1 /* Functions to handle multilingual characters.
2 Copyright (C) 1992, 1995 Free Software Foundation, Inc.
3 Copyright (C) 1995 Sun Microsystems, Inc.
4 Copyright (C) 1999,2000,2001,2002 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 /* Rewritten by Ben Wing <ben@xemacs.org>. */
25 /* Rewritten by MORIOKA Tomohiko <tomo@m17n.org> for XEmacs UTF-2000. */
41 /* The various pre-defined charsets. */
43 Lisp_Object Vcharset_ascii;
44 Lisp_Object Vcharset_control_1;
45 Lisp_Object Vcharset_latin_iso8859_1;
46 Lisp_Object Vcharset_latin_iso8859_2;
47 Lisp_Object Vcharset_latin_iso8859_3;
48 Lisp_Object Vcharset_latin_iso8859_4;
49 Lisp_Object Vcharset_thai_tis620;
50 Lisp_Object Vcharset_greek_iso8859_7;
51 Lisp_Object Vcharset_arabic_iso8859_6;
52 Lisp_Object Vcharset_hebrew_iso8859_8;
53 Lisp_Object Vcharset_katakana_jisx0201;
54 Lisp_Object Vcharset_latin_jisx0201;
55 Lisp_Object Vcharset_cyrillic_iso8859_5;
56 Lisp_Object Vcharset_latin_iso8859_9;
57 Lisp_Object Vcharset_japanese_jisx0208_1978;
58 Lisp_Object Vcharset_chinese_gb2312;
59 Lisp_Object Vcharset_chinese_gb12345;
60 Lisp_Object Vcharset_japanese_jisx0208;
61 Lisp_Object Vcharset_japanese_jisx0208_1990;
62 Lisp_Object Vcharset_korean_ksc5601;
63 Lisp_Object Vcharset_japanese_jisx0212;
64 Lisp_Object Vcharset_chinese_cns11643_1;
65 Lisp_Object Vcharset_chinese_cns11643_2;
67 Lisp_Object Vcharset_ucs;
68 Lisp_Object Vcharset_ucs_bmp;
69 Lisp_Object Vcharset_ucs_smp;
70 Lisp_Object Vcharset_ucs_sip;
71 Lisp_Object Vcharset_ucs_cns;
72 Lisp_Object Vcharset_ucs_jis;
73 Lisp_Object Vcharset_ucs_ks;
74 Lisp_Object Vcharset_ucs_big5;
75 Lisp_Object Vcharset_latin_viscii;
76 Lisp_Object Vcharset_latin_tcvn5712;
77 Lisp_Object Vcharset_latin_viscii_lower;
78 Lisp_Object Vcharset_latin_viscii_upper;
79 Lisp_Object Vcharset_jis_x0208;
80 Lisp_Object Vcharset_chinese_big5;
81 /* Lisp_Object Vcharset_chinese_big5_cdp; */
82 Lisp_Object Vcharset_ideograph_hanziku_1;
83 Lisp_Object Vcharset_ideograph_hanziku_2;
84 Lisp_Object Vcharset_ideograph_hanziku_3;
85 Lisp_Object Vcharset_ideograph_hanziku_4;
86 Lisp_Object Vcharset_ideograph_hanziku_5;
87 Lisp_Object Vcharset_ideograph_hanziku_6;
88 Lisp_Object Vcharset_ideograph_hanziku_7;
89 Lisp_Object Vcharset_ideograph_hanziku_8;
90 Lisp_Object Vcharset_ideograph_hanziku_9;
91 Lisp_Object Vcharset_ideograph_hanziku_10;
92 Lisp_Object Vcharset_ideograph_hanziku_11;
93 Lisp_Object Vcharset_ideograph_hanziku_12;
94 Lisp_Object Vcharset_china3_jef;
95 Lisp_Object Vcharset_ideograph_cbeta;
96 Lisp_Object Vcharset_ideograph_gt;
97 Lisp_Object Vcharset_ideograph_gt_pj_1;
98 Lisp_Object Vcharset_ideograph_gt_pj_2;
99 Lisp_Object Vcharset_ideograph_gt_pj_3;
100 Lisp_Object Vcharset_ideograph_gt_pj_4;
101 Lisp_Object Vcharset_ideograph_gt_pj_5;
102 Lisp_Object Vcharset_ideograph_gt_pj_6;
103 Lisp_Object Vcharset_ideograph_gt_pj_7;
104 Lisp_Object Vcharset_ideograph_gt_pj_8;
105 Lisp_Object Vcharset_ideograph_gt_pj_9;
106 Lisp_Object Vcharset_ideograph_gt_pj_10;
107 Lisp_Object Vcharset_ideograph_gt_pj_11;
108 Lisp_Object Vcharset_ideograph_daikanwa_2;
109 Lisp_Object Vcharset_ideograph_daikanwa;
110 Lisp_Object Vcharset_ethiopic_ucs;
112 Lisp_Object Vcharset_chinese_big5_1;
113 Lisp_Object Vcharset_chinese_big5_2;
115 #ifdef ENABLE_COMPOSITE_CHARS
116 Lisp_Object Vcharset_composite;
118 /* Hash tables for composite chars. One maps string representing
119 composed chars to their equivalent chars; one goes the
121 Lisp_Object Vcomposite_char_char2string_hash_table;
122 Lisp_Object Vcomposite_char_string2char_hash_table;
124 static int composite_char_row_next;
125 static int composite_char_col_next;
127 #endif /* ENABLE_COMPOSITE_CHARS */
129 struct charset_lookup *chlook;
131 static const struct lrecord_description charset_lookup_description_1[] = {
132 { XD_LISP_OBJECT_ARRAY, offsetof (struct charset_lookup, charset_by_leading_byte),
141 static const struct struct_description charset_lookup_description = {
142 sizeof (struct charset_lookup),
143 charset_lookup_description_1
147 /* Table of number of bytes in the string representation of a character
148 indexed by the first byte of that representation.
150 rep_bytes_by_first_byte(c) is more efficient than the equivalent
151 canonical computation:
153 XCHARSET_REP_BYTES (CHARSET_BY_LEADING_BYTE (c)) */
155 const Bytecount rep_bytes_by_first_byte[0xA0] =
156 { /* 0x00 - 0x7f are for straight ASCII */
157 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
158 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
159 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
160 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
161 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
162 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
163 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
164 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
165 /* 0x80 - 0x8f are for Dimension-1 official charsets */
167 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3,
169 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
171 /* 0x90 - 0x9d are for Dimension-2 official charsets */
172 /* 0x9e is for Dimension-1 private charsets */
173 /* 0x9f is for Dimension-2 private charsets */
174 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4
180 int decoding_table_check_elements (Lisp_Object v, int dim, int ccs_len);
182 decoding_table_check_elements (Lisp_Object v, int dim, int ccs_len)
186 if (XVECTOR_LENGTH (v) > ccs_len)
189 for (i = 0; i < XVECTOR_LENGTH (v); i++)
191 Lisp_Object c = XVECTOR_DATA(v)[i];
193 if (!NILP (c) && !CHARP (c))
197 int ret = decoding_table_check_elements (c, dim - 1, ccs_len);
209 put_char_ccs_code_point (Lisp_Object character,
210 Lisp_Object ccs, Lisp_Object value)
212 if (!EQ (XCHARSET_NAME (ccs), Qucs)
214 || (XCHAR (character) != XINT (value)))
216 Lisp_Object v = XCHARSET_DECODING_TABLE (ccs);
220 { /* obsolete representation: value must be a list of bytes */
221 Lisp_Object ret = Fcar (value);
225 signal_simple_error ("Invalid value for coded-charset", value);
226 code_point = XINT (ret);
227 if (XCHARSET_GRAPHIC (ccs) == 1)
235 signal_simple_error ("Invalid value for coded-charset",
239 signal_simple_error ("Invalid value for coded-charset",
242 if (XCHARSET_GRAPHIC (ccs) == 1)
244 code_point = (code_point << 8) | j;
247 value = make_int (code_point);
249 else if (INTP (value))
251 code_point = XINT (value);
252 if (XCHARSET_GRAPHIC (ccs) == 1)
254 code_point &= 0x7F7F7F7F;
255 value = make_int (code_point);
259 signal_simple_error ("Invalid value for coded-charset", value);
263 Lisp_Object cpos = Fget_char_attribute (character, ccs, Qnil);
266 decoding_table_remove_char (ccs, XINT (cpos));
269 decoding_table_put_char (ccs, code_point, character);
275 remove_char_ccs (Lisp_Object character, Lisp_Object ccs)
277 Lisp_Object decoding_table = XCHARSET_DECODING_TABLE (ccs);
278 Lisp_Object encoding_table = XCHARSET_ENCODING_TABLE (ccs);
280 if (VECTORP (decoding_table))
282 Lisp_Object cpos = Fget_char_attribute (character, ccs, Qnil);
286 decoding_table_remove_char (ccs, XINT (cpos));
289 if (CHAR_TABLEP (encoding_table))
291 put_char_id_table (XCHAR_TABLE(encoding_table), character, Qunbound);
299 int leading_code_private_11;
302 Lisp_Object Qcharsetp;
304 /* Qdoc_string, Qdimension, Qchars defined in general.c */
305 Lisp_Object Qregistry, Qfinal, Qgraphic;
306 Lisp_Object Qdirection;
307 Lisp_Object Qreverse_direction_charset;
308 Lisp_Object Qleading_byte;
309 Lisp_Object Qshort_name, Qlong_name;
311 Lisp_Object Qmin_code, Qmax_code, Qcode_offset;
312 Lisp_Object Qmother, Qconversion, Q94x60, Q94x94x60;
329 Qjapanese_jisx0208_1978,
333 Qjapanese_jisx0208_1990,
351 Qvietnamese_viscii_lower,
352 Qvietnamese_viscii_upper,
355 /* Qchinese_big5_cdp, */
356 Qideograph_hanziku_1,
357 Qideograph_hanziku_2,
358 Qideograph_hanziku_3,
359 Qideograph_hanziku_4,
360 Qideograph_hanziku_5,
361 Qideograph_hanziku_6,
362 Qideograph_hanziku_7,
363 Qideograph_hanziku_8,
364 Qideograph_hanziku_9,
365 Qideograph_hanziku_10,
366 Qideograph_hanziku_11,
367 Qideograph_hanziku_12,
370 Qideograph_daikanwa_2,
390 Lisp_Object Ql2r, Qr2l;
392 Lisp_Object Vcharset_hash_table;
394 /* Composite characters are characters constructed by overstriking two
395 or more regular characters.
397 1) The old Mule implementation involves storing composite characters
398 in a buffer as a tag followed by all of the actual characters
399 used to make up the composite character. I think this is a bad
400 idea; it greatly complicates code that wants to handle strings
401 one character at a time because it has to deal with the possibility
402 of great big ungainly characters. It's much more reasonable to
403 simply store an index into a table of composite characters.
405 2) The current implementation only allows for 16,384 separate
406 composite characters over the lifetime of the XEmacs process.
407 This could become a potential problem if the user
408 edited lots of different files that use composite characters.
409 Due to FSF bogosity, increasing the number of allowable
410 composite characters under Mule would decrease the number
411 of possible faces that can exist. Mule already has shrunk
412 this to 2048, and further shrinkage would become uncomfortable.
413 No such problems exist in XEmacs.
415 Composite characters could be represented as 0x80 C1 C2 C3,
416 where each C[1-3] is in the range 0xA0 - 0xFF. This allows
417 for slightly under 2^20 (one million) composite characters
418 over the XEmacs process lifetime, and you only need to
419 increase the size of a Mule character from 19 to 21 bits.
420 Or you could use 0x80 C1 C2 C3 C4, allowing for about
421 85 million (slightly over 2^26) composite characters. */
424 /************************************************************************/
425 /* Basic Emchar functions */
426 /************************************************************************/
428 /* Convert a non-ASCII Mule character C into a one-character Mule-encoded
429 string in STR. Returns the number of bytes stored.
430 Do not call this directly. Use the macro set_charptr_emchar() instead.
434 non_ascii_set_charptr_emchar (Bufbyte *str, Emchar c)
449 else if ( c <= 0x7ff )
451 *p++ = (c >> 6) | 0xc0;
452 *p++ = (c & 0x3f) | 0x80;
454 else if ( c <= 0xffff )
456 *p++ = (c >> 12) | 0xe0;
457 *p++ = ((c >> 6) & 0x3f) | 0x80;
458 *p++ = (c & 0x3f) | 0x80;
460 else if ( c <= 0x1fffff )
462 *p++ = (c >> 18) | 0xf0;
463 *p++ = ((c >> 12) & 0x3f) | 0x80;
464 *p++ = ((c >> 6) & 0x3f) | 0x80;
465 *p++ = (c & 0x3f) | 0x80;
467 else if ( c <= 0x3ffffff )
469 *p++ = (c >> 24) | 0xf8;
470 *p++ = ((c >> 18) & 0x3f) | 0x80;
471 *p++ = ((c >> 12) & 0x3f) | 0x80;
472 *p++ = ((c >> 6) & 0x3f) | 0x80;
473 *p++ = (c & 0x3f) | 0x80;
477 *p++ = (c >> 30) | 0xfc;
478 *p++ = ((c >> 24) & 0x3f) | 0x80;
479 *p++ = ((c >> 18) & 0x3f) | 0x80;
480 *p++ = ((c >> 12) & 0x3f) | 0x80;
481 *p++ = ((c >> 6) & 0x3f) | 0x80;
482 *p++ = (c & 0x3f) | 0x80;
485 BREAKUP_CHAR (c, charset, c1, c2);
486 lb = CHAR_LEADING_BYTE (c);
487 if (LEADING_BYTE_PRIVATE_P (lb))
488 *p++ = PRIVATE_LEADING_BYTE_PREFIX (lb);
490 if (EQ (charset, Vcharset_control_1))
499 /* Return the first character from a Mule-encoded string in STR,
500 assuming it's non-ASCII. Do not call this directly.
501 Use the macro charptr_emchar() instead. */
504 non_ascii_charptr_emchar (const Bufbyte *str)
517 else if ( b >= 0xf8 )
522 else if ( b >= 0xf0 )
527 else if ( b >= 0xe0 )
532 else if ( b >= 0xc0 )
542 for( ; len > 0; len-- )
545 ch = ( ch << 6 ) | ( b & 0x3f );
549 Bufbyte i0 = *str, i1, i2 = 0;
552 if (i0 == LEADING_BYTE_CONTROL_1)
553 return (Emchar) (*++str - 0x20);
555 if (LEADING_BYTE_PREFIX_P (i0))
560 charset = CHARSET_BY_LEADING_BYTE (i0);
561 if (XCHARSET_DIMENSION (charset) == 2)
564 return MAKE_CHAR (charset, i1, i2);
568 /* Return whether CH is a valid Emchar, assuming it's non-ASCII.
569 Do not call this directly. Use the macro valid_char_p() instead. */
573 non_ascii_valid_char_p (Emchar ch)
577 /* Must have only lowest 19 bits set */
581 f1 = CHAR_FIELD1 (ch);
582 f2 = CHAR_FIELD2 (ch);
583 f3 = CHAR_FIELD3 (ch);
589 if (f2 < MIN_CHAR_FIELD2_OFFICIAL ||
590 (f2 > MAX_CHAR_FIELD2_OFFICIAL && f2 < MIN_CHAR_FIELD2_PRIVATE) ||
591 f2 > MAX_CHAR_FIELD2_PRIVATE)
596 if (f3 != 0x20 && f3 != 0x7F && !(f2 >= MIN_CHAR_FIELD2_PRIVATE &&
597 f2 <= MAX_CHAR_FIELD2_PRIVATE))
601 NOTE: This takes advantage of the fact that
602 FIELD2_TO_OFFICIAL_LEADING_BYTE and
603 FIELD2_TO_PRIVATE_LEADING_BYTE are the same.
605 charset = CHARSET_BY_LEADING_BYTE (f2 + FIELD2_TO_OFFICIAL_LEADING_BYTE);
606 if (EQ (charset, Qnil))
608 return (XCHARSET_CHARS (charset) == 96);
614 if (f1 < MIN_CHAR_FIELD1_OFFICIAL ||
615 (f1 > MAX_CHAR_FIELD1_OFFICIAL && f1 < MIN_CHAR_FIELD1_PRIVATE) ||
616 f1 > MAX_CHAR_FIELD1_PRIVATE)
618 if (f2 < 0x20 || f3 < 0x20)
621 #ifdef ENABLE_COMPOSITE_CHARS
622 if (f1 + FIELD1_TO_OFFICIAL_LEADING_BYTE == LEADING_BYTE_COMPOSITE)
624 if (UNBOUNDP (Fgethash (make_int (ch),
625 Vcomposite_char_char2string_hash_table,
630 #endif /* ENABLE_COMPOSITE_CHARS */
632 if (f2 != 0x20 && f2 != 0x7F && f3 != 0x20 && f3 != 0x7F
633 && !(f1 >= MIN_CHAR_FIELD1_PRIVATE && f1 <= MAX_CHAR_FIELD1_PRIVATE))
636 if (f1 <= MAX_CHAR_FIELD1_OFFICIAL)
638 CHARSET_BY_LEADING_BYTE (f1 + FIELD1_TO_OFFICIAL_LEADING_BYTE);
641 CHARSET_BY_LEADING_BYTE (f1 + FIELD1_TO_PRIVATE_LEADING_BYTE);
643 if (EQ (charset, Qnil))
645 return (XCHARSET_CHARS (charset) == 96);
651 /************************************************************************/
652 /* Basic string functions */
653 /************************************************************************/
655 /* Copy the character pointed to by SRC into DST. Do not call this
656 directly. Use the macro charptr_copy_char() instead.
657 Return the number of bytes copied. */
660 non_ascii_charptr_copy_char (const Bufbyte *src, Bufbyte *dst)
662 unsigned int bytes = REP_BYTES_BY_FIRST_BYTE (*src);
664 for (i = bytes; i; i--, dst++, src++)
670 /************************************************************************/
671 /* streams of Emchars */
672 /************************************************************************/
674 /* Treat a stream as a stream of Emchar's rather than a stream of bytes.
675 The functions below are not meant to be called directly; use
676 the macros in insdel.h. */
679 Lstream_get_emchar_1 (Lstream *stream, int ch)
681 Bufbyte str[MAX_EMCHAR_LEN];
682 Bufbyte *strptr = str;
685 str[0] = (Bufbyte) ch;
687 for (bytes = REP_BYTES_BY_FIRST_BYTE (ch) - 1; bytes; bytes--)
689 int c = Lstream_getc (stream);
690 bufpos_checking_assert (c >= 0);
691 *++strptr = (Bufbyte) c;
693 return charptr_emchar (str);
697 Lstream_fput_emchar (Lstream *stream, Emchar ch)
699 Bufbyte str[MAX_EMCHAR_LEN];
700 Bytecount len = set_charptr_emchar (str, ch);
701 return Lstream_write (stream, str, len);
705 Lstream_funget_emchar (Lstream *stream, Emchar ch)
707 Bufbyte str[MAX_EMCHAR_LEN];
708 Bytecount len = set_charptr_emchar (str, ch);
709 Lstream_unread (stream, str, len);
713 /************************************************************************/
715 /************************************************************************/
718 mark_charset (Lisp_Object obj)
720 Lisp_Charset *cs = XCHARSET (obj);
722 mark_object (cs->short_name);
723 mark_object (cs->long_name);
724 mark_object (cs->doc_string);
725 mark_object (cs->registry);
726 mark_object (cs->ccl_program);
728 mark_object (cs->decoding_table);
729 mark_object (cs->mother);
735 print_charset (Lisp_Object obj, Lisp_Object printcharfun, int escapeflag)
737 Lisp_Charset *cs = XCHARSET (obj);
741 error ("printing unreadable object #<charset %s 0x%x>",
742 string_data (XSYMBOL (CHARSET_NAME (cs))->name),
745 write_c_string ("#<charset ", printcharfun);
746 print_internal (CHARSET_NAME (cs), printcharfun, 0);
747 write_c_string (" ", printcharfun);
748 print_internal (CHARSET_SHORT_NAME (cs), printcharfun, 1);
749 write_c_string (" ", printcharfun);
750 print_internal (CHARSET_LONG_NAME (cs), printcharfun, 1);
751 write_c_string (" ", printcharfun);
752 print_internal (CHARSET_DOC_STRING (cs), printcharfun, 1);
753 sprintf (buf, " %d^%d %s cols=%d g%d final='%c' reg=",
755 CHARSET_DIMENSION (cs),
756 CHARSET_DIRECTION (cs) == CHARSET_LEFT_TO_RIGHT ? "l2r" : "r2l",
757 CHARSET_COLUMNS (cs),
758 CHARSET_GRAPHIC (cs),
760 write_c_string (buf, printcharfun);
761 print_internal (CHARSET_REGISTRY (cs), printcharfun, 0);
762 sprintf (buf, " 0x%x>", cs->header.uid);
763 write_c_string (buf, printcharfun);
766 static const struct lrecord_description charset_description[] = {
767 { XD_LISP_OBJECT, offsetof (Lisp_Charset, name) },
768 { XD_LISP_OBJECT, offsetof (Lisp_Charset, doc_string) },
769 { XD_LISP_OBJECT, offsetof (Lisp_Charset, registry) },
770 { XD_LISP_OBJECT, offsetof (Lisp_Charset, short_name) },
771 { XD_LISP_OBJECT, offsetof (Lisp_Charset, long_name) },
772 { XD_LISP_OBJECT, offsetof (Lisp_Charset, reverse_direction_charset) },
773 { XD_LISP_OBJECT, offsetof (Lisp_Charset, ccl_program) },
775 { XD_LISP_OBJECT, offsetof (Lisp_Charset, decoding_table) },
776 { XD_LISP_OBJECT, offsetof (Lisp_Charset, mother) },
781 DEFINE_LRECORD_IMPLEMENTATION ("charset", charset,
782 mark_charset, print_charset, 0, 0, 0,
786 /* Make a new charset. */
787 /* #### SJT Should generic properties be allowed? */
789 make_charset (Charset_ID id, Lisp_Object name,
790 unsigned short chars, unsigned char dimension,
791 unsigned char columns, unsigned char graphic,
792 Bufbyte final, unsigned char direction, Lisp_Object short_name,
793 Lisp_Object long_name, Lisp_Object doc,
795 Lisp_Object decoding_table,
796 Emchar min_code, Emchar max_code,
797 Emchar code_offset, unsigned char byte_offset,
798 Lisp_Object mother, unsigned char conversion)
801 Lisp_Charset *cs = alloc_lcrecord_type (Lisp_Charset, &lrecord_charset);
805 XSETCHARSET (obj, cs);
807 CHARSET_ID (cs) = id;
808 CHARSET_NAME (cs) = name;
809 CHARSET_SHORT_NAME (cs) = short_name;
810 CHARSET_LONG_NAME (cs) = long_name;
811 CHARSET_CHARS (cs) = chars;
812 CHARSET_DIMENSION (cs) = dimension;
813 CHARSET_DIRECTION (cs) = direction;
814 CHARSET_COLUMNS (cs) = columns;
815 CHARSET_GRAPHIC (cs) = graphic;
816 CHARSET_FINAL (cs) = final;
817 CHARSET_DOC_STRING (cs) = doc;
818 CHARSET_REGISTRY (cs) = reg;
819 CHARSET_CCL_PROGRAM (cs) = Qnil;
820 CHARSET_REVERSE_DIRECTION_CHARSET (cs) = Qnil;
822 CHARSET_DECODING_TABLE(cs) = Qunbound;
823 CHARSET_MIN_CODE (cs) = min_code;
824 CHARSET_MAX_CODE (cs) = max_code;
825 CHARSET_CODE_OFFSET (cs) = code_offset;
826 CHARSET_BYTE_OFFSET (cs) = byte_offset;
827 CHARSET_MOTHER (cs) = mother;
828 CHARSET_CONVERSION (cs) = conversion;
832 if (id == LEADING_BYTE_ASCII)
833 CHARSET_REP_BYTES (cs) = 1;
835 CHARSET_REP_BYTES (cs) = CHARSET_DIMENSION (cs) + 1;
837 CHARSET_REP_BYTES (cs) = CHARSET_DIMENSION (cs) + 2;
842 /* some charsets do not have final characters. This includes
843 ASCII, Control-1, Composite, and the two faux private
845 unsigned char iso2022_type
846 = (dimension == 1 ? 0 : 2) + (chars == 94 ? 0 : 1);
848 if (code_offset == 0)
850 assert (NILP (chlook->charset_by_attributes[iso2022_type][final]));
851 chlook->charset_by_attributes[iso2022_type][final] = obj;
855 (chlook->charset_by_attributes[iso2022_type][final][direction]));
856 chlook->charset_by_attributes[iso2022_type][final][direction] = obj;
860 assert (NILP (chlook->charset_by_leading_byte[id - MIN_LEADING_BYTE]));
861 chlook->charset_by_leading_byte[id - MIN_LEADING_BYTE] = obj;
863 /* Some charsets are "faux" and don't have names or really exist at
864 all except in the leading-byte table. */
866 Fputhash (name, obj, Vcharset_hash_table);
871 get_unallocated_leading_byte (int dimension)
876 if (chlook->next_allocated_leading_byte > MAX_LEADING_BYTE_PRIVATE)
879 lb = chlook->next_allocated_leading_byte++;
883 if (chlook->next_allocated_1_byte_leading_byte > MAX_LEADING_BYTE_PRIVATE_1)
886 lb = chlook->next_allocated_1_byte_leading_byte++;
890 if (chlook->next_allocated_2_byte_leading_byte > MAX_LEADING_BYTE_PRIVATE_2)
893 lb = chlook->next_allocated_2_byte_leading_byte++;
899 ("No more character sets free for this dimension",
900 make_int (dimension));
906 /* Number of Big5 characters which have the same code in 1st byte. */
908 #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40)
911 decode_defined_char (Lisp_Object ccs, int code_point)
913 int dim = XCHARSET_DIMENSION (ccs);
914 Lisp_Object decoding_table = XCHARSET_DECODING_TABLE (ccs);
922 = get_ccs_octet_table (decoding_table, ccs,
923 (code_point >> (dim * 8)) & 255);
925 if (CHARP (decoding_table))
926 return XCHAR (decoding_table);
929 else if ( CHARSETP (mother = XCHARSET_MOTHER (ccs)) )
931 if ( XCHARSET_CONVERSION (ccs) == CONVERSION_IDENTICAL )
933 if ( EQ (mother, Vcharset_ucs) )
934 return DECODE_CHAR (mother, code_point);
936 return decode_defined_char (mother, code_point);
943 decode_builtin_char (Lisp_Object charset, int code_point)
945 Lisp_Object mother = XCHARSET_MOTHER (charset);
948 if ( XCHARSET_MAX_CODE (charset) > 0 )
950 if ( CHARSETP (mother) )
952 int code = code_point;
954 if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x60 )
956 int row = code_point >> 8;
957 int cell = code_point & 255;
961 else if (row < 16 + 32 + 30)
962 code = (row - (16 + 32)) * 94 + cell - 33;
963 else if (row < 18 + 32 + 30)
965 else if (row < 18 + 32 + 60)
966 code = (row - (18 + 32)) * 94 + cell - 33;
968 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x60 )
970 int plane = code_point >> 16;
971 int row = (code_point >> 8) & 255;
972 int cell = code_point & 255;
976 else if (row < 16 + 32 + 30)
978 = (plane - 33) * 94 * 60
979 + (row - (16 + 32)) * 94
981 else if (row < 18 + 32 + 30)
983 else if (row < 18 + 32 + 60)
985 = (plane - 33) * 94 * 60
986 + (row - (18 + 32)) * 94
990 decode_builtin_char (mother, code + XCHARSET_CODE_OFFSET(charset));
995 = (XCHARSET_DIMENSION (charset) == 1
997 code_point - XCHARSET_BYTE_OFFSET (charset)
999 ((code_point >> 8) - XCHARSET_BYTE_OFFSET (charset))
1000 * XCHARSET_CHARS (charset)
1001 + (code_point & 0xFF) - XCHARSET_BYTE_OFFSET (charset))
1002 + XCHARSET_CODE_OFFSET (charset);
1003 if ((cid < XCHARSET_MIN_CODE (charset))
1004 || (XCHARSET_MAX_CODE (charset) < cid))
1009 else if ((final = XCHARSET_FINAL (charset)) >= '0')
1011 if (XCHARSET_DIMENSION (charset) == 1)
1013 switch (XCHARSET_CHARS (charset))
1017 + (final - '0') * 94 + ((code_point & 0x7F) - 33);
1020 + (final - '0') * 96 + ((code_point & 0x7F) - 32);
1028 switch (XCHARSET_CHARS (charset))
1031 return MIN_CHAR_94x94
1032 + (final - '0') * 94 * 94
1033 + (((code_point >> 8) & 0x7F) - 33) * 94
1034 + ((code_point & 0x7F) - 33);
1036 return MIN_CHAR_96x96
1037 + (final - '0') * 96 * 96
1038 + (((code_point >> 8) & 0x7F) - 32) * 96
1039 + ((code_point & 0x7F) - 32);
1051 charset_code_point (Lisp_Object charset, Emchar ch, int defined_only)
1053 Lisp_Object encoding_table = XCHARSET_ENCODING_TABLE (charset);
1056 if ( CHAR_TABLEP (encoding_table)
1057 && INTP (ret = get_char_id_table (XCHAR_TABLE(encoding_table),
1062 Lisp_Object mother = XCHARSET_MOTHER (charset);
1063 int min = XCHARSET_MIN_CODE (charset);
1064 int max = XCHARSET_MAX_CODE (charset);
1067 if ( CHARSETP (mother) )
1068 code = charset_code_point (mother, ch, defined_only);
1069 else if (defined_only)
1073 if ( ((max == 0) && CHARSETP (mother)) ||
1074 ((min <= code) && (code <= max)) )
1076 int d = code - XCHARSET_CODE_OFFSET (charset);
1078 if ( XCHARSET_CONVERSION (charset) == CONVERSION_IDENTICAL )
1080 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94 )
1082 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96 )
1084 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x60 )
1087 int cell = d % 94 + 33;
1093 return (row << 8) | cell;
1095 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94 )
1096 return ((d / 94 + 33) << 8) | (d % 94 + 33);
1097 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96 )
1098 return ((d / 96 + 32) << 8) | (d % 96 + 32);
1099 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x60 )
1101 int plane = d / (94 * 60) + 33;
1102 int row = (d % (94 * 60)) / 94;
1103 int cell = d % 94 + 33;
1109 return (plane << 16) | (row << 8) | cell;
1111 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x94 )
1113 ( (d / (94 * 94) + 33) << 16)
1114 | ((d / 94 % 94 + 33) << 8)
1116 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96x96 )
1118 ( (d / (96 * 96) + 32) << 16)
1119 | ((d / 96 % 96 + 32) << 8)
1121 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x94x94 )
1123 ( (d / (94 * 94 * 94) + 33) << 24)
1124 | ((d / (94 * 94) % 94 + 33) << 16)
1125 | ((d / 94 % 94 + 33) << 8)
1127 else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96x96x96 )
1129 ( (d / (96 * 96 * 96) + 32) << 24)
1130 | ((d / (96 * 96) % 96 + 32) << 16)
1131 | ((d / 96 % 96 + 32) << 8)
1135 printf ("Unknown CCS-conversion %d is specified!",
1136 XCHARSET_CONVERSION (charset));
1140 else if ( ( XCHARSET_FINAL (charset) >= '0' ) &&
1141 ( XCHARSET_MIN_CODE (charset) == 0 )
1143 (XCHARSET_CODE_OFFSET (charset) == 0) ||
1144 (XCHARSET_CODE_OFFSET (charset)
1145 == XCHARSET_MIN_CODE (charset))
1150 if (XCHARSET_DIMENSION (charset) == 1)
1152 if (XCHARSET_CHARS (charset) == 94)
1154 if (((d = ch - (MIN_CHAR_94
1155 + (XCHARSET_FINAL (charset) - '0') * 94))
1160 else if (XCHARSET_CHARS (charset) == 96)
1162 if (((d = ch - (MIN_CHAR_96
1163 + (XCHARSET_FINAL (charset) - '0') * 96))
1171 else if (XCHARSET_DIMENSION (charset) == 2)
1173 if (XCHARSET_CHARS (charset) == 94)
1175 if (((d = ch - (MIN_CHAR_94x94
1177 (XCHARSET_FINAL (charset) - '0') * 94 * 94))
1180 return (((d / 94) + 33) << 8) | (d % 94 + 33);
1182 else if (XCHARSET_CHARS (charset) == 96)
1184 if (((d = ch - (MIN_CHAR_96x96
1186 (XCHARSET_FINAL (charset) - '0') * 96 * 96))
1189 return (((d / 96) + 32) << 8) | (d % 96 + 32);
1200 encode_builtin_char_1 (Emchar c, Lisp_Object* charset)
1202 if (c <= MAX_CHAR_BASIC_LATIN)
1204 *charset = Vcharset_ascii;
1209 *charset = Vcharset_control_1;
1214 *charset = Vcharset_latin_iso8859_1;
1218 else if ((MIN_CHAR_HEBREW <= c) && (c <= MAX_CHAR_HEBREW))
1220 *charset = Vcharset_hebrew_iso8859_8;
1221 return c - MIN_CHAR_HEBREW + 0x20;
1224 else if ((MIN_CHAR_THAI <= c) && (c <= MAX_CHAR_THAI))
1226 *charset = Vcharset_thai_tis620;
1227 return c - MIN_CHAR_THAI + 0x20;
1230 else if ((MIN_CHAR_HALFWIDTH_KATAKANA <= c)
1231 && (c <= MAX_CHAR_HALFWIDTH_KATAKANA))
1233 return list2 (Vcharset_katakana_jisx0201,
1234 make_int (c - MIN_CHAR_HALFWIDTH_KATAKANA + 33));
1237 else if (c <= MAX_CHAR_BMP)
1239 *charset = Vcharset_ucs_bmp;
1242 else if (c <= MAX_CHAR_SMP)
1244 *charset = Vcharset_ucs_smp;
1245 return c - MIN_CHAR_SMP;
1247 else if (c <= MAX_CHAR_SIP)
1249 *charset = Vcharset_ucs_sip;
1250 return c - MIN_CHAR_SIP;
1252 else if (c < MIN_CHAR_DAIKANWA)
1254 *charset = Vcharset_ucs;
1257 else if (c <= MAX_CHAR_DAIKANWA)
1259 *charset = Vcharset_ideograph_daikanwa;
1260 return c - MIN_CHAR_DAIKANWA;
1262 else if (c < MIN_CHAR_94)
1264 *charset = Vcharset_ucs;
1267 else if (c <= MAX_CHAR_94)
1269 *charset = CHARSET_BY_ATTRIBUTES (94, 1,
1270 ((c - MIN_CHAR_94) / 94) + '0',
1271 CHARSET_LEFT_TO_RIGHT);
1272 if (!NILP (*charset))
1273 return ((c - MIN_CHAR_94) % 94) + 33;
1276 *charset = Vcharset_ucs;
1280 else if (c <= MAX_CHAR_96)
1282 *charset = CHARSET_BY_ATTRIBUTES (96, 1,
1283 ((c - MIN_CHAR_96) / 96) + '0',
1284 CHARSET_LEFT_TO_RIGHT);
1285 if (!NILP (*charset))
1286 return ((c - MIN_CHAR_96) % 96) + 32;
1289 *charset = Vcharset_ucs;
1293 else if (c <= MAX_CHAR_94x94)
1296 = CHARSET_BY_ATTRIBUTES (94, 2,
1297 ((c - MIN_CHAR_94x94) / (94 * 94)) + '0',
1298 CHARSET_LEFT_TO_RIGHT);
1299 if (!NILP (*charset))
1300 return (((((c - MIN_CHAR_94x94) / 94) % 94) + 33) << 8)
1301 | (((c - MIN_CHAR_94x94) % 94) + 33);
1304 *charset = Vcharset_ucs;
1308 else if (c <= MAX_CHAR_96x96)
1311 = CHARSET_BY_ATTRIBUTES (96, 2,
1312 ((c - MIN_CHAR_96x96) / (96 * 96)) + '0',
1313 CHARSET_LEFT_TO_RIGHT);
1314 if (!NILP (*charset))
1315 return ((((c - MIN_CHAR_96x96) / 96) % 96) + 32) << 8
1316 | (((c - MIN_CHAR_96x96) % 96) + 32);
1319 *charset = Vcharset_ucs;
1325 *charset = Vcharset_ucs;
1330 Lisp_Object Vdefault_coded_charset_priority_list;
1334 /************************************************************************/
1335 /* Basic charset Lisp functions */
1336 /************************************************************************/
1338 DEFUN ("charsetp", Fcharsetp, 1, 1, 0, /*
1339 Return non-nil if OBJECT is a charset.
1343 return CHARSETP (object) ? Qt : Qnil;
1346 DEFUN ("find-charset", Ffind_charset, 1, 1, 0, /*
1347 Retrieve the charset of the given name.
1348 If CHARSET-OR-NAME is a charset object, it is simply returned.
1349 Otherwise, CHARSET-OR-NAME should be a symbol. If there is no such charset,
1350 nil is returned. Otherwise the associated charset object is returned.
1354 if (CHARSETP (charset_or_name))
1355 return charset_or_name;
1357 CHECK_SYMBOL (charset_or_name);
1358 return Fgethash (charset_or_name, Vcharset_hash_table, Qnil);
1361 DEFUN ("get-charset", Fget_charset, 1, 1, 0, /*
1362 Retrieve the charset of the given name.
1363 Same as `find-charset' except an error is signalled if there is no such
1364 charset instead of returning nil.
1368 Lisp_Object charset = Ffind_charset (name);
1371 signal_simple_error ("No such charset", name);
1375 /* We store the charsets in hash tables with the names as the key and the
1376 actual charset object as the value. Occasionally we need to use them
1377 in a list format. These routines provide us with that. */
1378 struct charset_list_closure
1380 Lisp_Object *charset_list;
1384 add_charset_to_list_mapper (Lisp_Object key, Lisp_Object value,
1385 void *charset_list_closure)
1387 /* This function can GC */
1388 struct charset_list_closure *chcl =
1389 (struct charset_list_closure*) charset_list_closure;
1390 Lisp_Object *charset_list = chcl->charset_list;
1392 *charset_list = Fcons (key /* XCHARSET_NAME (value) */, *charset_list);
1396 DEFUN ("charset-list", Fcharset_list, 0, 0, 0, /*
1397 Return a list of the names of all defined charsets.
1401 Lisp_Object charset_list = Qnil;
1402 struct gcpro gcpro1;
1403 struct charset_list_closure charset_list_closure;
1405 GCPRO1 (charset_list);
1406 charset_list_closure.charset_list = &charset_list;
1407 elisp_maphash (add_charset_to_list_mapper, Vcharset_hash_table,
1408 &charset_list_closure);
1411 return charset_list;
1414 DEFUN ("charset-name", Fcharset_name, 1, 1, 0, /*
1415 Return the name of charset CHARSET.
1419 return XCHARSET_NAME (Fget_charset (charset));
1422 /* #### SJT Should generic properties be allowed? */
1423 DEFUN ("make-charset", Fmake_charset, 3, 3, 0, /*
1424 Define a new character set.
1425 This function is for use with Mule support.
1426 NAME is a symbol, the name by which the character set is normally referred.
1427 DOC-STRING is a string describing the character set.
1428 PROPS is a property list, describing the specific nature of the
1429 character set. Recognized properties are:
1431 'short-name Short version of the charset name (ex: Latin-1)
1432 'long-name Long version of the charset name (ex: ISO8859-1 (Latin-1))
1433 'registry A regular expression matching the font registry field for
1435 'dimension Number of octets used to index a character in this charset.
1436 Either 1 or 2. Defaults to 1.
1437 If UTF-2000 feature is enabled, 3 or 4 are also available.
1438 'columns Number of columns used to display a character in this charset.
1439 Only used in TTY mode. (Under X, the actual width of a
1440 character can be derived from the font used to display the
1441 characters.) If unspecified, defaults to the dimension
1442 (this is almost always the correct value).
1443 'chars Number of characters in each dimension (94 or 96).
1444 Defaults to 94. Note that if the dimension is 2, the
1445 character set thus described is 94x94 or 96x96.
1446 If UTF-2000 feature is enabled, 128 or 256 are also available.
1447 'final Final byte of ISO 2022 escape sequence. Must be
1448 supplied. Each combination of (DIMENSION, CHARS) defines a
1449 separate namespace for final bytes. Note that ISO
1450 2022 restricts the final byte to the range
1451 0x30 - 0x7E if dimension == 1, and 0x30 - 0x5F if
1452 dimension == 2. Note also that final bytes in the range
1453 0x30 - 0x3F are reserved for user-defined (not official)
1455 'graphic 0 (use left half of font on output) or 1 (use right half
1456 of font on output). Defaults to 0. For example, for
1457 a font whose registry is ISO8859-1, the left half
1458 (octets 0x20 - 0x7F) is the `ascii' character set, while
1459 the right half (octets 0xA0 - 0xFF) is the `latin-1'
1460 character set. With 'graphic set to 0, the octets
1461 will have their high bit cleared; with it set to 1,
1462 the octets will have their high bit set.
1463 'direction 'l2r (left-to-right) or 'r2l (right-to-left).
1465 'ccl-program A compiled CCL program used to convert a character in
1466 this charset into an index into the font. This is in
1467 addition to the 'graphic property. The CCL program
1468 is passed the octets of the character, with the high
1469 bit cleared and set depending upon whether the value
1470 of the 'graphic property is 0 or 1.
1471 'mother [UTF-2000 only] Base coded-charset.
1472 'code-min [UTF-2000 only] Minimum code-point of a base coded-charset.
1473 'code-max [UTF-2000 only] Maximum code-point of a base coded-charset.
1474 'code-offset [UTF-2000 only] Offset for a code-point of a base
1476 'conversion [UTF-2000 only] Conversion for a code-point of a base
1477 coded-charset (94x60 or 94x94x60).
1479 (name, doc_string, props))
1481 int id, dimension = 1, chars = 94, graphic = 0, final = 0, columns = -1;
1482 int direction = CHARSET_LEFT_TO_RIGHT;
1483 Lisp_Object registry = Qnil;
1484 Lisp_Object charset;
1485 Lisp_Object ccl_program = Qnil;
1486 Lisp_Object short_name = Qnil, long_name = Qnil;
1487 Lisp_Object mother = Qnil;
1488 int min_code = 0, max_code = 0, code_offset = 0;
1489 int byte_offset = -1;
1492 CHECK_SYMBOL (name);
1493 if (!NILP (doc_string))
1494 CHECK_STRING (doc_string);
1496 charset = Ffind_charset (name);
1497 if (!NILP (charset))
1498 signal_simple_error ("Cannot redefine existing charset", name);
1501 EXTERNAL_PROPERTY_LIST_LOOP_3 (keyword, value, props)
1503 if (EQ (keyword, Qshort_name))
1505 CHECK_STRING (value);
1509 if (EQ (keyword, Qlong_name))
1511 CHECK_STRING (value);
1515 else if (EQ (keyword, Qdimension))
1518 dimension = XINT (value);
1519 if (dimension < 1 ||
1526 signal_simple_error ("Invalid value for 'dimension", value);
1529 else if (EQ (keyword, Qchars))
1532 chars = XINT (value);
1533 if (chars != 94 && chars != 96
1535 && chars != 128 && chars != 256
1538 signal_simple_error ("Invalid value for 'chars", value);
1541 else if (EQ (keyword, Qcolumns))
1544 columns = XINT (value);
1545 if (columns != 1 && columns != 2)
1546 signal_simple_error ("Invalid value for 'columns", value);
1549 else if (EQ (keyword, Qgraphic))
1552 graphic = XINT (value);
1560 signal_simple_error ("Invalid value for 'graphic", value);
1563 else if (EQ (keyword, Qregistry))
1565 CHECK_STRING (value);
1569 else if (EQ (keyword, Qdirection))
1571 if (EQ (value, Ql2r))
1572 direction = CHARSET_LEFT_TO_RIGHT;
1573 else if (EQ (value, Qr2l))
1574 direction = CHARSET_RIGHT_TO_LEFT;
1576 signal_simple_error ("Invalid value for 'direction", value);
1579 else if (EQ (keyword, Qfinal))
1581 CHECK_CHAR_COERCE_INT (value);
1582 final = XCHAR (value);
1583 if (final < '0' || final > '~')
1584 signal_simple_error ("Invalid value for 'final", value);
1588 else if (EQ (keyword, Qmother))
1590 mother = Fget_charset (value);
1593 else if (EQ (keyword, Qmin_code))
1596 min_code = XUINT (value);
1599 else if (EQ (keyword, Qmax_code))
1602 max_code = XUINT (value);
1605 else if (EQ (keyword, Qcode_offset))
1608 code_offset = XUINT (value);
1611 else if (EQ (keyword, Qconversion))
1613 if (EQ (value, Q94x60))
1614 conversion = CONVERSION_94x60;
1615 else if (EQ (value, Q94x94x60))
1616 conversion = CONVERSION_94x94x60;
1618 signal_simple_error ("Unrecognized conversion", value);
1622 else if (EQ (keyword, Qccl_program))
1624 struct ccl_program test_ccl;
1626 if (setup_ccl_program (&test_ccl, value) < 0)
1627 signal_simple_error ("Invalid value for 'ccl-program", value);
1628 ccl_program = value;
1632 signal_simple_error ("Unrecognized property", keyword);
1638 error ("'final must be specified");
1640 if (dimension == 2 && final > 0x5F)
1642 ("Final must be in the range 0x30 - 0x5F for dimension == 2",
1645 if (!NILP (CHARSET_BY_ATTRIBUTES (chars, dimension, final,
1646 CHARSET_LEFT_TO_RIGHT)) ||
1647 !NILP (CHARSET_BY_ATTRIBUTES (chars, dimension, final,
1648 CHARSET_RIGHT_TO_LEFT)))
1650 ("Character set already defined for this DIMENSION/CHARS/FINAL combo");
1652 id = get_unallocated_leading_byte (dimension);
1654 if (NILP (doc_string))
1655 doc_string = build_string ("");
1657 if (NILP (registry))
1658 registry = build_string ("");
1660 if (NILP (short_name))
1661 XSETSTRING (short_name, XSYMBOL (name)->name);
1663 if (NILP (long_name))
1664 long_name = doc_string;
1667 columns = dimension;
1669 if (byte_offset < 0)
1673 else if (chars == 96)
1679 charset = make_charset (id, name, chars, dimension, columns, graphic,
1680 final, direction, short_name, long_name,
1681 doc_string, registry,
1682 Qnil, min_code, max_code, code_offset, byte_offset,
1683 mother, conversion);
1684 if (!NILP (ccl_program))
1685 XCHARSET_CCL_PROGRAM (charset) = ccl_program;
1689 DEFUN ("make-reverse-direction-charset", Fmake_reverse_direction_charset,
1691 Make a charset equivalent to CHARSET but which goes in the opposite direction.
1692 NEW-NAME is the name of the new charset. Return the new charset.
1694 (charset, new_name))
1696 Lisp_Object new_charset = Qnil;
1697 int id, chars, dimension, columns, graphic, final;
1699 Lisp_Object registry, doc_string, short_name, long_name;
1702 charset = Fget_charset (charset);
1703 if (!NILP (XCHARSET_REVERSE_DIRECTION_CHARSET (charset)))
1704 signal_simple_error ("Charset already has reverse-direction charset",
1707 CHECK_SYMBOL (new_name);
1708 if (!NILP (Ffind_charset (new_name)))
1709 signal_simple_error ("Cannot redefine existing charset", new_name);
1711 cs = XCHARSET (charset);
1713 chars = CHARSET_CHARS (cs);
1714 dimension = CHARSET_DIMENSION (cs);
1715 columns = CHARSET_COLUMNS (cs);
1716 id = get_unallocated_leading_byte (dimension);
1718 graphic = CHARSET_GRAPHIC (cs);
1719 final = CHARSET_FINAL (cs);
1720 direction = CHARSET_RIGHT_TO_LEFT;
1721 if (CHARSET_DIRECTION (cs) == CHARSET_RIGHT_TO_LEFT)
1722 direction = CHARSET_LEFT_TO_RIGHT;
1723 doc_string = CHARSET_DOC_STRING (cs);
1724 short_name = CHARSET_SHORT_NAME (cs);
1725 long_name = CHARSET_LONG_NAME (cs);
1726 registry = CHARSET_REGISTRY (cs);
1728 new_charset = make_charset (id, new_name, chars, dimension, columns,
1729 graphic, final, direction, short_name, long_name,
1730 doc_string, registry,
1732 CHARSET_DECODING_TABLE(cs),
1733 CHARSET_MIN_CODE(cs),
1734 CHARSET_MAX_CODE(cs),
1735 CHARSET_CODE_OFFSET(cs),
1736 CHARSET_BYTE_OFFSET(cs),
1738 CHARSET_CONVERSION (cs)
1740 Qnil, 0, 0, 0, 0, Qnil, 0
1744 CHARSET_REVERSE_DIRECTION_CHARSET (cs) = new_charset;
1745 XCHARSET_REVERSE_DIRECTION_CHARSET (new_charset) = charset;
1750 DEFUN ("define-charset-alias", Fdefine_charset_alias, 2, 2, 0, /*
1751 Define symbol ALIAS as an alias for CHARSET.
1755 CHECK_SYMBOL (alias);
1756 charset = Fget_charset (charset);
1757 return Fputhash (alias, charset, Vcharset_hash_table);
1760 /* #### Reverse direction charsets not yet implemented. */
1762 DEFUN ("charset-reverse-direction-charset", Fcharset_reverse_direction_charset,
1764 Return the reverse-direction charset parallel to CHARSET, if any.
1765 This is the charset with the same properties (in particular, the same
1766 dimension, number of characters per dimension, and final byte) as
1767 CHARSET but whose characters are displayed in the opposite direction.
1771 charset = Fget_charset (charset);
1772 return XCHARSET_REVERSE_DIRECTION_CHARSET (charset);
1776 DEFUN ("charset-from-attributes", Fcharset_from_attributes, 3, 4, 0, /*
1777 Return a charset with the given DIMENSION, CHARS, FINAL, and DIRECTION.
1778 If DIRECTION is omitted, both directions will be checked (left-to-right
1779 will be returned if character sets exist for both directions).
1781 (dimension, chars, final, direction))
1783 int dm, ch, fi, di = -1;
1784 Lisp_Object obj = Qnil;
1786 CHECK_INT (dimension);
1787 dm = XINT (dimension);
1788 if (dm < 1 || dm > 2)
1789 signal_simple_error ("Invalid value for DIMENSION", dimension);
1793 if (ch != 94 && ch != 96)
1794 signal_simple_error ("Invalid value for CHARS", chars);
1796 CHECK_CHAR_COERCE_INT (final);
1798 if (fi < '0' || fi > '~')
1799 signal_simple_error ("Invalid value for FINAL", final);
1801 if (EQ (direction, Ql2r))
1802 di = CHARSET_LEFT_TO_RIGHT;
1803 else if (EQ (direction, Qr2l))
1804 di = CHARSET_RIGHT_TO_LEFT;
1805 else if (!NILP (direction))
1806 signal_simple_error ("Invalid value for DIRECTION", direction);
1808 if (dm == 2 && fi > 0x5F)
1810 ("Final must be in the range 0x30 - 0x5F for dimension == 2", final);
1814 obj = CHARSET_BY_ATTRIBUTES (ch, dm, fi, CHARSET_LEFT_TO_RIGHT);
1816 obj = CHARSET_BY_ATTRIBUTES (ch, dm, fi, CHARSET_RIGHT_TO_LEFT);
1819 obj = CHARSET_BY_ATTRIBUTES (ch, dm, fi, di);
1822 return XCHARSET_NAME (obj);
1826 DEFUN ("charset-short-name", Fcharset_short_name, 1, 1, 0, /*
1827 Return short name of CHARSET.
1831 return XCHARSET_SHORT_NAME (Fget_charset (charset));
1834 DEFUN ("charset-long-name", Fcharset_long_name, 1, 1, 0, /*
1835 Return long name of CHARSET.
1839 return XCHARSET_LONG_NAME (Fget_charset (charset));
1842 DEFUN ("charset-description", Fcharset_description, 1, 1, 0, /*
1843 Return description of CHARSET.
1847 return XCHARSET_DOC_STRING (Fget_charset (charset));
1850 DEFUN ("charset-dimension", Fcharset_dimension, 1, 1, 0, /*
1851 Return dimension of CHARSET.
1855 return make_int (XCHARSET_DIMENSION (Fget_charset (charset)));
1858 DEFUN ("charset-property", Fcharset_property, 2, 2, 0, /*
1859 Return property PROP of CHARSET, a charset object or symbol naming a charset.
1860 Recognized properties are those listed in `make-charset', as well as
1861 'name and 'doc-string.
1867 charset = Fget_charset (charset);
1868 cs = XCHARSET (charset);
1870 CHECK_SYMBOL (prop);
1871 if (EQ (prop, Qname)) return CHARSET_NAME (cs);
1872 if (EQ (prop, Qshort_name)) return CHARSET_SHORT_NAME (cs);
1873 if (EQ (prop, Qlong_name)) return CHARSET_LONG_NAME (cs);
1874 if (EQ (prop, Qdoc_string)) return CHARSET_DOC_STRING (cs);
1875 if (EQ (prop, Qdimension)) return make_int (CHARSET_DIMENSION (cs));
1876 if (EQ (prop, Qcolumns)) return make_int (CHARSET_COLUMNS (cs));
1877 if (EQ (prop, Qgraphic)) return make_int (CHARSET_GRAPHIC (cs));
1878 if (EQ (prop, Qfinal)) return CHARSET_FINAL (cs) == 0 ?
1879 Qnil : make_char (CHARSET_FINAL (cs));
1880 if (EQ (prop, Qchars)) return make_int (CHARSET_CHARS (cs));
1881 if (EQ (prop, Qregistry)) return CHARSET_REGISTRY (cs);
1882 if (EQ (prop, Qccl_program)) return CHARSET_CCL_PROGRAM (cs);
1883 if (EQ (prop, Qdirection))
1884 return CHARSET_DIRECTION (cs) == CHARSET_LEFT_TO_RIGHT ? Ql2r : Qr2l;
1885 if (EQ (prop, Qreverse_direction_charset))
1887 Lisp_Object obj = CHARSET_REVERSE_DIRECTION_CHARSET (cs);
1888 /* #### Is this translation OK? If so, error checking sufficient? */
1889 return CHARSETP (obj) ? XCHARSET_NAME (obj) : obj;
1892 if (EQ (prop, Qmother))
1893 return CHARSET_MOTHER (cs);
1894 if (EQ (prop, Qmin_code))
1895 return make_int (CHARSET_MIN_CODE (cs));
1896 if (EQ (prop, Qmax_code))
1897 return make_int (CHARSET_MAX_CODE (cs));
1899 signal_simple_error ("Unrecognized charset property name", prop);
1900 return Qnil; /* not reached */
1903 DEFUN ("charset-id", Fcharset_id, 1, 1, 0, /*
1904 Return charset identification number of CHARSET.
1908 return make_int(XCHARSET_LEADING_BYTE (Fget_charset (charset)));
1911 /* #### We need to figure out which properties we really want to
1914 DEFUN ("set-charset-ccl-program", Fset_charset_ccl_program, 2, 2, 0, /*
1915 Set the 'ccl-program property of CHARSET to CCL-PROGRAM.
1917 (charset, ccl_program))
1919 struct ccl_program test_ccl;
1921 charset = Fget_charset (charset);
1922 if (setup_ccl_program (&test_ccl, ccl_program) < 0)
1923 signal_simple_error ("Invalid ccl-program", ccl_program);
1924 XCHARSET_CCL_PROGRAM (charset) = ccl_program;
1929 invalidate_charset_font_caches (Lisp_Object charset)
1931 /* Invalidate font cache entries for charset on all devices. */
1932 Lisp_Object devcons, concons, hash_table;
1933 DEVICE_LOOP_NO_BREAK (devcons, concons)
1935 struct device *d = XDEVICE (XCAR (devcons));
1936 hash_table = Fgethash (charset, d->charset_font_cache, Qunbound);
1937 if (!UNBOUNDP (hash_table))
1938 Fclrhash (hash_table);
1942 DEFUN ("set-charset-registry", Fset_charset_registry, 2, 2, 0, /*
1943 Set the 'registry property of CHARSET to REGISTRY.
1945 (charset, registry))
1947 charset = Fget_charset (charset);
1948 CHECK_STRING (registry);
1949 XCHARSET_REGISTRY (charset) = registry;
1950 invalidate_charset_font_caches (charset);
1951 face_property_was_changed (Vdefault_face, Qfont, Qglobal);
1956 DEFUN ("charset-mapping-table", Fcharset_mapping_table, 1, 1, 0, /*
1957 Return mapping-table of CHARSET.
1961 return XCHARSET_DECODING_TABLE (Fget_charset (charset));
1964 DEFUN ("set-charset-mapping-table", Fset_charset_mapping_table, 2, 2, 0, /*
1965 Set mapping-table of CHARSET to TABLE.
1969 struct Lisp_Charset *cs;
1973 charset = Fget_charset (charset);
1974 cs = XCHARSET (charset);
1978 CHARSET_DECODING_TABLE(cs) = Qnil;
1981 else if (VECTORP (table))
1983 int ccs_len = CHARSET_BYTE_SIZE (cs);
1984 int ret = decoding_table_check_elements (table,
1985 CHARSET_DIMENSION (cs),
1990 signal_simple_error ("Too big table", table);
1992 signal_simple_error ("Invalid element is found", table);
1994 signal_simple_error ("Something wrong", table);
1996 CHARSET_DECODING_TABLE(cs) = Qnil;
1999 signal_error (Qwrong_type_argument,
2000 list2 (build_translated_string ("vector-or-nil-p"),
2003 byte_offset = CHARSET_BYTE_OFFSET (cs);
2004 switch (CHARSET_DIMENSION (cs))
2007 for (i = 0; i < XVECTOR_LENGTH (table); i++)
2009 Lisp_Object c = XVECTOR_DATA(table)[i];
2012 Fput_char_attribute (c, XCHARSET_NAME (charset),
2013 make_int (i + byte_offset));
2017 for (i = 0; i < XVECTOR_LENGTH (table); i++)
2019 Lisp_Object v = XVECTOR_DATA(table)[i];
2025 for (j = 0; j < XVECTOR_LENGTH (v); j++)
2027 Lisp_Object c = XVECTOR_DATA(v)[j];
2031 (c, XCHARSET_NAME (charset),
2032 make_int ( ( (i + byte_offset) << 8 )
2038 Fput_char_attribute (v, XCHARSET_NAME (charset),
2039 make_int (i + byte_offset));
2048 /************************************************************************/
2049 /* Lisp primitives for working with characters */
2050 /************************************************************************/
2053 DEFUN ("decode-char", Fdecode_char, 2, 3, 0, /*
2054 Make a character from CHARSET and code-point CODE.
2055 If DEFINED_ONLY is non-nil, builtin character is not returned.
2056 If corresponding character is not found, nil is returned.
2058 (charset, code, defined_only))
2062 charset = Fget_charset (charset);
2065 if (XCHARSET_GRAPHIC (charset) == 1)
2067 if (NILP (defined_only))
2068 c = DECODE_CHAR (charset, c);
2070 c = decode_defined_char (charset, c);
2071 return c >= 0 ? make_char (c) : Qnil;
2074 DEFUN ("decode-builtin-char", Fdecode_builtin_char, 2, 2, 0, /*
2075 Make a builtin character from CHARSET and code-point CODE.
2081 charset = Fget_charset (charset);
2083 if (EQ (charset, Vcharset_latin_viscii))
2085 Lisp_Object chr = Fdecode_char (charset, code, Qnil);
2091 (ret = Fget_char_attribute (chr,
2092 Vcharset_latin_viscii_lower,
2095 charset = Vcharset_latin_viscii_lower;
2099 (ret = Fget_char_attribute (chr,
2100 Vcharset_latin_viscii_upper,
2103 charset = Vcharset_latin_viscii_upper;
2110 if (XCHARSET_GRAPHIC (charset) == 1)
2113 c = decode_builtin_char (charset, c);
2114 return c >= 0 ? make_char (c) : Fdecode_char (charset, code, Qnil);
2118 DEFUN ("make-char", Fmake_char, 2, 3, 0, /*
2119 Make a character from CHARSET and octets ARG1 and ARG2.
2120 ARG2 is required only for characters from two-dimensional charsets.
2121 For example, (make-char 'latin-iso8859-2 185) will return the Latin 2
2122 character s with caron.
2124 (charset, arg1, arg2))
2128 int lowlim, highlim;
2130 charset = Fget_charset (charset);
2131 cs = XCHARSET (charset);
2133 if (EQ (charset, Vcharset_ascii)) lowlim = 0, highlim = 127;
2134 else if (EQ (charset, Vcharset_control_1)) lowlim = 0, highlim = 31;
2136 else if (CHARSET_CHARS (cs) == 256) lowlim = 0, highlim = 255;
2138 else if (CHARSET_CHARS (cs) == 94) lowlim = 33, highlim = 126;
2139 else /* CHARSET_CHARS (cs) == 96) */ lowlim = 32, highlim = 127;
2142 /* It is useful (and safe, according to Olivier Galibert) to strip
2143 the 8th bit off ARG1 and ARG2 because it allows programmers to
2144 write (make-char 'latin-iso8859-2 CODE) where code is the actual
2145 Latin 2 code of the character. */
2153 if (a1 < lowlim || a1 > highlim)
2154 args_out_of_range_3 (arg1, make_int (lowlim), make_int (highlim));
2156 if (CHARSET_DIMENSION (cs) == 1)
2160 ("Charset is of dimension one; second octet must be nil", arg2);
2161 return make_char (MAKE_CHAR (charset, a1, 0));
2170 a2 = XINT (arg2) & 0x7f;
2172 if (a2 < lowlim || a2 > highlim)
2173 args_out_of_range_3 (arg2, make_int (lowlim), make_int (highlim));
2175 return make_char (MAKE_CHAR (charset, a1, a2));
2178 DEFUN ("char-charset", Fchar_charset, 1, 1, 0, /*
2179 Return the character set of CHARACTER.
2183 CHECK_CHAR_COERCE_INT (character);
2185 return XCHARSET_NAME (CHAR_CHARSET (XCHAR (character)));
2188 DEFUN ("char-octet", Fchar_octet, 1, 2, 0, /*
2189 Return the octet numbered N (should be 0 or 1) of CHARACTER.
2190 N defaults to 0 if omitted.
2194 Lisp_Object charset;
2197 CHECK_CHAR_COERCE_INT (character);
2199 BREAKUP_CHAR (XCHAR (character), charset, octet0, octet1);
2201 if (NILP (n) || EQ (n, Qzero))
2202 return make_int (octet0);
2203 else if (EQ (n, make_int (1)))
2204 return make_int (octet1);
2206 signal_simple_error ("Octet number must be 0 or 1", n);
2210 DEFUN ("encode-char", Fencode_char, 2, 3, 0, /*
2211 Return code-point of CHARACTER in specified CHARSET.
2213 (character, charset, defined_only))
2217 CHECK_CHAR_COERCE_INT (character);
2218 charset = Fget_charset (charset);
2219 code_point = charset_code_point (charset, XCHAR (character),
2220 !NILP (defined_only));
2221 if (code_point >= 0)
2222 return make_int (code_point);
2228 DEFUN ("split-char", Fsplit_char, 1, 1, 0, /*
2229 Return list of charset and one or two position-codes of CHARACTER.
2233 /* This function can GC */
2234 struct gcpro gcpro1, gcpro2;
2235 Lisp_Object charset = Qnil;
2236 Lisp_Object rc = Qnil;
2244 GCPRO2 (charset, rc);
2245 CHECK_CHAR_COERCE_INT (character);
2248 code_point = ENCODE_CHAR (XCHAR (character), charset);
2249 dimension = XCHARSET_DIMENSION (charset);
2250 while (dimension > 0)
2252 rc = Fcons (make_int (code_point & 255), rc);
2256 rc = Fcons (XCHARSET_NAME (charset), rc);
2258 BREAKUP_CHAR (XCHAR (character), charset, c1, c2);
2260 if (XCHARSET_DIMENSION (Fget_charset (charset)) == 2)
2262 rc = list3 (XCHARSET_NAME (charset), make_int (c1), make_int (c2));
2266 rc = list2 (XCHARSET_NAME (charset), make_int (c1));
2275 #ifdef ENABLE_COMPOSITE_CHARS
2276 /************************************************************************/
2277 /* composite character functions */
2278 /************************************************************************/
2281 lookup_composite_char (Bufbyte *str, int len)
2283 Lisp_Object lispstr = make_string (str, len);
2284 Lisp_Object ch = Fgethash (lispstr,
2285 Vcomposite_char_string2char_hash_table,
2291 if (composite_char_row_next >= 128)
2292 signal_simple_error ("No more composite chars available", lispstr);
2293 emch = MAKE_CHAR (Vcharset_composite, composite_char_row_next,
2294 composite_char_col_next);
2295 Fputhash (make_char (emch), lispstr,
2296 Vcomposite_char_char2string_hash_table);
2297 Fputhash (lispstr, make_char (emch),
2298 Vcomposite_char_string2char_hash_table);
2299 composite_char_col_next++;
2300 if (composite_char_col_next >= 128)
2302 composite_char_col_next = 32;
2303 composite_char_row_next++;
2312 composite_char_string (Emchar ch)
2314 Lisp_Object str = Fgethash (make_char (ch),
2315 Vcomposite_char_char2string_hash_table,
2317 assert (!UNBOUNDP (str));
2321 xxDEFUN ("make-composite-char", Fmake_composite_char, 1, 1, 0, /*
2322 Convert a string into a single composite character.
2323 The character is the result of overstriking all the characters in
2328 CHECK_STRING (string);
2329 return make_char (lookup_composite_char (XSTRING_DATA (string),
2330 XSTRING_LENGTH (string)));
2333 xxDEFUN ("composite-char-string", Fcomposite_char_string, 1, 1, 0, /*
2334 Return a string of the characters comprising a composite character.
2342 if (CHAR_LEADING_BYTE (emch) != LEADING_BYTE_COMPOSITE)
2343 signal_simple_error ("Must be composite char", ch);
2344 return composite_char_string (emch);
2346 #endif /* ENABLE_COMPOSITE_CHARS */
2349 /************************************************************************/
2350 /* initialization */
2351 /************************************************************************/
2354 syms_of_mule_charset (void)
2356 INIT_LRECORD_IMPLEMENTATION (charset);
2358 DEFSUBR (Fcharsetp);
2359 DEFSUBR (Ffind_charset);
2360 DEFSUBR (Fget_charset);
2361 DEFSUBR (Fcharset_list);
2362 DEFSUBR (Fcharset_name);
2363 DEFSUBR (Fmake_charset);
2364 DEFSUBR (Fmake_reverse_direction_charset);
2365 /* DEFSUBR (Freverse_direction_charset); */
2366 DEFSUBR (Fdefine_charset_alias);
2367 DEFSUBR (Fcharset_from_attributes);
2368 DEFSUBR (Fcharset_short_name);
2369 DEFSUBR (Fcharset_long_name);
2370 DEFSUBR (Fcharset_description);
2371 DEFSUBR (Fcharset_dimension);
2372 DEFSUBR (Fcharset_property);
2373 DEFSUBR (Fcharset_id);
2374 DEFSUBR (Fset_charset_ccl_program);
2375 DEFSUBR (Fset_charset_registry);
2377 DEFSUBR (Fcharset_mapping_table);
2378 DEFSUBR (Fset_charset_mapping_table);
2382 DEFSUBR (Fdecode_char);
2383 DEFSUBR (Fdecode_builtin_char);
2384 DEFSUBR (Fencode_char);
2386 DEFSUBR (Fmake_char);
2387 DEFSUBR (Fchar_charset);
2388 DEFSUBR (Fchar_octet);
2389 DEFSUBR (Fsplit_char);
2391 #ifdef ENABLE_COMPOSITE_CHARS
2392 DEFSUBR (Fmake_composite_char);
2393 DEFSUBR (Fcomposite_char_string);
2396 defsymbol (&Qcharsetp, "charsetp");
2397 defsymbol (&Qregistry, "registry");
2398 defsymbol (&Qfinal, "final");
2399 defsymbol (&Qgraphic, "graphic");
2400 defsymbol (&Qdirection, "direction");
2401 defsymbol (&Qreverse_direction_charset, "reverse-direction-charset");
2402 defsymbol (&Qshort_name, "short-name");
2403 defsymbol (&Qlong_name, "long-name");
2405 defsymbol (&Qmother, "mother");
2406 defsymbol (&Qmin_code, "min-code");
2407 defsymbol (&Qmax_code, "max-code");
2408 defsymbol (&Qcode_offset, "code-offset");
2409 defsymbol (&Qconversion, "conversion");
2410 defsymbol (&Q94x60, "94x60");
2411 defsymbol (&Q94x94x60, "94x94x60");
2414 defsymbol (&Ql2r, "l2r");
2415 defsymbol (&Qr2l, "r2l");
2417 /* Charsets, compatible with FSF 20.3
2418 Naming convention is Script-Charset[-Edition] */
2419 defsymbol (&Qascii, "ascii");
2420 defsymbol (&Qcontrol_1, "control-1");
2421 defsymbol (&Qlatin_iso8859_1, "latin-iso8859-1");
2422 defsymbol (&Qlatin_iso8859_2, "latin-iso8859-2");
2423 defsymbol (&Qlatin_iso8859_3, "latin-iso8859-3");
2424 defsymbol (&Qlatin_iso8859_4, "latin-iso8859-4");
2425 defsymbol (&Qthai_tis620, "thai-tis620");
2426 defsymbol (&Qgreek_iso8859_7, "greek-iso8859-7");
2427 defsymbol (&Qarabic_iso8859_6, "arabic-iso8859-6");
2428 defsymbol (&Qhebrew_iso8859_8, "hebrew-iso8859-8");
2429 defsymbol (&Qkatakana_jisx0201, "katakana-jisx0201");
2430 defsymbol (&Qlatin_jisx0201, "latin-jisx0201");
2431 defsymbol (&Qcyrillic_iso8859_5, "cyrillic-iso8859-5");
2432 defsymbol (&Qlatin_iso8859_9, "latin-iso8859-9");
2433 defsymbol (&Qjapanese_jisx0208_1978, "japanese-jisx0208-1978");
2434 defsymbol (&Qchinese_gb2312, "chinese-gb2312");
2435 defsymbol (&Qchinese_gb12345, "chinese-gb12345");
2436 defsymbol (&Qjapanese_jisx0208, "japanese-jisx0208");
2437 defsymbol (&Qjapanese_jisx0208_1990, "japanese-jisx0208-1990");
2438 defsymbol (&Qkorean_ksc5601, "korean-ksc5601");
2439 defsymbol (&Qjapanese_jisx0212, "japanese-jisx0212");
2440 defsymbol (&Qchinese_cns11643_1, "chinese-cns11643-1");
2441 defsymbol (&Qchinese_cns11643_2, "chinese-cns11643-2");
2443 defsymbol (&Qucs, "ucs");
2444 defsymbol (&Qucs_bmp, "ucs-bmp");
2445 defsymbol (&Qucs_smp, "ucs-smp");
2446 defsymbol (&Qucs_sip, "ucs-sip");
2447 defsymbol (&Qucs_cns, "ucs-cns");
2448 defsymbol (&Qucs_jis, "ucs-jis");
2449 defsymbol (&Qucs_ks, "ucs-ks");
2450 defsymbol (&Qucs_big5, "ucs-big5");
2451 defsymbol (&Qlatin_viscii, "latin-viscii");
2452 defsymbol (&Qlatin_tcvn5712, "latin-tcvn5712");
2453 defsymbol (&Qlatin_viscii_lower, "latin-viscii-lower");
2454 defsymbol (&Qlatin_viscii_upper, "latin-viscii-upper");
2455 defsymbol (&Qvietnamese_viscii_lower, "vietnamese-viscii-lower");
2456 defsymbol (&Qvietnamese_viscii_upper, "vietnamese-viscii-upper");
2457 defsymbol (&Qjis_x0208, "=jis-x0208");
2458 defsymbol (&Qideograph_gt, "ideograph-gt");
2459 defsymbol (&Qideograph_gt_pj_1, "ideograph-gt-pj-1");
2460 defsymbol (&Qideograph_gt_pj_2, "ideograph-gt-pj-2");
2461 defsymbol (&Qideograph_gt_pj_3, "ideograph-gt-pj-3");
2462 defsymbol (&Qideograph_gt_pj_4, "ideograph-gt-pj-4");
2463 defsymbol (&Qideograph_gt_pj_5, "ideograph-gt-pj-5");
2464 defsymbol (&Qideograph_gt_pj_6, "ideograph-gt-pj-6");
2465 defsymbol (&Qideograph_gt_pj_7, "ideograph-gt-pj-7");
2466 defsymbol (&Qideograph_gt_pj_8, "ideograph-gt-pj-8");
2467 defsymbol (&Qideograph_gt_pj_9, "ideograph-gt-pj-9");
2468 defsymbol (&Qideograph_gt_pj_10, "ideograph-gt-pj-10");
2469 defsymbol (&Qideograph_gt_pj_11, "ideograph-gt-pj-11");
2470 defsymbol (&Qideograph_daikanwa_2, "ideograph-daikanwa-2");
2471 defsymbol (&Qideograph_daikanwa, "ideograph-daikanwa");
2472 defsymbol (&Qchinese_big5, "chinese-big5");
2473 /* defsymbol (&Qchinese_big5_cdp, "chinese-big5-cdp"); */
2474 defsymbol (&Qideograph_hanziku_1, "ideograph-hanziku-1");
2475 defsymbol (&Qideograph_hanziku_2, "ideograph-hanziku-2");
2476 defsymbol (&Qideograph_hanziku_3, "ideograph-hanziku-3");
2477 defsymbol (&Qideograph_hanziku_4, "ideograph-hanziku-4");
2478 defsymbol (&Qideograph_hanziku_5, "ideograph-hanziku-5");
2479 defsymbol (&Qideograph_hanziku_6, "ideograph-hanziku-6");
2480 defsymbol (&Qideograph_hanziku_7, "ideograph-hanziku-7");
2481 defsymbol (&Qideograph_hanziku_8, "ideograph-hanziku-8");
2482 defsymbol (&Qideograph_hanziku_9, "ideograph-hanziku-9");
2483 defsymbol (&Qideograph_hanziku_10, "ideograph-hanziku-10");
2484 defsymbol (&Qideograph_hanziku_11, "ideograph-hanziku-11");
2485 defsymbol (&Qideograph_hanziku_12, "ideograph-hanziku-12");
2486 defsymbol (&Qchina3_jef, "china3-jef");
2487 defsymbol (&Qideograph_cbeta, "ideograph-cbeta");
2488 defsymbol (&Qethiopic_ucs, "ethiopic-ucs");
2490 defsymbol (&Qchinese_big5_1, "chinese-big5-1");
2491 defsymbol (&Qchinese_big5_2, "chinese-big5-2");
2493 defsymbol (&Qcomposite, "composite");
2497 vars_of_mule_charset (void)
2504 chlook = xnew_and_zero (struct charset_lookup); /* zero for Purify. */
2505 dump_add_root_struct_ptr (&chlook, &charset_lookup_description);
2507 /* Table of charsets indexed by leading byte. */
2508 for (i = 0; i < countof (chlook->charset_by_leading_byte); i++)
2509 chlook->charset_by_leading_byte[i] = Qnil;
2512 /* Table of charsets indexed by type/final-byte. */
2513 for (i = 0; i < countof (chlook->charset_by_attributes); i++)
2514 for (j = 0; j < countof (chlook->charset_by_attributes[0]); j++)
2515 chlook->charset_by_attributes[i][j] = Qnil;
2517 /* Table of charsets indexed by type/final-byte/direction. */
2518 for (i = 0; i < countof (chlook->charset_by_attributes); i++)
2519 for (j = 0; j < countof (chlook->charset_by_attributes[0]); j++)
2520 for (k = 0; k < countof (chlook->charset_by_attributes[0][0]); k++)
2521 chlook->charset_by_attributes[i][j][k] = Qnil;
2525 chlook->next_allocated_leading_byte = MIN_LEADING_BYTE_PRIVATE;
2527 chlook->next_allocated_1_byte_leading_byte = MIN_LEADING_BYTE_PRIVATE_1;
2528 chlook->next_allocated_2_byte_leading_byte = MIN_LEADING_BYTE_PRIVATE_2;
2532 leading_code_private_11 = PRE_LEADING_BYTE_PRIVATE_1;
2533 DEFVAR_INT ("leading-code-private-11", &leading_code_private_11 /*
2534 Leading-code of private TYPE9N charset of column-width 1.
2536 leading_code_private_11 = PRE_LEADING_BYTE_PRIVATE_1;
2540 Vdefault_coded_charset_priority_list = Qnil;
2541 DEFVAR_LISP ("default-coded-charset-priority-list",
2542 &Vdefault_coded_charset_priority_list /*
2543 Default order of preferred coded-character-sets.
2549 complex_vars_of_mule_charset (void)
2551 staticpro (&Vcharset_hash_table);
2552 Vcharset_hash_table =
2553 make_lisp_hash_table (50, HASH_TABLE_NON_WEAK, HASH_TABLE_EQ);
2555 /* Predefined character sets. We store them into variables for
2559 staticpro (&Vcharset_ucs);
2561 make_charset (LEADING_BYTE_UCS, Qucs, 256, 4,
2562 1, 2, 0, CHARSET_LEFT_TO_RIGHT,
2563 build_string ("UCS"),
2564 build_string ("UCS"),
2565 build_string ("ISO/IEC 10646"),
2567 Qnil, 0, 0x7FFFFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL);
2568 staticpro (&Vcharset_ucs_bmp);
2570 make_charset (LEADING_BYTE_UCS_BMP, Qucs_bmp, 256, 2,
2571 1, 2, 0, CHARSET_LEFT_TO_RIGHT,
2572 build_string ("BMP"),
2573 build_string ("UCS-BMP"),
2574 build_string ("ISO/IEC 10646 Group 0 Plane 0 (BMP)"),
2576 ("\\(ISO10646.*-[01]\\|UCS00-0\\|UNICODE[23]?-0\\)"),
2577 Qnil, 0, 0xFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL);
2578 staticpro (&Vcharset_ucs_smp);
2580 make_charset (LEADING_BYTE_UCS_SMP, Qucs_smp, 256, 2,
2581 1, 2, 0, CHARSET_LEFT_TO_RIGHT,
2582 build_string ("SMP"),
2583 build_string ("UCS-SMP"),
2584 build_string ("ISO/IEC 10646 Group 0 Plane 1 (SMP)"),
2585 build_string ("UCS00-1"),
2586 Qnil, MIN_CHAR_SMP, MAX_CHAR_SMP,
2587 MIN_CHAR_SMP, 0, Qnil, CONVERSION_IDENTICAL);
2588 staticpro (&Vcharset_ucs_sip);
2590 make_charset (LEADING_BYTE_UCS_SIP, Qucs_sip, 256, 2,
2591 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2592 build_string ("SIP"),
2593 build_string ("UCS-SIP"),
2594 build_string ("ISO/IEC 10646 Group 0 Plane 2 (SIP)"),
2595 build_string ("\\(ISO10646.*-2\\|UCS00-2\\)"),
2596 Qnil, MIN_CHAR_SIP, MAX_CHAR_SIP,
2597 MIN_CHAR_SIP, 0, Qnil, CONVERSION_IDENTICAL);
2598 staticpro (&Vcharset_ucs_cns);
2600 make_charset (LEADING_BYTE_UCS_CNS, Qucs_cns, 256, 3,
2601 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2602 build_string ("UCS for CNS"),
2603 build_string ("UCS for CNS 11643"),
2604 build_string ("ISO/IEC 10646 for CNS 11643"),
2606 Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
2607 staticpro (&Vcharset_ucs_jis);
2609 make_charset (LEADING_BYTE_UCS_JIS, Qucs_jis, 256, 3,
2610 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2611 build_string ("UCS for JIS"),
2612 build_string ("UCS for JIS X 0208, 0212 and 0213"),
2614 ("ISO/IEC 10646 for JIS X 0208, 0212 and 0213"),
2616 Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
2617 staticpro (&Vcharset_ucs_ks);
2619 make_charset (LEADING_BYTE_UCS_KS, Qucs_ks, 256, 3,
2620 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2621 build_string ("UCS for KS"),
2622 build_string ("UCS for CCS defined by KS"),
2623 build_string ("ISO/IEC 10646 for Korean Standards"),
2625 Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
2626 staticpro (&Vcharset_ucs_big5);
2628 make_charset (LEADING_BYTE_UCS_BIG5, Qucs_big5, 256, 3,
2629 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2630 build_string ("UCS for Big5"),
2631 build_string ("UCS for Big5"),
2632 build_string ("ISO/IEC 10646 for Big5"),
2634 Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
2636 # define MIN_CHAR_THAI 0
2637 # define MAX_CHAR_THAI 0
2638 /* # define MIN_CHAR_HEBREW 0 */
2639 /* # define MAX_CHAR_HEBREW 0 */
2640 # define MIN_CHAR_HALFWIDTH_KATAKANA 0
2641 # define MAX_CHAR_HALFWIDTH_KATAKANA 0
2643 staticpro (&Vcharset_ascii);
2645 make_charset (LEADING_BYTE_ASCII, Qascii, 94, 1,
2646 1, 0, 'B', CHARSET_LEFT_TO_RIGHT,
2647 build_string ("ASCII"),
2648 build_string ("ASCII)"),
2649 build_string ("ASCII (ISO646 IRV)"),
2650 build_string ("\\(iso8859-[0-9]*\\|-ascii\\)"),
2651 Qnil, 0, 0x7F, 0, 0, Qnil, CONVERSION_IDENTICAL);
2652 staticpro (&Vcharset_control_1);
2653 Vcharset_control_1 =
2654 make_charset (LEADING_BYTE_CONTROL_1, Qcontrol_1, 94, 1,
2655 1, 1, 0, CHARSET_LEFT_TO_RIGHT,
2656 build_string ("C1"),
2657 build_string ("Control characters"),
2658 build_string ("Control characters 128-191"),
2660 Qnil, 0x80, 0x9F, 0x80, 0, Qnil, CONVERSION_IDENTICAL);
2661 staticpro (&Vcharset_latin_iso8859_1);
2662 Vcharset_latin_iso8859_1 =
2663 make_charset (LEADING_BYTE_LATIN_ISO8859_1, Qlatin_iso8859_1, 96, 1,
2664 1, 1, 'A', CHARSET_LEFT_TO_RIGHT,
2665 build_string ("Latin-1"),
2666 build_string ("ISO8859-1 (Latin-1)"),
2667 build_string ("ISO8859-1 (Latin-1)"),
2668 build_string ("iso8859-1"),
2669 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2670 staticpro (&Vcharset_latin_iso8859_2);
2671 Vcharset_latin_iso8859_2 =
2672 make_charset (LEADING_BYTE_LATIN_ISO8859_2, Qlatin_iso8859_2, 96, 1,
2673 1, 1, 'B', CHARSET_LEFT_TO_RIGHT,
2674 build_string ("Latin-2"),
2675 build_string ("ISO8859-2 (Latin-2)"),
2676 build_string ("ISO8859-2 (Latin-2)"),
2677 build_string ("iso8859-2"),
2678 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2679 staticpro (&Vcharset_latin_iso8859_3);
2680 Vcharset_latin_iso8859_3 =
2681 make_charset (LEADING_BYTE_LATIN_ISO8859_3, Qlatin_iso8859_3, 96, 1,
2682 1, 1, 'C', CHARSET_LEFT_TO_RIGHT,
2683 build_string ("Latin-3"),
2684 build_string ("ISO8859-3 (Latin-3)"),
2685 build_string ("ISO8859-3 (Latin-3)"),
2686 build_string ("iso8859-3"),
2687 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2688 staticpro (&Vcharset_latin_iso8859_4);
2689 Vcharset_latin_iso8859_4 =
2690 make_charset (LEADING_BYTE_LATIN_ISO8859_4, Qlatin_iso8859_4, 96, 1,
2691 1, 1, 'D', CHARSET_LEFT_TO_RIGHT,
2692 build_string ("Latin-4"),
2693 build_string ("ISO8859-4 (Latin-4)"),
2694 build_string ("ISO8859-4 (Latin-4)"),
2695 build_string ("iso8859-4"),
2696 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2697 staticpro (&Vcharset_thai_tis620);
2698 Vcharset_thai_tis620 =
2699 make_charset (LEADING_BYTE_THAI_TIS620, Qthai_tis620, 96, 1,
2700 1, 1, 'T', CHARSET_LEFT_TO_RIGHT,
2701 build_string ("TIS620"),
2702 build_string ("TIS620 (Thai)"),
2703 build_string ("TIS620.2529 (Thai)"),
2704 build_string ("tis620"),
2705 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2706 staticpro (&Vcharset_greek_iso8859_7);
2707 Vcharset_greek_iso8859_7 =
2708 make_charset (LEADING_BYTE_GREEK_ISO8859_7, Qgreek_iso8859_7, 96, 1,
2709 1, 1, 'F', CHARSET_LEFT_TO_RIGHT,
2710 build_string ("ISO8859-7"),
2711 build_string ("ISO8859-7 (Greek)"),
2712 build_string ("ISO8859-7 (Greek)"),
2713 build_string ("iso8859-7"),
2714 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2715 staticpro (&Vcharset_arabic_iso8859_6);
2716 Vcharset_arabic_iso8859_6 =
2717 make_charset (LEADING_BYTE_ARABIC_ISO8859_6, Qarabic_iso8859_6, 96, 1,
2718 1, 1, 'G', CHARSET_RIGHT_TO_LEFT,
2719 build_string ("ISO8859-6"),
2720 build_string ("ISO8859-6 (Arabic)"),
2721 build_string ("ISO8859-6 (Arabic)"),
2722 build_string ("iso8859-6"),
2723 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2724 staticpro (&Vcharset_hebrew_iso8859_8);
2725 Vcharset_hebrew_iso8859_8 =
2726 make_charset (LEADING_BYTE_HEBREW_ISO8859_8, Qhebrew_iso8859_8, 96, 1,
2727 1, 1, 'H', CHARSET_RIGHT_TO_LEFT,
2728 build_string ("ISO8859-8"),
2729 build_string ("ISO8859-8 (Hebrew)"),
2730 build_string ("ISO8859-8 (Hebrew)"),
2731 build_string ("iso8859-8"),
2733 0 /* MIN_CHAR_HEBREW */,
2734 0 /* MAX_CHAR_HEBREW */, 0, 32,
2735 Qnil, CONVERSION_IDENTICAL);
2736 staticpro (&Vcharset_katakana_jisx0201);
2737 Vcharset_katakana_jisx0201 =
2738 make_charset (LEADING_BYTE_KATAKANA_JISX0201, Qkatakana_jisx0201, 94, 1,
2739 1, 1, 'I', CHARSET_LEFT_TO_RIGHT,
2740 build_string ("JISX0201 Kana"),
2741 build_string ("JISX0201.1976 (Japanese Kana)"),
2742 build_string ("JISX0201.1976 Japanese Kana"),
2743 build_string ("jisx0201\\.1976"),
2744 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2745 staticpro (&Vcharset_latin_jisx0201);
2746 Vcharset_latin_jisx0201 =
2747 make_charset (LEADING_BYTE_LATIN_JISX0201, Qlatin_jisx0201, 94, 1,
2748 1, 0, 'J', CHARSET_LEFT_TO_RIGHT,
2749 build_string ("JISX0201 Roman"),
2750 build_string ("JISX0201.1976 (Japanese Roman)"),
2751 build_string ("JISX0201.1976 Japanese Roman"),
2752 build_string ("jisx0201\\.1976"),
2753 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2754 staticpro (&Vcharset_cyrillic_iso8859_5);
2755 Vcharset_cyrillic_iso8859_5 =
2756 make_charset (LEADING_BYTE_CYRILLIC_ISO8859_5, Qcyrillic_iso8859_5, 96, 1,
2757 1, 1, 'L', CHARSET_LEFT_TO_RIGHT,
2758 build_string ("ISO8859-5"),
2759 build_string ("ISO8859-5 (Cyrillic)"),
2760 build_string ("ISO8859-5 (Cyrillic)"),
2761 build_string ("iso8859-5"),
2762 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2763 staticpro (&Vcharset_latin_iso8859_9);
2764 Vcharset_latin_iso8859_9 =
2765 make_charset (LEADING_BYTE_LATIN_ISO8859_9, Qlatin_iso8859_9, 96, 1,
2766 1, 1, 'M', CHARSET_LEFT_TO_RIGHT,
2767 build_string ("Latin-5"),
2768 build_string ("ISO8859-9 (Latin-5)"),
2769 build_string ("ISO8859-9 (Latin-5)"),
2770 build_string ("iso8859-9"),
2771 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2773 staticpro (&Vcharset_jis_x0208);
2774 Vcharset_jis_x0208 =
2775 make_charset (LEADING_BYTE_JIS_X0208,
2777 2, 0, 'B', CHARSET_LEFT_TO_RIGHT,
2778 build_string ("JIS X0208"),
2779 build_string ("JIS X0208 Common"),
2780 build_string ("JIS X0208 Common part"),
2781 build_string ("jisx0208\\.1990"),
2783 MIN_CHAR_JIS_X0208_1990,
2784 MAX_CHAR_JIS_X0208_1990, MIN_CHAR_JIS_X0208_1990, 33,
2785 Qnil, CONVERSION_94x94);
2787 staticpro (&Vcharset_japanese_jisx0208_1978);
2788 Vcharset_japanese_jisx0208_1978 =
2789 make_charset (LEADING_BYTE_JAPANESE_JISX0208_1978,
2790 Qjapanese_jisx0208_1978, 94, 2,
2791 2, 0, '@', CHARSET_LEFT_TO_RIGHT,
2792 build_string ("JIS X0208:1978"),
2793 build_string ("JIS X0208:1978 (Japanese)"),
2795 ("JIS X0208:1978 Japanese Kanji (so called \"old JIS\")"),
2796 build_string ("\\(jisx0208\\|jisc6226\\)\\.1978"),
2803 CONVERSION_IDENTICAL);
2804 staticpro (&Vcharset_chinese_gb2312);
2805 Vcharset_chinese_gb2312 =
2806 make_charset (LEADING_BYTE_CHINESE_GB2312, Qchinese_gb2312, 94, 2,
2807 2, 0, 'A', CHARSET_LEFT_TO_RIGHT,
2808 build_string ("GB2312"),
2809 build_string ("GB2312)"),
2810 build_string ("GB2312 Chinese simplified"),
2811 build_string ("gb2312"),
2812 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2813 staticpro (&Vcharset_chinese_gb12345);
2814 Vcharset_chinese_gb12345 =
2815 make_charset (LEADING_BYTE_CHINESE_GB12345, Qchinese_gb12345, 94, 2,
2816 2, 0, 0, CHARSET_LEFT_TO_RIGHT,
2817 build_string ("G1"),
2818 build_string ("GB 12345)"),
2819 build_string ("GB 12345-1990"),
2820 build_string ("GB12345\\(\\.1990\\)?-0"),
2821 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2822 staticpro (&Vcharset_japanese_jisx0208);
2823 Vcharset_japanese_jisx0208 =
2824 make_charset (LEADING_BYTE_JAPANESE_JISX0208, Qjapanese_jisx0208, 94, 2,
2825 2, 0, 'B', CHARSET_LEFT_TO_RIGHT,
2826 build_string ("JISX0208"),
2827 build_string ("JIS X0208:1983 (Japanese)"),
2828 build_string ("JIS X0208:1983 Japanese Kanji"),
2829 build_string ("jisx0208\\.1983"),
2836 CONVERSION_IDENTICAL);
2838 staticpro (&Vcharset_japanese_jisx0208_1990);
2839 Vcharset_japanese_jisx0208_1990 =
2840 make_charset (LEADING_BYTE_JAPANESE_JISX0208_1990,
2841 Qjapanese_jisx0208_1990, 94, 2,
2842 2, 0, 0, CHARSET_LEFT_TO_RIGHT,
2843 build_string ("JISX0208-1990"),
2844 build_string ("JIS X0208:1990 (Japanese)"),
2845 build_string ("JIS X0208:1990 Japanese Kanji"),
2846 build_string ("jisx0208\\.1990"),
2848 0x2121 /* MIN_CHAR_JIS_X0208_1990 */,
2849 0x7426 /* MAX_CHAR_JIS_X0208_1990 */,
2850 0 /* MIN_CHAR_JIS_X0208_1990 */, 33,
2851 Vcharset_jis_x0208 /* Qnil */,
2852 CONVERSION_IDENTICAL /* CONVERSION_94x94 */);
2854 staticpro (&Vcharset_korean_ksc5601);
2855 Vcharset_korean_ksc5601 =
2856 make_charset (LEADING_BYTE_KOREAN_KSC5601, Qkorean_ksc5601, 94, 2,
2857 2, 0, 'C', CHARSET_LEFT_TO_RIGHT,
2858 build_string ("KSC5601"),
2859 build_string ("KSC5601 (Korean"),
2860 build_string ("KSC5601 Korean Hangul and Hanja"),
2861 build_string ("ksc5601"),
2862 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2863 staticpro (&Vcharset_japanese_jisx0212);
2864 Vcharset_japanese_jisx0212 =
2865 make_charset (LEADING_BYTE_JAPANESE_JISX0212, Qjapanese_jisx0212, 94, 2,
2866 2, 0, 'D', CHARSET_LEFT_TO_RIGHT,
2867 build_string ("JISX0212"),
2868 build_string ("JISX0212 (Japanese)"),
2869 build_string ("JISX0212 Japanese Supplement"),
2870 build_string ("jisx0212"),
2871 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2873 #define CHINESE_CNS_PLANE_RE(n) "cns11643[.-]\\(.*[.-]\\)?" n "$"
2874 staticpro (&Vcharset_chinese_cns11643_1);
2875 Vcharset_chinese_cns11643_1 =
2876 make_charset (LEADING_BYTE_CHINESE_CNS11643_1, Qchinese_cns11643_1, 94, 2,
2877 2, 0, 'G', CHARSET_LEFT_TO_RIGHT,
2878 build_string ("CNS11643-1"),
2879 build_string ("CNS11643-1 (Chinese traditional)"),
2881 ("CNS 11643 Plane 1 Chinese traditional"),
2882 build_string (CHINESE_CNS_PLANE_RE("1")),
2883 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2884 staticpro (&Vcharset_chinese_cns11643_2);
2885 Vcharset_chinese_cns11643_2 =
2886 make_charset (LEADING_BYTE_CHINESE_CNS11643_2, Qchinese_cns11643_2, 94, 2,
2887 2, 0, 'H', CHARSET_LEFT_TO_RIGHT,
2888 build_string ("CNS11643-2"),
2889 build_string ("CNS11643-2 (Chinese traditional)"),
2891 ("CNS 11643 Plane 2 Chinese traditional"),
2892 build_string (CHINESE_CNS_PLANE_RE("2")),
2893 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
2895 staticpro (&Vcharset_latin_tcvn5712);
2896 Vcharset_latin_tcvn5712 =
2897 make_charset (LEADING_BYTE_LATIN_TCVN5712, Qlatin_tcvn5712, 96, 1,
2898 1, 1, 'Z', CHARSET_LEFT_TO_RIGHT,
2899 build_string ("TCVN 5712"),
2900 build_string ("TCVN 5712 (VSCII-2)"),
2901 build_string ("Vietnamese TCVN 5712:1983 (VSCII-2)"),
2902 build_string ("tcvn5712\\(\\.1993\\)?-1"),
2903 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2904 staticpro (&Vcharset_latin_viscii_lower);
2905 Vcharset_latin_viscii_lower =
2906 make_charset (LEADING_BYTE_LATIN_VISCII_LOWER, Qlatin_viscii_lower, 96, 1,
2907 1, 1, '1', CHARSET_LEFT_TO_RIGHT,
2908 build_string ("VISCII lower"),
2909 build_string ("VISCII lower (Vietnamese)"),
2910 build_string ("VISCII lower (Vietnamese)"),
2911 build_string ("MULEVISCII-LOWER"),
2912 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2913 staticpro (&Vcharset_latin_viscii_upper);
2914 Vcharset_latin_viscii_upper =
2915 make_charset (LEADING_BYTE_LATIN_VISCII_UPPER, Qlatin_viscii_upper, 96, 1,
2916 1, 1, '2', CHARSET_LEFT_TO_RIGHT,
2917 build_string ("VISCII upper"),
2918 build_string ("VISCII upper (Vietnamese)"),
2919 build_string ("VISCII upper (Vietnamese)"),
2920 build_string ("MULEVISCII-UPPER"),
2921 Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL);
2922 staticpro (&Vcharset_latin_viscii);
2923 Vcharset_latin_viscii =
2924 make_charset (LEADING_BYTE_LATIN_VISCII, Qlatin_viscii, 256, 1,
2925 1, 2, 0, CHARSET_LEFT_TO_RIGHT,
2926 build_string ("VISCII"),
2927 build_string ("VISCII 1.1 (Vietnamese)"),
2928 build_string ("VISCII 1.1 (Vietnamese)"),
2929 build_string ("VISCII1\\.1"),
2930 Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL);
2931 staticpro (&Vcharset_chinese_big5);
2932 Vcharset_chinese_big5 =
2933 make_charset (LEADING_BYTE_CHINESE_BIG5, Qchinese_big5, 256, 2,
2934 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2935 build_string ("Big5"),
2936 build_string ("Big5"),
2937 build_string ("Big5 Chinese traditional"),
2938 build_string ("big5-0"),
2940 MIN_CHAR_BIG5_CDP, MAX_CHAR_BIG5_CDP,
2941 MIN_CHAR_BIG5_CDP, 0, Qnil, CONVERSION_IDENTICAL);
2943 staticpro (&Vcharset_chinese_big5_cdp);
2944 Vcharset_chinese_big5_cdp =
2945 make_charset (LEADING_BYTE_CHINESE_BIG5_CDP, Qchinese_big5_cdp, 256, 2,
2946 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2947 build_string ("Big5-CDP"),
2948 build_string ("Big5 + CDP extension"),
2949 build_string ("Big5 with CDP extension"),
2950 build_string ("big5\\.cdp-0"),
2951 Qnil, MIN_CHAR_BIG5_CDP, MAX_CHAR_BIG5_CDP,
2952 MIN_CHAR_BIG5_CDP, 0, Qnil, CONVERSION_IDENTICAL);
2954 #define DEF_HANZIKU(n) \
2955 staticpro (&Vcharset_ideograph_hanziku_##n); \
2956 Vcharset_ideograph_hanziku_##n = \
2957 make_charset (LEADING_BYTE_HANZIKU_##n, Qideograph_hanziku_##n, 256, 2, \
2958 2, 2, 0, CHARSET_LEFT_TO_RIGHT, \
2959 build_string ("HZK-"#n), \
2960 build_string ("HANZIKU-"#n), \
2961 build_string ("HANZIKU (pseudo BIG5 encoding) part "#n), \
2963 ("hanziku-"#n"$"), \
2964 Qnil, MIN_CHAR_HANZIKU_##n, MAX_CHAR_HANZIKU_##n, \
2965 MIN_CHAR_HANZIKU_##n, 0, Qnil, CONVERSION_IDENTICAL);
2978 staticpro (&Vcharset_china3_jef);
2979 Vcharset_china3_jef =
2980 make_charset (LEADING_BYTE_CHINA3_JEF, Qchina3_jef, 256, 2,
2981 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2982 build_string ("JC3"),
2983 build_string ("JEF + CHINA3"),
2984 build_string ("JEF + CHINA3 private characters"),
2985 build_string ("china3jef-0"),
2986 Qnil, MIN_CHAR_CHINA3_JEF, MAX_CHAR_CHINA3_JEF,
2987 MIN_CHAR_CHINA3_JEF, 0, Qnil, CONVERSION_IDENTICAL);
2988 staticpro (&Vcharset_ideograph_cbeta);
2989 Vcharset_ideograph_cbeta =
2990 make_charset (LEADING_BYTE_CBETA, Qideograph_cbeta, 256, 2,
2991 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
2992 build_string ("CB"),
2993 build_string ("CBETA"),
2994 build_string ("CBETA private characters"),
2995 build_string ("cbeta-0"),
2996 Qnil, MIN_CHAR_CBETA, MAX_CHAR_CBETA,
2997 MIN_CHAR_CBETA, 0, Qnil, CONVERSION_IDENTICAL);
2998 staticpro (&Vcharset_ideograph_gt);
2999 Vcharset_ideograph_gt =
3000 make_charset (LEADING_BYTE_GT, Qideograph_gt, 256, 3,
3001 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
3002 build_string ("GT"),
3003 build_string ("GT"),
3004 build_string ("GT"),
3006 Qnil, MIN_CHAR_GT, MAX_CHAR_GT,
3007 MIN_CHAR_GT, 0, Qnil, CONVERSION_IDENTICAL);
3008 #define DEF_GT_PJ(n) \
3009 staticpro (&Vcharset_ideograph_gt_pj_##n); \
3010 Vcharset_ideograph_gt_pj_##n = \
3011 make_charset (LEADING_BYTE_GT_PJ_##n, Qideograph_gt_pj_##n, 94, 2, \
3012 2, 0, 0, CHARSET_LEFT_TO_RIGHT, \
3013 build_string ("GT-PJ-"#n), \
3014 build_string ("GT (pseudo JIS encoding) part "#n), \
3015 build_string ("GT 2000 (pseudo JIS encoding) part "#n), \
3017 ("\\(GTpj-"#n "\\|jisx0208\\.GT-"#n "\\)$"), \
3018 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
3031 staticpro (&Vcharset_ideograph_daikanwa_2);
3032 Vcharset_ideograph_daikanwa_2 =
3033 make_charset (LEADING_BYTE_DAIKANWA_2, Qideograph_daikanwa_2, 256, 2,
3034 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
3035 build_string ("Daikanwa Rev."),
3036 build_string ("Morohashi's Daikanwa Rev."),
3038 ("Daikanwa dictionary (revised version)"),
3039 build_string ("Daikanwa\\(\\.[0-9]+\\)?-2"),
3040 Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL);
3041 staticpro (&Vcharset_ideograph_daikanwa);
3042 Vcharset_ideograph_daikanwa =
3043 make_charset (LEADING_BYTE_DAIKANWA_3, Qideograph_daikanwa, 256, 2,
3044 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
3045 build_string ("Daikanwa"),
3046 build_string ("Morohashi's Daikanwa Rev.2"),
3048 ("Daikanwa dictionary (second revised version)"),
3049 build_string ("Daikanwa\\(\\.[0-9]+\\)?-3"),
3050 Qnil, MIN_CHAR_DAIKANWA, MAX_CHAR_DAIKANWA,
3051 MIN_CHAR_DAIKANWA, 0, Qnil, CONVERSION_IDENTICAL);
3053 staticpro (&Vcharset_ethiopic_ucs);
3054 Vcharset_ethiopic_ucs =
3055 make_charset (LEADING_BYTE_ETHIOPIC_UCS, Qethiopic_ucs, 256, 2,
3056 2, 2, 0, CHARSET_LEFT_TO_RIGHT,
3057 build_string ("Ethiopic (UCS)"),
3058 build_string ("Ethiopic (UCS)"),
3059 build_string ("Ethiopic of UCS"),
3060 build_string ("Ethiopic-Unicode"),
3061 Qnil, 0x1200, 0x137F, 0, 0,
3062 Qnil, CONVERSION_IDENTICAL);
3064 staticpro (&Vcharset_chinese_big5_1);
3065 Vcharset_chinese_big5_1 =
3066 make_charset (LEADING_BYTE_CHINESE_BIG5_1, Qchinese_big5_1, 94, 2,
3067 2, 0, '0', CHARSET_LEFT_TO_RIGHT,
3068 build_string ("Big5"),
3069 build_string ("Big5 (Level-1)"),
3071 ("Big5 Level-1 Chinese traditional"),
3072 build_string ("big5"),
3073 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
3074 staticpro (&Vcharset_chinese_big5_2);
3075 Vcharset_chinese_big5_2 =
3076 make_charset (LEADING_BYTE_CHINESE_BIG5_2, Qchinese_big5_2, 94, 2,
3077 2, 0, '1', CHARSET_LEFT_TO_RIGHT,
3078 build_string ("Big5"),
3079 build_string ("Big5 (Level-2)"),
3081 ("Big5 Level-2 Chinese traditional"),
3082 build_string ("big5"),
3083 Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
3085 #ifdef ENABLE_COMPOSITE_CHARS
3086 /* #### For simplicity, we put composite chars into a 96x96 charset.
3087 This is going to lead to problems because you can run out of
3088 room, esp. as we don't yet recycle numbers. */
3089 staticpro (&Vcharset_composite);
3090 Vcharset_composite =
3091 make_charset (LEADING_BYTE_COMPOSITE, Qcomposite, 96, 2,
3092 2, 0, 0, CHARSET_LEFT_TO_RIGHT,
3093 build_string ("Composite"),
3094 build_string ("Composite characters"),
3095 build_string ("Composite characters"),
3098 /* #### not dumped properly */
3099 composite_char_row_next = 32;
3100 composite_char_col_next = 32;
3102 Vcomposite_char_string2char_hash_table =
3103 make_lisp_hash_table (500, HASH_TABLE_NON_WEAK, HASH_TABLE_EQUAL);
3104 Vcomposite_char_char2string_hash_table =
3105 make_lisp_hash_table (500, HASH_TABLE_NON_WEAK, HASH_TABLE_EQ);
3106 staticpro (&Vcomposite_char_string2char_hash_table);
3107 staticpro (&Vcomposite_char_char2string_hash_table);
3108 #endif /* ENABLE_COMPOSITE_CHARS */