1 /* Code conversion functions.
2 Copyright (C) 1991, 1995 Free Software Foundation, Inc.
3 Copyright (C) 1995 Sun Microsystems, Inc.
4 Copyright (C) 1999,2000 MORIOKA Tomohiko
6 This file is part of XEmacs.
8 XEmacs is free software; you can redistribute it and/or modify it
9 under the terms of the GNU General Public License as published by the
10 Free Software Foundation; either version 2, or (at your option) any
13 XEmacs is distributed in the hope that it will be useful, but WITHOUT
14 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
15 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
18 You should have received a copy of the GNU General Public License
19 along with XEmacs; 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. */
23 /* Synched up with: Mule 2.3. Not in FSF. */
25 /* Rewritten by Ben Wing <ben@xemacs.org>. */
39 #include "file-coding.h"
41 Lisp_Object Qcoding_system_error;
43 Lisp_Object Vkeyboard_coding_system;
44 Lisp_Object Vterminal_coding_system;
45 Lisp_Object Vcoding_system_for_read;
46 Lisp_Object Vcoding_system_for_write;
47 Lisp_Object Vfile_name_coding_system;
49 /* Table of symbols identifying each coding category. */
50 Lisp_Object coding_category_symbol[CODING_CATEGORY_LAST + 1];
54 struct file_coding_dump {
55 /* Coding system currently associated with each coding category. */
56 Lisp_Object coding_category_system[CODING_CATEGORY_LAST + 1];
58 /* Table of all coding categories in decreasing order of priority.
59 This describes a permutation of the possible coding categories. */
60 int coding_category_by_priority[CODING_CATEGORY_LAST + 1];
62 #if defined(MULE) && !defined(UTF2000)
63 Lisp_Object ucs_to_mule_table[65536];
67 static const struct lrecord_description fcd_description_1[] = {
68 { XD_LISP_OBJECT_ARRAY, offsetof (struct file_coding_dump, coding_category_system), CODING_CATEGORY_LAST + 1 },
69 #if defined(MULE) && !defined(UTF2000)
70 { XD_LISP_OBJECT_ARRAY, offsetof (struct file_coding_dump, ucs_to_mule_table), 65536 },
75 static const struct struct_description fcd_description = {
76 sizeof (struct file_coding_dump),
80 Lisp_Object mule_to_ucs_table;
82 Lisp_Object Qcoding_systemp;
84 Lisp_Object Qraw_text, Qno_conversion, Qccl, Qiso2022;
85 /* Qinternal in general.c */
87 Lisp_Object Qmnemonic, Qeol_type;
88 Lisp_Object Qcr, Qcrlf, Qlf;
89 Lisp_Object Qeol_cr, Qeol_crlf, Qeol_lf;
90 Lisp_Object Qpost_read_conversion;
91 Lisp_Object Qpre_write_conversion;
94 Lisp_Object Qucs4, Qutf8;
95 Lisp_Object Qbig5, Qshift_jis;
96 Lisp_Object Qcharset_g0, Qcharset_g1, Qcharset_g2, Qcharset_g3;
97 Lisp_Object Qforce_g0_on_output, Qforce_g1_on_output;
98 Lisp_Object Qforce_g2_on_output, Qforce_g3_on_output;
99 Lisp_Object Qno_iso6429;
100 Lisp_Object Qinput_charset_conversion, Qoutput_charset_conversion;
101 Lisp_Object Qescape_quoted;
102 Lisp_Object Qshort, Qno_ascii_eol, Qno_ascii_cntl, Qseven, Qlock_shift;
105 Lisp_Object Qdisable_composition;
107 Lisp_Object Qencode, Qdecode;
109 Lisp_Object Vcoding_system_hash_table;
111 int enable_multibyte_characters;
114 /* Additional information used by the ISO2022 decoder and detector. */
115 struct iso2022_decoder
117 /* CHARSET holds the character sets currently assigned to the G0
118 through G3 variables. It is initialized from the array
119 INITIAL_CHARSET in CODESYS. */
120 Lisp_Object charset[4];
122 /* Which registers are currently invoked into the left (GL) and
123 right (GR) halves of the 8-bit encoding space? */
124 int register_left, register_right;
126 /* ISO_ESC holds a value indicating part of an escape sequence
127 that has already been seen. */
128 enum iso_esc_flag esc;
130 /* This records the bytes we've seen so far in an escape sequence,
131 in case the sequence is invalid (we spit out the bytes unchanged). */
132 unsigned char esc_bytes[8];
134 /* Index for next byte to store in ISO escape sequence. */
137 #ifdef ENABLE_COMPOSITE_CHARS
138 /* Stuff seen so far when composing a string. */
139 unsigned_char_dynarr *composite_chars;
142 /* If we saw an invalid designation sequence for a particular
143 register, we flag it here and switch to ASCII. The next time we
144 see a valid designation for this register, we turn off the flag
145 and do the designation normally, but pretend the sequence was
146 invalid. The effect of all this is that (most of the time) the
147 escape sequences for both the switch to the unknown charset, and
148 the switch back to the known charset, get inserted literally into
149 the buffer and saved out as such. The hope is that we can
150 preserve the escape sequences so that the resulting written out
151 file makes sense. If we don't do any of this, the designation
152 to the invalid charset will be preserved but that switch back
153 to the known charset will probably get eaten because it was
154 the same charset that was already present in the register. */
155 unsigned char invalid_designated[4];
157 /* We try to do similar things as above for direction-switching
158 sequences. If we encountered a direction switch while an
159 invalid designation was present, or an invalid designation
160 just after a direction switch (i.e. no valid designation
161 encountered yet), we insert the direction-switch escape
162 sequence literally into the output stream, and later on
163 insert the corresponding direction-restoring escape sequence
165 unsigned int switched_dir_and_no_valid_charset_yet :1;
166 unsigned int invalid_switch_dir :1;
168 /* Tells the decoder to output the escape sequence literally
169 even though it was valid. Used in the games we play to
170 avoid lossage when we encounter invalid designations. */
171 unsigned int output_literally :1;
172 /* We encountered a direction switch followed by an invalid
173 designation. We didn't output the direction switch
174 literally because we didn't know about the invalid designation;
175 but we have to do so now. */
176 unsigned int output_direction_sequence :1;
179 EXFUN (Fcopy_coding_system, 2);
181 struct detection_state;
184 text_encode_generic (Lstream *encoding, const unsigned char *src,
185 unsigned_char_dynarr *dst, unsigned int n);
187 static int detect_coding_sjis (struct detection_state *st,
188 const unsigned char *src,
190 static void decode_coding_sjis (Lstream *decoding,
191 const unsigned char *src,
192 unsigned_char_dynarr *dst,
194 void char_encode_shift_jis (struct encoding_stream *str, Emchar c,
195 unsigned_char_dynarr *dst, unsigned int *flags);
196 void char_finish_shift_jis (struct encoding_stream *str,
197 unsigned_char_dynarr *dst, unsigned int *flags);
199 static int detect_coding_big5 (struct detection_state *st,
200 const unsigned char *src,
202 static void decode_coding_big5 (Lstream *decoding,
203 const unsigned char *src,
204 unsigned_char_dynarr *dst, unsigned int n);
205 static void encode_coding_big5 (Lstream *encoding,
206 const unsigned char *src,
207 unsigned_char_dynarr *dst, unsigned int n);
208 static int detect_coding_ucs4 (struct detection_state *st,
209 const unsigned char *src,
211 static void decode_coding_ucs4 (Lstream *decoding,
212 const unsigned char *src,
213 unsigned_char_dynarr *dst, unsigned int n);
214 void char_encode_ucs4 (struct encoding_stream *str, Emchar c,
215 unsigned_char_dynarr *dst, unsigned int *flags);
216 void char_finish_ucs4 (struct encoding_stream *str,
217 unsigned_char_dynarr *dst, unsigned int *flags);
219 static int detect_coding_utf8 (struct detection_state *st,
220 const unsigned char *src,
222 static void decode_coding_utf8 (Lstream *decoding,
223 const unsigned char *src,
224 unsigned_char_dynarr *dst, unsigned int n);
225 void char_encode_utf8 (struct encoding_stream *str, Emchar c,
226 unsigned_char_dynarr *dst, unsigned int *flags);
227 void char_finish_utf8 (struct encoding_stream *str,
228 unsigned_char_dynarr *dst, unsigned int *flags);
230 static int postprocess_iso2022_mask (int mask);
231 static void reset_iso2022 (Lisp_Object coding_system,
232 struct iso2022_decoder *iso);
233 static int detect_coding_iso2022 (struct detection_state *st,
234 const unsigned char *src,
236 static void decode_coding_iso2022 (Lstream *decoding,
237 const unsigned char *src,
238 unsigned_char_dynarr *dst, unsigned int n);
239 void char_encode_iso2022 (struct encoding_stream *str, Emchar c,
240 unsigned_char_dynarr *dst, unsigned int *flags);
241 void char_finish_iso2022 (struct encoding_stream *str,
242 unsigned_char_dynarr *dst, unsigned int *flags);
244 static void decode_coding_no_conversion (Lstream *decoding,
245 const unsigned char *src,
246 unsigned_char_dynarr *dst,
248 static void encode_coding_no_conversion (Lstream *encoding,
249 const unsigned char *src,
250 unsigned_char_dynarr *dst,
252 static void mule_decode (Lstream *decoding, const unsigned char *src,
253 unsigned_char_dynarr *dst, unsigned int n);
254 static void mule_encode (Lstream *encoding, const unsigned char *src,
255 unsigned_char_dynarr *dst, unsigned int n);
257 typedef struct codesys_prop codesys_prop;
266 Dynarr_declare (codesys_prop);
267 } codesys_prop_dynarr;
269 static const struct lrecord_description codesys_prop_description_1[] = {
270 { XD_LISP_OBJECT, offsetof (codesys_prop, sym) },
274 static const struct struct_description codesys_prop_description = {
275 sizeof (codesys_prop),
276 codesys_prop_description_1
279 static const struct lrecord_description codesys_prop_dynarr_description_1[] = {
280 XD_DYNARR_DESC (codesys_prop_dynarr, &codesys_prop_description),
284 static const struct struct_description codesys_prop_dynarr_description = {
285 sizeof (codesys_prop_dynarr),
286 codesys_prop_dynarr_description_1
289 codesys_prop_dynarr *the_codesys_prop_dynarr;
291 enum codesys_prop_enum
294 CODESYS_PROP_ISO2022,
299 /************************************************************************/
300 /* Coding system functions */
301 /************************************************************************/
303 static Lisp_Object mark_coding_system (Lisp_Object);
304 static void print_coding_system (Lisp_Object, Lisp_Object, int);
305 static void finalize_coding_system (void *header, int for_disksave);
308 static const struct lrecord_description ccs_description_1[] = {
309 { XD_LISP_OBJECT, offsetof (charset_conversion_spec, from_charset) },
310 { XD_LISP_OBJECT, offsetof (charset_conversion_spec, to_charset) },
314 static const struct struct_description ccs_description = {
315 sizeof (charset_conversion_spec),
319 static const struct lrecord_description ccsd_description_1[] = {
320 XD_DYNARR_DESC (charset_conversion_spec_dynarr, &ccs_description),
324 static const struct struct_description ccsd_description = {
325 sizeof (charset_conversion_spec_dynarr),
330 static const struct lrecord_description coding_system_description[] = {
331 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, name) },
332 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, doc_string) },
333 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, mnemonic) },
334 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, post_read_conversion) },
335 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, pre_write_conversion) },
336 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, eol_lf) },
337 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, eol_crlf) },
338 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, eol_cr) },
340 { XD_LISP_OBJECT_ARRAY, offsetof (Lisp_Coding_System, iso2022.initial_charset), 4 },
341 { XD_STRUCT_PTR, offsetof (Lisp_Coding_System, iso2022.input_conv), 1, &ccsd_description },
342 { XD_STRUCT_PTR, offsetof (Lisp_Coding_System, iso2022.output_conv), 1, &ccsd_description },
343 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, ccl.decode) },
344 { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, ccl.encode) },
349 DEFINE_LRECORD_IMPLEMENTATION ("coding-system", coding_system,
350 mark_coding_system, print_coding_system,
351 finalize_coding_system,
352 0, 0, coding_system_description,
356 mark_coding_system (Lisp_Object obj)
358 Lisp_Coding_System *codesys = XCODING_SYSTEM (obj);
360 mark_object (CODING_SYSTEM_NAME (codesys));
361 mark_object (CODING_SYSTEM_DOC_STRING (codesys));
362 mark_object (CODING_SYSTEM_MNEMONIC (codesys));
363 mark_object (CODING_SYSTEM_EOL_LF (codesys));
364 mark_object (CODING_SYSTEM_EOL_CRLF (codesys));
365 mark_object (CODING_SYSTEM_EOL_CR (codesys));
367 switch (CODING_SYSTEM_TYPE (codesys))
371 case CODESYS_ISO2022:
372 for (i = 0; i < 4; i++)
373 mark_object (CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i));
374 if (codesys->iso2022.input_conv)
376 for (i = 0; i < Dynarr_length (codesys->iso2022.input_conv); i++)
378 struct charset_conversion_spec *ccs =
379 Dynarr_atp (codesys->iso2022.input_conv, i);
380 mark_object (ccs->from_charset);
381 mark_object (ccs->to_charset);
384 if (codesys->iso2022.output_conv)
386 for (i = 0; i < Dynarr_length (codesys->iso2022.output_conv); i++)
388 struct charset_conversion_spec *ccs =
389 Dynarr_atp (codesys->iso2022.output_conv, i);
390 mark_object (ccs->from_charset);
391 mark_object (ccs->to_charset);
397 mark_object (CODING_SYSTEM_CCL_DECODE (codesys));
398 mark_object (CODING_SYSTEM_CCL_ENCODE (codesys));
405 mark_object (CODING_SYSTEM_PRE_WRITE_CONVERSION (codesys));
406 return CODING_SYSTEM_POST_READ_CONVERSION (codesys);
410 print_coding_system (Lisp_Object obj, Lisp_Object printcharfun,
413 Lisp_Coding_System *c = XCODING_SYSTEM (obj);
415 error ("printing unreadable object #<coding_system 0x%x>",
418 write_c_string ("#<coding_system ", printcharfun);
419 print_internal (c->name, printcharfun, 1);
420 write_c_string (">", printcharfun);
424 finalize_coding_system (void *header, int for_disksave)
426 Lisp_Coding_System *c = (Lisp_Coding_System *) header;
427 /* Since coding systems never go away, this function is not
428 necessary. But it would be necessary if we changed things
429 so that coding systems could go away. */
430 if (!for_disksave) /* see comment in lstream.c */
432 switch (CODING_SYSTEM_TYPE (c))
435 case CODESYS_ISO2022:
436 if (c->iso2022.input_conv)
438 Dynarr_free (c->iso2022.input_conv);
439 c->iso2022.input_conv = 0;
441 if (c->iso2022.output_conv)
443 Dynarr_free (c->iso2022.output_conv);
444 c->iso2022.output_conv = 0;
455 symbol_to_eol_type (Lisp_Object symbol)
457 CHECK_SYMBOL (symbol);
458 if (NILP (symbol)) return EOL_AUTODETECT;
459 if (EQ (symbol, Qlf)) return EOL_LF;
460 if (EQ (symbol, Qcrlf)) return EOL_CRLF;
461 if (EQ (symbol, Qcr)) return EOL_CR;
463 signal_simple_error ("Unrecognized eol type", symbol);
464 return EOL_AUTODETECT; /* not reached */
468 eol_type_to_symbol (eol_type_t type)
473 case EOL_LF: return Qlf;
474 case EOL_CRLF: return Qcrlf;
475 case EOL_CR: return Qcr;
476 case EOL_AUTODETECT: return Qnil;
481 setup_eol_coding_systems (Lisp_Coding_System *codesys)
483 Lisp_Object codesys_obj;
484 int len = string_length (XSYMBOL (CODING_SYSTEM_NAME (codesys))->name);
485 char *codesys_name = (char *) alloca (len + 7);
487 char *codesys_mnemonic=0;
489 Lisp_Object codesys_name_sym, sub_codesys_obj;
493 XSETCODING_SYSTEM (codesys_obj, codesys);
495 memcpy (codesys_name,
496 string_data (XSYMBOL (CODING_SYSTEM_NAME (codesys))->name), len);
498 if (STRINGP (CODING_SYSTEM_MNEMONIC (codesys)))
500 mlen = XSTRING_LENGTH (CODING_SYSTEM_MNEMONIC (codesys));
501 codesys_mnemonic = (char *) alloca (mlen + 7);
502 memcpy (codesys_mnemonic,
503 XSTRING_DATA (CODING_SYSTEM_MNEMONIC (codesys)), mlen);
506 #define DEFINE_SUB_CODESYS(op_sys, op_sys_abbr, Type) do { \
507 strcpy (codesys_name + len, "-" op_sys); \
509 strcpy (codesys_mnemonic + mlen, op_sys_abbr); \
510 codesys_name_sym = intern (codesys_name); \
511 sub_codesys_obj = Fcopy_coding_system (codesys_obj, codesys_name_sym); \
512 XCODING_SYSTEM_EOL_TYPE (sub_codesys_obj) = Type; \
514 XCODING_SYSTEM_MNEMONIC(sub_codesys_obj) = \
515 build_string (codesys_mnemonic); \
516 CODING_SYSTEM_##Type (codesys) = sub_codesys_obj; \
519 DEFINE_SUB_CODESYS("unix", "", EOL_LF);
520 DEFINE_SUB_CODESYS("dos", ":T", EOL_CRLF);
521 DEFINE_SUB_CODESYS("mac", ":t", EOL_CR);
524 DEFUN ("coding-system-p", Fcoding_system_p, 1, 1, 0, /*
525 Return t if OBJECT is a coding system.
526 A coding system is an object that defines how text containing multiple
527 character sets is encoded into a stream of (typically 8-bit) bytes.
528 The coding system is used to decode the stream into a series of
529 characters (which may be from multiple charsets) when the text is read
530 from a file or process, and is used to encode the text back into the
531 same format when it is written out to a file or process.
533 For example, many ISO2022-compliant coding systems (such as Compound
534 Text, which is used for inter-client data under the X Window System)
535 use escape sequences to switch between different charsets -- Japanese
536 Kanji, for example, is invoked with "ESC $ ( B"; ASCII is invoked
537 with "ESC ( B"; and Cyrillic is invoked with "ESC - L". See
538 `make-coding-system' for more information.
540 Coding systems are normally identified using a symbol, and the
541 symbol is accepted in place of the actual coding system object whenever
542 a coding system is called for. (This is similar to how faces work.)
546 return CODING_SYSTEMP (object) ? Qt : Qnil;
549 DEFUN ("find-coding-system", Ffind_coding_system, 1, 1, 0, /*
550 Retrieve the coding system of the given name.
552 If CODING-SYSTEM-OR-NAME is a coding-system object, it is simply
553 returned. Otherwise, CODING-SYSTEM-OR-NAME should be a symbol.
554 If there is no such coding system, nil is returned. Otherwise the
555 associated coding system object is returned.
557 (coding_system_or_name))
559 if (NILP (coding_system_or_name))
560 coding_system_or_name = Qbinary;
561 else if (CODING_SYSTEMP (coding_system_or_name))
562 return coding_system_or_name;
564 CHECK_SYMBOL (coding_system_or_name);
568 coding_system_or_name =
569 Fgethash (coding_system_or_name, Vcoding_system_hash_table, Qnil);
571 if (CODING_SYSTEMP (coding_system_or_name) || NILP (coding_system_or_name))
572 return coding_system_or_name;
576 DEFUN ("get-coding-system", Fget_coding_system, 1, 1, 0, /*
577 Retrieve the coding system of the given name.
578 Same as `find-coding-system' except that if there is no such
579 coding system, an error is signaled instead of returning nil.
583 Lisp_Object coding_system = Ffind_coding_system (name);
585 if (NILP (coding_system))
586 signal_simple_error ("No such coding system", name);
587 return coding_system;
590 /* We store the coding systems in hash tables with the names as the key and the
591 actual coding system object as the value. Occasionally we need to use them
592 in a list format. These routines provide us with that. */
593 struct coding_system_list_closure
595 Lisp_Object *coding_system_list;
599 add_coding_system_to_list_mapper (Lisp_Object key, Lisp_Object value,
600 void *coding_system_list_closure)
602 /* This function can GC */
603 struct coding_system_list_closure *cscl =
604 (struct coding_system_list_closure *) coding_system_list_closure;
605 Lisp_Object *coding_system_list = cscl->coding_system_list;
607 *coding_system_list = Fcons (key, *coding_system_list);
611 DEFUN ("coding-system-list", Fcoding_system_list, 0, 0, 0, /*
612 Return a list of the names of all defined coding systems.
616 Lisp_Object coding_system_list = Qnil;
618 struct coding_system_list_closure coding_system_list_closure;
620 GCPRO1 (coding_system_list);
621 coding_system_list_closure.coding_system_list = &coding_system_list;
622 elisp_maphash (add_coding_system_to_list_mapper, Vcoding_system_hash_table,
623 &coding_system_list_closure);
626 return coding_system_list;
629 DEFUN ("coding-system-name", Fcoding_system_name, 1, 1, 0, /*
630 Return the name of the given coding system.
634 coding_system = Fget_coding_system (coding_system);
635 return XCODING_SYSTEM_NAME (coding_system);
638 static Lisp_Coding_System *
639 allocate_coding_system (enum coding_system_type type, Lisp_Object name)
641 Lisp_Coding_System *codesys =
642 alloc_lcrecord_type (Lisp_Coding_System, &lrecord_coding_system);
644 zero_lcrecord (codesys);
645 CODING_SYSTEM_PRE_WRITE_CONVERSION (codesys) = Qnil;
646 CODING_SYSTEM_POST_READ_CONVERSION (codesys) = Qnil;
647 CODING_SYSTEM_EOL_TYPE (codesys) = EOL_AUTODETECT;
648 CODING_SYSTEM_EOL_CRLF (codesys) = Qnil;
649 CODING_SYSTEM_EOL_CR (codesys) = Qnil;
650 CODING_SYSTEM_EOL_LF (codesys) = Qnil;
651 CODING_SYSTEM_TYPE (codesys) = type;
652 CODING_SYSTEM_MNEMONIC (codesys) = Qnil;
654 if (type == CODESYS_ISO2022)
657 for (i = 0; i < 4; i++)
658 CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i) = Qnil;
660 else if (type == CODESYS_CCL)
662 CODING_SYSTEM_CCL_DECODE (codesys) = Qnil;
663 CODING_SYSTEM_CCL_ENCODE (codesys) = Qnil;
666 CODING_SYSTEM_NAME (codesys) = name;
672 /* Given a list of charset conversion specs as specified in a Lisp
673 program, parse it into STORE_HERE. */
676 parse_charset_conversion_specs (charset_conversion_spec_dynarr *store_here,
677 Lisp_Object spec_list)
681 EXTERNAL_LIST_LOOP (rest, spec_list)
683 Lisp_Object car = XCAR (rest);
684 Lisp_Object from, to;
685 struct charset_conversion_spec spec;
687 if (!CONSP (car) || !CONSP (XCDR (car)) || !NILP (XCDR (XCDR (car))))
688 signal_simple_error ("Invalid charset conversion spec", car);
689 from = Fget_charset (XCAR (car));
690 to = Fget_charset (XCAR (XCDR (car)));
691 if ( (XCHARSET_CHARS (from) != XCHARSET_CHARS (to)) ||
692 (XCHARSET_DIMENSION (from) != XCHARSET_DIMENSION (to)) )
693 signal_simple_error_2
694 ("Attempted conversion between different charset types",
696 spec.from_charset = from;
697 spec.to_charset = to;
699 Dynarr_add (store_here, spec);
703 /* Given a dynarr LOAD_HERE of internally-stored charset conversion
704 specs, return the equivalent as the Lisp programmer would see it.
706 If LOAD_HERE is 0, return Qnil. */
709 unparse_charset_conversion_specs (charset_conversion_spec_dynarr *load_here)
716 for (i = 0, result = Qnil; i < Dynarr_length (load_here); i++)
718 struct charset_conversion_spec *ccs = Dynarr_atp (load_here, i);
719 result = Fcons (list2 (ccs->from_charset, ccs->to_charset), result);
722 return Fnreverse (result);
727 DEFUN ("make-coding-system", Fmake_coding_system, 2, 4, 0, /*
728 Register symbol NAME as a coding system.
730 TYPE describes the conversion method used and should be one of
733 Automatic conversion. XEmacs attempts to detect the coding system
736 No conversion. Use this for binary files and such. On output,
737 graphic characters that are not in ASCII or Latin-1 will be
738 replaced by a ?. (For a no-conversion-encoded buffer, these
739 characters will only be present if you explicitly insert them.)
741 Shift-JIS (a Japanese encoding commonly used in PC operating systems).
743 ISO 10646 UCS-4 encoding.
745 ISO 10646 UTF-8 encoding.
747 Any ISO2022-compliant encoding. Among other things, this includes
748 JIS (the Japanese encoding commonly used for e-mail), EUC (the
749 standard Unix encoding for Japanese and other languages), and
750 Compound Text (the encoding used in X11). You can specify more
751 specific information about the conversion with the FLAGS argument.
753 Big5 (the encoding commonly used for Taiwanese).
755 The conversion is performed using a user-written pseudo-code
756 program. CCL (Code Conversion Language) is the name of this
759 Write out or read in the raw contents of the memory representing
760 the buffer's text. This is primarily useful for debugging
761 purposes, and is only enabled when XEmacs has been compiled with
762 DEBUG_XEMACS defined (via the --debug configure option).
763 WARNING: Reading in a file using 'internal conversion can result
764 in an internal inconsistency in the memory representing a
765 buffer's text, which will produce unpredictable results and may
766 cause XEmacs to crash. Under normal circumstances you should
767 never use 'internal conversion.
769 DOC-STRING is a string describing the coding system.
771 PROPS is a property list, describing the specific nature of the
772 character set. Recognized properties are:
775 String to be displayed in the modeline when this coding system is
779 End-of-line conversion to be used. It should be one of
782 Automatically detect the end-of-line type (LF, CRLF,
783 or CR). Also generate subsidiary coding systems named
784 `NAME-unix', `NAME-dos', and `NAME-mac', that are
785 identical to this coding system but have an EOL-TYPE
786 value of 'lf, 'crlf, and 'cr, respectively.
788 The end of a line is marked externally using ASCII LF.
789 Since this is also the way that XEmacs represents an
790 end-of-line internally, specifying this option results
791 in no end-of-line conversion. This is the standard
792 format for Unix text files.
794 The end of a line is marked externally using ASCII
795 CRLF. This is the standard format for MS-DOS text
798 The end of a line is marked externally using ASCII CR.
799 This is the standard format for Macintosh text files.
801 Automatically detect the end-of-line type but do not
802 generate subsidiary coding systems. (This value is
803 converted to nil when stored internally, and
804 `coding-system-property' will return nil.)
807 If non-nil, composition/decomposition for combining characters
810 'post-read-conversion
811 Function called after a file has been read in, to perform the
812 decoding. Called with two arguments, BEG and END, denoting
813 a region of the current buffer to be decoded.
815 'pre-write-conversion
816 Function called before a file is written out, to perform the
817 encoding. Called with two arguments, BEG and END, denoting
818 a region of the current buffer to be encoded.
821 The following additional properties are recognized if TYPE is 'iso2022:
827 The character set initially designated to the G0 - G3 registers.
828 The value should be one of
830 -- A charset object (designate that character set)
831 -- nil (do not ever use this register)
832 -- t (no character set is initially designated to
833 the register, but may be later on; this automatically
834 sets the corresponding `force-g*-on-output' property)
840 If non-nil, send an explicit designation sequence on output before
841 using the specified register.
844 If non-nil, use the short forms "ESC $ @", "ESC $ A", and
845 "ESC $ B" on output in place of the full designation sequences
846 "ESC $ ( @", "ESC $ ( A", and "ESC $ ( B".
849 If non-nil, don't designate ASCII to G0 at each end of line on output.
850 Setting this to non-nil also suppresses other state-resetting that
851 normally happens at the end of a line.
854 If non-nil, don't designate ASCII to G0 before control chars on output.
857 If non-nil, use 7-bit environment on output. Otherwise, use 8-bit
861 If non-nil, use locking-shift (SO/SI) instead of single-shift
862 or designation by escape sequence.
865 If non-nil, don't use ISO6429's direction specification.
868 If non-nil, literal control characters that are the same as
869 the beginning of a recognized ISO2022 or ISO6429 escape sequence
870 (in particular, ESC (0x1B), SO (0x0E), SI (0x0F), SS2 (0x8E),
871 SS3 (0x8F), and CSI (0x9B)) are "quoted" with an escape character
872 so that they can be properly distinguished from an escape sequence.
873 (Note that doing this results in a non-portable encoding.) This
874 encoding flag is used for byte-compiled files. Note that ESC
875 is a good choice for a quoting character because there are no
876 escape sequences whose second byte is a character from the Control-0
877 or Control-1 character sets; this is explicitly disallowed by the
880 'input-charset-conversion
881 A list of conversion specifications, specifying conversion of
882 characters in one charset to another when decoding is performed.
883 Each specification is a list of two elements: the source charset,
884 and the destination charset.
886 'output-charset-conversion
887 A list of conversion specifications, specifying conversion of
888 characters in one charset to another when encoding is performed.
889 The form of each specification is the same as for
890 'input-charset-conversion.
893 The following additional properties are recognized (and required)
897 CCL program used for decoding (converting to internal format).
900 CCL program used for encoding (converting to external format).
902 (name, type, doc_string, props))
904 Lisp_Coding_System *codesys;
905 Lisp_Object rest, key, value;
906 enum coding_system_type ty;
907 int need_to_setup_eol_systems = 1;
909 /* Convert type to constant */
910 if (NILP (type) || EQ (type, Qundecided))
911 { ty = CODESYS_AUTODETECT; }
913 else if (EQ (type, Qshift_jis)) { ty = CODESYS_SHIFT_JIS; }
914 else if (EQ (type, Qiso2022)) { ty = CODESYS_ISO2022; }
915 else if (EQ (type, Qbig5)) { ty = CODESYS_BIG5; }
916 else if (EQ (type, Qucs4)) { ty = CODESYS_UCS4; }
917 else if (EQ (type, Qutf8)) { ty = CODESYS_UTF8; }
918 else if (EQ (type, Qccl)) { ty = CODESYS_CCL; }
920 else if (EQ (type, Qno_conversion)) { ty = CODESYS_NO_CONVERSION; }
922 else if (EQ (type, Qinternal)) { ty = CODESYS_INTERNAL; }
925 signal_simple_error ("Invalid coding system type", type);
929 codesys = allocate_coding_system (ty, name);
931 if (NILP (doc_string))
932 doc_string = build_string ("");
934 CHECK_STRING (doc_string);
935 CODING_SYSTEM_DOC_STRING (codesys) = doc_string;
937 EXTERNAL_PROPERTY_LIST_LOOP (rest, key, value, props)
939 if (EQ (key, Qmnemonic))
942 CHECK_STRING (value);
943 CODING_SYSTEM_MNEMONIC (codesys) = value;
946 else if (EQ (key, Qeol_type))
948 need_to_setup_eol_systems = NILP (value);
951 CODING_SYSTEM_EOL_TYPE (codesys) = symbol_to_eol_type (value);
954 else if (EQ (key, Qpost_read_conversion))
955 CODING_SYSTEM_POST_READ_CONVERSION (codesys) = value;
956 else if (EQ (key, Qpre_write_conversion))
957 CODING_SYSTEM_PRE_WRITE_CONVERSION (codesys) = value;
959 else if (EQ (key, Qdisable_composition))
960 CODING_SYSTEM_DISABLE_COMPOSITION (codesys) = !NILP (value);
963 else if (ty == CODESYS_ISO2022)
965 #define FROB_INITIAL_CHARSET(charset_num) \
966 CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, charset_num) = \
967 ((EQ (value, Qt) || EQ (value, Qnil)) ? value : Fget_charset (value))
969 if (EQ (key, Qcharset_g0)) FROB_INITIAL_CHARSET (0);
970 else if (EQ (key, Qcharset_g1)) FROB_INITIAL_CHARSET (1);
971 else if (EQ (key, Qcharset_g2)) FROB_INITIAL_CHARSET (2);
972 else if (EQ (key, Qcharset_g3)) FROB_INITIAL_CHARSET (3);
974 #define FROB_FORCE_CHARSET(charset_num) \
975 CODING_SYSTEM_ISO2022_FORCE_CHARSET_ON_OUTPUT (codesys, charset_num) = !NILP (value)
977 else if (EQ (key, Qforce_g0_on_output)) FROB_FORCE_CHARSET (0);
978 else if (EQ (key, Qforce_g1_on_output)) FROB_FORCE_CHARSET (1);
979 else if (EQ (key, Qforce_g2_on_output)) FROB_FORCE_CHARSET (2);
980 else if (EQ (key, Qforce_g3_on_output)) FROB_FORCE_CHARSET (3);
982 #define FROB_BOOLEAN_PROPERTY(prop) \
983 CODING_SYSTEM_ISO2022_##prop (codesys) = !NILP (value)
985 else if (EQ (key, Qshort)) FROB_BOOLEAN_PROPERTY (SHORT);
986 else if (EQ (key, Qno_ascii_eol)) FROB_BOOLEAN_PROPERTY (NO_ASCII_EOL);
987 else if (EQ (key, Qno_ascii_cntl)) FROB_BOOLEAN_PROPERTY (NO_ASCII_CNTL);
988 else if (EQ (key, Qseven)) FROB_BOOLEAN_PROPERTY (SEVEN);
989 else if (EQ (key, Qlock_shift)) FROB_BOOLEAN_PROPERTY (LOCK_SHIFT);
990 else if (EQ (key, Qno_iso6429)) FROB_BOOLEAN_PROPERTY (NO_ISO6429);
991 else if (EQ (key, Qescape_quoted)) FROB_BOOLEAN_PROPERTY (ESCAPE_QUOTED);
993 else if (EQ (key, Qinput_charset_conversion))
995 codesys->iso2022.input_conv =
996 Dynarr_new (charset_conversion_spec);
997 parse_charset_conversion_specs (codesys->iso2022.input_conv,
1000 else if (EQ (key, Qoutput_charset_conversion))
1002 codesys->iso2022.output_conv =
1003 Dynarr_new (charset_conversion_spec);
1004 parse_charset_conversion_specs (codesys->iso2022.output_conv,
1008 signal_simple_error ("Unrecognized property", key);
1010 else if (EQ (type, Qccl))
1012 if (EQ (key, Qdecode))
1014 CHECK_VECTOR (value);
1015 CODING_SYSTEM_CCL_DECODE (codesys) = value;
1017 else if (EQ (key, Qencode))
1019 CHECK_VECTOR (value);
1020 CODING_SYSTEM_CCL_ENCODE (codesys) = value;
1023 signal_simple_error ("Unrecognized property", key);
1027 signal_simple_error ("Unrecognized property", key);
1030 if (need_to_setup_eol_systems)
1031 setup_eol_coding_systems (codesys);
1034 Lisp_Object codesys_obj;
1035 XSETCODING_SYSTEM (codesys_obj, codesys);
1036 Fputhash (name, codesys_obj, Vcoding_system_hash_table);
1041 DEFUN ("copy-coding-system", Fcopy_coding_system, 2, 2, 0, /*
1042 Copy OLD-CODING-SYSTEM to NEW-NAME.
1043 If NEW-NAME does not name an existing coding system, a new one will
1046 (old_coding_system, new_name))
1048 Lisp_Object new_coding_system;
1049 old_coding_system = Fget_coding_system (old_coding_system);
1050 new_coding_system = Ffind_coding_system (new_name);
1051 if (NILP (new_coding_system))
1053 XSETCODING_SYSTEM (new_coding_system,
1054 allocate_coding_system
1055 (XCODING_SYSTEM_TYPE (old_coding_system),
1057 Fputhash (new_name, new_coding_system, Vcoding_system_hash_table);
1061 Lisp_Coding_System *to = XCODING_SYSTEM (new_coding_system);
1062 Lisp_Coding_System *from = XCODING_SYSTEM (old_coding_system);
1063 memcpy (((char *) to ) + sizeof (to->header),
1064 ((char *) from) + sizeof (from->header),
1065 sizeof (*from) - sizeof (from->header));
1066 to->name = new_name;
1068 return new_coding_system;
1071 DEFUN ("coding-system-canonical-name-p", Fcoding_system_canonical_name_p, 1, 1, 0, /*
1072 Return t if OBJECT names a coding system, and is not a coding system alias.
1076 return CODING_SYSTEMP (Fgethash (object, Vcoding_system_hash_table, Qnil))
1080 DEFUN ("coding-system-alias-p", Fcoding_system_alias_p, 1, 1, 0, /*
1081 Return t if OBJECT is a coding system alias.
1082 All coding system aliases are created by `define-coding-system-alias'.
1086 return SYMBOLP (Fgethash (object, Vcoding_system_hash_table, Qzero))
1090 DEFUN ("coding-system-aliasee", Fcoding_system_aliasee, 1, 1, 0, /*
1091 Return the coding-system symbol for which symbol ALIAS is an alias.
1095 Lisp_Object aliasee = Fgethash (alias, Vcoding_system_hash_table, Qnil);
1096 if (SYMBOLP (aliasee))
1099 signal_simple_error ("Symbol is not a coding system alias", alias);
1103 append_suffix_to_symbol (Lisp_Object symbol, const char *ascii_string)
1105 return Fintern (concat2 (Fsymbol_name (symbol), build_string (ascii_string)),
1109 /* A maphash function, for removing dangling coding system aliases. */
1111 dangling_coding_system_alias_p (Lisp_Object alias,
1112 Lisp_Object aliasee,
1113 void *dangling_aliases)
1115 if (SYMBOLP (aliasee)
1116 && NILP (Fgethash (aliasee, Vcoding_system_hash_table, Qnil)))
1118 (*(int *) dangling_aliases)++;
1125 DEFUN ("define-coding-system-alias", Fdefine_coding_system_alias, 2, 2, 0, /*
1126 Define symbol ALIAS as an alias for coding system ALIASEE.
1128 You can use this function to redefine an alias that has already been defined,
1129 but you cannot redefine a name which is the canonical name for a coding system.
1130 \(a canonical name of a coding system is what is returned when you call
1131 `coding-system-name' on a coding system).
1133 ALIASEE itself can be an alias, which allows you to define nested aliases.
1135 You are forbidden, however, from creating alias loops or `dangling' aliases.
1136 These will be detected, and an error will be signaled if you attempt to do so.
1138 If ALIASEE is nil, then ALIAS will simply be undefined.
1140 See also `coding-system-alias-p', `coding-system-aliasee',
1141 and `coding-system-canonical-name-p'.
1145 Lisp_Object real_coding_system, probe;
1147 CHECK_SYMBOL (alias);
1149 if (!NILP (Fcoding_system_canonical_name_p (alias)))
1151 ("Symbol is the canonical name of a coding system and cannot be redefined",
1156 Lisp_Object subsidiary_unix = append_suffix_to_symbol (alias, "-unix");
1157 Lisp_Object subsidiary_dos = append_suffix_to_symbol (alias, "-dos");
1158 Lisp_Object subsidiary_mac = append_suffix_to_symbol (alias, "-mac");
1160 Fremhash (alias, Vcoding_system_hash_table);
1162 /* Undefine subsidiary aliases,
1163 presumably created by a previous call to this function */
1164 if (! NILP (Fcoding_system_alias_p (subsidiary_unix)) &&
1165 ! NILP (Fcoding_system_alias_p (subsidiary_dos)) &&
1166 ! NILP (Fcoding_system_alias_p (subsidiary_mac)))
1168 Fdefine_coding_system_alias (subsidiary_unix, Qnil);
1169 Fdefine_coding_system_alias (subsidiary_dos, Qnil);
1170 Fdefine_coding_system_alias (subsidiary_mac, Qnil);
1173 /* Undefine dangling coding system aliases. */
1175 int dangling_aliases;
1178 dangling_aliases = 0;
1179 elisp_map_remhash (dangling_coding_system_alias_p,
1180 Vcoding_system_hash_table,
1182 } while (dangling_aliases > 0);
1188 if (CODING_SYSTEMP (aliasee))
1189 aliasee = XCODING_SYSTEM_NAME (aliasee);
1191 /* Checks that aliasee names a coding-system */
1192 real_coding_system = Fget_coding_system (aliasee);
1194 /* Check for coding system alias loops */
1195 if (EQ (alias, aliasee))
1196 alias_loop: signal_simple_error_2
1197 ("Attempt to create a coding system alias loop", alias, aliasee);
1199 for (probe = aliasee;
1201 probe = Fgethash (probe, Vcoding_system_hash_table, Qzero))
1203 if (EQ (probe, alias))
1207 Fputhash (alias, aliasee, Vcoding_system_hash_table);
1209 /* Set up aliases for subsidiaries.
1210 #### There must be a better way to handle subsidiary coding systems. */
1212 static const char *suffixes[] = { "-unix", "-dos", "-mac" };
1214 for (i = 0; i < countof (suffixes); i++)
1216 Lisp_Object alias_subsidiary =
1217 append_suffix_to_symbol (alias, suffixes[i]);
1218 Lisp_Object aliasee_subsidiary =
1219 append_suffix_to_symbol (aliasee, suffixes[i]);
1221 if (! NILP (Ffind_coding_system (aliasee_subsidiary)))
1222 Fdefine_coding_system_alias (alias_subsidiary, aliasee_subsidiary);
1225 /* FSF return value is a vector of [ALIAS-unix ALIAS-dos ALIAS-mac],
1226 but it doesn't look intentional, so I'd rather return something
1227 meaningful or nothing at all. */
1232 subsidiary_coding_system (Lisp_Object coding_system, eol_type_t type)
1234 Lisp_Coding_System *cs = XCODING_SYSTEM (coding_system);
1235 Lisp_Object new_coding_system;
1237 if (CODING_SYSTEM_EOL_TYPE (cs) != EOL_AUTODETECT)
1238 return coding_system;
1242 case EOL_AUTODETECT: return coding_system;
1243 case EOL_LF: new_coding_system = CODING_SYSTEM_EOL_LF (cs); break;
1244 case EOL_CR: new_coding_system = CODING_SYSTEM_EOL_CR (cs); break;
1245 case EOL_CRLF: new_coding_system = CODING_SYSTEM_EOL_CRLF (cs); break;
1249 return NILP (new_coding_system) ? coding_system : new_coding_system;
1252 DEFUN ("subsidiary-coding-system", Fsubsidiary_coding_system, 2, 2, 0, /*
1253 Return the subsidiary coding system of CODING-SYSTEM with eol type EOL-TYPE.
1255 (coding_system, eol_type))
1257 coding_system = Fget_coding_system (coding_system);
1259 return subsidiary_coding_system (coding_system,
1260 symbol_to_eol_type (eol_type));
1264 /************************************************************************/
1265 /* Coding system accessors */
1266 /************************************************************************/
1268 DEFUN ("coding-system-doc-string", Fcoding_system_doc_string, 1, 1, 0, /*
1269 Return the doc string for CODING-SYSTEM.
1273 coding_system = Fget_coding_system (coding_system);
1274 return XCODING_SYSTEM_DOC_STRING (coding_system);
1277 DEFUN ("coding-system-type", Fcoding_system_type, 1, 1, 0, /*
1278 Return the type of CODING-SYSTEM.
1282 switch (XCODING_SYSTEM_TYPE (Fget_coding_system (coding_system)))
1285 case CODESYS_AUTODETECT: return Qundecided;
1287 case CODESYS_SHIFT_JIS: return Qshift_jis;
1288 case CODESYS_ISO2022: return Qiso2022;
1289 case CODESYS_BIG5: return Qbig5;
1290 case CODESYS_UCS4: return Qucs4;
1291 case CODESYS_UTF8: return Qutf8;
1292 case CODESYS_CCL: return Qccl;
1294 case CODESYS_NO_CONVERSION: return Qno_conversion;
1296 case CODESYS_INTERNAL: return Qinternal;
1303 Lisp_Object coding_system_charset (Lisp_Object coding_system, int gnum)
1306 = XCODING_SYSTEM_ISO2022_INITIAL_CHARSET (coding_system, gnum);
1308 return CHARSETP (cs) ? XCHARSET_NAME (cs) : Qnil;
1311 DEFUN ("coding-system-charset", Fcoding_system_charset, 2, 2, 0, /*
1312 Return initial charset of CODING-SYSTEM designated to GNUM.
1315 (coding_system, gnum))
1317 coding_system = Fget_coding_system (coding_system);
1320 return coding_system_charset (coding_system, XINT (gnum));
1324 DEFUN ("coding-system-property", Fcoding_system_property, 2, 2, 0, /*
1325 Return the PROP property of CODING-SYSTEM.
1327 (coding_system, prop))
1330 enum coding_system_type type;
1332 coding_system = Fget_coding_system (coding_system);
1333 CHECK_SYMBOL (prop);
1334 type = XCODING_SYSTEM_TYPE (coding_system);
1336 for (i = 0; !ok && i < Dynarr_length (the_codesys_prop_dynarr); i++)
1337 if (EQ (Dynarr_at (the_codesys_prop_dynarr, i).sym, prop))
1340 switch (Dynarr_at (the_codesys_prop_dynarr, i).prop_type)
1342 case CODESYS_PROP_ALL_OK:
1345 case CODESYS_PROP_ISO2022:
1346 if (type != CODESYS_ISO2022)
1348 ("Property only valid in ISO2022 coding systems",
1352 case CODESYS_PROP_CCL:
1353 if (type != CODESYS_CCL)
1355 ("Property only valid in CCL coding systems",
1365 signal_simple_error ("Unrecognized property", prop);
1367 if (EQ (prop, Qname))
1368 return XCODING_SYSTEM_NAME (coding_system);
1369 else if (EQ (prop, Qtype))
1370 return Fcoding_system_type (coding_system);
1371 else if (EQ (prop, Qdoc_string))
1372 return XCODING_SYSTEM_DOC_STRING (coding_system);
1373 else if (EQ (prop, Qmnemonic))
1374 return XCODING_SYSTEM_MNEMONIC (coding_system);
1375 else if (EQ (prop, Qeol_type))
1376 return eol_type_to_symbol (XCODING_SYSTEM_EOL_TYPE (coding_system));
1377 else if (EQ (prop, Qeol_lf))
1378 return XCODING_SYSTEM_EOL_LF (coding_system);
1379 else if (EQ (prop, Qeol_crlf))
1380 return XCODING_SYSTEM_EOL_CRLF (coding_system);
1381 else if (EQ (prop, Qeol_cr))
1382 return XCODING_SYSTEM_EOL_CR (coding_system);
1383 else if (EQ (prop, Qpost_read_conversion))
1384 return XCODING_SYSTEM_POST_READ_CONVERSION (coding_system);
1385 else if (EQ (prop, Qpre_write_conversion))
1386 return XCODING_SYSTEM_PRE_WRITE_CONVERSION (coding_system);
1388 else if (type == CODESYS_ISO2022)
1390 if (EQ (prop, Qcharset_g0))
1391 return coding_system_charset (coding_system, 0);
1392 else if (EQ (prop, Qcharset_g1))
1393 return coding_system_charset (coding_system, 1);
1394 else if (EQ (prop, Qcharset_g2))
1395 return coding_system_charset (coding_system, 2);
1396 else if (EQ (prop, Qcharset_g3))
1397 return coding_system_charset (coding_system, 3);
1399 #define FORCE_CHARSET(charset_num) \
1400 (XCODING_SYSTEM_ISO2022_FORCE_CHARSET_ON_OUTPUT \
1401 (coding_system, charset_num) ? Qt : Qnil)
1403 else if (EQ (prop, Qforce_g0_on_output)) return FORCE_CHARSET (0);
1404 else if (EQ (prop, Qforce_g1_on_output)) return FORCE_CHARSET (1);
1405 else if (EQ (prop, Qforce_g2_on_output)) return FORCE_CHARSET (2);
1406 else if (EQ (prop, Qforce_g3_on_output)) return FORCE_CHARSET (3);
1408 #define LISP_BOOLEAN(prop) \
1409 (XCODING_SYSTEM_ISO2022_##prop (coding_system) ? Qt : Qnil)
1411 else if (EQ (prop, Qshort)) return LISP_BOOLEAN (SHORT);
1412 else if (EQ (prop, Qno_ascii_eol)) return LISP_BOOLEAN (NO_ASCII_EOL);
1413 else if (EQ (prop, Qno_ascii_cntl)) return LISP_BOOLEAN (NO_ASCII_CNTL);
1414 else if (EQ (prop, Qseven)) return LISP_BOOLEAN (SEVEN);
1415 else if (EQ (prop, Qlock_shift)) return LISP_BOOLEAN (LOCK_SHIFT);
1416 else if (EQ (prop, Qno_iso6429)) return LISP_BOOLEAN (NO_ISO6429);
1417 else if (EQ (prop, Qescape_quoted)) return LISP_BOOLEAN (ESCAPE_QUOTED);
1419 else if (EQ (prop, Qinput_charset_conversion))
1421 unparse_charset_conversion_specs
1422 (XCODING_SYSTEM (coding_system)->iso2022.input_conv);
1423 else if (EQ (prop, Qoutput_charset_conversion))
1425 unparse_charset_conversion_specs
1426 (XCODING_SYSTEM (coding_system)->iso2022.output_conv);
1430 else if (type == CODESYS_CCL)
1432 if (EQ (prop, Qdecode))
1433 return XCODING_SYSTEM_CCL_DECODE (coding_system);
1434 else if (EQ (prop, Qencode))
1435 return XCODING_SYSTEM_CCL_ENCODE (coding_system);
1443 return Qnil; /* not reached */
1447 /************************************************************************/
1448 /* Coding category functions */
1449 /************************************************************************/
1452 decode_coding_category (Lisp_Object symbol)
1456 CHECK_SYMBOL (symbol);
1457 for (i = 0; i <= CODING_CATEGORY_LAST; i++)
1458 if (EQ (coding_category_symbol[i], symbol))
1461 signal_simple_error ("Unrecognized coding category", symbol);
1462 return 0; /* not reached */
1465 DEFUN ("coding-category-list", Fcoding_category_list, 0, 0, 0, /*
1466 Return a list of all recognized coding categories.
1471 Lisp_Object list = Qnil;
1473 for (i = CODING_CATEGORY_LAST; i >= 0; i--)
1474 list = Fcons (coding_category_symbol[i], list);
1478 DEFUN ("set-coding-priority-list", Fset_coding_priority_list, 1, 1, 0, /*
1479 Change the priority order of the coding categories.
1480 LIST should be list of coding categories, in descending order of
1481 priority. Unspecified coding categories will be lower in priority
1482 than all specified ones, in the same relative order they were in
1487 int category_to_priority[CODING_CATEGORY_LAST + 1];
1491 /* First generate a list that maps coding categories to priorities. */
1493 for (i = 0; i <= CODING_CATEGORY_LAST; i++)
1494 category_to_priority[i] = -1;
1496 /* Highest priority comes from the specified list. */
1498 EXTERNAL_LIST_LOOP (rest, list)
1500 int cat = decode_coding_category (XCAR (rest));
1502 if (category_to_priority[cat] >= 0)
1503 signal_simple_error ("Duplicate coding category in list", XCAR (rest));
1504 category_to_priority[cat] = i++;
1507 /* Now go through the existing categories by priority to retrieve
1508 the categories not yet specified and preserve their priority
1510 for (j = 0; j <= CODING_CATEGORY_LAST; j++)
1512 int cat = fcd->coding_category_by_priority[j];
1513 if (category_to_priority[cat] < 0)
1514 category_to_priority[cat] = i++;
1517 /* Now we need to construct the inverse of the mapping we just
1520 for (i = 0; i <= CODING_CATEGORY_LAST; i++)
1521 fcd->coding_category_by_priority[category_to_priority[i]] = i;
1523 /* Phew! That was confusing. */
1527 DEFUN ("coding-priority-list", Fcoding_priority_list, 0, 0, 0, /*
1528 Return a list of coding categories in descending order of priority.
1533 Lisp_Object list = Qnil;
1535 for (i = CODING_CATEGORY_LAST; i >= 0; i--)
1536 list = Fcons (coding_category_symbol[fcd->coding_category_by_priority[i]],
1541 DEFUN ("set-coding-category-system", Fset_coding_category_system, 2, 2, 0, /*
1542 Change the coding system associated with a coding category.
1544 (coding_category, coding_system))
1546 int cat = decode_coding_category (coding_category);
1548 coding_system = Fget_coding_system (coding_system);
1549 fcd->coding_category_system[cat] = coding_system;
1553 DEFUN ("coding-category-system", Fcoding_category_system, 1, 1, 0, /*
1554 Return the coding system associated with a coding category.
1558 int cat = decode_coding_category (coding_category);
1559 Lisp_Object sys = fcd->coding_category_system[cat];
1562 return XCODING_SYSTEM_NAME (sys);
1567 /************************************************************************/
1568 /* Detecting the encoding of data */
1569 /************************************************************************/
1571 struct detection_state
1573 eol_type_t eol_type;
1609 struct iso2022_decoder iso;
1611 int high_byte_count;
1612 unsigned int saw_single_shift:1;
1625 acceptable_control_char_p (int c)
1629 /* Allow and ignore control characters that you might
1630 reasonably see in a text file */
1635 case 8: /* backspace */
1636 case 11: /* vertical tab */
1637 case 12: /* form feed */
1638 case 26: /* MS-DOS C-z junk */
1639 case 31: /* '^_' -- for info */
1647 mask_has_at_most_one_bit_p (int mask)
1649 /* Perhaps the only thing useful you learn from intensive Microsoft
1650 technical interviews */
1651 return (mask & (mask - 1)) == 0;
1655 detect_eol_type (struct detection_state *st, const unsigned char *src,
1665 if (st->eol.just_saw_cr)
1667 else if (st->eol.seen_anything)
1670 else if (st->eol.just_saw_cr)
1673 st->eol.just_saw_cr = 1;
1675 st->eol.just_saw_cr = 0;
1676 st->eol.seen_anything = 1;
1679 return EOL_AUTODETECT;
1682 /* Attempt to determine the encoding and EOL type of the given text.
1683 Before calling this function for the first type, you must initialize
1684 st->eol_type as appropriate and initialize st->mask to ~0.
1686 st->eol_type holds the determined EOL type, or EOL_AUTODETECT if
1689 st->mask holds the determined coding category mask, or ~0 if only
1690 ASCII has been seen so far.
1694 0 == st->eol_type is EOL_AUTODETECT and/or more than coding category
1695 is present in st->mask
1696 1 == definitive answers are here for both st->eol_type and st->mask
1700 detect_coding_type (struct detection_state *st, const Extbyte *src,
1701 unsigned int n, int just_do_eol)
1705 if (st->eol_type == EOL_AUTODETECT)
1706 st->eol_type = detect_eol_type (st, src, n);
1709 return st->eol_type != EOL_AUTODETECT;
1711 if (!st->seen_non_ascii)
1713 for (; n; n--, src++)
1716 if ((c < 0x20 && !acceptable_control_char_p (c)) || c >= 0x80)
1718 st->seen_non_ascii = 1;
1720 st->shift_jis.mask = ~0;
1724 st->iso2022.mask = ~0;
1734 if (!mask_has_at_most_one_bit_p (st->iso2022.mask))
1735 st->iso2022.mask = detect_coding_iso2022 (st, src, n);
1736 if (!mask_has_at_most_one_bit_p (st->shift_jis.mask))
1737 st->shift_jis.mask = detect_coding_sjis (st, src, n);
1738 if (!mask_has_at_most_one_bit_p (st->big5.mask))
1739 st->big5.mask = detect_coding_big5 (st, src, n);
1740 if (!mask_has_at_most_one_bit_p (st->utf8.mask))
1741 st->utf8.mask = detect_coding_utf8 (st, src, n);
1742 if (!mask_has_at_most_one_bit_p (st->ucs4.mask))
1743 st->ucs4.mask = detect_coding_ucs4 (st, src, n);
1746 = st->iso2022.mask | st->shift_jis.mask | st->big5.mask
1747 | st->utf8.mask | st->ucs4.mask;
1750 int retval = mask_has_at_most_one_bit_p (st->mask);
1751 st->mask |= CODING_CATEGORY_NO_CONVERSION_MASK;
1752 return retval && st->eol_type != EOL_AUTODETECT;
1757 coding_system_from_mask (int mask)
1761 /* If the file was entirely or basically ASCII, use the
1762 default value of `buffer-file-coding-system'. */
1763 Lisp_Object retval =
1764 XBUFFER (Vbuffer_defaults)->buffer_file_coding_system;
1767 retval = Ffind_coding_system (retval);
1771 (Qbad_variable, Qwarning,
1772 "Invalid `default-buffer-file-coding-system', set to nil");
1773 XBUFFER (Vbuffer_defaults)->buffer_file_coding_system = Qnil;
1777 retval = Fget_coding_system (Qraw_text);
1785 mask = postprocess_iso2022_mask (mask);
1787 /* Look through the coding categories by priority and find
1788 the first one that is allowed. */
1789 for (i = 0; i <= CODING_CATEGORY_LAST; i++)
1791 cat = fcd->coding_category_by_priority[i];
1792 if ((mask & (1 << cat)) &&
1793 !NILP (fcd->coding_category_system[cat]))
1797 return fcd->coding_category_system[cat];
1799 return Fget_coding_system (Qraw_text);
1803 /* Given a seekable read stream and potential coding system and EOL type
1804 as specified, do any autodetection that is called for. If the
1805 coding system and/or EOL type are not `autodetect', they will be left
1806 alone; but this function will never return an autodetect coding system
1809 This function does not automatically fetch subsidiary coding systems;
1810 that should be unnecessary with the explicit eol-type argument. */
1812 #define LENGTH(string_constant) (sizeof (string_constant) - 1)
1815 determine_real_coding_system (Lstream *stream, Lisp_Object *codesys_in_out,
1816 eol_type_t *eol_type_in_out)
1818 struct detection_state decst;
1820 if (*eol_type_in_out == EOL_AUTODETECT)
1821 *eol_type_in_out = XCODING_SYSTEM_EOL_TYPE (*codesys_in_out);
1824 decst.eol_type = *eol_type_in_out;
1827 /* If autodetection is called for, do it now. */
1828 if (XCODING_SYSTEM_TYPE (*codesys_in_out) == CODESYS_AUTODETECT
1829 || *eol_type_in_out == EOL_AUTODETECT)
1832 Lisp_Object coding_system = Qnil;
1834 ssize_t nread = Lstream_read (stream, buf, sizeof (buf));
1837 /* Look for initial "-*-"; mode line prefix */
1839 scan_end = buf + nread - LENGTH ("-*-coding:?-*-");
1844 if (*p == '-' && *(p+1) == '*' && *(p+2) == '-')
1846 Extbyte *local_vars_beg = p + 3;
1847 /* Look for final "-*-"; mode line suffix */
1848 for (p = local_vars_beg,
1849 scan_end = buf + nread - LENGTH ("-*-");
1854 if (*p == '-' && *(p+1) == '*' && *(p+2) == '-')
1856 Extbyte *suffix = p;
1857 /* Look for "coding:" */
1858 for (p = local_vars_beg,
1859 scan_end = suffix - LENGTH ("coding:?");
1862 if (memcmp ("coding:", p, LENGTH ("coding:")) == 0
1863 && (p == local_vars_beg
1864 || (*(p-1) == ' ' ||
1870 p += LENGTH ("coding:");
1871 while (*p == ' ' || *p == '\t') p++;
1873 /* Get coding system name */
1874 save = *suffix; *suffix = '\0';
1875 /* Characters valid in a MIME charset name (rfc 1521),
1876 and in a Lisp symbol name. */
1877 n = strspn ( (char *) p,
1878 "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
1879 "abcdefghijklmnopqrstuvwxyz"
1885 save = p[n]; p[n] = '\0';
1887 Ffind_coding_system (intern ((char *) p));
1897 if (NILP (coding_system))
1900 if (detect_coding_type (&decst, buf, nread,
1901 XCODING_SYSTEM_TYPE (*codesys_in_out)
1902 != CODESYS_AUTODETECT))
1904 nread = Lstream_read (stream, buf, sizeof (buf));
1910 else if (XCODING_SYSTEM_TYPE (*codesys_in_out) == CODESYS_AUTODETECT
1911 && XCODING_SYSTEM_EOL_TYPE (coding_system) == EOL_AUTODETECT)
1914 if (detect_coding_type (&decst, buf, nread, 1))
1916 nread = Lstream_read (stream, buf, sizeof (buf));
1922 *eol_type_in_out = decst.eol_type;
1923 if (XCODING_SYSTEM_TYPE (*codesys_in_out) == CODESYS_AUTODETECT)
1925 if (NILP (coding_system))
1926 *codesys_in_out = coding_system_from_mask (decst.mask);
1928 *codesys_in_out = coding_system;
1932 /* If we absolutely can't determine the EOL type, just assume LF. */
1933 if (*eol_type_in_out == EOL_AUTODETECT)
1934 *eol_type_in_out = EOL_LF;
1936 Lstream_rewind (stream);
1939 DEFUN ("detect-coding-region", Fdetect_coding_region, 2, 3, 0, /*
1940 Detect coding system of the text in the region between START and END.
1941 Returned a list of possible coding systems ordered by priority.
1942 If only ASCII characters are found, it returns 'undecided or one of
1943 its subsidiary coding systems according to a detected end-of-line
1944 type. Optional arg BUFFER defaults to the current buffer.
1946 (start, end, buffer))
1948 Lisp_Object val = Qnil;
1949 struct buffer *buf = decode_buffer (buffer, 0);
1951 Lisp_Object instream, lb_instream;
1952 Lstream *istr, *lb_istr;
1953 struct detection_state decst;
1954 struct gcpro gcpro1, gcpro2;
1956 get_buffer_range_char (buf, start, end, &b, &e, 0);
1957 lb_instream = make_lisp_buffer_input_stream (buf, b, e, 0);
1958 lb_istr = XLSTREAM (lb_instream);
1959 instream = make_encoding_input_stream (lb_istr, Fget_coding_system (Qbinary));
1960 istr = XLSTREAM (instream);
1961 GCPRO2 (instream, lb_instream);
1963 decst.eol_type = EOL_AUTODETECT;
1967 unsigned char random_buffer[4096];
1968 ssize_t nread = Lstream_read (istr, random_buffer, sizeof (random_buffer));
1972 if (detect_coding_type (&decst, random_buffer, nread, 0))
1976 if (decst.mask == ~0)
1977 val = subsidiary_coding_system (Fget_coding_system (Qundecided),
1985 decst.mask = postprocess_iso2022_mask (decst.mask);
1987 for (i = CODING_CATEGORY_LAST; i >= 0; i--)
1989 int sys = fcd->coding_category_by_priority[i];
1990 if (decst.mask & (1 << sys))
1992 Lisp_Object codesys = fcd->coding_category_system[sys];
1993 if (!NILP (codesys))
1994 codesys = subsidiary_coding_system (codesys, decst.eol_type);
1995 val = Fcons (codesys, val);
1999 Lstream_close (istr);
2001 Lstream_delete (istr);
2002 Lstream_delete (lb_istr);
2007 /************************************************************************/
2008 /* Converting to internal Mule format ("decoding") */
2009 /************************************************************************/
2011 /* A decoding stream is a stream used for decoding text (i.e.
2012 converting from some external format to internal format).
2013 The decoding-stream object keeps track of the actual coding
2014 stream, the stream that is at the other end, and data that
2015 needs to be persistent across the lifetime of the stream. */
2017 /* Handle the EOL stuff related to just-read-in character C.
2018 EOL_TYPE is the EOL type of the coding stream.
2019 FLAGS is the current value of FLAGS in the coding stream, and may
2020 be modified by this macro. (The macro only looks at the
2021 CODING_STATE_CR flag.) DST is the Dynarr to which the decoded
2022 bytes are to be written. You need to also define a local goto
2023 label "label_continue_loop" that is at the end of the main
2024 character-reading loop.
2026 If C is a CR character, then this macro handles it entirely and
2027 jumps to label_continue_loop. Otherwise, this macro does not add
2028 anything to DST, and continues normally. You should continue
2029 processing C normally after this macro. */
2031 #define DECODE_HANDLE_EOL_TYPE(eol_type, c, flags, dst) \
2035 if (eol_type == EOL_CR) \
2036 Dynarr_add (dst, '\n'); \
2037 else if (eol_type != EOL_CRLF || flags & CODING_STATE_CR) \
2038 Dynarr_add (dst, c); \
2040 flags |= CODING_STATE_CR; \
2041 goto label_continue_loop; \
2043 else if (flags & CODING_STATE_CR) \
2044 { /* eol_type == CODING_SYSTEM_EOL_CRLF */ \
2046 Dynarr_add (dst, '\r'); \
2047 flags &= ~CODING_STATE_CR; \
2051 /* C should be a binary character in the range 0 - 255; convert
2052 to internal format and add to Dynarr DST. */
2055 #define DECODE_ADD_BINARY_CHAR(c, dst) \
2057 if (BYTE_ASCII_P (c)) \
2058 Dynarr_add (dst, c); \
2061 Dynarr_add (dst, (c >> 6) | 0xc0); \
2062 Dynarr_add (dst, (c & 0x3f) | 0x80); \
2066 INLINE_HEADER void DECODE_ADD_UCS_CHAR(Emchar c, unsigned_char_dynarr* dst);
2068 DECODE_ADD_UCS_CHAR(Emchar c, unsigned_char_dynarr* dst)
2072 Dynarr_add (dst, c);
2074 else if ( c <= 0x7ff )
2076 Dynarr_add (dst, (c >> 6) | 0xc0);
2077 Dynarr_add (dst, (c & 0x3f) | 0x80);
2079 else if ( c <= 0xffff )
2081 Dynarr_add (dst, (c >> 12) | 0xe0);
2082 Dynarr_add (dst, ((c >> 6) & 0x3f) | 0x80);
2083 Dynarr_add (dst, (c & 0x3f) | 0x80);
2085 else if ( c <= 0x1fffff )
2087 Dynarr_add (dst, (c >> 18) | 0xf0);
2088 Dynarr_add (dst, ((c >> 12) & 0x3f) | 0x80);
2089 Dynarr_add (dst, ((c >> 6) & 0x3f) | 0x80);
2090 Dynarr_add (dst, (c & 0x3f) | 0x80);
2092 else if ( c <= 0x3ffffff )
2094 Dynarr_add (dst, (c >> 24) | 0xf8);
2095 Dynarr_add (dst, ((c >> 18) & 0x3f) | 0x80);
2096 Dynarr_add (dst, ((c >> 12) & 0x3f) | 0x80);
2097 Dynarr_add (dst, ((c >> 6) & 0x3f) | 0x80);
2098 Dynarr_add (dst, (c & 0x3f) | 0x80);
2102 Dynarr_add (dst, (c >> 30) | 0xfc);
2103 Dynarr_add (dst, ((c >> 24) & 0x3f) | 0x80);
2104 Dynarr_add (dst, ((c >> 18) & 0x3f) | 0x80);
2105 Dynarr_add (dst, ((c >> 12) & 0x3f) | 0x80);
2106 Dynarr_add (dst, ((c >> 6) & 0x3f) | 0x80);
2107 Dynarr_add (dst, (c & 0x3f) | 0x80);
2111 #define DECODE_ADD_BINARY_CHAR(c, dst) \
2113 if (BYTE_ASCII_P (c)) \
2114 Dynarr_add (dst, c); \
2115 else if (BYTE_C1_P (c)) \
2117 Dynarr_add (dst, LEADING_BYTE_CONTROL_1); \
2118 Dynarr_add (dst, c + 0x20); \
2122 Dynarr_add (dst, LEADING_BYTE_LATIN_ISO8859_1); \
2123 Dynarr_add (dst, c); \
2128 #define DECODE_OUTPUT_PARTIAL_CHAR(ch) \
2132 DECODE_ADD_BINARY_CHAR (ch, dst); \
2137 #define DECODE_HANDLE_END_OF_CONVERSION(flags, ch, dst) \
2139 if (flags & CODING_STATE_END) \
2141 DECODE_OUTPUT_PARTIAL_CHAR (ch); \
2142 if (flags & CODING_STATE_CR) \
2143 Dynarr_add (dst, '\r'); \
2147 #define DECODING_STREAM_DATA(stream) LSTREAM_TYPE_DATA (stream, decoding)
2149 struct decoding_stream
2151 /* Coding system that governs the conversion. */
2152 Lisp_Coding_System *codesys;
2154 /* Stream that we read the encoded data from or
2155 write the decoded data to. */
2158 /* If we are reading, then we can return only a fixed amount of
2159 data, so if the conversion resulted in too much data, we store it
2160 here for retrieval the next time around. */
2161 unsigned_char_dynarr *runoff;
2163 /* FLAGS holds flags indicating the current state of the decoding.
2164 Some of these flags are dependent on the coding system. */
2167 /* CPOS holds a partially built-up code-point of character. */
2170 /* EOL_TYPE specifies the type of end-of-line conversion that
2171 currently applies. We need to keep this separate from the
2172 EOL type stored in CODESYS because the latter might indicate
2173 automatic EOL-type detection while the former will always
2174 indicate a particular EOL type. */
2175 eol_type_t eol_type;
2177 /* Additional ISO2022 information. We define the structure above
2178 because it's also needed by the detection routines. */
2179 struct iso2022_decoder iso2022;
2181 /* Additional information (the state of the running CCL program)
2182 used by the CCL decoder. */
2183 struct ccl_program ccl;
2185 /* counter for UTF-8 or UCS-4 */
2186 unsigned char counter;
2189 unsigned combined_char_count;
2190 Emchar combined_chars[16];
2191 Lisp_Object combining_table;
2193 struct detection_state decst;
2197 extern Lisp_Object Vcharacter_composition_table;
2200 COMPOSE_FLUSH_CHARS (struct decoding_stream *str, unsigned_char_dynarr* dst);
2202 COMPOSE_FLUSH_CHARS (struct decoding_stream *str, unsigned_char_dynarr* dst)
2206 for (i = 0; i < str->combined_char_count; i++)
2207 DECODE_ADD_UCS_CHAR (str->combined_chars[i], dst);
2208 str->combined_char_count = 0;
2209 str->combining_table = Qnil;
2212 void COMPOSE_ADD_CHAR(struct decoding_stream *str, Emchar character,
2213 unsigned_char_dynarr* dst);
2215 COMPOSE_ADD_CHAR(struct decoding_stream *str,
2216 Emchar character, unsigned_char_dynarr* dst)
2218 if (CODING_SYSTEM_DISABLE_COMPOSITION (str->codesys))
2219 DECODE_ADD_UCS_CHAR (character, dst);
2220 else if (!CHAR_ID_TABLE_P (str->combining_table))
2223 = get_char_id_table (character, Vcharacter_composition_table);
2226 DECODE_ADD_UCS_CHAR (character, dst);
2229 str->combined_chars[0] = character;
2230 str->combined_char_count = 1;
2231 str->combining_table = ret;
2237 = get_char_id_table (character, str->combining_table);
2241 Emchar char2 = XCHARVAL (ret);
2242 ret = get_char_id_table (char2, Vcharacter_composition_table);
2245 DECODE_ADD_UCS_CHAR (char2, dst);
2246 str->combined_char_count = 0;
2247 str->combining_table = Qnil;
2251 str->combined_chars[0] = char2;
2252 str->combined_char_count = 1;
2253 str->combining_table = ret;
2256 else if (CHAR_ID_TABLE_P (ret))
2258 str->combined_chars[str->combined_char_count++] = character;
2259 str->combining_table = ret;
2263 COMPOSE_FLUSH_CHARS (str, dst);
2264 DECODE_ADD_UCS_CHAR (character, dst);
2268 #else /* not UTF2000 */
2269 #define COMPOSE_FLUSH_CHARS(str, dst)
2270 #define COMPOSE_ADD_CHAR(str, ch, dst) DECODE_ADD_UCS_CHAR (ch, dst)
2271 #endif /* UTF2000 */
2273 static ssize_t decoding_reader (Lstream *stream,
2274 unsigned char *data, size_t size);
2275 static ssize_t decoding_writer (Lstream *stream,
2276 const unsigned char *data, size_t size);
2277 static int decoding_rewinder (Lstream *stream);
2278 static int decoding_seekable_p (Lstream *stream);
2279 static int decoding_flusher (Lstream *stream);
2280 static int decoding_closer (Lstream *stream);
2282 static Lisp_Object decoding_marker (Lisp_Object stream);
2284 DEFINE_LSTREAM_IMPLEMENTATION ("decoding", lstream_decoding,
2285 sizeof (struct decoding_stream));
2288 decoding_marker (Lisp_Object stream)
2290 Lstream *str = DECODING_STREAM_DATA (XLSTREAM (stream))->other_end;
2291 Lisp_Object str_obj;
2293 /* We do not need to mark the coding systems or charsets stored
2294 within the stream because they are stored in a global list
2295 and automatically marked. */
2297 XSETLSTREAM (str_obj, str);
2298 mark_object (str_obj);
2299 if (str->imp->marker)
2300 return (str->imp->marker) (str_obj);
2305 /* Read SIZE bytes of data and store it into DATA. We are a decoding stream
2306 so we read data from the other end, decode it, and store it into DATA. */
2309 decoding_reader (Lstream *stream, unsigned char *data, size_t size)
2311 struct decoding_stream *str = DECODING_STREAM_DATA (stream);
2312 unsigned char *orig_data = data;
2314 int error_occurred = 0;
2316 /* We need to interface to mule_decode(), which expects to take some
2317 amount of data and store the result into a Dynarr. We have
2318 mule_decode() store into str->runoff, and take data from there
2321 /* We loop until we have enough data, reading chunks from the other
2322 end and decoding it. */
2325 /* Take data from the runoff if we can. Make sure to take at
2326 most SIZE bytes, and delete the data from the runoff. */
2327 if (Dynarr_length (str->runoff) > 0)
2329 size_t chunk = min (size, (size_t) Dynarr_length (str->runoff));
2330 memcpy (data, Dynarr_atp (str->runoff, 0), chunk);
2331 Dynarr_delete_many (str->runoff, 0, chunk);
2337 break; /* No more room for data */
2339 if (str->flags & CODING_STATE_END)
2340 /* This means that on the previous iteration, we hit the EOF on
2341 the other end. We loop once more so that mule_decode() can
2342 output any final stuff it may be holding, or any "go back
2343 to a sane state" escape sequences. (This latter makes sense
2344 during encoding.) */
2347 /* Exhausted the runoff, so get some more. DATA has at least
2348 SIZE bytes left of storage in it, so it's OK to read directly
2349 into it. (We'll be overwriting above, after we've decoded it
2350 into the runoff.) */
2351 read_size = Lstream_read (str->other_end, data, size);
2358 /* There might be some more end data produced in the translation.
2359 See the comment above. */
2360 str->flags |= CODING_STATE_END;
2361 mule_decode (stream, data, str->runoff, read_size);
2364 if (data - orig_data == 0)
2365 return error_occurred ? -1 : 0;
2367 return data - orig_data;
2371 decoding_writer (Lstream *stream, const unsigned char *data, size_t size)
2373 struct decoding_stream *str = DECODING_STREAM_DATA (stream);
2376 /* Decode all our data into the runoff, and then attempt to write
2377 it all out to the other end. Remove whatever chunk we succeeded
2379 mule_decode (stream, data, str->runoff, size);
2380 retval = Lstream_write (str->other_end, Dynarr_atp (str->runoff, 0),
2381 Dynarr_length (str->runoff));
2383 Dynarr_delete_many (str->runoff, 0, retval);
2384 /* Do NOT return retval. The return value indicates how much
2385 of the incoming data was written, not how many bytes were
2391 reset_decoding_stream (struct decoding_stream *str)
2394 if (CODING_SYSTEM_TYPE (str->codesys) == CODESYS_ISO2022)
2396 Lisp_Object coding_system;
2397 XSETCODING_SYSTEM (coding_system, str->codesys);
2398 reset_iso2022 (coding_system, &str->iso2022);
2400 else if (CODING_SYSTEM_TYPE (str->codesys) == CODESYS_CCL)
2402 setup_ccl_program (&str->ccl, CODING_SYSTEM_CCL_DECODE (str->codesys));
2407 str->combined_char_count = 0;
2408 str->combining_table = Qnil;
2410 str->flags = str->cpos = 0;
2414 decoding_rewinder (Lstream *stream)
2416 struct decoding_stream *str = DECODING_STREAM_DATA (stream);
2417 reset_decoding_stream (str);
2418 Dynarr_reset (str->runoff);
2419 return Lstream_rewind (str->other_end);
2423 decoding_seekable_p (Lstream *stream)
2425 struct decoding_stream *str = DECODING_STREAM_DATA (stream);
2426 return Lstream_seekable_p (str->other_end);
2430 decoding_flusher (Lstream *stream)
2432 struct decoding_stream *str = DECODING_STREAM_DATA (stream);
2433 return Lstream_flush (str->other_end);
2437 decoding_closer (Lstream *stream)
2439 struct decoding_stream *str = DECODING_STREAM_DATA (stream);
2440 if (stream->flags & LSTREAM_FL_WRITE)
2442 str->flags |= CODING_STATE_END;
2443 decoding_writer (stream, 0, 0);
2445 Dynarr_free (str->runoff);
2447 #ifdef ENABLE_COMPOSITE_CHARS
2448 if (str->iso2022.composite_chars)
2449 Dynarr_free (str->iso2022.composite_chars);
2452 return Lstream_close (str->other_end);
2456 decoding_stream_coding_system (Lstream *stream)
2458 Lisp_Object coding_system;
2459 struct decoding_stream *str = DECODING_STREAM_DATA (stream);
2461 XSETCODING_SYSTEM (coding_system, str->codesys);
2462 return subsidiary_coding_system (coding_system, str->eol_type);
2466 set_decoding_stream_coding_system (Lstream *lstr, Lisp_Object codesys)
2468 Lisp_Coding_System *cs = XCODING_SYSTEM (codesys);
2469 struct decoding_stream *str = DECODING_STREAM_DATA (lstr);
2471 if (CODING_SYSTEM_EOL_TYPE (cs) != EOL_AUTODETECT)
2472 str->eol_type = CODING_SYSTEM_EOL_TYPE (cs);
2473 reset_decoding_stream (str);
2476 /* WARNING WARNING WARNING WARNING!!!!! If you open up a decoding
2477 stream for writing, no automatic code detection will be performed.
2478 The reason for this is that automatic code detection requires a
2479 seekable input. Things will also fail if you open a decoding
2480 stream for reading using a non-fully-specified coding system and
2481 a non-seekable input stream. */
2484 make_decoding_stream_1 (Lstream *stream, Lisp_Object codesys,
2487 Lstream *lstr = Lstream_new (lstream_decoding, mode);
2488 struct decoding_stream *str = DECODING_STREAM_DATA (lstr);
2492 str->other_end = stream;
2493 str->runoff = (unsigned_char_dynarr *) Dynarr_new (unsigned_char);
2494 str->eol_type = EOL_AUTODETECT;
2495 if (!strcmp (mode, "r")
2496 && Lstream_seekable_p (stream))
2497 /* We can determine the coding system now. */
2498 determine_real_coding_system (stream, &codesys, &str->eol_type);
2499 set_decoding_stream_coding_system (lstr, codesys);
2500 str->decst.eol_type = str->eol_type;
2501 str->decst.mask = ~0;
2502 XSETLSTREAM (obj, lstr);
2507 make_decoding_input_stream (Lstream *stream, Lisp_Object codesys)
2509 return make_decoding_stream_1 (stream, codesys, "r");
2513 make_decoding_output_stream (Lstream *stream, Lisp_Object codesys)
2515 return make_decoding_stream_1 (stream, codesys, "w");
2518 /* Note: the decode_coding_* functions all take the same
2519 arguments as mule_decode(), which is to say some SRC data of
2520 size N, which is to be stored into dynamic array DST.
2521 DECODING is the stream within which the decoding is
2522 taking place, but no data is actually read from or
2523 written to that stream; that is handled in decoding_reader()
2524 or decoding_writer(). This allows the same functions to
2525 be used for both reading and writing. */
2528 mule_decode (Lstream *decoding, const unsigned char *src,
2529 unsigned_char_dynarr *dst, unsigned int n)
2531 struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
2533 /* If necessary, do encoding-detection now. We do this when
2534 we're a writing stream or a non-seekable reading stream,
2535 meaning that we can't just process the whole input,
2536 rewind, and start over. */
2538 if (CODING_SYSTEM_TYPE (str->codesys) == CODESYS_AUTODETECT ||
2539 str->eol_type == EOL_AUTODETECT)
2541 Lisp_Object codesys;
2543 XSETCODING_SYSTEM (codesys, str->codesys);
2544 detect_coding_type (&str->decst, src, n,
2545 CODING_SYSTEM_TYPE (str->codesys) !=
2546 CODESYS_AUTODETECT);
2547 if (CODING_SYSTEM_TYPE (str->codesys) == CODESYS_AUTODETECT &&
2548 str->decst.mask != ~0)
2549 /* #### This is cheesy. What we really ought to do is
2550 buffer up a certain amount of data so as to get a
2551 less random result. */
2552 codesys = coding_system_from_mask (str->decst.mask);
2553 str->eol_type = str->decst.eol_type;
2554 if (XCODING_SYSTEM (codesys) != str->codesys)
2556 /* Preserve the CODING_STATE_END flag in case it was set.
2557 If we erase it, bad things might happen. */
2558 int was_end = str->flags & CODING_STATE_END;
2559 set_decoding_stream_coding_system (decoding, codesys);
2561 str->flags |= CODING_STATE_END;
2565 switch (CODING_SYSTEM_TYPE (str->codesys))
2568 case CODESYS_INTERNAL:
2569 Dynarr_add_many (dst, src, n);
2572 case CODESYS_AUTODETECT:
2573 /* If we got this far and still haven't decided on the coding
2574 system, then do no conversion. */
2575 case CODESYS_NO_CONVERSION:
2576 decode_coding_no_conversion (decoding, src, dst, n);
2579 case CODESYS_SHIFT_JIS:
2580 decode_coding_sjis (decoding, src, dst, n);
2583 decode_coding_big5 (decoding, src, dst, n);
2586 decode_coding_ucs4 (decoding, src, dst, n);
2589 decode_coding_utf8 (decoding, src, dst, n);
2592 str->ccl.last_block = str->flags & CODING_STATE_END;
2593 ccl_driver (&str->ccl, src, dst, n, 0, CCL_MODE_DECODING);
2595 case CODESYS_ISO2022:
2596 decode_coding_iso2022 (decoding, src, dst, n);
2604 DEFUN ("decode-coding-region", Fdecode_coding_region, 3, 4, 0, /*
2605 Decode the text between START and END which is encoded in CODING-SYSTEM.
2606 This is useful if you've read in encoded text from a file without decoding
2607 it (e.g. you read in a JIS-formatted file but used the `binary' or
2608 `no-conversion' coding system, so that it shows up as "^[$B!<!+^[(B").
2609 Return length of decoded text.
2610 BUFFER defaults to the current buffer if unspecified.
2612 (start, end, coding_system, buffer))
2615 struct buffer *buf = decode_buffer (buffer, 0);
2616 Lisp_Object instream, lb_outstream, de_outstream, outstream;
2617 Lstream *istr, *ostr;
2618 struct gcpro gcpro1, gcpro2, gcpro3, gcpro4;
2620 get_buffer_range_char (buf, start, end, &b, &e, 0);
2622 barf_if_buffer_read_only (buf, b, e);
2624 coding_system = Fget_coding_system (coding_system);
2625 instream = make_lisp_buffer_input_stream (buf, b, e, 0);
2626 lb_outstream = make_lisp_buffer_output_stream (buf, b, 0);
2627 de_outstream = make_decoding_output_stream (XLSTREAM (lb_outstream),
2629 outstream = make_encoding_output_stream (XLSTREAM (de_outstream),
2630 Fget_coding_system (Qbinary));
2631 istr = XLSTREAM (instream);
2632 ostr = XLSTREAM (outstream);
2633 GCPRO4 (instream, lb_outstream, de_outstream, outstream);
2635 /* The chain of streams looks like this:
2637 [BUFFER] <----- send through
2638 ------> [ENCODE AS BINARY]
2639 ------> [DECODE AS SPECIFIED]
2645 char tempbuf[1024]; /* some random amount */
2646 Bufpos newpos, even_newer_pos;
2647 Bufpos oldpos = lisp_buffer_stream_startpos (istr);
2648 ssize_t size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
2652 newpos = lisp_buffer_stream_startpos (istr);
2653 Lstream_write (ostr, tempbuf, size_in_bytes);
2654 even_newer_pos = lisp_buffer_stream_startpos (istr);
2655 buffer_delete_range (buf, even_newer_pos - (newpos - oldpos),
2658 Lstream_close (istr);
2659 Lstream_close (ostr);
2661 Lstream_delete (istr);
2662 Lstream_delete (ostr);
2663 Lstream_delete (XLSTREAM (de_outstream));
2664 Lstream_delete (XLSTREAM (lb_outstream));
2669 /************************************************************************/
2670 /* Converting to an external encoding ("encoding") */
2671 /************************************************************************/
2673 /* An encoding stream is an output stream. When you create the
2674 stream, you specify the coding system that governs the encoding
2675 and another stream that the resulting encoded data is to be
2676 sent to, and then start sending data to it. */
2678 #define ENCODING_STREAM_DATA(stream) LSTREAM_TYPE_DATA (stream, encoding)
2680 struct encoding_stream
2682 /* Coding system that governs the conversion. */
2683 Lisp_Coding_System *codesys;
2685 /* Stream that we read the encoded data from or
2686 write the decoded data to. */
2689 /* If we are reading, then we can return only a fixed amount of
2690 data, so if the conversion resulted in too much data, we store it
2691 here for retrieval the next time around. */
2692 unsigned_char_dynarr *runoff;
2694 /* FLAGS holds flags indicating the current state of the encoding.
2695 Some of these flags are dependent on the coding system. */
2698 /* CH holds a partially built-up character. Since we only deal
2699 with one- and two-byte characters at the moment, we only use
2700 this to store the first byte of a two-byte character. */
2703 /* Additional information used by the ISO2022 encoder. */
2706 /* CHARSET holds the character sets currently assigned to the G0
2707 through G3 registers. It is initialized from the array
2708 INITIAL_CHARSET in CODESYS. */
2709 Lisp_Object charset[4];
2711 /* Which registers are currently invoked into the left (GL) and
2712 right (GR) halves of the 8-bit encoding space? */
2713 int register_left, register_right;
2715 /* Whether we need to explicitly designate the charset in the
2716 G? register before using it. It is initialized from the
2717 array FORCE_CHARSET_ON_OUTPUT in CODESYS. */
2718 unsigned char force_charset_on_output[4];
2720 /* Other state variables that need to be preserved across
2722 Lisp_Object current_charset;
2724 int current_char_boundary;
2727 void (*encode_char) (struct encoding_stream *str, Emchar c,
2728 unsigned_char_dynarr *dst, unsigned int *flags);
2729 void (*finish) (struct encoding_stream *str,
2730 unsigned_char_dynarr *dst, unsigned int *flags);
2732 /* Additional information (the state of the running CCL program)
2733 used by the CCL encoder. */
2734 struct ccl_program ccl;
2738 static ssize_t encoding_reader (Lstream *stream, unsigned char *data, size_t size);
2739 static ssize_t encoding_writer (Lstream *stream, const unsigned char *data,
2741 static int encoding_rewinder (Lstream *stream);
2742 static int encoding_seekable_p (Lstream *stream);
2743 static int encoding_flusher (Lstream *stream);
2744 static int encoding_closer (Lstream *stream);
2746 static Lisp_Object encoding_marker (Lisp_Object stream);
2748 DEFINE_LSTREAM_IMPLEMENTATION ("encoding", lstream_encoding,
2749 sizeof (struct encoding_stream));
2752 encoding_marker (Lisp_Object stream)
2754 Lstream *str = ENCODING_STREAM_DATA (XLSTREAM (stream))->other_end;
2755 Lisp_Object str_obj;
2757 /* We do not need to mark the coding systems or charsets stored
2758 within the stream because they are stored in a global list
2759 and automatically marked. */
2761 XSETLSTREAM (str_obj, str);
2762 mark_object (str_obj);
2763 if (str->imp->marker)
2764 return (str->imp->marker) (str_obj);
2769 /* Read SIZE bytes of data and store it into DATA. We are a encoding stream
2770 so we read data from the other end, encode it, and store it into DATA. */
2773 encoding_reader (Lstream *stream, unsigned char *data, size_t size)
2775 struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
2776 unsigned char *orig_data = data;
2778 int error_occurred = 0;
2780 /* We need to interface to mule_encode(), which expects to take some
2781 amount of data and store the result into a Dynarr. We have
2782 mule_encode() store into str->runoff, and take data from there
2785 /* We loop until we have enough data, reading chunks from the other
2786 end and encoding it. */
2789 /* Take data from the runoff if we can. Make sure to take at
2790 most SIZE bytes, and delete the data from the runoff. */
2791 if (Dynarr_length (str->runoff) > 0)
2793 int chunk = min ((int) size, Dynarr_length (str->runoff));
2794 memcpy (data, Dynarr_atp (str->runoff, 0), chunk);
2795 Dynarr_delete_many (str->runoff, 0, chunk);
2801 break; /* No more room for data */
2803 if (str->flags & CODING_STATE_END)
2804 /* This means that on the previous iteration, we hit the EOF on
2805 the other end. We loop once more so that mule_encode() can
2806 output any final stuff it may be holding, or any "go back
2807 to a sane state" escape sequences. (This latter makes sense
2808 during encoding.) */
2811 /* Exhausted the runoff, so get some more. DATA at least SIZE bytes
2812 left of storage in it, so it's OK to read directly into it.
2813 (We'll be overwriting above, after we've encoded it into the
2815 read_size = Lstream_read (str->other_end, data, size);
2822 /* There might be some more end data produced in the translation.
2823 See the comment above. */
2824 str->flags |= CODING_STATE_END;
2825 mule_encode (stream, data, str->runoff, read_size);
2828 if (data == orig_data)
2829 return error_occurred ? -1 : 0;
2831 return data - orig_data;
2835 encoding_writer (Lstream *stream, const unsigned char *data, size_t size)
2837 struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
2840 /* Encode all our data into the runoff, and then attempt to write
2841 it all out to the other end. Remove whatever chunk we succeeded
2843 mule_encode (stream, data, str->runoff, size);
2844 retval = Lstream_write (str->other_end, Dynarr_atp (str->runoff, 0),
2845 Dynarr_length (str->runoff));
2847 Dynarr_delete_many (str->runoff, 0, retval);
2848 /* Do NOT return retval. The return value indicates how much
2849 of the incoming data was written, not how many bytes were
2855 reset_encoding_stream (struct encoding_stream *str)
2858 switch (CODING_SYSTEM_TYPE (str->codesys))
2860 case CODESYS_ISO2022:
2864 str->encode_char = &char_encode_iso2022;
2865 str->finish = &char_finish_iso2022;
2866 for (i = 0; i < 4; i++)
2868 str->iso2022.charset[i] =
2869 CODING_SYSTEM_ISO2022_INITIAL_CHARSET (str->codesys, i);
2870 str->iso2022.force_charset_on_output[i] =
2871 CODING_SYSTEM_ISO2022_FORCE_CHARSET_ON_OUTPUT (str->codesys, i);
2873 str->iso2022.register_left = 0;
2874 str->iso2022.register_right = 1;
2875 str->iso2022.current_charset = Qnil;
2876 str->iso2022.current_half = 0;
2880 setup_ccl_program (&str->ccl, CODING_SYSTEM_CCL_ENCODE (str->codesys));
2883 str->encode_char = &char_encode_utf8;
2884 str->finish = &char_finish_utf8;
2887 str->encode_char = &char_encode_ucs4;
2888 str->finish = &char_finish_ucs4;
2890 case CODESYS_SHIFT_JIS:
2891 str->encode_char = &char_encode_shift_jis;
2892 str->finish = &char_finish_shift_jis;
2898 str->iso2022.current_char_boundary = 0;
2899 str->flags = str->ch = 0;
2903 encoding_rewinder (Lstream *stream)
2905 struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
2906 reset_encoding_stream (str);
2907 Dynarr_reset (str->runoff);
2908 return Lstream_rewind (str->other_end);
2912 encoding_seekable_p (Lstream *stream)
2914 struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
2915 return Lstream_seekable_p (str->other_end);
2919 encoding_flusher (Lstream *stream)
2921 struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
2922 return Lstream_flush (str->other_end);
2926 encoding_closer (Lstream *stream)
2928 struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
2929 if (stream->flags & LSTREAM_FL_WRITE)
2931 str->flags |= CODING_STATE_END;
2932 encoding_writer (stream, 0, 0);
2934 Dynarr_free (str->runoff);
2935 return Lstream_close (str->other_end);
2939 encoding_stream_coding_system (Lstream *stream)
2941 Lisp_Object coding_system;
2942 struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
2944 XSETCODING_SYSTEM (coding_system, str->codesys);
2945 return coding_system;
2949 set_encoding_stream_coding_system (Lstream *lstr, Lisp_Object codesys)
2951 Lisp_Coding_System *cs = XCODING_SYSTEM (codesys);
2952 struct encoding_stream *str = ENCODING_STREAM_DATA (lstr);
2954 reset_encoding_stream (str);
2958 make_encoding_stream_1 (Lstream *stream, Lisp_Object codesys,
2961 Lstream *lstr = Lstream_new (lstream_encoding, mode);
2962 struct encoding_stream *str = ENCODING_STREAM_DATA (lstr);
2966 str->runoff = Dynarr_new (unsigned_char);
2967 str->other_end = stream;
2968 set_encoding_stream_coding_system (lstr, codesys);
2969 XSETLSTREAM (obj, lstr);
2974 make_encoding_input_stream (Lstream *stream, Lisp_Object codesys)
2976 return make_encoding_stream_1 (stream, codesys, "r");
2980 make_encoding_output_stream (Lstream *stream, Lisp_Object codesys)
2982 return make_encoding_stream_1 (stream, codesys, "w");
2985 /* Convert N bytes of internally-formatted data stored in SRC to an
2986 external format, according to the encoding stream ENCODING.
2987 Store the encoded data into DST. */
2990 mule_encode (Lstream *encoding, const unsigned char *src,
2991 unsigned_char_dynarr *dst, unsigned int n)
2993 struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
2995 switch (CODING_SYSTEM_TYPE (str->codesys))
2998 case CODESYS_INTERNAL:
2999 Dynarr_add_many (dst, src, n);
3002 case CODESYS_AUTODETECT:
3003 /* If we got this far and still haven't decided on the coding
3004 system, then do no conversion. */
3005 case CODESYS_NO_CONVERSION:
3006 encode_coding_no_conversion (encoding, src, dst, n);
3010 encode_coding_big5 (encoding, src, dst, n);
3013 str->ccl.last_block = str->flags & CODING_STATE_END;
3014 ccl_driver (&str->ccl, src, dst, n, 0, CCL_MODE_ENCODING);
3018 text_encode_generic (encoding, src, dst, n);
3022 DEFUN ("encode-coding-region", Fencode_coding_region, 3, 4, 0, /*
3023 Encode the text between START and END using CODING-SYSTEM.
3024 This will, for example, convert Japanese characters into stuff such as
3025 "^[$B!<!+^[(B" if you use the JIS encoding. Return length of encoded
3026 text. BUFFER defaults to the current buffer if unspecified.
3028 (start, end, coding_system, buffer))
3031 struct buffer *buf = decode_buffer (buffer, 0);
3032 Lisp_Object instream, lb_outstream, de_outstream, outstream;
3033 Lstream *istr, *ostr;
3034 struct gcpro gcpro1, gcpro2, gcpro3, gcpro4;
3036 get_buffer_range_char (buf, start, end, &b, &e, 0);
3038 barf_if_buffer_read_only (buf, b, e);
3040 coding_system = Fget_coding_system (coding_system);
3041 instream = make_lisp_buffer_input_stream (buf, b, e, 0);
3042 lb_outstream = make_lisp_buffer_output_stream (buf, b, 0);
3043 de_outstream = make_decoding_output_stream (XLSTREAM (lb_outstream),
3044 Fget_coding_system (Qbinary));
3045 outstream = make_encoding_output_stream (XLSTREAM (de_outstream),
3047 istr = XLSTREAM (instream);
3048 ostr = XLSTREAM (outstream);
3049 GCPRO4 (instream, outstream, de_outstream, lb_outstream);
3050 /* The chain of streams looks like this:
3052 [BUFFER] <----- send through
3053 ------> [ENCODE AS SPECIFIED]
3054 ------> [DECODE AS BINARY]
3059 char tempbuf[1024]; /* some random amount */
3060 Bufpos newpos, even_newer_pos;
3061 Bufpos oldpos = lisp_buffer_stream_startpos (istr);
3062 ssize_t size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
3066 newpos = lisp_buffer_stream_startpos (istr);
3067 Lstream_write (ostr, tempbuf, size_in_bytes);
3068 even_newer_pos = lisp_buffer_stream_startpos (istr);
3069 buffer_delete_range (buf, even_newer_pos - (newpos - oldpos),
3075 lisp_buffer_stream_startpos (XLSTREAM (instream)) - b;
3076 Lstream_close (istr);
3077 Lstream_close (ostr);
3079 Lstream_delete (istr);
3080 Lstream_delete (ostr);
3081 Lstream_delete (XLSTREAM (de_outstream));
3082 Lstream_delete (XLSTREAM (lb_outstream));
3083 return make_int (retlen);
3090 text_encode_generic (Lstream *encoding, const unsigned char *src,
3091 unsigned_char_dynarr *dst, unsigned int n)
3094 unsigned char char_boundary;
3095 struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
3096 unsigned int flags = str->flags;
3097 Emchar ch = str->ch;
3099 char_boundary = str->iso2022.current_char_boundary;
3105 if (char_boundary == 0)
3133 (*str->encode_char) (str, c, dst, &flags);
3135 else if (char_boundary == 1)
3137 (*str->encode_char) (str, (ch << 6) | (c & 0x3f), dst, &flags);
3143 ch = (ch << 6) | (c & 0x3f);
3148 if ((char_boundary == 0) && (flags & CODING_STATE_END))
3150 (*str->finish) (str, dst, &flags);
3155 str->iso2022.current_char_boundary = char_boundary;
3159 /************************************************************************/
3160 /* Shift-JIS methods */
3161 /************************************************************************/
3163 /* Shift-JIS is a coding system encoding three character sets: ASCII, right
3164 half of JISX0201-Kana, and JISX0208. An ASCII character is encoded
3165 as is. A character of JISX0201-Kana (DIMENSION1_CHARS94 character set) is
3166 encoded by "position-code + 0x80". A character of JISX0208
3167 (DIMENSION2_CHARS94 character set) is encoded in 2-byte but two
3168 position-codes are divided and shifted so that it fit in the range
3171 --- CODE RANGE of Shift-JIS ---
3172 (character set) (range)
3174 JISX0201-Kana 0xA0 .. 0xDF
3175 JISX0208 (1st byte) 0x80 .. 0x9F and 0xE0 .. 0xEF
3176 (2nd byte) 0x40 .. 0x7E and 0x80 .. 0xFC
3177 -------------------------------
3181 /* Is this the first byte of a Shift-JIS two-byte char? */
3183 #define BYTE_SJIS_TWO_BYTE_1_P(c) \
3184 (((c) >= 0x81 && (c) <= 0x9F) || ((c) >= 0xE0 && (c) <= 0xEF))
3186 /* Is this the second byte of a Shift-JIS two-byte char? */
3188 #define BYTE_SJIS_TWO_BYTE_2_P(c) \
3189 (((c) >= 0x40 && (c) <= 0x7E) || ((c) >= 0x80 && (c) <= 0xFC))
3191 #define BYTE_SJIS_KATAKANA_P(c) \
3192 ((c) >= 0xA1 && (c) <= 0xDF)
3195 detect_coding_sjis (struct detection_state *st, const unsigned char *src,
3203 if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO)
3205 if (st->shift_jis.in_second_byte)
3207 st->shift_jis.in_second_byte = 0;
3211 else if ((c >= 0x80 && c < 0xA0) || c >= 0xE0)
3212 st->shift_jis.in_second_byte = 1;
3214 return CODING_CATEGORY_SHIFT_JIS_MASK;
3217 /* Convert Shift-JIS data to internal format. */
3220 decode_coding_sjis (Lstream *decoding, const unsigned char *src,
3221 unsigned_char_dynarr *dst, unsigned int n)
3224 struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
3225 unsigned int flags = str->flags;
3226 unsigned int cpos = str->cpos;
3227 eol_type_t eol_type = str->eol_type;
3235 /* Previous character was first byte of Shift-JIS Kanji char. */
3236 if (BYTE_SJIS_TWO_BYTE_2_P (c))
3238 unsigned char e1, e2;
3240 DECODE_SJIS (cpos, c, e1, e2);
3242 DECODE_ADD_UCS_CHAR(MAKE_CHAR(Vcharset_japanese_jisx0208,
3246 Dynarr_add (dst, LEADING_BYTE_JAPANESE_JISX0208);
3247 Dynarr_add (dst, e1);
3248 Dynarr_add (dst, e2);
3253 DECODE_ADD_BINARY_CHAR (cpos, dst);
3254 DECODE_ADD_BINARY_CHAR (c, dst);
3260 DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
3261 if (BYTE_SJIS_TWO_BYTE_1_P (c))
3263 else if (BYTE_SJIS_KATAKANA_P (c))
3266 DECODE_ADD_UCS_CHAR(MAKE_CHAR(Vcharset_katakana_jisx0201,
3269 Dynarr_add (dst, LEADING_BYTE_KATAKANA_JISX0201);
3270 Dynarr_add (dst, c);
3275 DECODE_ADD_UCS_CHAR(MAKE_CHAR(Vcharset_latin_jisx0201,
3279 DECODE_ADD_BINARY_CHAR (c, dst);
3281 label_continue_loop:;
3284 DECODE_HANDLE_END_OF_CONVERSION (flags, cpos, dst);
3290 /* Convert internal character representation to Shift_JIS. */
3293 char_encode_shift_jis (struct encoding_stream *str, Emchar ch,
3294 unsigned_char_dynarr *dst, unsigned int *flags)
3296 eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
3300 if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT)
3301 Dynarr_add (dst, '\r');
3302 if (eol_type != EOL_CR)
3303 Dynarr_add (dst, ch);
3307 Lisp_Object charset;
3308 unsigned int c1, c2, s1, s2;
3310 int code_point = charset_code_point (Vcharset_latin_jisx0201, ch);
3312 if (code_point >= 0)
3314 charset = Vcharset_latin_jisx0201;
3320 BREAKUP_CHAR (ch, charset, c1, c2);
3322 if (EQ(charset, Vcharset_katakana_jisx0201))
3324 Dynarr_add (dst, c1 | 0x80);
3328 Dynarr_add (dst, c1);
3330 else if (EQ(charset, Vcharset_japanese_jisx0208))
3332 ENCODE_SJIS (c1 | 0x80, c2 | 0x80, s1, s2);
3333 Dynarr_add (dst, s1);
3334 Dynarr_add (dst, s2);
3337 Dynarr_add (dst, '?');
3342 char_finish_shift_jis (struct encoding_stream *str, unsigned_char_dynarr *dst,
3343 unsigned int *flags)
3347 DEFUN ("decode-shift-jis-char", Fdecode_shift_jis_char, 1, 1, 0, /*
3348 Decode a JISX0208 character of Shift-JIS coding-system.
3349 CODE is the character code in Shift-JIS as a cons of type bytes.
3350 Return the corresponding character.
3354 unsigned char c1, c2, s1, s2;
3357 CHECK_INT (XCAR (code));
3358 CHECK_INT (XCDR (code));
3359 s1 = XINT (XCAR (code));
3360 s2 = XINT (XCDR (code));
3361 if (BYTE_SJIS_TWO_BYTE_1_P (s1) &&
3362 BYTE_SJIS_TWO_BYTE_2_P (s2))
3364 DECODE_SJIS (s1, s2, c1, c2);
3365 return make_char (MAKE_CHAR (Vcharset_japanese_jisx0208,
3366 c1 & 0x7F, c2 & 0x7F));
3372 DEFUN ("encode-shift-jis-char", Fencode_shift_jis_char, 1, 1, 0, /*
3373 Encode a JISX0208 character CHAR to SHIFT-JIS coding-system.
3374 Return the corresponding character code in SHIFT-JIS as a cons of two bytes.
3378 Lisp_Object charset;
3381 CHECK_CHAR_COERCE_INT (ch);
3382 BREAKUP_CHAR (XCHAR (ch), charset, c1, c2);
3383 if (EQ (charset, Vcharset_japanese_jisx0208))
3385 ENCODE_SJIS (c1 | 0x80, c2 | 0x80, s1, s2);
3386 return Fcons (make_int (s1), make_int (s2));
3393 /************************************************************************/
3395 /************************************************************************/
3397 /* BIG5 is a coding system encoding two character sets: ASCII and
3398 Big5. An ASCII character is encoded as is. Big5 is a two-byte
3399 character set and is encoded in two-byte.
3401 --- CODE RANGE of BIG5 ---
3402 (character set) (range)
3404 Big5 (1st byte) 0xA1 .. 0xFE
3405 (2nd byte) 0x40 .. 0x7E and 0xA1 .. 0xFE
3406 --------------------------
3408 Since the number of characters in Big5 is larger than maximum
3409 characters in Emacs' charset (96x96), it can't be handled as one
3410 charset. So, in Emacs, Big5 is divided into two: `charset-big5-1'
3411 and `charset-big5-2'. Both <type>s are DIMENSION2_CHARS94. The former
3412 contains frequently used characters and the latter contains less
3413 frequently used characters. */
3415 #define BYTE_BIG5_TWO_BYTE_1_P(c) \
3416 ((c) >= 0xA1 && (c) <= 0xFE)
3418 /* Is this the second byte of a Shift-JIS two-byte char? */
3420 #define BYTE_BIG5_TWO_BYTE_2_P(c) \
3421 (((c) >= 0x40 && (c) <= 0x7E) || ((c) >= 0xA1 && (c) <= 0xFE))
3423 /* Number of Big5 characters which have the same code in 1st byte. */
3425 #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40)
3427 /* Code conversion macros. These are macros because they are used in
3428 inner loops during code conversion.
3430 Note that temporary variables in macros introduce the classic
3431 dynamic-scoping problems with variable names. We use capital-
3432 lettered variables in the assumption that XEmacs does not use
3433 capital letters in variables except in a very formalized way
3436 /* Convert Big5 code (b1, b2) into its internal string representation
3439 /* There is a much simpler way to split the Big5 charset into two.
3440 For the moment I'm going to leave the algorithm as-is because it
3441 claims to separate out the most-used characters into a single
3442 charset, which perhaps will lead to optimizations in various
3445 The way the algorithm works is something like this:
3447 Big5 can be viewed as a 94x157 charset, where the row is
3448 encoded into the bytes 0xA1 .. 0xFE and the column is encoded
3449 into the bytes 0x40 .. 0x7E and 0xA1 .. 0xFE. As for frequency,
3450 the split between low and high column numbers is apparently
3451 meaningless; ascending rows produce less and less frequent chars.
3452 Therefore, we assign the lower half of rows (0xA1 .. 0xC8) to
3453 the first charset, and the upper half (0xC9 .. 0xFE) to the
3454 second. To do the conversion, we convert the character into
3455 a single number where 0 .. 156 is the first row, 157 .. 313
3456 is the second, etc. That way, the characters are ordered by
3457 decreasing frequency. Then we just chop the space in two
3458 and coerce the result into a 94x94 space.
3461 #define DECODE_BIG5(b1, b2, lb, c1, c2) do \
3463 int B1 = b1, B2 = b2; \
3465 = (B1 - 0xA1) * BIG5_SAME_ROW + B2 - (B2 < 0x7F ? 0x40 : 0x62); \
3469 lb = LEADING_BYTE_CHINESE_BIG5_1; \
3473 lb = LEADING_BYTE_CHINESE_BIG5_2; \
3474 I -= (BIG5_SAME_ROW) * (0xC9 - 0xA1); \
3476 c1 = I / (0xFF - 0xA1) + 0xA1; \
3477 c2 = I % (0xFF - 0xA1) + 0xA1; \
3480 /* Convert the internal string representation of a Big5 character
3481 (lb, c1, c2) into Big5 code (b1, b2). */
3483 #define ENCODE_BIG5(lb, c1, c2, b1, b2) do \
3485 unsigned int I = ((c1) - 0xA1) * (0xFF - 0xA1) + ((c2) - 0xA1); \
3487 if (lb == LEADING_BYTE_CHINESE_BIG5_2) \
3489 I += BIG5_SAME_ROW * (0xC9 - 0xA1); \
3491 b1 = I / BIG5_SAME_ROW + 0xA1; \
3492 b2 = I % BIG5_SAME_ROW; \
3493 b2 += b2 < 0x3F ? 0x40 : 0x62; \
3497 detect_coding_big5 (struct detection_state *st, const unsigned char *src,
3505 if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO ||
3506 (c >= 0x80 && c <= 0xA0))
3508 if (st->big5.in_second_byte)
3510 st->big5.in_second_byte = 0;
3511 if (c < 0x40 || (c >= 0x80 && c <= 0xA0))
3515 st->big5.in_second_byte = 1;
3517 return CODING_CATEGORY_BIG5_MASK;
3520 /* Convert Big5 data to internal format. */
3523 decode_coding_big5 (Lstream *decoding, const unsigned char *src,
3524 unsigned_char_dynarr *dst, unsigned int n)
3527 struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
3528 unsigned int flags = str->flags;
3529 unsigned int cpos = str->cpos;
3530 eol_type_t eol_type = str->eol_type;
3537 /* Previous character was first byte of Big5 char. */
3538 if (BYTE_BIG5_TWO_BYTE_2_P (c))
3542 unsigned char b2, b3;
3543 DECODE_BIG5 (cpos, c, b1, b2, b3);
3544 DECODE_ADD_UCS_CHAR (MAKE_CHAR (CHARSET_BY_LEADING_BYTE (b1),
3545 b2 & 0x7F, b3 & 0x7F),
3548 unsigned char b1, b2, b3;
3549 DECODE_BIG5 (cpos, c, b1, b2, b3);
3550 Dynarr_add (dst, b1);
3551 Dynarr_add (dst, b2);
3552 Dynarr_add (dst, b3);
3557 DECODE_ADD_BINARY_CHAR (cpos, dst);
3558 DECODE_ADD_BINARY_CHAR (c, dst);
3564 DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
3565 if (BYTE_BIG5_TWO_BYTE_1_P (c))
3568 DECODE_ADD_BINARY_CHAR (c, dst);
3570 label_continue_loop:;
3573 DECODE_HANDLE_END_OF_CONVERSION (flags, cpos, dst);
3579 /* Convert internally-formatted data to Big5. */
3582 encode_coding_big5 (Lstream *encoding, const unsigned char *src,
3583 unsigned_char_dynarr *dst, unsigned int n)
3587 struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
3588 unsigned int flags = str->flags;
3589 unsigned int ch = str->ch;
3590 eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
3597 if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT)
3598 Dynarr_add (dst, '\r');
3599 if (eol_type != EOL_CR)
3600 Dynarr_add (dst, '\n');
3602 else if (BYTE_ASCII_P (c))
3605 Dynarr_add (dst, c);
3607 else if (BUFBYTE_LEADING_BYTE_P (c))
3609 if (c == LEADING_BYTE_CHINESE_BIG5_1 ||
3610 c == LEADING_BYTE_CHINESE_BIG5_2)
3612 /* A recognized leading byte. */
3614 continue; /* not done with this character. */
3616 /* otherwise just ignore this character. */
3618 else if (ch == LEADING_BYTE_CHINESE_BIG5_1 ||
3619 ch == LEADING_BYTE_CHINESE_BIG5_2)
3621 /* Previous char was a recognized leading byte. */
3623 continue; /* not done with this character. */
3627 /* Encountering second byte of a Big5 character. */
3628 unsigned char b1, b2;
3630 ENCODE_BIG5 (ch >> 8, ch & 0xFF, c, b1, b2);
3631 Dynarr_add (dst, b1);
3632 Dynarr_add (dst, b2);
3644 DEFUN ("decode-big5-char", Fdecode_big5_char, 1, 1, 0, /*
3645 Decode a Big5 character CODE of BIG5 coding-system.
3646 CODE is the character code in BIG5, a cons of two integers.
3647 Return the corresponding character.
3651 unsigned char c1, c2, b1, b2;
3654 CHECK_INT (XCAR (code));
3655 CHECK_INT (XCDR (code));
3656 b1 = XINT (XCAR (code));
3657 b2 = XINT (XCDR (code));
3658 if (BYTE_BIG5_TWO_BYTE_1_P (b1) &&
3659 BYTE_BIG5_TWO_BYTE_2_P (b2))
3661 Charset_ID leading_byte;
3662 Lisp_Object charset;
3663 DECODE_BIG5 (b1, b2, leading_byte, c1, c2);
3664 charset = CHARSET_BY_LEADING_BYTE (leading_byte);
3665 return make_char (MAKE_CHAR (charset, c1 & 0x7F, c2 & 0x7F));
3671 DEFUN ("encode-big5-char", Fencode_big5_char, 1, 1, 0, /*
3672 Encode the Big5 character CH to BIG5 coding-system.
3673 Return the corresponding character code in Big5.
3677 Lisp_Object charset;
3680 CHECK_CHAR_COERCE_INT (ch);
3681 BREAKUP_CHAR (XCHAR (ch), charset, c1, c2);
3682 if (EQ (charset, Vcharset_chinese_big5_1) ||
3683 EQ (charset, Vcharset_chinese_big5_2))
3685 ENCODE_BIG5 (XCHARSET_LEADING_BYTE (charset), c1 | 0x80, c2 | 0x80,
3687 return Fcons (make_int (b1), make_int (b2));
3694 /************************************************************************/
3696 /************************************************************************/
3699 detect_coding_ucs4 (struct detection_state *st, const unsigned char *src,
3705 switch (st->ucs4.in_byte)
3714 st->ucs4.in_byte = 0;
3720 return CODING_CATEGORY_UCS4_MASK;
3724 decode_coding_ucs4 (Lstream *decoding, const unsigned char *src,
3725 unsigned_char_dynarr *dst, unsigned int n)
3727 struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
3728 unsigned int flags = str->flags;
3729 unsigned int cpos = str->cpos;
3730 unsigned char counter = str->counter;
3734 unsigned char c = *src++;
3742 DECODE_ADD_UCS_CHAR ((cpos << 8) | c, dst);
3747 cpos = ( cpos << 8 ) | c;
3751 if (counter & CODING_STATE_END)
3752 DECODE_OUTPUT_PARTIAL_CHAR (cpos);
3756 str->counter = counter;
3760 char_encode_ucs4 (struct encoding_stream *str, Emchar ch,
3761 unsigned_char_dynarr *dst, unsigned int *flags)
3763 Dynarr_add (dst, ch >> 24);
3764 Dynarr_add (dst, ch >> 16);
3765 Dynarr_add (dst, ch >> 8);
3766 Dynarr_add (dst, ch );
3770 char_finish_ucs4 (struct encoding_stream *str, unsigned_char_dynarr *dst,
3771 unsigned int *flags)
3776 /************************************************************************/
3778 /************************************************************************/
3781 detect_coding_utf8 (struct detection_state *st, const unsigned char *src,
3786 unsigned char c = *src++;
3787 switch (st->utf8.in_byte)
3790 if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO)
3793 st->utf8.in_byte = 5;
3795 st->utf8.in_byte = 4;
3797 st->utf8.in_byte = 3;
3799 st->utf8.in_byte = 2;
3801 st->utf8.in_byte = 1;
3806 if ((c & 0xc0) != 0x80)
3812 return CODING_CATEGORY_UTF8_MASK;
3816 decode_coding_utf8 (Lstream *decoding, const unsigned char *src,
3817 unsigned_char_dynarr *dst, unsigned int n)
3819 struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
3820 unsigned int flags = str->flags;
3821 unsigned int cpos = str->cpos;
3822 eol_type_t eol_type = str->eol_type;
3823 unsigned char counter = str->counter;
3827 unsigned char c = *src++;
3836 else if ( c >= 0xf8 )
3841 else if ( c >= 0xf0 )
3846 else if ( c >= 0xe0 )
3851 else if ( c >= 0xc0 )
3858 DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
3859 DECODE_ADD_UCS_CHAR (c, dst);
3863 cpos = ( cpos << 6 ) | ( c & 0x3f );
3864 DECODE_ADD_UCS_CHAR (cpos, dst);
3869 cpos = ( cpos << 6 ) | ( c & 0x3f );
3872 label_continue_loop:;
3875 if (flags & CODING_STATE_END)
3876 DECODE_OUTPUT_PARTIAL_CHAR (cpos);
3880 str->counter = counter;
3884 char_encode_utf8 (struct encoding_stream *str, Emchar ch,
3885 unsigned_char_dynarr *dst, unsigned int *flags)
3887 eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
3891 if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT)
3892 Dynarr_add (dst, '\r');
3893 if (eol_type != EOL_CR)
3894 Dynarr_add (dst, ch);
3896 else if (ch <= 0x7f)
3898 Dynarr_add (dst, ch);
3900 else if (ch <= 0x7ff)
3902 Dynarr_add (dst, (ch >> 6) | 0xc0);
3903 Dynarr_add (dst, (ch & 0x3f) | 0x80);
3905 else if (ch <= 0xffff)
3907 Dynarr_add (dst, (ch >> 12) | 0xe0);
3908 Dynarr_add (dst, ((ch >> 6) & 0x3f) | 0x80);
3909 Dynarr_add (dst, (ch & 0x3f) | 0x80);
3911 else if (ch <= 0x1fffff)
3913 Dynarr_add (dst, (ch >> 18) | 0xf0);
3914 Dynarr_add (dst, ((ch >> 12) & 0x3f) | 0x80);
3915 Dynarr_add (dst, ((ch >> 6) & 0x3f) | 0x80);
3916 Dynarr_add (dst, (ch & 0x3f) | 0x80);
3918 else if (ch <= 0x3ffffff)
3920 Dynarr_add (dst, (ch >> 24) | 0xf8);
3921 Dynarr_add (dst, ((ch >> 18) & 0x3f) | 0x80);
3922 Dynarr_add (dst, ((ch >> 12) & 0x3f) | 0x80);
3923 Dynarr_add (dst, ((ch >> 6) & 0x3f) | 0x80);
3924 Dynarr_add (dst, (ch & 0x3f) | 0x80);
3928 Dynarr_add (dst, (ch >> 30) | 0xfc);
3929 Dynarr_add (dst, ((ch >> 24) & 0x3f) | 0x80);
3930 Dynarr_add (dst, ((ch >> 18) & 0x3f) | 0x80);
3931 Dynarr_add (dst, ((ch >> 12) & 0x3f) | 0x80);
3932 Dynarr_add (dst, ((ch >> 6) & 0x3f) | 0x80);
3933 Dynarr_add (dst, (ch & 0x3f) | 0x80);
3938 char_finish_utf8 (struct encoding_stream *str, unsigned_char_dynarr *dst,
3939 unsigned int *flags)
3944 /************************************************************************/
3945 /* ISO2022 methods */
3946 /************************************************************************/
3948 /* The following note describes the coding system ISO2022 briefly.
3949 Since the intention of this note is to help understand the
3950 functions in this file, some parts are NOT ACCURATE or OVERLY
3951 SIMPLIFIED. For thorough understanding, please refer to the
3952 original document of ISO2022.
3954 ISO2022 provides many mechanisms to encode several character sets
3955 in 7-bit and 8-bit environments. For 7-bit environments, all text
3956 is encoded using bytes less than 128. This may make the encoded
3957 text a little bit longer, but the text passes more easily through
3958 several gateways, some of which strip off MSB (Most Signigant Bit).
3960 There are two kinds of character sets: control character set and
3961 graphic character set. The former contains control characters such
3962 as `newline' and `escape' to provide control functions (control
3963 functions are also provided by escape sequences). The latter
3964 contains graphic characters such as 'A' and '-'. Emacs recognizes
3965 two control character sets and many graphic character sets.
3967 Graphic character sets are classified into one of the following
3968 four classes, according to the number of bytes (DIMENSION) and
3969 number of characters in one dimension (CHARS) of the set:
3970 - DIMENSION1_CHARS94
3971 - DIMENSION1_CHARS96
3972 - DIMENSION2_CHARS94
3973 - DIMENSION2_CHARS96
3975 In addition, each character set is assigned an identification tag,
3976 unique for each set, called "final character" (denoted as <F>
3977 hereafter). The <F> of each character set is decided by ECMA(*)
3978 when it is registered in ISO. The code range of <F> is 0x30..0x7F
3979 (0x30..0x3F are for private use only).
3981 Note (*): ECMA = European Computer Manufacturers Association
3983 Here are examples of graphic character set [NAME(<F>)]:
3984 o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ...
3985 o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ...
3986 o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ...
3987 o DIMENSION2_CHARS96 -- none for the moment
3989 A code area (1 byte = 8 bits) is divided into 4 areas, C0, GL, C1, and GR.
3990 C0 [0x00..0x1F] -- control character plane 0
3991 GL [0x20..0x7F] -- graphic character plane 0
3992 C1 [0x80..0x9F] -- control character plane 1
3993 GR [0xA0..0xFF] -- graphic character plane 1
3995 A control character set is directly designated and invoked to C0 or
3996 C1 by an escape sequence. The most common case is that:
3997 - ISO646's control character set is designated/invoked to C0, and
3998 - ISO6429's control character set is designated/invoked to C1,
3999 and usually these designations/invocations are omitted in encoded
4000 text. In a 7-bit environment, only C0 can be used, and a control
4001 character for C1 is encoded by an appropriate escape sequence to
4002 fit into the environment. All control characters for C1 are
4003 defined to have corresponding escape sequences.
4005 A graphic character set is at first designated to one of four
4006 graphic registers (G0 through G3), then these graphic registers are
4007 invoked to GL or GR. These designations and invocations can be
4008 done independently. The most common case is that G0 is invoked to
4009 GL, G1 is invoked to GR, and ASCII is designated to G0. Usually
4010 these invocations and designations are omitted in encoded text.
4011 In a 7-bit environment, only GL can be used.
4013 When a graphic character set of CHARS94 is invoked to GL, codes
4014 0x20 and 0x7F of the GL area work as control characters SPACE and
4015 DEL respectively, and codes 0xA0 and 0xFF of the GR area should not
4018 There are two ways of invocation: locking-shift and single-shift.
4019 With locking-shift, the invocation lasts until the next different
4020 invocation, whereas with single-shift, the invocation affects the
4021 following character only and doesn't affect the locking-shift
4022 state. Invocations are done by the following control characters or
4025 ----------------------------------------------------------------------
4026 abbrev function cntrl escape seq description
4027 ----------------------------------------------------------------------
4028 SI/LS0 (shift-in) 0x0F none invoke G0 into GL
4029 SO/LS1 (shift-out) 0x0E none invoke G1 into GL
4030 LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL
4031 LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL
4032 LS1R (locking-shift-1 right) none ESC '~' invoke G1 into GR (*)
4033 LS2R (locking-shift-2 right) none ESC '}' invoke G2 into GR (*)
4034 LS3R (locking-shift 3 right) none ESC '|' invoke G3 into GR (*)
4035 SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 for one char
4036 SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 for one char
4037 ----------------------------------------------------------------------
4038 (*) These are not used by any known coding system.
4040 Control characters for these functions are defined by macros
4041 ISO_CODE_XXX in `coding.h'.
4043 Designations are done by the following escape sequences:
4044 ----------------------------------------------------------------------
4045 escape sequence description
4046 ----------------------------------------------------------------------
4047 ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0
4048 ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1
4049 ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2
4050 ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3
4051 ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*)
4052 ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1
4053 ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2
4054 ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3
4055 ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**)
4056 ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1
4057 ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2
4058 ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3
4059 ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*)
4060 ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1
4061 ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2
4062 ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3
4063 ----------------------------------------------------------------------
4065 In this list, "DIMENSION1_CHARS94<F>" means a graphic character set
4066 of dimension 1, chars 94, and final character <F>, etc...
4068 Note (*): Although these designations are not allowed in ISO2022,
4069 Emacs accepts them on decoding, and produces them on encoding
4070 CHARS96 character sets in a coding system which is characterized as
4071 7-bit environment, non-locking-shift, and non-single-shift.
4073 Note (**): If <F> is '@', 'A', or 'B', the intermediate character
4074 '(' can be omitted. We refer to this as "short-form" hereafter.
4076 Now you may notice that there are a lot of ways for encoding the
4077 same multilingual text in ISO2022. Actually, there exist many
4078 coding systems such as Compound Text (used in X11's inter client
4079 communication, ISO-2022-JP (used in Japanese internet), ISO-2022-KR
4080 (used in Korean internet), EUC (Extended UNIX Code, used in Asian
4081 localized platforms), and all of these are variants of ISO2022.
4083 In addition to the above, Emacs handles two more kinds of escape
4084 sequences: ISO6429's direction specification and Emacs' private
4085 sequence for specifying character composition.
4087 ISO6429's direction specification takes the following form:
4088 o CSI ']' -- end of the current direction
4089 o CSI '0' ']' -- end of the current direction
4090 o CSI '1' ']' -- start of left-to-right text
4091 o CSI '2' ']' -- start of right-to-left text
4092 The control character CSI (0x9B: control sequence introducer) is
4093 abbreviated to the escape sequence ESC '[' in a 7-bit environment.
4095 Character composition specification takes the following form:
4096 o ESC '0' -- start character composition
4097 o ESC '1' -- end character composition
4098 Since these are not standard escape sequences of any ISO standard,
4099 their use with these meanings is restricted to Emacs only. */
4102 reset_iso2022 (Lisp_Object coding_system, struct iso2022_decoder *iso)
4106 for (i = 0; i < 4; i++)
4108 if (!NILP (coding_system))
4110 XCODING_SYSTEM_ISO2022_INITIAL_CHARSET (coding_system, i);
4112 iso->charset[i] = Qt;
4113 iso->invalid_designated[i] = 0;
4115 iso->esc = ISO_ESC_NOTHING;
4116 iso->esc_bytes_index = 0;
4117 iso->register_left = 0;
4118 iso->register_right = 1;
4119 iso->switched_dir_and_no_valid_charset_yet = 0;
4120 iso->invalid_switch_dir = 0;
4121 iso->output_direction_sequence = 0;
4122 iso->output_literally = 0;
4123 #ifdef ENABLE_COMPOSITE_CHARS
4124 if (iso->composite_chars)
4125 Dynarr_reset (iso->composite_chars);
4130 fit_to_be_escape_quoted (unsigned char c)
4147 /* Parse one byte of an ISO2022 escape sequence.
4148 If the result is an invalid escape sequence, return 0 and
4149 do not change anything in STR. Otherwise, if the result is
4150 an incomplete escape sequence, update ISO2022.ESC and
4151 ISO2022.ESC_BYTES and return -1. Otherwise, update
4152 all the state variables (but not ISO2022.ESC_BYTES) and
4155 If CHECK_INVALID_CHARSETS is non-zero, check for designation
4156 or invocation of an invalid character set and treat that as
4157 an unrecognized escape sequence. */
4160 parse_iso2022_esc (Lisp_Object codesys, struct iso2022_decoder *iso,
4161 unsigned char c, unsigned int *flags,
4162 int check_invalid_charsets)
4164 /* (1) If we're at the end of a designation sequence, CS is the
4165 charset being designated and REG is the register to designate
4168 (2) If we're at the end of a locking-shift sequence, REG is
4169 the register to invoke and HALF (0 == left, 1 == right) is
4170 the half to invoke it into.
4172 (3) If we're at the end of a single-shift sequence, REG is
4173 the register to invoke. */
4174 Lisp_Object cs = Qnil;
4177 /* NOTE: This code does goto's all over the fucking place.
4178 The reason for this is that we're basically implementing
4179 a state machine here, and hierarchical languages like C
4180 don't really provide a clean way of doing this. */
4182 if (! (*flags & CODING_STATE_ESCAPE))
4183 /* At beginning of escape sequence; we need to reset our
4184 escape-state variables. */
4185 iso->esc = ISO_ESC_NOTHING;
4187 iso->output_literally = 0;
4188 iso->output_direction_sequence = 0;
4192 case ISO_ESC_NOTHING:
4193 iso->esc_bytes_index = 0;
4196 case ISO_CODE_ESC: /* Start escape sequence */
4197 *flags |= CODING_STATE_ESCAPE;
4201 case ISO_CODE_CSI: /* ISO6429 (specifying directionality) */
4202 *flags |= CODING_STATE_ESCAPE;
4203 iso->esc = ISO_ESC_5_11;
4206 case ISO_CODE_SO: /* locking shift 1 */
4209 case ISO_CODE_SI: /* locking shift 0 */
4213 case ISO_CODE_SS2: /* single shift */
4216 case ISO_CODE_SS3: /* single shift */
4220 default: /* Other control characters */
4227 /**** single shift ****/
4229 case 'N': /* single shift 2 */
4232 case 'O': /* single shift 3 */
4236 /**** locking shift ****/
4238 case '~': /* locking shift 1 right */
4241 case 'n': /* locking shift 2 */
4244 case '}': /* locking shift 2 right */
4247 case 'o': /* locking shift 3 */
4250 case '|': /* locking shift 3 right */
4254 #ifdef ENABLE_COMPOSITE_CHARS
4255 /**** composite ****/
4258 iso->esc = ISO_ESC_START_COMPOSITE;
4259 *flags = (*flags & CODING_STATE_ISO2022_LOCK) |
4260 CODING_STATE_COMPOSITE;
4264 iso->esc = ISO_ESC_END_COMPOSITE;
4265 *flags = (*flags & CODING_STATE_ISO2022_LOCK) &
4266 ~CODING_STATE_COMPOSITE;
4268 #endif /* ENABLE_COMPOSITE_CHARS */
4270 /**** directionality ****/
4273 iso->esc = ISO_ESC_5_11;
4276 /**** designation ****/
4278 case '$': /* multibyte charset prefix */
4279 iso->esc = ISO_ESC_2_4;
4283 if (0x28 <= c && c <= 0x2F)
4285 iso->esc = (enum iso_esc_flag) (c - 0x28 + ISO_ESC_2_8);
4289 /* This function is called with CODESYS equal to nil when
4290 doing coding-system detection. */
4292 && XCODING_SYSTEM_ISO2022_ESCAPE_QUOTED (codesys)
4293 && fit_to_be_escape_quoted (c))
4295 iso->esc = ISO_ESC_LITERAL;
4296 *flags &= CODING_STATE_ISO2022_LOCK;
4306 /**** directionality ****/
4308 case ISO_ESC_5_11: /* ISO6429 direction control */
4311 *flags &= (CODING_STATE_ISO2022_LOCK & ~CODING_STATE_R2L);
4312 goto directionality;
4314 if (c == '0') iso->esc = ISO_ESC_5_11_0;
4315 else if (c == '1') iso->esc = ISO_ESC_5_11_1;
4316 else if (c == '2') iso->esc = ISO_ESC_5_11_2;
4320 case ISO_ESC_5_11_0:
4323 *flags &= (CODING_STATE_ISO2022_LOCK & ~CODING_STATE_R2L);
4324 goto directionality;
4328 case ISO_ESC_5_11_1:
4331 *flags = (CODING_STATE_ISO2022_LOCK & ~CODING_STATE_R2L);
4332 goto directionality;
4336 case ISO_ESC_5_11_2:
4339 *flags = (*flags & CODING_STATE_ISO2022_LOCK) | CODING_STATE_R2L;
4340 goto directionality;
4345 iso->esc = ISO_ESC_DIRECTIONALITY;
4346 /* Various junk here to attempt to preserve the direction sequences
4347 literally in the text if they would otherwise be swallowed due
4348 to invalid designations that don't show up as actual charset
4349 changes in the text. */
4350 if (iso->invalid_switch_dir)
4352 /* We already inserted a direction switch literally into the
4353 text. We assume (#### this may not be right) that the
4354 next direction switch is the one going the other way,
4355 and we need to output that literally as well. */
4356 iso->output_literally = 1;
4357 iso->invalid_switch_dir = 0;
4363 /* If we are in the thrall of an invalid designation,
4364 then stick the directionality sequence literally into the
4365 output stream so it ends up in the original text again. */
4366 for (jj = 0; jj < 4; jj++)
4367 if (iso->invalid_designated[jj])
4371 iso->output_literally = 1;
4372 iso->invalid_switch_dir = 1;
4375 /* Indicate that we haven't yet seen a valid designation,
4376 so that if a switch-dir is directly followed by an
4377 invalid designation, both get inserted literally. */
4378 iso->switched_dir_and_no_valid_charset_yet = 1;
4383 /**** designation ****/
4386 if (0x28 <= c && c <= 0x2F)
4388 iso->esc = (enum iso_esc_flag) (c - 0x28 + ISO_ESC_2_4_8);
4391 if (0x40 <= c && c <= 0x42)
4394 cs = CHARSET_BY_ATTRIBUTES (94, -1, c,
4395 *flags & CODING_STATE_R2L ?
4396 CHARSET_RIGHT_TO_LEFT :
4397 CHARSET_LEFT_TO_RIGHT);
4408 if (c < '0' || c > '~')
4409 return 0; /* bad final byte */
4411 if (iso->esc >= ISO_ESC_2_8 &&
4412 iso->esc <= ISO_ESC_2_15)
4414 chars = (iso->esc >= ISO_ESC_2_12) ? 96 : 94;
4415 single = 1; /* single-byte */
4416 reg = (iso->esc - ISO_ESC_2_8) & 3;
4418 else if (iso->esc >= ISO_ESC_2_4_8 &&
4419 iso->esc <= ISO_ESC_2_4_15)
4421 chars = (iso->esc >= ISO_ESC_2_4_12) ? 96 : 94;
4422 single = -1; /* multi-byte */
4423 reg = (iso->esc - ISO_ESC_2_4_8) & 3;
4427 /* Can this ever be reached? -slb */
4431 cs = CHARSET_BY_ATTRIBUTES (chars, single, c,
4432 *flags & CODING_STATE_R2L ?
4433 CHARSET_RIGHT_TO_LEFT :
4434 CHARSET_LEFT_TO_RIGHT);
4440 iso->esc_bytes[iso->esc_bytes_index++] = (unsigned char) c;
4444 if (check_invalid_charsets && !CHARSETP (iso->charset[reg]))
4445 /* can't invoke something that ain't there. */
4447 iso->esc = ISO_ESC_SINGLE_SHIFT;
4448 *flags &= CODING_STATE_ISO2022_LOCK;
4450 *flags |= CODING_STATE_SS2;
4452 *flags |= CODING_STATE_SS3;
4456 if (check_invalid_charsets &&
4457 !CHARSETP (iso->charset[reg]))
4458 /* can't invoke something that ain't there. */
4461 iso->register_right = reg;
4463 iso->register_left = reg;
4464 *flags &= CODING_STATE_ISO2022_LOCK;
4465 iso->esc = ISO_ESC_LOCKING_SHIFT;
4469 if (NILP (cs) && check_invalid_charsets)
4471 iso->invalid_designated[reg] = 1;
4472 iso->charset[reg] = Vcharset_ascii;
4473 iso->esc = ISO_ESC_DESIGNATE;
4474 *flags &= CODING_STATE_ISO2022_LOCK;
4475 iso->output_literally = 1;
4476 if (iso->switched_dir_and_no_valid_charset_yet)
4478 /* We encountered a switch-direction followed by an
4479 invalid designation. Ensure that the switch-direction
4480 gets outputted; otherwise it will probably get eaten
4481 when the text is written out again. */
4482 iso->switched_dir_and_no_valid_charset_yet = 0;
4483 iso->output_direction_sequence = 1;
4484 /* And make sure that the switch-dir going the other
4485 way gets outputted, as well. */
4486 iso->invalid_switch_dir = 1;
4490 /* This function is called with CODESYS equal to nil when
4491 doing coding-system detection. */
4492 if (!NILP (codesys))
4494 charset_conversion_spec_dynarr *dyn =
4495 XCODING_SYSTEM (codesys)->iso2022.input_conv;
4501 for (i = 0; i < Dynarr_length (dyn); i++)
4503 struct charset_conversion_spec *spec = Dynarr_atp (dyn, i);
4504 if (EQ (cs, spec->from_charset))
4505 cs = spec->to_charset;
4510 iso->charset[reg] = cs;
4511 iso->esc = ISO_ESC_DESIGNATE;
4512 *flags &= CODING_STATE_ISO2022_LOCK;
4513 if (iso->invalid_designated[reg])
4515 iso->invalid_designated[reg] = 0;
4516 iso->output_literally = 1;
4518 if (iso->switched_dir_and_no_valid_charset_yet)
4519 iso->switched_dir_and_no_valid_charset_yet = 0;
4524 detect_coding_iso2022 (struct detection_state *st, const unsigned char *src,
4529 /* #### There are serious deficiencies in the recognition mechanism
4530 here. This needs to be much smarter if it's going to cut it.
4531 The sequence "\xff\x0f" is currently detected as LOCK_SHIFT while
4532 it should be detected as Latin-1.
4533 All the ISO2022 stuff in this file should be synced up with the
4534 code from FSF Emacs-20.4, in which Mule should be more or less stable.
4535 Perhaps we should wait till R2L works in FSF Emacs? */
4537 if (!st->iso2022.initted)
4539 reset_iso2022 (Qnil, &st->iso2022.iso);
4540 st->iso2022.mask = (CODING_CATEGORY_ISO_7_MASK |
4541 CODING_CATEGORY_ISO_8_DESIGNATE_MASK |
4542 CODING_CATEGORY_ISO_8_1_MASK |
4543 CODING_CATEGORY_ISO_8_2_MASK |
4544 CODING_CATEGORY_ISO_LOCK_SHIFT_MASK);
4545 st->iso2022.flags = 0;
4546 st->iso2022.high_byte_count = 0;
4547 st->iso2022.saw_single_shift = 0;
4548 st->iso2022.initted = 1;
4551 mask = st->iso2022.mask;
4558 mask &= ~CODING_CATEGORY_ISO_7_MASK;
4559 st->iso2022.high_byte_count++;
4563 if (st->iso2022.high_byte_count && !st->iso2022.saw_single_shift)
4565 if (st->iso2022.high_byte_count & 1)
4566 /* odd number of high bytes; assume not iso-8-2 */
4567 mask &= ~CODING_CATEGORY_ISO_8_2_MASK;
4569 st->iso2022.high_byte_count = 0;
4570 st->iso2022.saw_single_shift = 0;
4572 mask &= ~CODING_CATEGORY_ISO_7_MASK;
4574 if (!(st->iso2022.flags & CODING_STATE_ESCAPE)
4575 && (BYTE_C0_P (c) || BYTE_C1_P (c)))
4576 { /* control chars */
4579 /* Allow and ignore control characters that you might
4580 reasonably see in a text file */
4585 case 8: /* backspace */
4586 case 11: /* vertical tab */
4587 case 12: /* form feed */
4588 case 26: /* MS-DOS C-z junk */
4589 case 31: /* '^_' -- for info */
4590 goto label_continue_loop;
4597 if ((st->iso2022.flags & CODING_STATE_ESCAPE) || BYTE_C0_P (c)
4600 if (parse_iso2022_esc (Qnil, &st->iso2022.iso, c,
4601 &st->iso2022.flags, 0))
4603 switch (st->iso2022.iso.esc)
4605 case ISO_ESC_DESIGNATE:
4606 mask &= ~CODING_CATEGORY_ISO_8_1_MASK;
4607 mask &= ~CODING_CATEGORY_ISO_8_2_MASK;
4609 case ISO_ESC_LOCKING_SHIFT:
4610 mask = CODING_CATEGORY_ISO_LOCK_SHIFT_MASK;
4611 goto ran_out_of_chars;
4612 case ISO_ESC_SINGLE_SHIFT:
4613 mask &= ~CODING_CATEGORY_ISO_8_DESIGNATE_MASK;
4614 st->iso2022.saw_single_shift = 1;
4623 goto ran_out_of_chars;
4626 label_continue_loop:;
4635 postprocess_iso2022_mask (int mask)
4637 /* #### kind of cheesy */
4638 /* If seven-bit ISO is allowed, then assume that the encoding is
4639 entirely seven-bit and turn off the eight-bit ones. */
4640 if (mask & CODING_CATEGORY_ISO_7_MASK)
4641 mask &= ~ (CODING_CATEGORY_ISO_8_DESIGNATE_MASK |
4642 CODING_CATEGORY_ISO_8_1_MASK |
4643 CODING_CATEGORY_ISO_8_2_MASK);
4647 /* If FLAGS is a null pointer or specifies right-to-left motion,
4648 output a switch-dir-to-left-to-right sequence to DST.
4649 Also update FLAGS if it is not a null pointer.
4650 If INTERNAL_P is set, we are outputting in internal format and
4651 need to handle the CSI differently. */
4654 restore_left_to_right_direction (Lisp_Coding_System *codesys,
4655 unsigned_char_dynarr *dst,
4656 unsigned int *flags,
4659 if (!flags || (*flags & CODING_STATE_R2L))
4661 if (CODING_SYSTEM_ISO2022_SEVEN (codesys))
4663 Dynarr_add (dst, ISO_CODE_ESC);
4664 Dynarr_add (dst, '[');
4666 else if (internal_p)
4667 DECODE_ADD_BINARY_CHAR (ISO_CODE_CSI, dst);
4669 Dynarr_add (dst, ISO_CODE_CSI);
4670 Dynarr_add (dst, '0');
4671 Dynarr_add (dst, ']');
4673 *flags &= ~CODING_STATE_R2L;
4677 /* If FLAGS is a null pointer or specifies a direction different from
4678 DIRECTION (which should be either CHARSET_RIGHT_TO_LEFT or
4679 CHARSET_LEFT_TO_RIGHT), output the appropriate switch-dir escape
4680 sequence to DST. Also update FLAGS if it is not a null pointer.
4681 If INTERNAL_P is set, we are outputting in internal format and
4682 need to handle the CSI differently. */
4685 ensure_correct_direction (int direction, Lisp_Coding_System *codesys,
4686 unsigned_char_dynarr *dst, unsigned int *flags,
4689 if ((!flags || (*flags & CODING_STATE_R2L)) &&
4690 direction == CHARSET_LEFT_TO_RIGHT)
4691 restore_left_to_right_direction (codesys, dst, flags, internal_p);
4692 else if (!CODING_SYSTEM_ISO2022_NO_ISO6429 (codesys)
4693 && (!flags || !(*flags & CODING_STATE_R2L)) &&
4694 direction == CHARSET_RIGHT_TO_LEFT)
4696 if (CODING_SYSTEM_ISO2022_SEVEN (codesys))
4698 Dynarr_add (dst, ISO_CODE_ESC);
4699 Dynarr_add (dst, '[');
4701 else if (internal_p)
4702 DECODE_ADD_BINARY_CHAR (ISO_CODE_CSI, dst);
4704 Dynarr_add (dst, ISO_CODE_CSI);
4705 Dynarr_add (dst, '2');
4706 Dynarr_add (dst, ']');
4708 *flags |= CODING_STATE_R2L;
4712 /* Convert ISO2022-format data to internal format. */
4715 decode_coding_iso2022 (Lstream *decoding, const unsigned char *src,
4716 unsigned_char_dynarr *dst, unsigned int n)
4718 struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
4719 unsigned int flags = str->flags;
4720 unsigned int cpos = str->cpos;
4721 unsigned char counter = str->counter;
4722 eol_type_t eol_type = str->eol_type;
4723 #ifdef ENABLE_COMPOSITE_CHARS
4724 unsigned_char_dynarr *real_dst = dst;
4726 Lisp_Object coding_system;
4728 XSETCODING_SYSTEM (coding_system, str->codesys);
4730 #ifdef ENABLE_COMPOSITE_CHARS
4731 if (flags & CODING_STATE_COMPOSITE)
4732 dst = str->iso2022.composite_chars;
4733 #endif /* ENABLE_COMPOSITE_CHARS */
4737 unsigned char c = *src++;
4738 if (flags & CODING_STATE_ESCAPE)
4739 { /* Within ESC sequence */
4740 int retval = parse_iso2022_esc (coding_system, &str->iso2022,
4745 switch (str->iso2022.esc)
4747 #ifdef ENABLE_COMPOSITE_CHARS
4748 case ISO_ESC_START_COMPOSITE:
4749 if (str->iso2022.composite_chars)
4750 Dynarr_reset (str->iso2022.composite_chars);
4752 str->iso2022.composite_chars = Dynarr_new (unsigned_char);
4753 dst = str->iso2022.composite_chars;
4755 case ISO_ESC_END_COMPOSITE:
4757 Bufbyte comstr[MAX_EMCHAR_LEN];
4759 Emchar emch = lookup_composite_char (Dynarr_atp (dst, 0),
4760 Dynarr_length (dst));
4762 len = set_charptr_emchar (comstr, emch);
4763 Dynarr_add_many (dst, comstr, len);
4766 #endif /* ENABLE_COMPOSITE_CHARS */
4768 case ISO_ESC_LITERAL:
4769 COMPOSE_FLUSH_CHARS (str, dst);
4770 DECODE_ADD_BINARY_CHAR (c, dst);
4774 /* Everything else handled already */
4779 /* Attempted error recovery. */
4780 if (str->iso2022.output_direction_sequence)
4781 ensure_correct_direction (flags & CODING_STATE_R2L ?
4782 CHARSET_RIGHT_TO_LEFT :
4783 CHARSET_LEFT_TO_RIGHT,
4784 str->codesys, dst, 0, 1);
4785 /* More error recovery. */
4786 if (!retval || str->iso2022.output_literally)
4788 /* Output the (possibly invalid) sequence */
4790 COMPOSE_FLUSH_CHARS (str, dst);
4791 for (i = 0; i < str->iso2022.esc_bytes_index; i++)
4792 DECODE_ADD_BINARY_CHAR (str->iso2022.esc_bytes[i], dst);
4793 flags &= CODING_STATE_ISO2022_LOCK;
4795 n++, src--;/* Repeat the loop with the same character. */
4798 /* No sense in reprocessing the final byte of the
4799 escape sequence; it could mess things up anyway.
4801 COMPOSE_FLUSH_CHARS (str, dst);
4802 DECODE_ADD_BINARY_CHAR (c, dst);
4808 else if (BYTE_C0_P (c) || BYTE_C1_P (c))
4809 { /* Control characters */
4811 /***** Error-handling *****/
4813 /* If we were in the middle of a character, dump out the
4814 partial character. */
4817 COMPOSE_FLUSH_CHARS (str, dst);
4821 DECODE_ADD_BINARY_CHAR
4822 ((unsigned char)(cpos >> (counter * 8)), dst);
4827 /* If we just saw a single-shift character, dump it out.
4828 This may dump out the wrong sort of single-shift character,
4829 but least it will give an indication that something went
4831 if (flags & CODING_STATE_SS2)
4833 COMPOSE_FLUSH_CHARS (str, dst);
4834 DECODE_ADD_BINARY_CHAR (ISO_CODE_SS2, dst);
4835 flags &= ~CODING_STATE_SS2;
4837 if (flags & CODING_STATE_SS3)
4839 COMPOSE_FLUSH_CHARS (str, dst);
4840 DECODE_ADD_BINARY_CHAR (ISO_CODE_SS3, dst);
4841 flags &= ~CODING_STATE_SS3;
4844 /***** Now handle the control characters. *****/
4850 COMPOSE_FLUSH_CHARS (str, dst);
4851 if (eol_type == EOL_CR)
4852 Dynarr_add (dst, '\n');
4853 else if (eol_type != EOL_CRLF || flags & CODING_STATE_CR)
4854 Dynarr_add (dst, c);
4856 flags |= CODING_STATE_CR;
4857 goto label_continue_loop;
4859 else if (flags & CODING_STATE_CR)
4860 { /* eol_type == CODING_SYSTEM_EOL_CRLF */
4862 Dynarr_add (dst, '\r');
4863 flags &= ~CODING_STATE_CR;
4866 DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
4869 flags &= CODING_STATE_ISO2022_LOCK;
4871 if (!parse_iso2022_esc (coding_system, &str->iso2022, c, &flags, 1))
4873 COMPOSE_FLUSH_CHARS (str, dst);
4874 DECODE_ADD_BINARY_CHAR (c, dst);
4878 { /* Graphic characters */
4879 Lisp_Object charset;
4888 COMPOSE_FLUSH_CHARS (str, dst);
4889 if (eol_type == EOL_CR)
4890 Dynarr_add (dst, '\n');
4891 else if (eol_type != EOL_CRLF || flags & CODING_STATE_CR)
4892 Dynarr_add (dst, c);
4894 flags |= CODING_STATE_CR;
4895 goto label_continue_loop;
4897 else if (flags & CODING_STATE_CR)
4898 { /* eol_type == CODING_SYSTEM_EOL_CRLF */
4900 Dynarr_add (dst, '\r');
4901 flags &= ~CODING_STATE_CR;
4904 DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
4907 /* Now determine the charset. */
4908 reg = ((flags & CODING_STATE_SS2) ? 2
4909 : (flags & CODING_STATE_SS3) ? 3
4910 : !BYTE_ASCII_P (c) ? str->iso2022.register_right
4911 : str->iso2022.register_left);
4912 charset = str->iso2022.charset[reg];
4914 /* Error checking: */
4915 if (! CHARSETP (charset)
4916 || str->iso2022.invalid_designated[reg]
4917 || (((c & 0x7F) == ' ' || (c & 0x7F) == ISO_CODE_DEL)
4918 && XCHARSET_CHARS (charset) == 94))
4919 /* Mrmph. We are trying to invoke a register that has no
4920 or an invalid charset in it, or trying to add a character
4921 outside the range of the charset. Insert that char literally
4922 to preserve it for the output. */
4924 COMPOSE_FLUSH_CHARS (str, dst);
4928 DECODE_ADD_BINARY_CHAR
4929 ((unsigned char)(cpos >> (counter * 8)), dst);
4932 DECODE_ADD_BINARY_CHAR (c, dst);
4937 /* Things are probably hunky-dorey. */
4939 /* Fetch reverse charset, maybe. */
4940 if (((flags & CODING_STATE_R2L) &&
4941 XCHARSET_DIRECTION (charset) == CHARSET_LEFT_TO_RIGHT)
4943 (!(flags & CODING_STATE_R2L) &&
4944 XCHARSET_DIRECTION (charset) == CHARSET_RIGHT_TO_LEFT))
4946 Lisp_Object new_charset =
4947 XCHARSET_REVERSE_DIRECTION_CHARSET (charset);
4948 if (!NILP (new_charset))
4949 charset = new_charset;
4954 if (XCHARSET_DIMENSION (charset) == counter)
4956 COMPOSE_ADD_CHAR (str,
4957 DECODE_CHAR (charset,
4958 ((cpos & 0x7F7F7F) << 8)
4965 cpos = (cpos << 8) | c;
4967 lb = XCHARSET_LEADING_BYTE (charset);
4968 switch (XCHARSET_REP_BYTES (charset))
4971 DECODE_OUTPUT_PARTIAL_CHAR (ch);
4972 Dynarr_add (dst, c & 0x7F);
4975 case 2: /* one-byte official */
4976 DECODE_OUTPUT_PARTIAL_CHAR (ch);
4977 Dynarr_add (dst, lb);
4978 Dynarr_add (dst, c | 0x80);
4981 case 3: /* one-byte private or two-byte official */
4982 if (XCHARSET_PRIVATE_P (charset))
4984 DECODE_OUTPUT_PARTIAL_CHAR (ch);
4985 Dynarr_add (dst, PRE_LEADING_BYTE_PRIVATE_1);
4986 Dynarr_add (dst, lb);
4987 Dynarr_add (dst, c | 0x80);
4993 Dynarr_add (dst, lb);
4994 Dynarr_add (dst, ch | 0x80);
4995 Dynarr_add (dst, c | 0x80);
5003 default: /* two-byte private */
5006 Dynarr_add (dst, PRE_LEADING_BYTE_PRIVATE_2);
5007 Dynarr_add (dst, lb);
5008 Dynarr_add (dst, ch | 0x80);
5009 Dynarr_add (dst, c | 0x80);
5019 flags &= CODING_STATE_ISO2022_LOCK;
5022 label_continue_loop:;
5025 if (flags & CODING_STATE_END)
5027 COMPOSE_FLUSH_CHARS (str, dst);
5028 DECODE_OUTPUT_PARTIAL_CHAR (cpos);
5032 str->counter = counter;
5036 /***** ISO2022 encoder *****/
5038 /* Designate CHARSET into register REG. */
5041 iso2022_designate (Lisp_Object charset, unsigned char reg,
5042 struct encoding_stream *str, unsigned_char_dynarr *dst)
5044 static const char inter94[] = "()*+";
5045 static const char inter96[] = ",-./";
5046 unsigned short chars;
5047 unsigned char dimension;
5048 unsigned char final;
5049 Lisp_Object old_charset = str->iso2022.charset[reg];
5051 str->iso2022.charset[reg] = charset;
5052 if (!CHARSETP (charset))
5053 /* charset might be an initial nil or t. */
5055 chars = XCHARSET_CHARS (charset);
5056 dimension = XCHARSET_DIMENSION (charset);
5057 final = XCHARSET_FINAL (charset);
5058 if (!str->iso2022.force_charset_on_output[reg] &&
5059 CHARSETP (old_charset) &&
5060 XCHARSET_CHARS (old_charset) == chars &&
5061 XCHARSET_DIMENSION (old_charset) == dimension &&
5062 XCHARSET_FINAL (old_charset) == final)
5065 str->iso2022.force_charset_on_output[reg] = 0;
5068 charset_conversion_spec_dynarr *dyn =
5069 str->codesys->iso2022.output_conv;
5075 for (i = 0; i < Dynarr_length (dyn); i++)
5077 struct charset_conversion_spec *spec = Dynarr_atp (dyn, i);
5078 if (EQ (charset, spec->from_charset))
5079 charset = spec->to_charset;
5084 Dynarr_add (dst, ISO_CODE_ESC);
5089 Dynarr_add (dst, inter94[reg]);
5092 Dynarr_add (dst, '$');
5094 || !(CODING_SYSTEM_ISO2022_SHORT (str->codesys))
5097 Dynarr_add (dst, inter94[reg]);
5102 Dynarr_add (dst, inter96[reg]);
5105 Dynarr_add (dst, '$');
5106 Dynarr_add (dst, inter96[reg]);
5110 Dynarr_add (dst, final);
5114 ensure_normal_shift (struct encoding_stream *str, unsigned_char_dynarr *dst)
5116 if (str->iso2022.register_left != 0)
5118 Dynarr_add (dst, ISO_CODE_SI);
5119 str->iso2022.register_left = 0;
5124 ensure_shift_out (struct encoding_stream *str, unsigned_char_dynarr *dst)
5126 if (str->iso2022.register_left != 1)
5128 Dynarr_add (dst, ISO_CODE_SO);
5129 str->iso2022.register_left = 1;
5134 char_encode_iso2022 (struct encoding_stream *str, Emchar ch,
5135 unsigned_char_dynarr *dst, unsigned int *flags)
5137 unsigned char charmask;
5138 Lisp_Coding_System* codesys = str->codesys;
5139 eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
5141 Lisp_Object charset = str->iso2022.current_charset;
5142 int half = str->iso2022.current_half;
5143 int code_point = -1;
5147 restore_left_to_right_direction (codesys, dst, flags, 0);
5149 /* Make sure G0 contains ASCII */
5150 if ((ch > ' ' && ch < ISO_CODE_DEL)
5151 || !CODING_SYSTEM_ISO2022_NO_ASCII_CNTL (codesys))
5153 ensure_normal_shift (str, dst);
5154 iso2022_designate (Vcharset_ascii, 0, str, dst);
5157 /* If necessary, restore everything to the default state
5159 if (ch == '\n' && !(CODING_SYSTEM_ISO2022_NO_ASCII_EOL (codesys)))
5161 restore_left_to_right_direction (codesys, dst, flags, 0);
5163 ensure_normal_shift (str, dst);
5165 for (i = 0; i < 4; i++)
5167 Lisp_Object initial_charset =
5168 CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i);
5169 iso2022_designate (initial_charset, i, str, dst);
5174 if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT)
5175 Dynarr_add (dst, '\r');
5176 if (eol_type != EOL_CR)
5177 Dynarr_add (dst, ch);
5181 if (CODING_SYSTEM_ISO2022_ESCAPE_QUOTED (codesys)
5182 && fit_to_be_escape_quoted (ch))
5183 Dynarr_add (dst, ISO_CODE_ESC);
5184 Dynarr_add (dst, ch);
5187 else if ( (0x80 <= ch) && (ch <= 0x9f) )
5189 charmask = (half == 0 ? 0x00 : 0x80);
5191 if (CODING_SYSTEM_ISO2022_ESCAPE_QUOTED (codesys)
5192 && fit_to_be_escape_quoted (ch))
5193 Dynarr_add (dst, ISO_CODE_ESC);
5194 /* you asked for it ... */
5195 Dynarr_add (dst, ch);
5201 /* Now determine which register to use. */
5203 for (i = 0; i < 4; i++)
5205 if ((CHARSETP (charset = str->iso2022.charset[i])
5206 && ((code_point = charset_code_point (charset, ch)) >= 0))
5210 = CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i))
5211 && ((code_point = charset_code_point (charset, ch)) >= 0)))
5219 Lisp_Object original_default_coded_charset_priority_list
5220 = Vdefault_coded_charset_priority_list;
5222 while (!EQ (Vdefault_coded_charset_priority_list, Qnil))
5224 code_point = ENCODE_CHAR (ch, charset);
5225 if (XCHARSET_FINAL (charset))
5227 Vdefault_coded_charset_priority_list
5228 = Fcdr (Fmemq (XCHARSET_NAME (charset),
5229 Vdefault_coded_charset_priority_list));
5231 code_point = ENCODE_CHAR (ch, charset);
5232 if (!XCHARSET_FINAL (charset))
5234 charset = Vcharset_ascii;
5238 Vdefault_coded_charset_priority_list
5239 = original_default_coded_charset_priority_list;
5241 ensure_correct_direction (XCHARSET_DIRECTION (charset),
5242 codesys, dst, flags, 0);
5246 if (XCHARSET_GRAPHIC (charset) != 0)
5248 if (!NILP (str->iso2022.charset[1]) &&
5249 (!CODING_SYSTEM_ISO2022_SEVEN (codesys)
5250 || CODING_SYSTEM_ISO2022_LOCK_SHIFT (codesys)))
5252 else if (!NILP (str->iso2022.charset[2]))
5254 else if (!NILP (str->iso2022.charset[3]))
5263 iso2022_designate (charset, reg, str, dst);
5265 /* Now invoke that register. */
5269 ensure_normal_shift (str, dst);
5273 if (CODING_SYSTEM_ISO2022_SEVEN (codesys))
5275 ensure_shift_out (str, dst);
5282 if (CODING_SYSTEM_ISO2022_SEVEN (str->codesys))
5284 Dynarr_add (dst, ISO_CODE_ESC);
5285 Dynarr_add (dst, 'N');
5290 Dynarr_add (dst, ISO_CODE_SS2);
5295 if (CODING_SYSTEM_ISO2022_SEVEN (str->codesys))
5297 Dynarr_add (dst, ISO_CODE_ESC);
5298 Dynarr_add (dst, 'O');
5303 Dynarr_add (dst, ISO_CODE_SS3);
5311 charmask = (half == 0 ? 0x00 : 0x80);
5313 switch (XCHARSET_DIMENSION (charset))
5316 Dynarr_add (dst, (code_point & 0xFF) | charmask);
5319 Dynarr_add (dst, ((code_point >> 8) & 0xFF) | charmask);
5320 Dynarr_add (dst, ( code_point & 0xFF) | charmask);
5323 Dynarr_add (dst, ((code_point >> 16) & 0xFF) | charmask);
5324 Dynarr_add (dst, ((code_point >> 8) & 0xFF) | charmask);
5325 Dynarr_add (dst, ( code_point & 0xFF) | charmask);
5328 Dynarr_add (dst, ((code_point >> 24) & 0xFF) | charmask);
5329 Dynarr_add (dst, ((code_point >> 16) & 0xFF) | charmask);
5330 Dynarr_add (dst, ((code_point >> 8) & 0xFF) | charmask);
5331 Dynarr_add (dst, ( code_point & 0xFF) | charmask);
5337 str->iso2022.current_charset = charset;
5338 str->iso2022.current_half = half;
5342 char_finish_iso2022 (struct encoding_stream *str, unsigned_char_dynarr *dst,
5343 unsigned int *flags)
5345 Lisp_Coding_System* codesys = str->codesys;
5348 restore_left_to_right_direction (codesys, dst, flags, 0);
5349 ensure_normal_shift (str, dst);
5350 for (i = 0; i < 4; i++)
5352 Lisp_Object initial_charset
5353 = CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i);
5354 iso2022_designate (initial_charset, i, str, dst);
5359 /************************************************************************/
5360 /* No-conversion methods */
5361 /************************************************************************/
5363 /* This is used when reading in "binary" files -- i.e. files that may
5364 contain all 256 possible byte values and that are not to be
5365 interpreted as being in any particular decoding. */
5367 decode_coding_no_conversion (Lstream *decoding, const unsigned char *src,
5368 unsigned_char_dynarr *dst, unsigned int n)
5371 struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
5372 unsigned int flags = str->flags;
5373 unsigned int cpos = str->cpos;
5374 eol_type_t eol_type = str->eol_type;
5380 DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
5381 DECODE_ADD_BINARY_CHAR (c, dst);
5382 label_continue_loop:;
5385 DECODE_HANDLE_END_OF_CONVERSION (flags, cpos, dst);
5392 encode_coding_no_conversion (Lstream *encoding, const unsigned char *src,
5393 unsigned_char_dynarr *dst, unsigned int n)
5396 struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
5397 unsigned int flags = str->flags;
5398 unsigned int ch = str->ch;
5399 eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
5401 unsigned char char_boundary = str->iso2022.current_char_boundary;
5408 if (char_boundary == 0)
5414 else if ( c >= 0xf8 )
5419 else if ( c >= 0xf0 )
5424 else if ( c >= 0xe0 )
5429 else if ( c >= 0xc0 )
5439 if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT)
5440 Dynarr_add (dst, '\r');
5441 if (eol_type != EOL_CR)
5442 Dynarr_add (dst, c);
5445 Dynarr_add (dst, c);
5448 else if (char_boundary == 1)
5450 ch = ( ch << 6 ) | ( c & 0x3f );
5451 Dynarr_add (dst, ch & 0xff);
5456 ch = ( ch << 6 ) | ( c & 0x3f );
5459 #else /* not UTF2000 */
5462 if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT)
5463 Dynarr_add (dst, '\r');
5464 if (eol_type != EOL_CR)
5465 Dynarr_add (dst, '\n');
5468 else if (BYTE_ASCII_P (c))
5471 Dynarr_add (dst, c);
5473 else if (BUFBYTE_LEADING_BYTE_P (c))
5476 if (c == LEADING_BYTE_LATIN_ISO8859_1 ||
5477 c == LEADING_BYTE_CONTROL_1)
5480 Dynarr_add (dst, '~'); /* untranslatable character */
5484 if (ch == LEADING_BYTE_LATIN_ISO8859_1)
5485 Dynarr_add (dst, c);
5486 else if (ch == LEADING_BYTE_CONTROL_1)
5489 Dynarr_add (dst, c - 0x20);
5491 /* else it should be the second or third byte of an
5492 untranslatable character, so ignore it */
5495 #endif /* not UTF2000 */
5501 str->iso2022.current_char_boundary = char_boundary;
5507 /************************************************************************/
5508 /* Initialization */
5509 /************************************************************************/
5512 syms_of_file_coding (void)
5514 INIT_LRECORD_IMPLEMENTATION (coding_system);
5516 deferror (&Qcoding_system_error, "coding-system-error",
5517 "Coding-system error", Qio_error);
5519 DEFSUBR (Fcoding_system_p);
5520 DEFSUBR (Ffind_coding_system);
5521 DEFSUBR (Fget_coding_system);
5522 DEFSUBR (Fcoding_system_list);
5523 DEFSUBR (Fcoding_system_name);
5524 DEFSUBR (Fmake_coding_system);
5525 DEFSUBR (Fcopy_coding_system);
5526 DEFSUBR (Fcoding_system_canonical_name_p);
5527 DEFSUBR (Fcoding_system_alias_p);
5528 DEFSUBR (Fcoding_system_aliasee);
5529 DEFSUBR (Fdefine_coding_system_alias);
5530 DEFSUBR (Fsubsidiary_coding_system);
5532 DEFSUBR (Fcoding_system_type);
5533 DEFSUBR (Fcoding_system_doc_string);
5535 DEFSUBR (Fcoding_system_charset);
5537 DEFSUBR (Fcoding_system_property);
5539 DEFSUBR (Fcoding_category_list);
5540 DEFSUBR (Fset_coding_priority_list);
5541 DEFSUBR (Fcoding_priority_list);
5542 DEFSUBR (Fset_coding_category_system);
5543 DEFSUBR (Fcoding_category_system);
5545 DEFSUBR (Fdetect_coding_region);
5546 DEFSUBR (Fdecode_coding_region);
5547 DEFSUBR (Fencode_coding_region);
5549 DEFSUBR (Fdecode_shift_jis_char);
5550 DEFSUBR (Fencode_shift_jis_char);
5551 DEFSUBR (Fdecode_big5_char);
5552 DEFSUBR (Fencode_big5_char);
5554 defsymbol (&Qcoding_systemp, "coding-system-p");
5555 defsymbol (&Qno_conversion, "no-conversion");
5556 defsymbol (&Qraw_text, "raw-text");
5558 defsymbol (&Qbig5, "big5");
5559 defsymbol (&Qshift_jis, "shift-jis");
5560 defsymbol (&Qucs4, "ucs-4");
5561 defsymbol (&Qutf8, "utf-8");
5562 defsymbol (&Qccl, "ccl");
5563 defsymbol (&Qiso2022, "iso2022");
5565 defsymbol (&Qmnemonic, "mnemonic");
5566 defsymbol (&Qeol_type, "eol-type");
5567 defsymbol (&Qpost_read_conversion, "post-read-conversion");
5568 defsymbol (&Qpre_write_conversion, "pre-write-conversion");
5570 defsymbol (&Qcr, "cr");
5571 defsymbol (&Qlf, "lf");
5572 defsymbol (&Qcrlf, "crlf");
5573 defsymbol (&Qeol_cr, "eol-cr");
5574 defsymbol (&Qeol_lf, "eol-lf");
5575 defsymbol (&Qeol_crlf, "eol-crlf");
5577 defsymbol (&Qcharset_g0, "charset-g0");
5578 defsymbol (&Qcharset_g1, "charset-g1");
5579 defsymbol (&Qcharset_g2, "charset-g2");
5580 defsymbol (&Qcharset_g3, "charset-g3");
5581 defsymbol (&Qforce_g0_on_output, "force-g0-on-output");
5582 defsymbol (&Qforce_g1_on_output, "force-g1-on-output");
5583 defsymbol (&Qforce_g2_on_output, "force-g2-on-output");
5584 defsymbol (&Qforce_g3_on_output, "force-g3-on-output");
5585 defsymbol (&Qno_iso6429, "no-iso6429");
5586 defsymbol (&Qinput_charset_conversion, "input-charset-conversion");
5587 defsymbol (&Qoutput_charset_conversion, "output-charset-conversion");
5589 defsymbol (&Qshort, "short");
5590 defsymbol (&Qno_ascii_eol, "no-ascii-eol");
5591 defsymbol (&Qno_ascii_cntl, "no-ascii-cntl");
5592 defsymbol (&Qseven, "seven");
5593 defsymbol (&Qlock_shift, "lock-shift");
5594 defsymbol (&Qescape_quoted, "escape-quoted");
5597 defsymbol (&Qdisable_composition, "disable-composition");
5599 defsymbol (&Qencode, "encode");
5600 defsymbol (&Qdecode, "decode");
5603 defsymbol (&coding_category_symbol[CODING_CATEGORY_SHIFT_JIS],
5605 defsymbol (&coding_category_symbol[CODING_CATEGORY_BIG5],
5607 defsymbol (&coding_category_symbol[CODING_CATEGORY_UCS4],
5609 defsymbol (&coding_category_symbol[CODING_CATEGORY_UTF8],
5611 defsymbol (&coding_category_symbol[CODING_CATEGORY_ISO_7],
5613 defsymbol (&coding_category_symbol[CODING_CATEGORY_ISO_8_DESIGNATE],
5615 defsymbol (&coding_category_symbol[CODING_CATEGORY_ISO_8_1],
5617 defsymbol (&coding_category_symbol[CODING_CATEGORY_ISO_8_2],
5619 defsymbol (&coding_category_symbol[CODING_CATEGORY_ISO_LOCK_SHIFT],
5622 defsymbol (&coding_category_symbol[CODING_CATEGORY_NO_CONVERSION],
5627 lstream_type_create_file_coding (void)
5629 LSTREAM_HAS_METHOD (decoding, reader);
5630 LSTREAM_HAS_METHOD (decoding, writer);
5631 LSTREAM_HAS_METHOD (decoding, rewinder);
5632 LSTREAM_HAS_METHOD (decoding, seekable_p);
5633 LSTREAM_HAS_METHOD (decoding, flusher);
5634 LSTREAM_HAS_METHOD (decoding, closer);
5635 LSTREAM_HAS_METHOD (decoding, marker);
5637 LSTREAM_HAS_METHOD (encoding, reader);
5638 LSTREAM_HAS_METHOD (encoding, writer);
5639 LSTREAM_HAS_METHOD (encoding, rewinder);
5640 LSTREAM_HAS_METHOD (encoding, seekable_p);
5641 LSTREAM_HAS_METHOD (encoding, flusher);
5642 LSTREAM_HAS_METHOD (encoding, closer);
5643 LSTREAM_HAS_METHOD (encoding, marker);
5647 vars_of_file_coding (void)
5651 fcd = xnew (struct file_coding_dump);
5652 dumpstruct (&fcd, &fcd_description);
5654 /* Initialize to something reasonable ... */
5655 for (i = 0; i <= CODING_CATEGORY_LAST; i++)
5657 fcd->coding_category_system[i] = Qnil;
5658 fcd->coding_category_by_priority[i] = i;
5661 Fprovide (intern ("file-coding"));
5663 DEFVAR_LISP ("keyboard-coding-system", &Vkeyboard_coding_system /*
5664 Coding system used for TTY keyboard input.
5665 Not used under a windowing system.
5667 Vkeyboard_coding_system = Qnil;
5669 DEFVAR_LISP ("terminal-coding-system", &Vterminal_coding_system /*
5670 Coding system used for TTY display output.
5671 Not used under a windowing system.
5673 Vterminal_coding_system = Qnil;
5675 DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read /*
5676 Overriding coding system used when reading from a file or process.
5677 You should bind this variable with `let', but do not set it globally.
5678 If this is non-nil, it specifies the coding system that will be used
5679 to decode input on read operations, such as from a file or process.
5680 It overrides `buffer-file-coding-system-for-read',
5681 `insert-file-contents-pre-hook', etc. Use those variables instead of
5682 this one for permanent changes to the environment. */ );
5683 Vcoding_system_for_read = Qnil;
5685 DEFVAR_LISP ("coding-system-for-write",
5686 &Vcoding_system_for_write /*
5687 Overriding coding system used when writing to a file or process.
5688 You should bind this variable with `let', but do not set it globally.
5689 If this is non-nil, it specifies the coding system that will be used
5690 to encode output for write operations, such as to a file or process.
5691 It overrides `buffer-file-coding-system', `write-region-pre-hook', etc.
5692 Use those variables instead of this one for permanent changes to the
5694 Vcoding_system_for_write = Qnil;
5696 DEFVAR_LISP ("file-name-coding-system", &Vfile_name_coding_system /*
5697 Coding system used to convert pathnames when accessing files.
5699 Vfile_name_coding_system = Qnil;
5701 DEFVAR_BOOL ("enable-multibyte-characters", &enable_multibyte_characters /*
5702 Non-nil means the buffer contents are regarded as multi-byte form
5703 of characters, not a binary code. This affects the display, file I/O,
5704 and behaviors of various editing commands.
5706 Setting this to nil does not do anything.
5708 enable_multibyte_characters = 1;
5712 complex_vars_of_file_coding (void)
5714 staticpro (&Vcoding_system_hash_table);
5715 Vcoding_system_hash_table =
5716 make_lisp_hash_table (50, HASH_TABLE_NON_WEAK, HASH_TABLE_EQ);
5718 the_codesys_prop_dynarr = Dynarr_new (codesys_prop);
5719 dumpstruct (&the_codesys_prop_dynarr, &codesys_prop_dynarr_description);
5721 #define DEFINE_CODESYS_PROP(Prop_Type, Sym) do \
5723 struct codesys_prop csp; \
5725 csp.prop_type = (Prop_Type); \
5726 Dynarr_add (the_codesys_prop_dynarr, csp); \
5729 DEFINE_CODESYS_PROP (CODESYS_PROP_ALL_OK, Qmnemonic);
5730 DEFINE_CODESYS_PROP (CODESYS_PROP_ALL_OK, Qeol_type);
5731 DEFINE_CODESYS_PROP (CODESYS_PROP_ALL_OK, Qeol_cr);
5732 DEFINE_CODESYS_PROP (CODESYS_PROP_ALL_OK, Qeol_crlf);
5733 DEFINE_CODESYS_PROP (CODESYS_PROP_ALL_OK, Qeol_lf);
5734 DEFINE_CODESYS_PROP (CODESYS_PROP_ALL_OK, Qpost_read_conversion);
5735 DEFINE_CODESYS_PROP (CODESYS_PROP_ALL_OK, Qpre_write_conversion);
5737 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qcharset_g0);
5738 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qcharset_g1);
5739 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qcharset_g2);
5740 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qcharset_g3);
5741 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qforce_g0_on_output);
5742 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qforce_g1_on_output);
5743 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qforce_g2_on_output);
5744 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qforce_g3_on_output);
5745 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qshort);
5746 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qno_ascii_eol);
5747 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qno_ascii_cntl);
5748 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qseven);
5749 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qlock_shift);
5750 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qno_iso6429);
5751 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qescape_quoted);
5752 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qinput_charset_conversion);
5753 DEFINE_CODESYS_PROP (CODESYS_PROP_ISO2022, Qoutput_charset_conversion);
5755 DEFINE_CODESYS_PROP (CODESYS_PROP_CCL, Qencode);
5756 DEFINE_CODESYS_PROP (CODESYS_PROP_CCL, Qdecode);
5758 /* Need to create this here or we're really screwed. */
5760 (Qraw_text, Qno_conversion,
5761 build_string ("Raw text, which means it converts only line-break-codes."),
5762 list2 (Qmnemonic, build_string ("Raw")));
5765 (Qbinary, Qno_conversion,
5766 build_string ("Binary, which means it does not convert anything."),
5767 list4 (Qeol_type, Qlf,
5768 Qmnemonic, build_string ("Binary")));
5773 build_string ("Coding-system of ISO/IEC 10646 UTF-8."),
5774 list2 (Qmnemonic, build_string ("UTF8")));
5777 Fdefine_coding_system_alias (Qno_conversion, Qraw_text);
5779 Fdefine_coding_system_alias (Qfile_name, Qbinary);
5781 Fdefine_coding_system_alias (Qterminal, Qbinary);
5782 Fdefine_coding_system_alias (Qkeyboard, Qbinary);
5784 /* Need this for bootstrapping */
5785 fcd->coding_category_system[CODING_CATEGORY_NO_CONVERSION] =
5786 Fget_coding_system (Qraw_text);
5789 fcd->coding_category_system[CODING_CATEGORY_UTF8]
5790 = Fget_coding_system (Qutf8);
5793 #if defined(MULE) && !defined(UTF2000)
5797 for (i = 0; i < 65536; i++)
5798 fcd->ucs_to_mule_table[i] = Qnil;
5800 staticpro (&mule_to_ucs_table);
5801 mule_to_ucs_table = Fmake_char_table(Qgeneric);
5802 #endif /* defined(MULE) && !defined(UTF2000) */