1 /* GNU Emacs routines to deal with category tables.
2 Copyright (C) 1995, 1997 Electrotechnical Laboratory, JAPAN.
3 Licensed to the Free Software Foundation.
5 This file is part of GNU Emacs.
7 GNU Emacs is free software; you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation; either version 2, or (at your option)
12 GNU Emacs is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
17 You should have received a copy of the GNU General Public License
18 along with GNU Emacs; see the file COPYING. If not, write to
19 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
20 Boston, MA 02111-1307, USA. */
23 /* Here we handle three objects: category, category set, and category
24 table. Read comments in the file category.h to understand them. */
33 /* The version number of the latest category table. Each category
34 table has a unique version number. It is assigned a new number
35 also when it is modified. When a regular expression is compiled
36 into the struct re_pattern_buffer, the version number of the
37 category table (of the current buffer) at that moment is also
38 embedded in the structure.
40 For the moment, we are not using this feature. */
41 static int category_table_version
;
43 Lisp_Object Qcategory_table
, Qcategoryp
, Qcategorysetp
, Qcategory_table_p
;
45 /* Variables to determine word boundary. */
46 Lisp_Object Vword_combining_categories
, Vword_separating_categories
;
48 /* Temporary internal variable used in macro CHAR_HAS_CATEGORY. */
49 Lisp_Object _temp_category_set
;
52 /* Category set staff. */
54 DEFUN ("make-category-set", Fmake_category_set
, Smake_category_set
, 1, 1, 0,
55 "Return a newly created category-set which contains CATEGORIES.\n\
56 CATEGORIES is a string of category mnemonics.\n\
57 The value is a bool-vector which has t at the indices corresponding to\n\
60 Lisp_Object categories
;
65 CHECK_STRING (categories
, 0);
66 val
= MAKE_CATEGORY_SET
;
68 if (STRING_MULTIBYTE (categories
))
69 error ("Multibyte string in make-category-set");
71 len
= XSTRING (categories
)->size
;
76 XSETFASTINT (category
, XSTRING (categories
)->data
[len
]);
77 CHECK_CATEGORY (category
, 0);
78 SET_CATEGORY_SET (val
, category
, Qt
);
86 Lisp_Object
check_category_table ();
88 DEFUN ("define-category", Fdefine_category
, Sdefine_category
, 2, 3, 0,
89 "Define CHAR as a category which is described by DOCSTRING.\n\
90 CHAR should be an ASCII printing character in the range ` ' to `~'.\n\
91 DOCSTRING is a documentation string of the category.\n\
92 The category is defined only in category table TABLE, which defaults to\n\
93 the current buffer's category table.")
94 (category
, docstring
, table
)
95 Lisp_Object category
, docstring
, table
;
97 CHECK_CATEGORY (category
, 0);
98 CHECK_STRING (docstring
, 1);
99 table
= check_category_table (table
);
101 if (!NILP (CATEGORY_DOCSTRING (table
, XFASTINT (category
))))
102 error ("Category `%c' is already defined", XFASTINT (category
));
103 CATEGORY_DOCSTRING (table
, XFASTINT (category
)) = docstring
;
108 DEFUN ("category-docstring", Fcategory_docstring
, Scategory_docstring
, 1, 2, 0,
109 "Return the documentation string of CATEGORY, as defined in CATEGORY-TABLE.")
111 Lisp_Object category
, table
;
115 CHECK_CATEGORY (category
, 0);
116 table
= check_category_table (table
);
118 return CATEGORY_DOCSTRING (table
, XFASTINT (category
));
121 DEFUN ("get-unused-category", Fget_unused_category
, Sget_unused_category
,
123 "Return a category which is not yet defined in CATEGORY-TABLE.\n\
124 If no category remains available, return nil.\n\
125 The optional argument CATEGORY-TABLE specifies which category table\n\
126 to modify; it defaults to the current buffer's category table.")
131 Lisp_Object docstring_vector
;
133 table
= check_category_table (table
);
135 for (i
= ' '; i
<= '~'; i
++)
136 if (NILP (CATEGORY_DOCSTRING (table
, i
)))
137 return make_number (i
);
143 /* Category-table staff. */
145 DEFUN ("category-table-p", Fcategory_table_p
, Scategory_table_p
, 1, 1, 0,
146 "Return t if ARG is a category table.")
150 if (CHAR_TABLE_P (arg
)
151 && EQ (XCHAR_TABLE (arg
)->purpose
, Qcategory_table
))
156 /* If TABLE is nil, return the current category table. If TABLE is
157 not nil, check the validity of TABLE as a category table. If
158 valid, return TABLE itself, but if not valid, signal an error of
159 wrong-type-argument. */
162 check_category_table (table
)
165 register Lisp_Object tem
;
167 return current_buffer
->category_table
;
168 while (tem
= Fcategory_table_p (table
), NILP (tem
))
169 table
= wrong_type_argument (Qcategory_table_p
, table
);
173 DEFUN ("category-table", Fcategory_table
, Scategory_table
, 0, 0, 0,
174 "Return the current category table.\n\
175 This is the one specified by the current buffer.")
178 return current_buffer
->category_table
;
181 DEFUN ("standard-category-table", Fstandard_category_table
,
182 Sstandard_category_table
, 0, 0, 0,
183 "Return the standard category table.\n\
184 This is the one used for new buffers.")
187 return Vstandard_category_table
;
190 /* Return a copy of category table TABLE. We can't simply use the
191 function copy-sequence because no contents should be shared between
192 the original and the copy. This function is called recursively by
193 binding TABLE to a sub char table. */
196 copy_category_table (table
)
202 if (!NILP (XCHAR_TABLE (table
)->top
))
204 /* TABLE is a top level char table.
205 At first, make a copy of tree structure of the table. */
206 table
= Fcopy_sequence (table
);
208 /* Then, copy elements for single byte characters one by one. */
209 for (i
= 0; i
< CHAR_TABLE_SINGLE_BYTE_SLOTS
; i
++)
210 if (!NILP (tmp
= XCHAR_TABLE (table
)->contents
[i
]))
211 XCHAR_TABLE (table
)->contents
[i
] = Fcopy_sequence (tmp
);
212 to
= CHAR_TABLE_ORDINARY_SLOTS
;
214 /* Also copy the first (and sole) extra slot. It is a vector
215 containing docstring of each category. */
216 Fset_char_table_extra_slot
217 (table
, make_number (0),
218 Fcopy_sequence (Fchar_table_extra_slot (table
, make_number (0))));
223 to
= SUB_CHAR_TABLE_ORDINARY_SLOTS
;
226 /* If the table has non-nil default value, copy it. */
227 if (!NILP (tmp
= XCHAR_TABLE (table
)->defalt
))
228 XCHAR_TABLE (table
)->defalt
= Fcopy_sequence (tmp
);
230 /* At last, copy the remaining elements while paying attention to a
233 if (!NILP (tmp
= XCHAR_TABLE (table
)->contents
[i
]))
234 XCHAR_TABLE (table
)->contents
[i
]
235 = (SUB_CHAR_TABLE_P (tmp
)
236 ? copy_category_table (tmp
) : Fcopy_sequence (tmp
));
241 DEFUN ("copy-category-table", Fcopy_category_table
, Scopy_category_table
,
243 "Construct a new category table and return it.\n\
244 It is a copy of the TABLE, which defaults to the standard category table.")
249 check_category_table (table
);
251 table
= Vstandard_category_table
;
253 return copy_category_table (table
);
256 DEFUN ("make-category-table", Fmake_category_table
, Smake_category_table
,
258 "Construct a new and empty category table and return it.")
263 val
= Fmake_char_table (Qcategory_table
, Qnil
);
264 XCHAR_TABLE (val
)->defalt
= MAKE_CATEGORY_SET
;
265 Fset_char_table_extra_slot (val
, make_number (0),
266 Fmake_vector (make_number (95), Qnil
));
270 DEFUN ("set-category-table", Fset_category_table
, Sset_category_table
, 1, 1, 0,
271 "Specify TABLE as the category table for the current buffer.")
276 table
= check_category_table (table
);
277 current_buffer
->category_table
= table
;
278 /* Indicate that this buffer now has a specified category table. */
279 idx
= PER_BUFFER_VAR_IDX (category_table
);
280 SET_PER_BUFFER_VALUE_P (current_buffer
, idx
, 1);
285 DEFUN ("char-category-set", Fchar_category_set
, Schar_category_set
, 1, 1, 0,
286 "Return the category set of CHAR.")
292 unsigned char c1
, c2
;
294 CHECK_NUMBER (ch
, 0);
295 return CATEGORY_SET (XFASTINT (ch
));
298 DEFUN ("category-set-mnemonics", Fcategory_set_mnemonics
,
299 Scategory_set_mnemonics
, 1, 1, 0,
300 "Return a string containing mnemonics of the categories in CATEGORY-SET.\n\
301 CATEGORY-SET is a bool-vector, and the categories \"in\" it are those\n\
302 that are indexes where t occurs the bool-vector.\n\
303 The return value is a string containing those same categories.")
305 Lisp_Object category_set
;
310 CHECK_CATEGORY_SET (category_set
, 0);
313 for (i
= 32; i
< 127; i
++)
314 if (CATEGORY_MEMBER (i
, category_set
))
318 return build_string (str
);
321 /* Modify all category sets stored under sub char-table TABLE so that
322 they contain (SET_VALUE is t) or don't contain (SET_VALUE is nil)
326 modify_lower_category_set (table
, category
, set_value
)
327 Lisp_Object table
, category
, set_value
;
332 val
= XCHAR_TABLE (table
)->defalt
;
333 if (!CATEGORY_SET_P (val
))
334 val
= MAKE_CATEGORY_SET
;
335 SET_CATEGORY_SET (val
, category
, set_value
);
336 XCHAR_TABLE (table
)->defalt
= val
;
338 for (i
= 32; i
< SUB_CHAR_TABLE_ORDINARY_SLOTS
; i
++)
340 val
= XCHAR_TABLE (table
)->contents
[i
];
342 if (CATEGORY_SET_P (val
))
343 SET_CATEGORY_SET (val
, category
, set_value
);
344 else if (SUB_CHAR_TABLE_P (val
))
345 modify_lower_category_set (val
, category
, set_value
);
350 set_category_set (category_set
, category
, val
)
351 Lisp_Object category_set
, category
, val
;
354 int idx
= XINT (category
) / 8;
355 unsigned char bits
= 1 << (XINT (category
) % 8);
358 XCATEGORY_SET (category_set
)->data
[idx
] &= ~bits
;
360 XCATEGORY_SET (category_set
)->data
[idx
] |= bits
;
364 DEFUN ("modify-category-entry", Fmodify_category_entry
,
365 Smodify_category_entry
, 2, 4, 0,
366 "Modify the category set of CHARACTER by adding CATEGORY to it.\n\
367 The category is changed only for table TABLE, which defaults to\n\
368 the current buffer's category table.\n\
369 If optional fourth argument RESET is non-nil,\n\
370 then delete CATEGORY from the category set instead of adding it.")
371 (character
, category
, table
, reset
)
372 Lisp_Object character
, category
, table
, reset
;
374 int c
, charset
, c1
, c2
;
375 Lisp_Object set_value
; /* Actual value to be set in category sets. */
376 Lisp_Object val
, category_set
;
378 CHECK_NUMBER (character
, 0);
379 c
= XINT (character
);
380 CHECK_CATEGORY (category
, 1);
381 table
= check_category_table (table
);
383 if (NILP (CATEGORY_DOCSTRING (table
, XFASTINT (category
))))
384 error ("Undefined category: %c", XFASTINT (category
));
386 set_value
= NILP (reset
) ? Qt
: Qnil
;
388 if (c
< CHAR_TABLE_SINGLE_BYTE_SLOTS
)
390 val
= XCHAR_TABLE (table
)->contents
[c
];
391 if (!CATEGORY_SET_P (val
))
392 XCHAR_TABLE (table
)->contents
[c
] = (val
= MAKE_CATEGORY_SET
);
393 SET_CATEGORY_SET (val
, category
, set_value
);
397 SPLIT_CHAR (c
, charset
, c1
, c2
);
399 /* The top level table. */
400 val
= XCHAR_TABLE (table
)->contents
[charset
+ 128];
401 if (CATEGORY_SET_P (val
))
403 else if (!SUB_CHAR_TABLE_P (val
))
405 category_set
= val
= MAKE_CATEGORY_SET
;
406 XCHAR_TABLE (table
)->contents
[charset
+ 128] = category_set
;
411 /* Only a charset is specified. */
412 if (SUB_CHAR_TABLE_P (val
))
413 /* All characters in CHARSET should be the same as for having
415 modify_lower_category_set (val
, category
, set_value
);
417 SET_CATEGORY_SET (category_set
, category
, set_value
);
421 /* The second level table. */
422 if (!SUB_CHAR_TABLE_P (val
))
424 val
= make_sub_char_table (Qnil
);
425 XCHAR_TABLE (table
)->contents
[charset
+ 128] = val
;
426 /* We must set default category set of CHARSET in `defalt' slot. */
427 XCHAR_TABLE (val
)->defalt
= category_set
;
431 val
= XCHAR_TABLE (table
)->contents
[c1
];
432 if (CATEGORY_SET_P (val
))
434 else if (!SUB_CHAR_TABLE_P (val
))
436 category_set
= val
= Fcopy_sequence (XCHAR_TABLE (table
)->defalt
);
437 XCHAR_TABLE (table
)->contents
[c1
] = category_set
;
442 if (SUB_CHAR_TABLE_P (val
))
443 /* All characters in C1 group of CHARSET should be the same as
445 modify_lower_category_set (val
, category
, set_value
);
447 SET_CATEGORY_SET (category_set
, category
, set_value
);
451 /* The third (bottom) level table. */
452 if (!SUB_CHAR_TABLE_P (val
))
454 val
= make_sub_char_table (Qnil
);
455 XCHAR_TABLE (table
)->contents
[c1
] = val
;
456 /* We must set default category set of CHARSET and C1 in
458 XCHAR_TABLE (val
)->defalt
= category_set
;
462 val
= XCHAR_TABLE (table
)->contents
[c2
];
463 if (CATEGORY_SET_P (val
))
465 else if (!SUB_CHAR_TABLE_P (val
))
467 category_set
= Fcopy_sequence (XCHAR_TABLE (table
)->defalt
);
468 XCHAR_TABLE (table
)->contents
[c2
] = category_set
;
471 /* This should never happen. */
472 error ("Invalid category table");
474 SET_CATEGORY_SET (category_set
, category
, set_value
);
479 /* Dump category table to buffer in human-readable format */
482 describe_category (value
)
485 Lisp_Object mnemonics
;
487 Findent_to (make_number (16), make_number (1));
491 insert_string ("default\n");
495 if (CHAR_TABLE_P (value
))
497 insert_string ("deeper char-table ...\n");
501 if (!CATEGORY_SET_P (value
))
503 insert_string ("invalid\n");
507 mnemonics
= Fcategory_set_mnemonics (value
);
508 insert_from_string (mnemonics
, 0, 0, XSTRING (mnemonics
)->size
,
509 STRING_BYTES (XSTRING (mnemonics
)), 0);
510 insert_string ("\n");
515 describe_category_1 (vector
)
518 struct buffer
*old
= current_buffer
;
519 set_buffer_internal (XBUFFER (Vstandard_output
));
520 describe_vector (vector
, Qnil
, describe_category
, 0, Qnil
, Qnil
,
524 Lisp_Object docs
= XCHAR_TABLE (vector
)->extras
[0];
527 if (!VECTORP (docs
) || XVECTOR (docs
)->size
!= 95)
529 insert_string ("Invalid first extra slot in this char table\n");
533 insert_string ("Meanings of mnemonice characters are:\n");
534 for (i
= 0; i
< 95; i
++)
536 elt
= XVECTOR (docs
)->contents
[i
];
540 insert_char (i
+ 32);
542 insert_from_string (elt
, 0, 0, XSTRING (elt
)->size
,
543 STRING_BYTES (XSTRING (elt
)), 0);
548 while (! NILP (XCHAR_TABLE (vector
)->parent
))
550 vector
= XCHAR_TABLE (vector
)->parent
;
551 insert_string ("\nThe parent category table is:");
552 describe_vector (vector
, Qnil
, describe_category
, 0, Qnil
, Qnil
,
556 call0 (intern ("help-mode"));
557 set_buffer_internal (old
);
561 DEFUN ("describe-categories", Fdescribe_categories
, Sdescribe_categories
, 0, 0, "",
562 "Describe the category specifications in the current category table.\n\
563 The descriptions are inserted in a buffer, which is then displayed.")
566 internal_with_output_to_temp_buffer
567 ("*Help*", describe_category_1
, current_buffer
->category_table
);
572 /* Return 1 if there is a word boundary between two word-constituent
573 characters C1 and C2 if they appear in this order, else return 0.
574 Use the macro WORD_BOUNDARY_P instead of calling this function
578 word_boundary_p (c1
, c2
)
581 Lisp_Object category_set1
, category_set2
;
585 if (CHAR_CHARSET (c1
) == CHAR_CHARSET (c2
))
587 tail
= Vword_separating_categories
;
592 tail
= Vword_combining_categories
;
596 category_set1
= CATEGORY_SET (c1
);
597 if (NILP (category_set1
))
598 return default_result
;
599 category_set2
= CATEGORY_SET (c2
);
600 if (NILP (category_set2
))
601 return default_result
;
603 for (; CONSP (tail
); tail
= XCDR (tail
))
605 Lisp_Object elt
= XCAR (tail
);
608 && CATEGORYP (XCAR (elt
))
609 && CATEGORYP (XCDR (elt
))
610 && CATEGORY_MEMBER (XFASTINT (XCAR (elt
)), category_set1
)
611 && CATEGORY_MEMBER (XFASTINT (XCDR (elt
)), category_set2
))
612 return !default_result
;
614 return default_result
;
619 init_category_once ()
621 /* This has to be done here, before we call Fmake_char_table. */
622 Qcategory_table
= intern ("category-table");
623 staticpro (&Qcategory_table
);
625 /* Intern this now in case it isn't already done.
626 Setting this variable twice is harmless.
627 But don't staticpro it here--that is done in alloc.c. */
628 Qchar_table_extra_slots
= intern ("char-table-extra-slots");
630 /* Now we are ready to set up this property, so we can
631 create category tables. */
632 Fput (Qcategory_table
, Qchar_table_extra_slots
, make_number (2));
634 Vstandard_category_table
= Fmake_char_table (Qcategory_table
, Qnil
);
635 /* Set a category set which contains nothing to the default. */
636 XCHAR_TABLE (Vstandard_category_table
)->defalt
= MAKE_CATEGORY_SET
;
637 Fset_char_table_extra_slot (Vstandard_category_table
, make_number (0),
638 Fmake_vector (make_number (95), Qnil
));
644 Qcategoryp
= intern ("categoryp");
645 staticpro (&Qcategoryp
);
646 Qcategorysetp
= intern ("categorysetp");
647 staticpro (&Qcategorysetp
);
648 Qcategory_table_p
= intern ("category-table-p");
649 staticpro (&Qcategory_table_p
);
651 DEFVAR_LISP ("word-combining-categories", &Vword_combining_categories
,
652 "List of pair (cons) of categories to determine word boundary.\n\
654 Emacs treats a sequence of word constituent characters as a single\n\
655 word (i.e. finds no word boundary between them) iff they belongs to\n\
656 the same charset. But, exceptions are allowed in the following cases.\n\
658 \(1) The case that characters are in different charsets is controlled\n\
659 by the variable `word-combining-categories'.\n\
661 Emacs finds no word boundary between characters of different charsets\n\
662 if they have categories matching some element of this list.\n\
664 More precisely, if an element of this list is a cons of category CAT1\n\
665 and CAT2, and a multibyte character C1 which has CAT1 is followed by\n\
666 C2 which has CAT2, there's no word boundary between C1 and C2.\n\
668 For instance, to tell that ASCII characters and Latin-1 characters can\n\
669 form a single word, the element `(?l . ?l)' should be in this list\n\
670 because both characters have the category `l' (Latin characters).\n\
672 \(2) The case that character are in the same charset is controlled by\n\
673 the variable `word-separating-categories'.\n\
675 Emacs find a word boundary between characters of the same charset\n\
676 if they have categories matching some element of this list.\n\
678 More precisely, if an element of this list is a cons of category CAT1\n\
679 and CAT2, and a multibyte character C1 which has CAT1 is followed by\n\
680 C2 which has CAT2, there's a word boundary between C1 and C2.\n\
682 For instance, to tell that there's a word boundary between Japanese\n\
683 Hiragana and Japanese Kanji (both are in the same charset), the\n\
684 element `(?H . ?C) should be in this list.");
686 Vword_combining_categories
= Qnil
;
688 DEFVAR_LISP ("word-separating-categories", &Vword_separating_categories
,
689 "List of pair (cons) of categories to determine word boundary.\n\
690 See the documentation of the variable `word-combining-categories'.");
692 Vword_separating_categories
= Qnil
;
694 defsubr (&Smake_category_set
);
695 defsubr (&Sdefine_category
);
696 defsubr (&Scategory_docstring
);
697 defsubr (&Sget_unused_category
);
698 defsubr (&Scategory_table_p
);
699 defsubr (&Scategory_table
);
700 defsubr (&Sstandard_category_table
);
701 defsubr (&Scopy_category_table
);
702 defsubr (&Smake_category_table
);
703 defsubr (&Sset_category_table
);
704 defsubr (&Schar_category_set
);
705 defsubr (&Scategory_set_mnemonics
);
706 defsubr (&Smodify_category_entry
);
707 defsubr (&Sdescribe_categories
);
709 category_table_version
= 0;