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. */
34 /* The version number of the latest category table. Each category
35 table has a unique version number. It is assigned a new number
36 also when it is modified. When a regular expression is compiled
37 into the struct re_pattern_buffer, the version number of the
38 category table (of the current buffer) at that moment is also
39 embedded in the structure.
41 For the moment, we are not using this feature. */
42 static int category_table_version
;
44 Lisp_Object Qcategory_table
, Qcategoryp
, Qcategorysetp
, Qcategory_table_p
;
46 /* Variables to determine word boundary. */
47 Lisp_Object Vword_combining_categories
, Vword_separating_categories
;
49 /* Temporary internal variable used in macro CHAR_HAS_CATEGORY. */
50 Lisp_Object _temp_category_set
;
53 /* Category set staff. */
55 DEFUN ("make-category-set", Fmake_category_set
, Smake_category_set
, 1, 1, 0,
56 doc
: /* Return a newly created category-set which contains CATEGORIES.
57 CATEGORIES is a string of category mnemonics.
58 The value is a bool-vector which has t at the indices corresponding to
61 Lisp_Object categories
;
66 CHECK_STRING (categories
);
67 val
= MAKE_CATEGORY_SET
;
69 if (STRING_MULTIBYTE (categories
))
70 error ("Multibyte string in make-category-set");
72 len
= XSTRING (categories
)->size
;
77 XSETFASTINT (category
, XSTRING (categories
)->data
[len
]);
78 CHECK_CATEGORY (category
);
79 SET_CATEGORY_SET (val
, category
, Qt
);
87 Lisp_Object
check_category_table ();
89 DEFUN ("define-category", Fdefine_category
, Sdefine_category
, 2, 3, 0,
90 doc
: /* Define CHAR as a category which is described by DOCSTRING.
91 CHAR should be an ASCII printing character in the range ` ' to `~'.
92 DOCSTRING is a documentation string of the category.
93 The category is defined only in category table TABLE, which defaults to
94 the current buffer's category table. */)
95 (category
, docstring
, table
)
96 Lisp_Object category
, docstring
, table
;
98 CHECK_CATEGORY (category
);
99 CHECK_STRING (docstring
);
100 table
= check_category_table (table
);
102 if (!NILP (CATEGORY_DOCSTRING (table
, XFASTINT (category
))))
103 error ("Category `%c' is already defined", XFASTINT (category
));
104 CATEGORY_DOCSTRING (table
, XFASTINT (category
)) = docstring
;
109 DEFUN ("category-docstring", Fcategory_docstring
, Scategory_docstring
, 1, 2, 0,
110 doc
: /* Return the documentation string of CATEGORY, as defined in CATEGORY-TABLE. */)
112 Lisp_Object category
, table
;
114 CHECK_CATEGORY (category
);
115 table
= check_category_table (table
);
117 return CATEGORY_DOCSTRING (table
, XFASTINT (category
));
120 DEFUN ("get-unused-category", Fget_unused_category
, Sget_unused_category
,
122 doc
: /* Return a category which is not yet defined in CATEGORY-TABLE.
123 If no category remains available, return nil.
124 The optional argument CATEGORY-TABLE
125 specifies which category table to modify;
126 it defaults to the current buffer's category table. */)
132 table
= check_category_table (table
);
134 for (i
= ' '; i
<= '~'; i
++)
135 if (NILP (CATEGORY_DOCSTRING (table
, i
)))
136 return make_number (i
);
142 /* Category-table staff. */
144 DEFUN ("category-table-p", Fcategory_table_p
, Scategory_table_p
, 1, 1, 0,
145 doc
: /* Return t if ARG is a category table. */)
149 if (CHAR_TABLE_P (arg
)
150 && EQ (XCHAR_TABLE (arg
)->purpose
, Qcategory_table
))
155 /* If TABLE is nil, return the current category table. If TABLE is
156 not nil, check the validity of TABLE as a category table. If
157 valid, return TABLE itself, but if not valid, signal an error of
158 wrong-type-argument. */
161 check_category_table (table
)
164 register Lisp_Object tem
;
166 return current_buffer
->category_table
;
167 while (tem
= Fcategory_table_p (table
), NILP (tem
))
168 table
= wrong_type_argument (Qcategory_table_p
, table
);
172 DEFUN ("category-table", Fcategory_table
, Scategory_table
, 0, 0, 0,
173 doc
: /* Return the current category table.
174 This is the one specified by the current buffer. */)
177 return current_buffer
->category_table
;
180 DEFUN ("standard-category-table", Fstandard_category_table
,
181 Sstandard_category_table
, 0, 0, 0,
182 doc
: /* Return the standard category table.
183 This is the one used for new buffers. */)
186 return Vstandard_category_table
;
189 /* Return a copy of category table TABLE. We can't simply use the
190 function copy-sequence because no contents should be shared between
191 the original and the copy. This function is called recursively by
192 binding TABLE to a sub char table. */
195 copy_category_table (table
)
201 if (!NILP (XCHAR_TABLE (table
)->top
))
203 /* TABLE is a top level char table.
204 At first, make a copy of tree structure of the table. */
205 table
= Fcopy_sequence (table
);
207 /* Then, copy elements for single byte characters one by one. */
208 for (i
= 0; i
< CHAR_TABLE_SINGLE_BYTE_SLOTS
; i
++)
209 if (!NILP (tmp
= XCHAR_TABLE (table
)->contents
[i
]))
210 XCHAR_TABLE (table
)->contents
[i
] = Fcopy_sequence (tmp
);
211 to
= CHAR_TABLE_ORDINARY_SLOTS
;
213 /* Also copy the first (and sole) extra slot. It is a vector
214 containing docstring of each category. */
215 Fset_char_table_extra_slot
216 (table
, make_number (0),
217 Fcopy_sequence (Fchar_table_extra_slot (table
, make_number (0))));
222 to
= SUB_CHAR_TABLE_ORDINARY_SLOTS
;
225 /* If the table has non-nil default value, copy it. */
226 if (!NILP (tmp
= XCHAR_TABLE (table
)->defalt
))
227 XCHAR_TABLE (table
)->defalt
= Fcopy_sequence (tmp
);
229 /* At last, copy the remaining elements while paying attention to a
232 if (!NILP (tmp
= XCHAR_TABLE (table
)->contents
[i
]))
233 XCHAR_TABLE (table
)->contents
[i
]
234 = (SUB_CHAR_TABLE_P (tmp
)
235 ? copy_category_table (tmp
) : Fcopy_sequence (tmp
));
240 DEFUN ("copy-category-table", Fcopy_category_table
, Scopy_category_table
,
242 doc
: /* Construct a new category table and return it.
243 It is a copy of the TABLE, which defaults to the standard category table. */)
248 check_category_table (table
);
250 table
= Vstandard_category_table
;
252 return copy_category_table (table
);
255 DEFUN ("make-category-table", Fmake_category_table
, Smake_category_table
,
257 doc
: /* Construct a new and empty category table and return it. */)
262 val
= Fmake_char_table (Qcategory_table
, Qnil
);
263 XCHAR_TABLE (val
)->defalt
= MAKE_CATEGORY_SET
;
264 Fset_char_table_extra_slot (val
, make_number (0),
265 Fmake_vector (make_number (95), Qnil
));
269 DEFUN ("set-category-table", Fset_category_table
, Sset_category_table
, 1, 1, 0,
270 doc
: /* Specify TABLE as the category table for the current buffer. */)
275 table
= check_category_table (table
);
276 current_buffer
->category_table
= table
;
277 /* Indicate that this buffer now has a specified category table. */
278 idx
= PER_BUFFER_VAR_IDX (category_table
);
279 SET_PER_BUFFER_VALUE_P (current_buffer
, idx
, 1);
284 DEFUN ("char-category-set", Fchar_category_set
, Schar_category_set
, 1, 1, 0,
285 doc
: /* Return the category set of CHAR. */)
290 return CATEGORY_SET (XFASTINT (ch
));
293 DEFUN ("category-set-mnemonics", Fcategory_set_mnemonics
,
294 Scategory_set_mnemonics
, 1, 1, 0,
295 doc
: /* Return a string containing mnemonics of the categories in CATEGORY-SET.
296 CATEGORY-SET is a bool-vector, and the categories \"in\" it are those
297 that are indexes where t occurs the bool-vector.
298 The return value is a string containing those same categories. */)
300 Lisp_Object category_set
;
305 CHECK_CATEGORY_SET (category_set
);
308 for (i
= 32; i
< 127; i
++)
309 if (CATEGORY_MEMBER (i
, category_set
))
313 return build_string (str
);
316 /* Modify all category sets stored under sub char-table TABLE so that
317 they contain (SET_VALUE is t) or don't contain (SET_VALUE is nil)
321 modify_lower_category_set (table
, category
, set_value
)
322 Lisp_Object table
, category
, set_value
;
327 val
= XCHAR_TABLE (table
)->defalt
;
328 if (!CATEGORY_SET_P (val
))
329 val
= MAKE_CATEGORY_SET
;
330 SET_CATEGORY_SET (val
, category
, set_value
);
331 XCHAR_TABLE (table
)->defalt
= val
;
333 for (i
= 32; i
< SUB_CHAR_TABLE_ORDINARY_SLOTS
; i
++)
335 val
= XCHAR_TABLE (table
)->contents
[i
];
337 if (CATEGORY_SET_P (val
))
338 SET_CATEGORY_SET (val
, category
, set_value
);
339 else if (SUB_CHAR_TABLE_P (val
))
340 modify_lower_category_set (val
, category
, set_value
);
345 set_category_set (category_set
, category
, val
)
346 Lisp_Object category_set
, category
, val
;
349 int idx
= XINT (category
) / 8;
350 unsigned char bits
= 1 << (XINT (category
) % 8);
353 XCATEGORY_SET (category_set
)->data
[idx
] &= ~bits
;
355 XCATEGORY_SET (category_set
)->data
[idx
] |= bits
;
359 DEFUN ("modify-category-entry", Fmodify_category_entry
,
360 Smodify_category_entry
, 2, 4, 0,
361 doc
: /* Modify the category set of CHARACTER by adding CATEGORY to it.
362 The category is changed only for table TABLE, which defaults to
363 the current buffer's category table.
364 If optional fourth argument RESET is non-nil,
365 then delete CATEGORY from the category set instead of adding it. */)
366 (character
, category
, table
, reset
)
367 Lisp_Object character
, category
, table
, reset
;
369 int c
, charset
, c1
, c2
;
370 Lisp_Object set_value
; /* Actual value to be set in category sets. */
371 Lisp_Object val
, category_set
;
373 CHECK_NUMBER (character
);
374 c
= XINT (character
);
375 CHECK_CATEGORY (category
);
376 table
= check_category_table (table
);
378 if (NILP (CATEGORY_DOCSTRING (table
, XFASTINT (category
))))
379 error ("Undefined category: %c", XFASTINT (category
));
381 set_value
= NILP (reset
) ? Qt
: Qnil
;
383 if (c
< CHAR_TABLE_SINGLE_BYTE_SLOTS
)
385 val
= XCHAR_TABLE (table
)->contents
[c
];
386 if (!CATEGORY_SET_P (val
))
387 XCHAR_TABLE (table
)->contents
[c
] = (val
= MAKE_CATEGORY_SET
);
388 SET_CATEGORY_SET (val
, category
, set_value
);
392 SPLIT_CHAR (c
, charset
, c1
, c2
);
394 /* The top level table. */
395 val
= XCHAR_TABLE (table
)->contents
[charset
+ 128];
396 if (CATEGORY_SET_P (val
))
398 else if (!SUB_CHAR_TABLE_P (val
))
400 category_set
= val
= MAKE_CATEGORY_SET
;
401 XCHAR_TABLE (table
)->contents
[charset
+ 128] = category_set
;
406 /* Only a charset is specified. */
407 if (SUB_CHAR_TABLE_P (val
))
408 /* All characters in CHARSET should be the same as for having
410 modify_lower_category_set (val
, category
, set_value
);
412 SET_CATEGORY_SET (category_set
, category
, set_value
);
416 /* The second level table. */
417 if (!SUB_CHAR_TABLE_P (val
))
419 val
= make_sub_char_table (Qnil
);
420 XCHAR_TABLE (table
)->contents
[charset
+ 128] = val
;
421 /* We must set default category set of CHARSET in `defalt' slot. */
422 XCHAR_TABLE (val
)->defalt
= category_set
;
426 val
= XCHAR_TABLE (table
)->contents
[c1
];
427 if (CATEGORY_SET_P (val
))
429 else if (!SUB_CHAR_TABLE_P (val
))
431 category_set
= val
= Fcopy_sequence (XCHAR_TABLE (table
)->defalt
);
432 XCHAR_TABLE (table
)->contents
[c1
] = category_set
;
437 if (SUB_CHAR_TABLE_P (val
))
438 /* All characters in C1 group of CHARSET should be the same as
440 modify_lower_category_set (val
, category
, set_value
);
442 SET_CATEGORY_SET (category_set
, category
, set_value
);
446 /* The third (bottom) level table. */
447 if (!SUB_CHAR_TABLE_P (val
))
449 val
= make_sub_char_table (Qnil
);
450 XCHAR_TABLE (table
)->contents
[c1
] = val
;
451 /* We must set default category set of CHARSET and C1 in
453 XCHAR_TABLE (val
)->defalt
= category_set
;
457 val
= XCHAR_TABLE (table
)->contents
[c2
];
458 if (CATEGORY_SET_P (val
))
460 else if (!SUB_CHAR_TABLE_P (val
))
462 category_set
= Fcopy_sequence (XCHAR_TABLE (table
)->defalt
);
463 XCHAR_TABLE (table
)->contents
[c2
] = category_set
;
466 /* This should never happen. */
467 error ("Invalid category table");
469 SET_CATEGORY_SET (category_set
, category
, set_value
);
474 /* Dump category table to buffer in human-readable format */
477 describe_category (value
)
480 Lisp_Object mnemonics
;
482 Findent_to (make_number (16), make_number (1));
486 insert_string ("default\n");
490 if (CHAR_TABLE_P (value
))
492 insert_string ("deeper char-table ...\n");
496 if (!CATEGORY_SET_P (value
))
498 insert_string ("invalid\n");
502 mnemonics
= Fcategory_set_mnemonics (value
);
503 insert_from_string (mnemonics
, 0, 0, XSTRING (mnemonics
)->size
,
504 STRING_BYTES (XSTRING (mnemonics
)), 0);
505 insert_string ("\n");
510 describe_category_1 (vector
)
513 struct buffer
*old
= current_buffer
;
514 set_buffer_internal (XBUFFER (Vstandard_output
));
515 describe_vector (vector
, Qnil
, describe_category
, 0, Qnil
, Qnil
,
519 Lisp_Object docs
= XCHAR_TABLE (vector
)->extras
[0];
522 if (!VECTORP (docs
) || XVECTOR (docs
)->size
!= 95)
524 insert_string ("Invalid first extra slot in this char table\n");
528 insert_string ("Meanings of mnemonice characters are:\n");
529 for (i
= 0; i
< 95; i
++)
531 elt
= XVECTOR (docs
)->contents
[i
];
535 insert_char (i
+ 32);
537 insert_from_string (elt
, 0, 0, XSTRING (elt
)->size
,
538 STRING_BYTES (XSTRING (elt
)), 0);
543 while (! NILP (XCHAR_TABLE (vector
)->parent
))
545 vector
= XCHAR_TABLE (vector
)->parent
;
546 insert_string ("\nThe parent category table is:");
547 describe_vector (vector
, Qnil
, describe_category
, 0, Qnil
, Qnil
,
551 call0 (intern ("help-mode"));
552 set_buffer_internal (old
);
556 DEFUN ("describe-categories", Fdescribe_categories
, Sdescribe_categories
, 0, 0, "",
557 doc
: /* Describe the category specifications in the current category table.
558 The descriptions are inserted in a buffer, which is then displayed. */)
561 internal_with_output_to_temp_buffer
562 ("*Help*", describe_category_1
, current_buffer
->category_table
);
567 /* Return 1 if there is a word boundary between two word-constituent
568 characters C1 and C2 if they appear in this order, else return 0.
569 Use the macro WORD_BOUNDARY_P instead of calling this function
573 word_boundary_p (c1
, c2
)
576 Lisp_Object category_set1
, category_set2
;
580 if (CHAR_CHARSET (c1
) == CHAR_CHARSET (c2
))
582 tail
= Vword_separating_categories
;
587 tail
= Vword_combining_categories
;
591 category_set1
= CATEGORY_SET (c1
);
592 if (NILP (category_set1
))
593 return default_result
;
594 category_set2
= CATEGORY_SET (c2
);
595 if (NILP (category_set2
))
596 return default_result
;
598 for (; CONSP (tail
); tail
= XCDR (tail
))
600 Lisp_Object elt
= XCAR (tail
);
603 && CATEGORYP (XCAR (elt
))
604 && CATEGORYP (XCDR (elt
))
605 && CATEGORY_MEMBER (XFASTINT (XCAR (elt
)), category_set1
)
606 && CATEGORY_MEMBER (XFASTINT (XCDR (elt
)), category_set2
))
607 return !default_result
;
609 return default_result
;
614 init_category_once ()
616 /* This has to be done here, before we call Fmake_char_table. */
617 Qcategory_table
= intern ("category-table");
618 staticpro (&Qcategory_table
);
620 /* Intern this now in case it isn't already done.
621 Setting this variable twice is harmless.
622 But don't staticpro it here--that is done in alloc.c. */
623 Qchar_table_extra_slots
= intern ("char-table-extra-slots");
625 /* Now we are ready to set up this property, so we can
626 create category tables. */
627 Fput (Qcategory_table
, Qchar_table_extra_slots
, make_number (2));
629 Vstandard_category_table
= Fmake_char_table (Qcategory_table
, Qnil
);
630 /* Set a category set which contains nothing to the default. */
631 XCHAR_TABLE (Vstandard_category_table
)->defalt
= MAKE_CATEGORY_SET
;
632 Fset_char_table_extra_slot (Vstandard_category_table
, make_number (0),
633 Fmake_vector (make_number (95), Qnil
));
639 Qcategoryp
= intern ("categoryp");
640 staticpro (&Qcategoryp
);
641 Qcategorysetp
= intern ("categorysetp");
642 staticpro (&Qcategorysetp
);
643 Qcategory_table_p
= intern ("category-table-p");
644 staticpro (&Qcategory_table_p
);
646 DEFVAR_LISP ("word-combining-categories", &Vword_combining_categories
,
647 doc
: /* List of pair (cons) of categories to determine word boundary.
649 Emacs treats a sequence of word constituent characters as a single
650 word (i.e. finds no word boundary between them) iff they belongs to
651 the same charset. But, exceptions are allowed in the following cases.
653 \(1) The case that characters are in different charsets is controlled
654 by the variable `word-combining-categories'.
656 Emacs finds no word boundary between characters of different charsets
657 if they have categories matching some element of this list.
659 More precisely, if an element of this list is a cons of category CAT1
660 and CAT2, and a multibyte character C1 which has CAT1 is followed by
661 C2 which has CAT2, there's no word boundary between C1 and C2.
663 For instance, to tell that ASCII characters and Latin-1 characters can
664 form a single word, the element `(?l . ?l)' should be in this list
665 because both characters have the category `l' (Latin characters).
667 \(2) The case that character are in the same charset is controlled by
668 the variable `word-separating-categories'.
670 Emacs find a word boundary between characters of the same charset
671 if they have categories matching some element of this list.
673 More precisely, if an element of this list is a cons of category CAT1
674 and CAT2, and a multibyte character C1 which has CAT1 is followed by
675 C2 which has CAT2, there's a word boundary between C1 and C2.
677 For instance, to tell that there's a word boundary between Japanese
678 Hiragana and Japanese Kanji (both are in the same charset), the
679 element `(?H . ?C) should be in this list. */);
681 Vword_combining_categories
= Qnil
;
683 DEFVAR_LISP ("word-separating-categories", &Vword_separating_categories
,
684 doc
: /* List of pair (cons) of categories to determine word boundary.
685 See the documentation of the variable `word-combining-categories'. */);
687 Vword_separating_categories
= Qnil
;
689 defsubr (&Smake_category_set
);
690 defsubr (&Sdefine_category
);
691 defsubr (&Scategory_docstring
);
692 defsubr (&Sget_unused_category
);
693 defsubr (&Scategory_table_p
);
694 defsubr (&Scategory_table
);
695 defsubr (&Sstandard_category_table
);
696 defsubr (&Scopy_category_table
);
697 defsubr (&Smake_category_table
);
698 defsubr (&Sset_category_table
);
699 defsubr (&Schar_category_set
);
700 defsubr (&Scategory_set_mnemonics
);
701 defsubr (&Smodify_category_entry
);
702 defsubr (&Sdescribe_categories
);
704 category_table_version
= 0;