encoding a text. |
| 7377 | |
| 7378 | If set, this function is called to force a user to select a proper |
| 7379 | coding system which can encode the text in the case that a default |
| 7380 | coding system used in each operation can't encode the text. |
| 7381 | |
| 7382 | The default value is `select-safe-coding-system' (which see). */); |
| 7383 | Vselect_safe_coding_system_function = Qnil; |
| 7384 | |
| 7385 | DEFVAR_LISP ("char-coding-system-table", &Vchar_coding_system_table, |
| 7386 | doc: /* Char-table containing safe coding systems of each characters. |
| 7387 | Each element doesn't include such generic coding systems that can |
| 7388 | encode any characters. They are in the first extra slot. */); |
| 7389 | Vchar_coding_system_table = Fmake_char_table (Qchar_coding_system, Qnil); |
| 7390 | |
| 7391 | DEFVAR_BOOL ("inhibit-iso-escape-detection", |
| 7392 | &inhibit_iso_escape_detection, |
| 7393 | doc: /* If non-nil, Emacs ignores ISO2022's escape sequence on code detection. |
| 7394 | |
| 7395 | By default, on reading a file, Emacs tries to detect how the text is |
| 7396 | encoded. This code detection is sensitive to escape sequences. If |
| 7397 | the sequence is valid as ISO2022, the code is determined as one of |
| 7398 | the ISO2022 encodings, and the file is decoded by the corresponding |
| 7399 | coding system (e.g. `iso-2022-7bit'). |
| 7400 | |
| 7401 | However, there may be a case that you want to read escape sequences in |
| 7402 | a file as is. In such a case, you can set this variable to non-nil. |
| 7403 | Then, as the code detection ignores any escape sequences, no file is |
| 7404 | detected as encoded in some ISO2022 encoding. The result is that all |
| 7405 | escape sequences become visible in a buffer. |
| 7406 | |
| 7407 | The default value is nil, and it is strongly recommended not to change |
| 7408 | it. That is because many Emacs Lisp source files that contain |
| 7409 | non-ASCII characters are encoded by the coding system `iso-2022-7bit' |
| 7410 | in Emacs's distribution, and they won't be decoded correctly on |
| 7411 | reading if you suppress escape sequence detection. |
| 7412 | |
| 7413 | The other way to read escape sequences in a file without decoding is |
| 7414 | to explicitly specify some coding system that doesn't use ISO2022's |
| 7415 | escape sequence (e.g `latin-1') on reading by \\[universal-coding-system-argument]. */); |
| 7416 | inhibit_iso_escape_detection = 0; |
| 7417 | } |
| 7418 | |
| 7419 | char * |
| 7420 | emacs_strerror (error_number) |
| 7421 | int error_number; |
| 7422 | { |
| 7423 | char *str; |
| 7424 | |
| 7425 | synchronize_system_messages_locale (); |
| 7426 | str = strerror (error_number); |
| 7427 | |
| 7428 | if (! NILP (Vlocale_coding_system)) |
| 7429 | { |
| 7430 | Lisp_Object dec = code_convert_string_norecord (build_string (str), |
| 7431 | Vlocale_coding_system, |
| 7432 | 0); |
| 7433 | str = (char *) XSTRING (dec)->data; |
| 7434 | } |
| 7435 | |
| 7436 | return str; |
| 7437 | } |
| 7438 | |
| 7439 | #endif /* emacs */ |
| 7440 | |