1 /* XEmacs routines to deal with char tables.
2 Copyright (C) 1992, 1995 Free Software Foundation, Inc.
3 Copyright (C) 1995 Sun Microsystems, Inc.
4 Copyright (C) 1995, 1996 Ben Wing.
5 Copyright (C) 1995, 1997, 1999 Electrotechnical Laboratory, JAPAN.
6 Licensed to the Free Software Foundation.
8 This file is part of XEmacs.
10 XEmacs is free software; you can redistribute it and/or modify it
11 under the terms of the GNU General Public License as published by the
12 Free Software Foundation; either version 2, or (at your option) any
15 XEmacs is distributed in the hope that it will be useful, but WITHOUT
16 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
17 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
20 You should have received a copy of the GNU General Public License
21 along with XEmacs; see the file COPYING. If not, write to
22 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 Boston, MA 02111-1307, USA. */
25 /* Synched up with: Mule 2.3. Not synched with FSF.
27 This file was written independently of the FSF implementation,
28 and is not compatible. */
32 Ben Wing: wrote, for 19.13 (Mule). Some category table stuff
33 loosely based on the original Mule.
34 Jareth Hein: fixed a couple of bugs in the implementation, and
35 added regex support for categories with check_category_at
45 Lisp_Object Qchar_tablep, Qchar_table;
47 Lisp_Object Vall_syntax_tables;
50 Lisp_Object Qcategory_table_p;
51 Lisp_Object Qcategory_designator_p;
52 Lisp_Object Qcategory_table_value_p;
54 Lisp_Object Vstandard_category_table;
56 /* Variables to determine word boundary. */
57 Lisp_Object Vword_combining_categories, Vword_separating_categories;
61 /* A char table maps from ranges of characters to values.
63 Implementing a general data structure that maps from arbitrary
64 ranges of numbers to values is tricky to do efficiently. As it
65 happens, it should suffice (and is usually more convenient, anyway)
66 when dealing with characters to restrict the sorts of ranges that
67 can be assigned values, as follows:
70 2) All characters in a charset.
71 3) All characters in a particular row of a charset, where a "row"
72 means all characters with the same first byte.
73 4) A particular character in a charset.
75 We use char tables to generalize the 256-element vectors now
76 littering the Emacs code.
78 Possible uses (all should be converted at some point):
84 5) keyboard-translate-table?
87 abstract type to generalize the Emacs vectors and Mule
88 vectors-of-vectors goo.
91 /************************************************************************/
92 /* Char Table object */
93 /************************************************************************/
98 mark_char_table_entry (Lisp_Object obj, void (*markobj) (Lisp_Object))
100 struct Lisp_Char_Table_Entry *cte = XCHAR_TABLE_ENTRY (obj);
103 for (i = 0; i < 96; i++)
105 markobj (cte->level2[i]);
111 char_table_entry_equal (Lisp_Object obj1, Lisp_Object obj2, int depth)
113 struct Lisp_Char_Table_Entry *cte1 = XCHAR_TABLE_ENTRY (obj1);
114 struct Lisp_Char_Table_Entry *cte2 = XCHAR_TABLE_ENTRY (obj2);
117 for (i = 0; i < 96; i++)
118 if (!internal_equal (cte1->level2[i], cte2->level2[i], depth + 1))
125 char_table_entry_hash (Lisp_Object obj, int depth)
127 struct Lisp_Char_Table_Entry *cte = XCHAR_TABLE_ENTRY (obj);
129 return internal_array_hash (cte->level2, 96, depth);
132 static const struct lrecord_description char_table_entry_description[] = {
133 { XD_LISP_OBJECT, offsetof(struct Lisp_Char_Table_Entry, level2), 96 },
137 DEFINE_LRECORD_IMPLEMENTATION ("char-table-entry", char_table_entry,
138 mark_char_table_entry, internal_object_printer,
139 0, char_table_entry_equal,
140 char_table_entry_hash,
141 char_table_entry_description,
142 struct Lisp_Char_Table_Entry);
146 mark_char_table (Lisp_Object obj, void (*markobj) (Lisp_Object))
148 struct Lisp_Char_Table *ct = XCHAR_TABLE (obj);
151 for (i = 0; i < NUM_ASCII_CHARS; i++)
152 markobj (ct->ascii[i]);
154 for (i = 0; i < NUM_LEADING_BYTES; i++)
155 markobj (ct->level1[i]);
157 return ct->mirror_table;
160 /* WARNING: All functions of this nature need to be written extremely
161 carefully to avoid crashes during GC. Cf. prune_specifiers()
162 and prune_weak_hash_tables(). */
165 prune_syntax_tables (int (*obj_marked_p) (Lisp_Object))
167 Lisp_Object rest, prev = Qnil;
169 for (rest = Vall_syntax_tables;
171 rest = XCHAR_TABLE (rest)->next_table)
173 if (! obj_marked_p (rest))
175 /* This table is garbage. Remove it from the list. */
177 Vall_syntax_tables = XCHAR_TABLE (rest)->next_table;
179 XCHAR_TABLE (prev)->next_table =
180 XCHAR_TABLE (rest)->next_table;
186 char_table_type_to_symbol (enum char_table_type type)
191 case CHAR_TABLE_TYPE_GENERIC: return Qgeneric;
192 case CHAR_TABLE_TYPE_SYNTAX: return Qsyntax;
193 case CHAR_TABLE_TYPE_DISPLAY: return Qdisplay;
194 case CHAR_TABLE_TYPE_CHAR: return Qchar;
196 case CHAR_TABLE_TYPE_CATEGORY: return Qcategory;
201 static enum char_table_type
202 symbol_to_char_table_type (Lisp_Object symbol)
204 CHECK_SYMBOL (symbol);
206 if (EQ (symbol, Qgeneric)) return CHAR_TABLE_TYPE_GENERIC;
207 if (EQ (symbol, Qsyntax)) return CHAR_TABLE_TYPE_SYNTAX;
208 if (EQ (symbol, Qdisplay)) return CHAR_TABLE_TYPE_DISPLAY;
209 if (EQ (symbol, Qchar)) return CHAR_TABLE_TYPE_CHAR;
211 if (EQ (symbol, Qcategory)) return CHAR_TABLE_TYPE_CATEGORY;
214 signal_simple_error ("Unrecognized char table type", symbol);
215 return CHAR_TABLE_TYPE_GENERIC; /* not reached */
219 print_chartab_range (Emchar first, Emchar last, Lisp_Object val,
220 Lisp_Object printcharfun)
224 write_c_string (" (", printcharfun);
225 print_internal (make_char (first), printcharfun, 0);
226 write_c_string (" ", printcharfun);
227 print_internal (make_char (last), printcharfun, 0);
228 write_c_string (") ", printcharfun);
232 write_c_string (" ", printcharfun);
233 print_internal (make_char (first), printcharfun, 0);
234 write_c_string (" ", printcharfun);
236 print_internal (val, printcharfun, 1);
242 print_chartab_charset_row (Lisp_Object charset,
244 struct Lisp_Char_Table_Entry *cte,
245 Lisp_Object printcharfun)
248 Lisp_Object cat = Qunbound;
251 for (i = 32; i < 128; i++)
253 Lisp_Object pam = cte->level2[i - 32];
265 print_chartab_range (MAKE_CHAR (charset, first, 0),
266 MAKE_CHAR (charset, i - 1, 0),
269 print_chartab_range (MAKE_CHAR (charset, row, first),
270 MAKE_CHAR (charset, row, i - 1),
280 print_chartab_range (MAKE_CHAR (charset, first, 0),
281 MAKE_CHAR (charset, i - 1, 0),
284 print_chartab_range (MAKE_CHAR (charset, row, first),
285 MAKE_CHAR (charset, row, i - 1),
291 print_chartab_two_byte_charset (Lisp_Object charset,
292 struct Lisp_Char_Table_Entry *cte,
293 Lisp_Object printcharfun)
297 for (i = 32; i < 128; i++)
299 Lisp_Object jen = cte->level2[i - 32];
301 if (!CHAR_TABLE_ENTRYP (jen))
305 write_c_string (" [", printcharfun);
306 print_internal (XCHARSET_NAME (charset), printcharfun, 0);
307 sprintf (buf, " %d] ", i);
308 write_c_string (buf, printcharfun);
309 print_internal (jen, printcharfun, 0);
312 print_chartab_charset_row (charset, i, XCHAR_TABLE_ENTRY (jen),
320 print_char_table (Lisp_Object obj, Lisp_Object printcharfun, int escapeflag)
322 struct Lisp_Char_Table *ct = XCHAR_TABLE (obj);
325 sprintf (buf, "#s(char-table type %s data (",
326 string_data (symbol_name (XSYMBOL
327 (char_table_type_to_symbol (ct->type)))));
328 write_c_string (buf, printcharfun);
330 /* Now write out the ASCII/Control-1 stuff. */
334 Lisp_Object val = Qunbound;
336 for (i = 0; i < NUM_ASCII_CHARS; i++)
345 if (!EQ (ct->ascii[i], val))
347 print_chartab_range (first, i - 1, val, printcharfun);
354 print_chartab_range (first, i - 1, val, printcharfun);
361 for (i = MIN_LEADING_BYTE; i < MIN_LEADING_BYTE + NUM_LEADING_BYTES;
364 Lisp_Object ann = ct->level1[i - MIN_LEADING_BYTE];
365 Lisp_Object charset = CHARSET_BY_LEADING_BYTE (i);
367 if (!CHARSETP (charset) || i == LEADING_BYTE_ASCII
368 || i == LEADING_BYTE_CONTROL_1)
370 if (!CHAR_TABLE_ENTRYP (ann))
372 write_c_string (" ", printcharfun);
373 print_internal (XCHARSET_NAME (charset),
375 write_c_string (" ", printcharfun);
376 print_internal (ann, printcharfun, 0);
380 struct Lisp_Char_Table_Entry *cte = XCHAR_TABLE_ENTRY (ann);
381 if (XCHARSET_DIMENSION (charset) == 1)
382 print_chartab_charset_row (charset, -1, cte, printcharfun);
384 print_chartab_two_byte_charset (charset, cte, printcharfun);
390 write_c_string ("))", printcharfun);
394 char_table_equal (Lisp_Object obj1, Lisp_Object obj2, int depth)
396 struct Lisp_Char_Table *ct1 = XCHAR_TABLE (obj1);
397 struct Lisp_Char_Table *ct2 = XCHAR_TABLE (obj2);
400 if (CHAR_TABLE_TYPE (ct1) != CHAR_TABLE_TYPE (ct2))
403 for (i = 0; i < NUM_ASCII_CHARS; i++)
404 if (!internal_equal (ct1->ascii[i], ct2->ascii[i], depth + 1))
408 for (i = 0; i < NUM_LEADING_BYTES; i++)
409 if (!internal_equal (ct1->level1[i], ct2->level1[i], depth + 1))
417 char_table_hash (Lisp_Object obj, int depth)
419 struct Lisp_Char_Table *ct = XCHAR_TABLE (obj);
420 unsigned long hashval = internal_array_hash (ct->ascii, NUM_ASCII_CHARS,
423 hashval = HASH2 (hashval,
424 internal_array_hash (ct->level1, NUM_LEADING_BYTES, depth));
429 static const struct lrecord_description char_table_description[] = {
430 { XD_LISP_OBJECT, offsetof(struct Lisp_Char_Table, ascii), NUM_ASCII_CHARS },
432 { XD_LISP_OBJECT, offsetof(struct Lisp_Char_Table, level1), NUM_LEADING_BYTES },
437 DEFINE_LRECORD_IMPLEMENTATION ("char-table", char_table,
438 mark_char_table, print_char_table, 0,
439 char_table_equal, char_table_hash,
440 char_table_description,
441 struct Lisp_Char_Table);
443 DEFUN ("char-table-p", Fchar_table_p, 1, 1, 0, /*
444 Return non-nil if OBJECT is a char table.
446 A char table is a table that maps characters (or ranges of characters)
447 to values. Char tables are specialized for characters, only allowing
448 particular sorts of ranges to be assigned values. Although this
449 loses in generality, it makes for extremely fast (constant-time)
450 lookups, and thus is feasible for applications that do an extremely
451 large number of lookups (e.g. scanning a buffer for a character in
452 a particular syntax, where a lookup in the syntax table must occur
455 When Mule support exists, the types of ranges that can be assigned
460 -- a single row in a two-octet charset
461 -- a single character
463 When Mule support is not present, the types of ranges that can be
467 -- a single character
469 To create a char table, use `make-char-table'. To modify a char
470 table, use `put-char-table' or `remove-char-table'. To retrieve the
471 value for a particular character, use `get-char-table'. See also
472 `map-char-table', `clear-char-table', `copy-char-table',
473 `valid-char-table-type-p', `char-table-type-list', `valid-char-table-value-p',
474 and `check-char-table-value'.
478 return CHAR_TABLEP (object) ? Qt : Qnil;
481 DEFUN ("char-table-type-list", Fchar_table_type_list, 0, 0, 0, /*
482 Return a list of the recognized char table types.
483 See `valid-char-table-type-p'.
488 return list5 (Qchar, Qcategory, Qdisplay, Qgeneric, Qsyntax);
490 return list4 (Qchar, Qdisplay, Qgeneric, Qsyntax);
494 DEFUN ("valid-char-table-type-p", Fvalid_char_table_type_p, 1, 1, 0, /*
495 Return t if TYPE if a recognized char table type.
497 Each char table type is used for a different purpose and allows different
498 sorts of values. The different char table types are
501 Used for category tables, which specify the regexp categories
502 that a character is in. The valid values are nil or a
503 bit vector of 95 elements. Higher-level Lisp functions are
504 provided for working with category tables. Currently categories
505 and category tables only exist when Mule support is present.
507 A generalized char table, for mapping from one character to
508 another. Used for case tables, syntax matching tables,
509 `keyboard-translate-table', etc. The valid values are characters.
511 An even more generalized char table, for mapping from a
512 character to anything.
514 Used for display tables, which specify how a particular character
515 is to appear when displayed. #### Not yet implemented.
517 Used for syntax tables, which specify the syntax of a particular
518 character. Higher-level Lisp functions are provided for
519 working with syntax tables. The valid values are integers.
524 return (EQ (type, Qchar) ||
526 EQ (type, Qcategory) ||
528 EQ (type, Qdisplay) ||
529 EQ (type, Qgeneric) ||
530 EQ (type, Qsyntax)) ? Qt : Qnil;
533 DEFUN ("char-table-type", Fchar_table_type, 1, 1, 0, /*
534 Return the type of char table TABLE.
535 See `valid-char-table-type-p'.
539 CHECK_CHAR_TABLE (table);
540 return char_table_type_to_symbol (XCHAR_TABLE (table)->type);
544 fill_char_table (struct Lisp_Char_Table *ct, Lisp_Object value)
548 for (i = 0; i < NUM_ASCII_CHARS; i++)
549 ct->ascii[i] = value;
551 for (i = 0; i < NUM_LEADING_BYTES; i++)
552 ct->level1[i] = value;
555 if (ct->type == CHAR_TABLE_TYPE_SYNTAX)
556 update_syntax_table (ct);
559 DEFUN ("reset-char-table", Freset_char_table, 1, 1, 0, /*
560 Reset a char table to its default state.
564 struct Lisp_Char_Table *ct;
566 CHECK_CHAR_TABLE (table);
567 ct = XCHAR_TABLE (table);
571 case CHAR_TABLE_TYPE_CHAR:
572 fill_char_table (ct, make_char (0));
574 case CHAR_TABLE_TYPE_DISPLAY:
575 case CHAR_TABLE_TYPE_GENERIC:
577 case CHAR_TABLE_TYPE_CATEGORY:
579 fill_char_table (ct, Qnil);
582 case CHAR_TABLE_TYPE_SYNTAX:
583 fill_char_table (ct, make_int (Sinherit));
593 DEFUN ("make-char-table", Fmake_char_table, 1, 1, 0, /*
594 Return a new, empty char table of type TYPE.
595 Currently recognized types are 'char, 'category, 'display, 'generic,
596 and 'syntax. See `valid-char-table-type-p'.
600 struct Lisp_Char_Table *ct;
602 enum char_table_type ty = symbol_to_char_table_type (type);
604 ct = alloc_lcrecord_type (struct Lisp_Char_Table, &lrecord_char_table);
606 if (ty == CHAR_TABLE_TYPE_SYNTAX)
608 ct->mirror_table = Fmake_char_table (Qgeneric);
609 fill_char_table (XCHAR_TABLE (ct->mirror_table),
613 ct->mirror_table = Qnil;
614 ct->next_table = Qnil;
615 XSETCHAR_TABLE (obj, ct);
616 if (ty == CHAR_TABLE_TYPE_SYNTAX)
618 ct->next_table = Vall_syntax_tables;
619 Vall_syntax_tables = obj;
621 Freset_char_table (obj);
628 make_char_table_entry (Lisp_Object initval)
632 struct Lisp_Char_Table_Entry *cte =
633 alloc_lcrecord_type (struct Lisp_Char_Table_Entry,
634 &lrecord_char_table_entry);
636 for (i = 0; i < 96; i++)
637 cte->level2[i] = initval;
639 XSETCHAR_TABLE_ENTRY (obj, cte);
644 copy_char_table_entry (Lisp_Object entry)
646 struct Lisp_Char_Table_Entry *cte = XCHAR_TABLE_ENTRY (entry);
649 struct Lisp_Char_Table_Entry *ctenew =
650 alloc_lcrecord_type (struct Lisp_Char_Table_Entry,
651 &lrecord_char_table_entry);
653 for (i = 0; i < 96; i++)
655 Lisp_Object new = cte->level2[i];
656 if (CHAR_TABLE_ENTRYP (new))
657 ctenew->level2[i] = copy_char_table_entry (new);
659 ctenew->level2[i] = new;
662 XSETCHAR_TABLE_ENTRY (obj, ctenew);
668 DEFUN ("copy-char-table", Fcopy_char_table, 1, 1, 0, /*
669 Make a new char table which is a copy of OLD-TABLE.
670 It will contain the same values for the same characters and ranges
671 as OLD-TABLE. The values will not themselves be copied.
675 struct Lisp_Char_Table *ct, *ctnew;
679 CHECK_CHAR_TABLE (old_table);
680 ct = XCHAR_TABLE (old_table);
681 ctnew = alloc_lcrecord_type (struct Lisp_Char_Table, &lrecord_char_table);
682 ctnew->type = ct->type;
684 for (i = 0; i < NUM_ASCII_CHARS; i++)
686 Lisp_Object new = ct->ascii[i];
688 assert (! (CHAR_TABLE_ENTRYP (new)));
690 ctnew->ascii[i] = new;
695 for (i = 0; i < NUM_LEADING_BYTES; i++)
697 Lisp_Object new = ct->level1[i];
698 if (CHAR_TABLE_ENTRYP (new))
699 ctnew->level1[i] = copy_char_table_entry (new);
701 ctnew->level1[i] = new;
706 if (CHAR_TABLEP (ct->mirror_table))
707 ctnew->mirror_table = Fcopy_char_table (ct->mirror_table);
709 ctnew->mirror_table = ct->mirror_table;
710 XSETCHAR_TABLE (obj, ctnew);
715 decode_char_table_range (Lisp_Object range, struct chartab_range *outrange)
718 outrange->type = CHARTAB_RANGE_ALL;
719 else if (CHAR_OR_CHAR_INTP (range))
721 outrange->type = CHARTAB_RANGE_CHAR;
722 outrange->ch = XCHAR_OR_CHAR_INT (range);
726 signal_simple_error ("Range must be t or a character", range);
728 else if (VECTORP (range))
730 struct Lisp_Vector *vec = XVECTOR (range);
731 Lisp_Object *elts = vector_data (vec);
732 if (vector_length (vec) != 2)
733 signal_simple_error ("Length of charset row vector must be 2",
735 outrange->type = CHARTAB_RANGE_ROW;
736 outrange->charset = Fget_charset (elts[0]);
738 outrange->row = XINT (elts[1]);
739 switch (XCHARSET_TYPE (outrange->charset))
741 case CHARSET_TYPE_94:
742 case CHARSET_TYPE_96:
743 signal_simple_error ("Charset in row vector must be multi-byte",
745 case CHARSET_TYPE_94X94:
746 check_int_range (outrange->row, 33, 126);
748 case CHARSET_TYPE_96X96:
749 check_int_range (outrange->row, 32, 127);
757 if (!CHARSETP (range) && !SYMBOLP (range))
759 ("Char table range must be t, charset, char, or vector", range);
760 outrange->type = CHARTAB_RANGE_CHARSET;
761 outrange->charset = Fget_charset (range);
768 /* called from CHAR_TABLE_VALUE(). */
770 get_non_ascii_char_table_value (struct Lisp_Char_Table *ct, int leading_byte,
774 Lisp_Object charset = CHARSET_BY_LEADING_BYTE (leading_byte);
777 BREAKUP_CHAR_1_UNSAFE (c, charset, byte1, byte2);
778 val = ct->level1[leading_byte - MIN_LEADING_BYTE];
779 if (CHAR_TABLE_ENTRYP (val))
781 struct Lisp_Char_Table_Entry *cte = XCHAR_TABLE_ENTRY (val);
782 val = cte->level2[byte1 - 32];
783 if (CHAR_TABLE_ENTRYP (val))
785 cte = XCHAR_TABLE_ENTRY (val);
786 assert (byte2 >= 32);
787 val = cte->level2[byte2 - 32];
788 assert (!CHAR_TABLE_ENTRYP (val));
798 get_char_table (Emchar ch, struct Lisp_Char_Table *ct)
806 BREAKUP_CHAR (ch, charset, byte1, byte2);
808 if (EQ (charset, Vcharset_ascii))
809 val = ct->ascii[byte1];
810 else if (EQ (charset, Vcharset_control_1))
811 val = ct->ascii[byte1 + 128];
814 int lb = XCHARSET_LEADING_BYTE (charset) - MIN_LEADING_BYTE;
815 val = ct->level1[lb];
816 if (CHAR_TABLE_ENTRYP (val))
818 struct Lisp_Char_Table_Entry *cte = XCHAR_TABLE_ENTRY (val);
819 val = cte->level2[byte1 - 32];
820 if (CHAR_TABLE_ENTRYP (val))
822 cte = XCHAR_TABLE_ENTRY (val);
823 assert (byte2 >= 32);
824 val = cte->level2[byte2 - 32];
825 assert (!CHAR_TABLE_ENTRYP (val));
833 return ct->ascii[(unsigned char)ch];
834 #endif /* not MULE */
838 DEFUN ("get-char-table", Fget_char_table, 2, 2, 0, /*
839 Find value for char CH in TABLE.
843 struct Lisp_Char_Table *ct;
845 CHECK_CHAR_TABLE (table);
846 ct = XCHAR_TABLE (table);
847 CHECK_CHAR_COERCE_INT (ch);
849 return get_char_table (XCHAR (ch), ct);
852 DEFUN ("get-range-char-table", Fget_range_char_table, 2, 3, 0, /*
853 Find value for a range in TABLE.
854 If there is more than one value, return MULTI (defaults to nil).
856 (range, table, multi))
858 struct Lisp_Char_Table *ct;
859 struct chartab_range rainj;
861 if (CHAR_OR_CHAR_INTP (range))
862 return Fget_char_table (range, table);
863 CHECK_CHAR_TABLE (table);
864 ct = XCHAR_TABLE (table);
866 decode_char_table_range (range, &rainj);
869 case CHARTAB_RANGE_ALL:
872 Lisp_Object first = ct->ascii[0];
874 for (i = 1; i < NUM_ASCII_CHARS; i++)
875 if (!EQ (first, ct->ascii[i]))
879 for (i = MIN_LEADING_BYTE; i < MIN_LEADING_BYTE + NUM_LEADING_BYTES;
882 if (!CHARSETP (CHARSET_BY_LEADING_BYTE (i))
883 || i == LEADING_BYTE_ASCII
884 || i == LEADING_BYTE_CONTROL_1)
886 if (!EQ (first, ct->level1[i - MIN_LEADING_BYTE]))
895 case CHARTAB_RANGE_CHARSET:
896 if (EQ (rainj.charset, Vcharset_ascii))
899 Lisp_Object first = ct->ascii[0];
901 for (i = 1; i < 128; i++)
902 if (!EQ (first, ct->ascii[i]))
907 if (EQ (rainj.charset, Vcharset_control_1))
910 Lisp_Object first = ct->ascii[128];
912 for (i = 129; i < 160; i++)
913 if (!EQ (first, ct->ascii[i]))
919 Lisp_Object val = ct->level1[XCHARSET_LEADING_BYTE (rainj.charset) -
921 if (CHAR_TABLE_ENTRYP (val))
926 case CHARTAB_RANGE_ROW:
928 Lisp_Object val = ct->level1[XCHARSET_LEADING_BYTE (rainj.charset) -
930 if (!CHAR_TABLE_ENTRYP (val))
932 val = XCHAR_TABLE_ENTRY (val)->level2[rainj.row - 32];
933 if (CHAR_TABLE_ENTRYP (val))
937 #endif /* not MULE */
943 return Qnil; /* not reached */
947 check_valid_char_table_value (Lisp_Object value, enum char_table_type type,
952 case CHAR_TABLE_TYPE_SYNTAX:
953 if (!ERRB_EQ (errb, ERROR_ME))
954 return INTP (value) || (CONSP (value) && INTP (XCAR (value))
955 && CHAR_OR_CHAR_INTP (XCDR (value)));
958 Lisp_Object cdr = XCDR (value);
959 CHECK_INT (XCAR (value));
960 CHECK_CHAR_COERCE_INT (cdr);
967 case CHAR_TABLE_TYPE_CATEGORY:
968 if (!ERRB_EQ (errb, ERROR_ME))
969 return CATEGORY_TABLE_VALUEP (value);
970 CHECK_CATEGORY_TABLE_VALUE (value);
974 case CHAR_TABLE_TYPE_GENERIC:
977 case CHAR_TABLE_TYPE_DISPLAY:
979 maybe_signal_simple_error ("Display char tables not yet implemented",
980 value, Qchar_table, errb);
983 case CHAR_TABLE_TYPE_CHAR:
984 if (!ERRB_EQ (errb, ERROR_ME))
985 return CHAR_OR_CHAR_INTP (value);
986 CHECK_CHAR_COERCE_INT (value);
993 return 0; /* not reached */
997 canonicalize_char_table_value (Lisp_Object value, enum char_table_type type)
1001 case CHAR_TABLE_TYPE_SYNTAX:
1004 Lisp_Object car = XCAR (value);
1005 Lisp_Object cdr = XCDR (value);
1006 CHECK_CHAR_COERCE_INT (cdr);
1007 return Fcons (car, cdr);
1010 case CHAR_TABLE_TYPE_CHAR:
1011 CHECK_CHAR_COERCE_INT (value);
1019 DEFUN ("valid-char-table-value-p", Fvalid_char_table_value_p, 2, 2, 0, /*
1020 Return non-nil if VALUE is a valid value for CHAR-TABLE-TYPE.
1022 (value, char_table_type))
1024 enum char_table_type type = symbol_to_char_table_type (char_table_type);
1026 return check_valid_char_table_value (value, type, ERROR_ME_NOT) ? Qt : Qnil;
1029 DEFUN ("check-valid-char-table-value", Fcheck_valid_char_table_value, 2, 2, 0, /*
1030 Signal an error if VALUE is not a valid value for CHAR-TABLE-TYPE.
1032 (value, char_table_type))
1034 enum char_table_type type = symbol_to_char_table_type (char_table_type);
1036 check_valid_char_table_value (value, type, ERROR_ME);
1040 /* Assign VAL to all characters in RANGE in char table CT. */
1043 put_char_table (struct Lisp_Char_Table *ct, struct chartab_range *range,
1046 switch (range->type)
1048 case CHARTAB_RANGE_ALL:
1049 fill_char_table (ct, val);
1050 return; /* avoid the duplicate call to update_syntax_table() below,
1051 since fill_char_table() also did that. */
1054 case CHARTAB_RANGE_CHARSET:
1055 if (EQ (range->charset, Vcharset_ascii))
1058 for (i = 0; i < 128; i++)
1061 else if (EQ (range->charset, Vcharset_control_1))
1064 for (i = 128; i < 160; i++)
1069 int lb = XCHARSET_LEADING_BYTE (range->charset) - MIN_LEADING_BYTE;
1070 ct->level1[lb] = val;
1074 case CHARTAB_RANGE_ROW:
1076 struct Lisp_Char_Table_Entry *cte;
1077 int lb = XCHARSET_LEADING_BYTE (range->charset) - MIN_LEADING_BYTE;
1078 /* make sure that there is a separate entry for the row. */
1079 if (!CHAR_TABLE_ENTRYP (ct->level1[lb]))
1080 ct->level1[lb] = make_char_table_entry (ct->level1[lb]);
1081 cte = XCHAR_TABLE_ENTRY (ct->level1[lb]);
1082 cte->level2[range->row - 32] = val;
1087 case CHARTAB_RANGE_CHAR:
1090 Lisp_Object charset;
1093 BREAKUP_CHAR (range->ch, charset, byte1, byte2);
1094 if (EQ (charset, Vcharset_ascii))
1095 ct->ascii[byte1] = val;
1096 else if (EQ (charset, Vcharset_control_1))
1097 ct->ascii[byte1 + 128] = val;
1100 struct Lisp_Char_Table_Entry *cte;
1101 int lb = XCHARSET_LEADING_BYTE (charset) - MIN_LEADING_BYTE;
1102 /* make sure that there is a separate entry for the row. */
1103 if (!CHAR_TABLE_ENTRYP (ct->level1[lb]))
1104 ct->level1[lb] = make_char_table_entry (ct->level1[lb]);
1105 cte = XCHAR_TABLE_ENTRY (ct->level1[lb]);
1106 /* now CTE is a char table entry for the charset;
1107 each entry is for a single row (or character of
1108 a one-octet charset). */
1109 if (XCHARSET_DIMENSION (charset) == 1)
1110 cte->level2[byte1 - 32] = val;
1113 /* assigning to one character in a two-octet charset. */
1114 /* make sure that the charset row contains a separate
1115 entry for each character. */
1116 if (!CHAR_TABLE_ENTRYP (cte->level2[byte1 - 32]))
1117 cte->level2[byte1 - 32] =
1118 make_char_table_entry (cte->level2[byte1 - 32]);
1119 cte = XCHAR_TABLE_ENTRY (cte->level2[byte1 - 32]);
1120 cte->level2[byte2 - 32] = val;
1124 #else /* not MULE */
1125 ct->ascii[(unsigned char) (range->ch)] = val;
1127 #endif /* not MULE */
1130 if (ct->type == CHAR_TABLE_TYPE_SYNTAX)
1131 update_syntax_table (ct);
1134 DEFUN ("put-char-table", Fput_char_table, 3, 3, 0, /*
1135 Set the value for chars in RANGE to be VAL in TABLE.
1137 RANGE specifies one or more characters to be affected and should be
1138 one of the following:
1140 -- t (all characters are affected)
1141 -- A charset (only allowed when Mule support is present)
1142 -- A vector of two elements: a two-octet charset and a row number
1143 (only allowed when Mule support is present)
1144 -- A single character
1146 VAL must be a value appropriate for the type of TABLE.
1147 See `valid-char-table-type-p'.
1149 (range, val, table))
1151 struct Lisp_Char_Table *ct;
1152 struct chartab_range rainj;
1154 CHECK_CHAR_TABLE (table);
1155 ct = XCHAR_TABLE (table);
1156 check_valid_char_table_value (val, ct->type, ERROR_ME);
1157 decode_char_table_range (range, &rainj);
1158 val = canonicalize_char_table_value (val, ct->type);
1159 put_char_table (ct, &rainj, val);
1163 /* Map FN over the ASCII chars in CT. */
1166 map_over_charset_ascii (struct Lisp_Char_Table *ct,
1167 int (*fn) (struct chartab_range *range,
1168 Lisp_Object val, void *arg),
1171 struct chartab_range rainj;
1180 rainj.type = CHARTAB_RANGE_CHAR;
1182 for (i = start, retval = 0; i < stop && retval == 0; i++)
1184 rainj.ch = (Emchar) i;
1185 retval = (fn) (&rainj, ct->ascii[i], arg);
1193 /* Map FN over the Control-1 chars in CT. */
1196 map_over_charset_control_1 (struct Lisp_Char_Table *ct,
1197 int (*fn) (struct chartab_range *range,
1198 Lisp_Object val, void *arg),
1201 struct chartab_range rainj;
1204 int stop = start + 32;
1206 rainj.type = CHARTAB_RANGE_CHAR;
1208 for (i = start, retval = 0; i < stop && retval == 0; i++)
1210 rainj.ch = (Emchar) (i);
1211 retval = (fn) (&rainj, ct->ascii[i], arg);
1217 /* Map FN over the row ROW of two-byte charset CHARSET.
1218 There must be a separate value for that row in the char table.
1219 CTE specifies the char table entry for CHARSET. */
1222 map_over_charset_row (struct Lisp_Char_Table_Entry *cte,
1223 Lisp_Object charset, int row,
1224 int (*fn) (struct chartab_range *range,
1225 Lisp_Object val, void *arg),
1228 Lisp_Object val = cte->level2[row - 32];
1230 if (!CHAR_TABLE_ENTRYP (val))
1232 struct chartab_range rainj;
1234 rainj.type = CHARTAB_RANGE_ROW;
1235 rainj.charset = charset;
1237 return (fn) (&rainj, val, arg);
1241 struct chartab_range rainj;
1243 int charset94_p = (XCHARSET_CHARS (charset) == 94);
1244 int start = charset94_p ? 33 : 32;
1245 int stop = charset94_p ? 127 : 128;
1247 cte = XCHAR_TABLE_ENTRY (val);
1249 rainj.type = CHARTAB_RANGE_CHAR;
1251 for (i = start, retval = 0; i < stop && retval == 0; i++)
1253 rainj.ch = MAKE_CHAR (charset, row, i);
1254 retval = (fn) (&rainj, cte->level2[i - 32], arg);
1262 map_over_other_charset (struct Lisp_Char_Table *ct, int lb,
1263 int (*fn) (struct chartab_range *range,
1264 Lisp_Object val, void *arg),
1267 Lisp_Object val = ct->level1[lb - MIN_LEADING_BYTE];
1268 Lisp_Object charset = CHARSET_BY_LEADING_BYTE (lb);
1270 if (!CHARSETP (charset)
1271 || lb == LEADING_BYTE_ASCII
1272 || lb == LEADING_BYTE_CONTROL_1)
1275 if (!CHAR_TABLE_ENTRYP (val))
1277 struct chartab_range rainj;
1279 rainj.type = CHARTAB_RANGE_CHARSET;
1280 rainj.charset = charset;
1281 return (fn) (&rainj, val, arg);
1285 struct Lisp_Char_Table_Entry *cte = XCHAR_TABLE_ENTRY (val);
1286 int charset94_p = (XCHARSET_CHARS (charset) == 94);
1287 int start = charset94_p ? 33 : 32;
1288 int stop = charset94_p ? 127 : 128;
1291 if (XCHARSET_DIMENSION (charset) == 1)
1293 struct chartab_range rainj;
1294 rainj.type = CHARTAB_RANGE_CHAR;
1296 for (i = start, retval = 0; i < stop && retval == 0; i++)
1298 rainj.ch = MAKE_CHAR (charset, i, 0);
1299 retval = (fn) (&rainj, cte->level2[i - 32], arg);
1304 for (i = start, retval = 0; i < stop && retval == 0; i++)
1305 retval = map_over_charset_row (cte, charset, i, fn, arg);
1314 /* Map FN (with client data ARG) over range RANGE in char table CT.
1315 Mapping stops the first time FN returns non-zero, and that value
1316 becomes the return value of map_char_table(). */
1319 map_char_table (struct Lisp_Char_Table *ct,
1320 struct chartab_range *range,
1321 int (*fn) (struct chartab_range *range,
1322 Lisp_Object val, void *arg),
1325 switch (range->type)
1327 case CHARTAB_RANGE_ALL:
1331 retval = map_over_charset_ascii (ct, fn, arg);
1335 retval = map_over_charset_control_1 (ct, fn, arg);
1340 int start = MIN_LEADING_BYTE;
1341 int stop = start + NUM_LEADING_BYTES;
1343 for (i = start, retval = 0; i < stop && retval == 0; i++)
1345 retval = map_over_other_charset (ct, i, fn, arg);
1353 case CHARTAB_RANGE_CHARSET:
1354 return map_over_other_charset (ct,
1355 XCHARSET_LEADING_BYTE (range->charset),
1358 case CHARTAB_RANGE_ROW:
1360 Lisp_Object val = ct->level1[XCHARSET_LEADING_BYTE (range->charset) - MIN_LEADING_BYTE];
1361 if (!CHAR_TABLE_ENTRYP (val))
1363 struct chartab_range rainj;
1365 rainj.type = CHARTAB_RANGE_ROW;
1366 rainj.charset = range->charset;
1367 rainj.row = range->row;
1368 return (fn) (&rainj, val, arg);
1371 return map_over_charset_row (XCHAR_TABLE_ENTRY (val),
1372 range->charset, range->row,
1377 case CHARTAB_RANGE_CHAR:
1379 Emchar ch = range->ch;
1380 Lisp_Object val = CHAR_TABLE_VALUE_UNSAFE (ct, ch);
1381 struct chartab_range rainj;
1383 rainj.type = CHARTAB_RANGE_CHAR;
1385 return (fn) (&rainj, val, arg);
1395 struct slow_map_char_table_arg
1397 Lisp_Object function;
1402 slow_map_char_table_fun (struct chartab_range *range,
1403 Lisp_Object val, void *arg)
1405 Lisp_Object ranjarg = Qnil;
1406 struct slow_map_char_table_arg *closure =
1407 (struct slow_map_char_table_arg *) arg;
1409 switch (range->type)
1411 case CHARTAB_RANGE_ALL:
1416 case CHARTAB_RANGE_CHARSET:
1417 ranjarg = XCHARSET_NAME (range->charset);
1420 case CHARTAB_RANGE_ROW:
1421 ranjarg = vector2 (XCHARSET_NAME (range->charset),
1422 make_int (range->row));
1425 case CHARTAB_RANGE_CHAR:
1426 ranjarg = make_char (range->ch);
1432 closure->retval = call2 (closure->function, ranjarg, val);
1433 return !NILP (closure->retval);
1436 DEFUN ("map-char-table", Fmap_char_table, 2, 3, 0, /*
1437 Map FUNCTION over entries in TABLE, calling it with two args,
1438 each key and value in the table.
1440 RANGE specifies a subrange to map over and is in the same format as
1441 the RANGE argument to `put-range-table'. If omitted or t, it defaults to
1444 (function, table, range))
1446 struct Lisp_Char_Table *ct;
1447 struct slow_map_char_table_arg slarg;
1448 struct gcpro gcpro1, gcpro2;
1449 struct chartab_range rainj;
1451 CHECK_CHAR_TABLE (table);
1452 ct = XCHAR_TABLE (table);
1455 decode_char_table_range (range, &rainj);
1456 slarg.function = function;
1457 slarg.retval = Qnil;
1458 GCPRO2 (slarg.function, slarg.retval);
1459 map_char_table (ct, &rainj, slow_map_char_table_fun, &slarg);
1462 return slarg.retval;
1467 /************************************************************************/
1468 /* Char table read syntax */
1469 /************************************************************************/
1472 chartab_type_validate (Lisp_Object keyword, Lisp_Object value,
1473 Error_behavior errb)
1475 /* #### should deal with ERRB */
1476 symbol_to_char_table_type (value);
1481 chartab_data_validate (Lisp_Object keyword, Lisp_Object value,
1482 Error_behavior errb)
1486 /* #### should deal with ERRB */
1487 EXTERNAL_LIST_LOOP (rest, value)
1489 Lisp_Object range = XCAR (rest);
1490 struct chartab_range dummy;
1494 signal_simple_error ("Invalid list format", value);
1497 if (!CONSP (XCDR (range))
1498 || !NILP (XCDR (XCDR (range))))
1499 signal_simple_error ("Invalid range format", range);
1500 decode_char_table_range (XCAR (range), &dummy);
1501 decode_char_table_range (XCAR (XCDR (range)), &dummy);
1504 decode_char_table_range (range, &dummy);
1511 chartab_instantiate (Lisp_Object data)
1513 Lisp_Object chartab;
1514 Lisp_Object type = Qgeneric;
1515 Lisp_Object dataval = Qnil;
1517 while (!NILP (data))
1519 Lisp_Object keyw = Fcar (data);
1525 if (EQ (keyw, Qtype))
1527 else if (EQ (keyw, Qdata))
1531 chartab = Fmake_char_table (type);
1534 while (!NILP (data))
1536 Lisp_Object range = Fcar (data);
1537 Lisp_Object val = Fcar (Fcdr (data));
1539 data = Fcdr (Fcdr (data));
1542 if (CHAR_OR_CHAR_INTP (XCAR (range)))
1544 Emchar first = XCHAR_OR_CHAR_INT (Fcar (range));
1545 Emchar last = XCHAR_OR_CHAR_INT (Fcar (Fcdr (range)));
1548 for (i = first; i <= last; i++)
1549 Fput_char_table (make_char (i), val, chartab);
1555 Fput_char_table (range, val, chartab);
1564 /************************************************************************/
1565 /* Category Tables, specifically */
1566 /************************************************************************/
1568 DEFUN ("category-table-p", Fcategory_table_p, 1, 1, 0, /*
1569 Return t if ARG is a category table.
1570 A category table is a type of char table used for keeping track of
1571 categories. Categories are used for classifying characters for use
1572 in regexps -- you can refer to a category rather than having to use
1573 a complicated [] expression (and category lookups are significantly
1576 There are 95 different categories available, one for each printable
1577 character (including space) in the ASCII charset. Each category
1578 is designated by one such character, called a "category designator".
1579 They are specified in a regexp using the syntax "\\cX", where X is
1580 a category designator.
1582 A category table specifies, for each character, the categories that
1583 the character is in. Note that a character can be in more than one
1584 category. More specifically, a category table maps from a character
1585 to either the value nil (meaning the character is in no categories)
1586 or a 95-element bit vector, specifying for each of the 95 categories
1587 whether the character is in that category.
1589 Special Lisp functions are provided that abstract this, so you do not
1590 have to directly manipulate bit vectors.
1594 return (CHAR_TABLEP (obj) &&
1595 XCHAR_TABLE_TYPE (obj) == CHAR_TABLE_TYPE_CATEGORY) ?
1600 check_category_table (Lisp_Object obj, Lisp_Object def)
1604 while (NILP (Fcategory_table_p (obj)))
1605 obj = wrong_type_argument (Qcategory_table_p, obj);
1610 check_category_char (Emchar ch, Lisp_Object table,
1611 unsigned int designator, unsigned int not)
1613 REGISTER Lisp_Object temp;
1614 struct Lisp_Char_Table *ctbl;
1615 #ifdef ERROR_CHECK_TYPECHECK
1616 if (NILP (Fcategory_table_p (table)))
1617 signal_simple_error ("Expected category table", table);
1619 ctbl = XCHAR_TABLE (table);
1620 temp = get_char_table (ch, ctbl);
1625 return bit_vector_bit (XBIT_VECTOR (temp), designator) ? !not : not;
1628 DEFUN ("check-category-at", Fcheck_category_at, 2, 4, 0, /*
1629 Return t if category of a character at POS includes DESIGNATOR,
1630 else return nil. Optional third arg specifies which buffer
1631 \(defaulting to current), and fourth specifies the CATEGORY-TABLE,
1632 \(defaulting to the buffer's category table).
1634 (pos, designator, buffer, category_table))
1639 struct buffer *buf = decode_buffer (buffer, 0);
1642 CHECK_CATEGORY_DESIGNATOR (designator);
1643 des = XCHAR (designator);
1644 ctbl = check_category_table (category_table, Vstandard_category_table);
1645 ch = BUF_FETCH_CHAR (buf, XINT (pos));
1646 return check_category_char (ch, ctbl, des, 0) ? Qt : Qnil;
1649 DEFUN ("char-in-category-p", Fchar_in_category_p, 2, 3, 0, /*
1650 Return t if category of character CHR includes DESIGNATOR, else nil.
1651 Optional third arg specifies the CATEGORY-TABLE to use,
1652 which defaults to the system default table.
1654 (chr, designator, category_table))
1660 CHECK_CATEGORY_DESIGNATOR (designator);
1661 des = XCHAR (designator);
1664 ctbl = check_category_table (category_table, Vstandard_category_table);
1665 return check_category_char (ch, ctbl, des, 0) ? Qt : Qnil;
1668 DEFUN ("category-table", Fcategory_table, 0, 1, 0, /*
1669 Return the current category table.
1670 This is the one specified by the current buffer, or by BUFFER if it
1675 return decode_buffer (buffer, 0)->category_table;
1678 DEFUN ("standard-category-table", Fstandard_category_table, 0, 0, 0, /*
1679 Return the standard category table.
1680 This is the one used for new buffers.
1684 return Vstandard_category_table;
1687 DEFUN ("copy-category-table", Fcopy_category_table, 0, 1, 0, /*
1688 Construct a new category table and return it.
1689 It is a copy of the TABLE, which defaults to the standard category table.
1693 if (NILP (Vstandard_category_table))
1694 return Fmake_char_table (Qcategory);
1696 table = check_category_table (table, Vstandard_category_table);
1697 return Fcopy_char_table (table);
1700 DEFUN ("set-category-table", Fset_category_table, 1, 2, 0, /*
1701 Select a new category table for BUFFER.
1702 One argument, a category table.
1703 BUFFER defaults to the current buffer if omitted.
1707 struct buffer *buf = decode_buffer (buffer, 0);
1708 table = check_category_table (table, Qnil);
1709 buf->category_table = table;
1710 /* Indicate that this buffer now has a specified category table. */
1711 buf->local_var_flags |= XINT (buffer_local_flags.category_table);
1715 DEFUN ("category-designator-p", Fcategory_designator_p, 1, 1, 0, /*
1716 Return t if ARG is a category designator (a char in the range ' ' to '~').
1720 return CATEGORY_DESIGNATORP (obj) ? Qt : Qnil;
1723 DEFUN ("category-table-value-p", Fcategory_table_value_p, 1, 1, 0, /*
1724 Return t if ARG is a category table value.
1725 Valid values are nil or a bit vector of size 95.
1729 return CATEGORY_TABLE_VALUEP (obj) ? Qt : Qnil;
1733 #define CATEGORYP(x) \
1734 (CHARP ((x)) && XCHAR ((x)) >= 0x20 && XCHAR ((x)) <= 0x7E)
1736 #define CATEGORY_SET(c) \
1737 (get_char_table(c, XCHAR_TABLE(current_buffer->category_table)))
1739 /* Return 1 if CATEGORY_SET contains CATEGORY, else return 0.
1740 The faster version of `!NILP (Faref (category_set, category))'. */
1741 #define CATEGORY_MEMBER(category, category_set) \
1742 (bit_vector_bit(XBIT_VECTOR (category_set), category - 32))
1744 /* Return 1 if there is a word boundary between two word-constituent
1745 characters C1 and C2 if they appear in this order, else return 0.
1746 Use the macro WORD_BOUNDARY_P instead of calling this function
1750 word_boundary_p (Emchar c1, Emchar c2)
1752 Lisp_Object category_set1, category_set2;
1757 if (COMPOSITE_CHAR_P (c1))
1758 c1 = cmpchar_component (c1, 0, 1);
1759 if (COMPOSITE_CHAR_P (c2))
1760 c2 = cmpchar_component (c2, 0, 1);
1763 if (EQ (CHAR_CHARSET (c1), CHAR_CHARSET (c2)))
1765 tail = Vword_separating_categories;
1770 tail = Vword_combining_categories;
1774 category_set1 = CATEGORY_SET (c1);
1775 if (NILP (category_set1))
1776 return default_result;
1777 category_set2 = CATEGORY_SET (c2);
1778 if (NILP (category_set2))
1779 return default_result;
1781 for (; CONSP (tail); tail = XCONS (tail)->cdr)
1783 Lisp_Object elt = XCONS(tail)->car;
1786 && CATEGORYP (XCONS (elt)->car)
1787 && CATEGORYP (XCONS (elt)->cdr)
1788 && CATEGORY_MEMBER (XCHAR (XCONS (elt)->car), category_set1)
1789 && CATEGORY_MEMBER (XCHAR (XCONS (elt)->cdr), category_set2))
1790 return !default_result;
1792 return default_result;
1798 syms_of_chartab (void)
1801 defsymbol (&Qcategory_table_p, "category-table-p");
1802 defsymbol (&Qcategory_designator_p, "category-designator-p");
1803 defsymbol (&Qcategory_table_value_p, "category-table-value-p");
1806 defsymbol (&Qchar_table, "char-table");
1807 defsymbol (&Qchar_tablep, "char-table-p");
1809 DEFSUBR (Fchar_table_p);
1810 DEFSUBR (Fchar_table_type_list);
1811 DEFSUBR (Fvalid_char_table_type_p);
1812 DEFSUBR (Fchar_table_type);
1813 DEFSUBR (Freset_char_table);
1814 DEFSUBR (Fmake_char_table);
1815 DEFSUBR (Fcopy_char_table);
1816 DEFSUBR (Fget_char_table);
1817 DEFSUBR (Fget_range_char_table);
1818 DEFSUBR (Fvalid_char_table_value_p);
1819 DEFSUBR (Fcheck_valid_char_table_value);
1820 DEFSUBR (Fput_char_table);
1821 DEFSUBR (Fmap_char_table);
1824 DEFSUBR (Fcategory_table_p);
1825 DEFSUBR (Fcategory_table);
1826 DEFSUBR (Fstandard_category_table);
1827 DEFSUBR (Fcopy_category_table);
1828 DEFSUBR (Fset_category_table);
1829 DEFSUBR (Fcheck_category_at);
1830 DEFSUBR (Fchar_in_category_p);
1831 DEFSUBR (Fcategory_designator_p);
1832 DEFSUBR (Fcategory_table_value_p);
1838 vars_of_chartab (void)
1840 /* DO NOT staticpro this. It works just like Vweak_hash_tables. */
1841 Vall_syntax_tables = Qnil;
1845 structure_type_create_chartab (void)
1847 struct structure_type *st;
1849 st = define_structure_type (Qchar_table, 0, chartab_instantiate);
1851 define_structure_type_keyword (st, Qtype, chartab_type_validate);
1852 define_structure_type_keyword (st, Qdata, chartab_data_validate);
1856 complex_vars_of_chartab (void)
1859 /* Set this now, so first buffer creation can refer to it. */
1860 /* Make it nil before calling copy-category-table
1861 so that copy-category-table will know not to try to copy from garbage */
1862 Vstandard_category_table = Qnil;
1863 Vstandard_category_table = Fcopy_category_table (Qnil);
1864 staticpro (&Vstandard_category_table);
1866 DEFVAR_LISP ("word-combining-categories", &Vword_combining_categories /*
1867 List of pair (cons) of categories to determine word boundary.
1869 Emacs treats a sequence of word constituent characters as a single
1870 word (i.e. finds no word boundary between them) iff they belongs to
1871 the same charset. But, exceptions are allowed in the following cases.
1873 (1) The case that characters are in different charsets is controlled
1874 by the variable `word-combining-categories'.
1876 Emacs finds no word boundary between characters of different charsets
1877 if they have categories matching some element of this list.
1879 More precisely, if an element of this list is a cons of category CAT1
1880 and CAT2, and a multibyte character C1 which has CAT1 is followed by
1881 C2 which has CAT2, there's no word boundary between C1 and C2.
1883 For instance, to tell that ASCII characters and Latin-1 characters can
1884 form a single word, the element `(?l . ?l)' should be in this list
1885 because both characters have the category `l' (Latin characters).
1887 (2) The case that character are in the same charset is controlled by
1888 the variable `word-separating-categories'.
1890 Emacs find a word boundary between characters of the same charset
1891 if they have categories matching some element of this list.
1893 More precisely, if an element of this list is a cons of category CAT1
1894 and CAT2, and a multibyte character C1 which has CAT1 is followed by
1895 C2 which has CAT2, there's a word boundary between C1 and C2.
1897 For instance, to tell that there's a word boundary between Japanese
1898 Hiragana and Japanese Kanji (both are in the same charset), the
1899 element `(?H . ?C) should be in this list.
1902 Vword_combining_categories = Qnil;
1904 DEFVAR_LISP ("word-separating-categories", &Vword_separating_categories /*
1905 List of pair (cons) of categories to determine word boundary.
1906 See the documentation of the variable `word-combining-categories'.
1909 Vword_separating_categories = Qnil;