1 /* Basic character support.
3 Copyright (C) 2001-2012 Free Software Foundation, Inc.
4 Copyright (C) 1995, 1997, 1998, 2001 Electrotechnical Laboratory, JAPAN.
5 Licensed to the Free Software Foundation.
6 Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
7 National Institute of Advanced Industrial Science and Technology (AIST)
8 Registration Number H13PRO009
10 This file is part of GNU Emacs.
12 GNU Emacs is free software: you can redistribute it and/or modify
13 it under the terms of the GNU General Public License as published by
14 the Free Software Foundation, either version 3 of the License, or
15 (at your option) any later version.
17 GNU Emacs is distributed in the hope that it will be useful,
18 but WITHOUT ANY WARRANTY; without even the implied warranty of
19 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
20 GNU General Public License for more details.
22 You should have received a copy of the GNU General Public License
23 along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */
25 /* At first, see the document in `character.h' to understand the code
32 #define CHARACTER_INLINE EXTERN_INLINE
38 #include <sys/types.h>
42 #include "character.h"
45 #include "composite.h"
54 Lisp_Object Qcharacterp
;
56 static Lisp_Object Qauto_fill_chars
;
58 /* Char-table of information about which character to unify to which
59 Unicode character. Mainly used by the macro MAYBE_UNIFY_CHAR. */
60 Lisp_Object Vchar_unify_table
;
62 static Lisp_Object Qchar_script_table
;
66 /* If character code C has modifier masks, reflect them to the
67 character code if possible. Return the resulting code. */
70 char_resolve_modifier_mask (EMACS_INT c
)
72 /* A non-ASCII character can't reflect modifier bits to the code. */
73 if (! ASCII_CHAR_P ((c
& ~CHAR_MODIFIER_MASK
)))
76 /* For Meta, Shift, and Control modifiers, we need special care. */
79 /* Shift modifier is valid only with [A-Za-z]. */
80 if ((c
& 0377) >= 'A' && (c
& 0377) <= 'Z')
82 else if ((c
& 0377) >= 'a' && (c
& 0377) <= 'z')
83 c
= (c
& ~CHAR_SHIFT
) - ('a' - 'A');
84 /* Shift modifier for control characters and SPC is ignored. */
85 else if ((c
& ~CHAR_MODIFIER_MASK
) <= 0x20)
90 /* Simulate the code in lread.c. */
91 /* Allow `\C- ' and `\C-?'. */
92 if ((c
& 0377) == ' ')
93 c
&= ~0177 & ~ CHAR_CTL
;
94 else if ((c
& 0377) == '?')
95 c
= 0177 | (c
& ~0177 & ~CHAR_CTL
);
96 /* ASCII control chars are made from letters (both cases),
97 as well as the non-letters within 0100...0137. */
98 else if ((c
& 0137) >= 0101 && (c
& 0137) <= 0132)
99 c
&= (037 | (~0177 & ~CHAR_CTL
));
100 else if ((c
& 0177) >= 0100 && (c
& 0177) <= 0137)
101 c
&= (037 | (~0177 & ~CHAR_CTL
));
103 #if 0 /* This is outside the scope of this function. (bug#4751) */
106 /* Move the meta bit to the right place for a string. */
107 c
= (c
& ~CHAR_META
) | 0x80;
115 /* Store multibyte form of character C at P. If C has modifier bits,
116 handle them appropriately. */
119 char_string (unsigned int c
, unsigned char *p
)
123 if (c
& CHAR_MODIFIER_MASK
)
125 c
= char_resolve_modifier_mask (c
);
126 /* If C still has any modifier bits, just ignore it. */
127 c
&= ~CHAR_MODIFIER_MASK
;
130 MAYBE_UNIFY_CHAR (c
);
132 if (c
<= MAX_3_BYTE_CHAR
)
134 bytes
= CHAR_STRING (c
, p
);
136 else if (c
<= MAX_4_BYTE_CHAR
)
138 p
[0] = (0xF0 | (c
>> 18));
139 p
[1] = (0x80 | ((c
>> 12) & 0x3F));
140 p
[2] = (0x80 | ((c
>> 6) & 0x3F));
141 p
[3] = (0x80 | (c
& 0x3F));
144 else if (c
<= MAX_5_BYTE_CHAR
)
147 p
[1] = (0x80 | ((c
>> 18) & 0x0F));
148 p
[2] = (0x80 | ((c
>> 12) & 0x3F));
149 p
[3] = (0x80 | ((c
>> 6) & 0x3F));
150 p
[4] = (0x80 | (c
& 0x3F));
153 else if (c
<= MAX_CHAR
)
155 c
= CHAR_TO_BYTE8 (c
);
156 bytes
= BYTE8_STRING (c
, p
);
159 error ("Invalid character: %x", c
);
165 /* Return a character whose multibyte form is at P. If LEN is not
166 NULL, it must be a pointer to integer. In that case, set *LEN to
167 the byte length of the multibyte form. If ADVANCED is not NULL, it
168 must be a pointer to unsigned char. In that case, set *ADVANCED to
169 the ending address (i.e., the starting address of the next
170 character) of the multibyte form. */
173 string_char (const unsigned char *p
, const unsigned char **advanced
, int *len
)
176 const unsigned char *saved_p
= p
;
178 if (*p
< 0x80 || ! (*p
& 0x20) || ! (*p
& 0x10))
180 c
= STRING_CHAR_ADVANCE (p
);
182 else if (! (*p
& 0x08))
184 c
= ((((p
)[0] & 0xF) << 18)
185 | (((p
)[1] & 0x3F) << 12)
186 | (((p
)[2] & 0x3F) << 6)
192 c
= ((((p
)[1] & 0x3F) << 18)
193 | (((p
)[2] & 0x3F) << 12)
194 | (((p
)[3] & 0x3F) << 6)
199 MAYBE_UNIFY_CHAR (c
);
209 /* Translate character C by translation table TABLE. If no translation is
210 found in TABLE, return the untranslated character. If TABLE is a list,
211 elements are char tables. In that case, recursively translate C by all the
212 tables in the list. */
215 translate_char (Lisp_Object table
, int c
)
217 if (CHAR_TABLE_P (table
))
221 ch
= CHAR_TABLE_REF (table
, c
);
227 for (; CONSP (table
); table
= XCDR (table
))
228 c
= translate_char (XCAR (table
), c
);
233 /* Convert ASCII or 8-bit character C to unibyte. If C is none of
234 them, return (C & 0xFF). */
237 multibyte_char_to_unibyte (int c
)
241 if (CHAR_BYTE8_P (c
))
242 return CHAR_TO_BYTE8 (c
);
246 /* Like multibyte_char_to_unibyte, but return -1 if C is not supported
247 by charset_unibyte. */
250 multibyte_char_to_unibyte_safe (int c
)
254 if (CHAR_BYTE8_P (c
))
255 return CHAR_TO_BYTE8 (c
);
259 DEFUN ("characterp", Fcharacterp
, Scharacterp
, 1, 2, 0,
260 doc
: /* Return non-nil if OBJECT is a character.
261 usage: (characterp OBJECT) */)
262 (Lisp_Object object
, Lisp_Object ignore
)
264 return (CHARACTERP (object
) ? Qt
: Qnil
);
267 DEFUN ("max-char", Fmax_char
, Smax_char
, 0, 0, 0,
268 doc
: /* Return the character of the maximum code. */)
271 return make_number (MAX_CHAR
);
274 DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte
,
275 Sunibyte_char_to_multibyte
, 1, 1, 0,
276 doc
: /* Convert the byte CH to multibyte character. */)
281 CHECK_CHARACTER (ch
);
284 error ("Not a unibyte character: %d", c
);
285 MAKE_CHAR_MULTIBYTE (c
);
286 return make_number (c
);
289 DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte
,
290 Smultibyte_char_to_unibyte
, 1, 1, 0,
291 doc
: /* Convert the multibyte character CH to a byte.
292 If the multibyte character does not represent a byte, return -1. */)
297 CHECK_CHARACTER (ch
);
300 /* Can't distinguish a byte read from a unibyte buffer from
301 a latin1 char, so let's let it slide. */
305 int cu
= CHAR_TO_BYTE_SAFE (cm
);
306 return make_number (cu
);
311 /* Return width (columns) of C considering the buffer display table DP. */
314 char_width (int c
, struct Lisp_Char_Table
*dp
)
316 ptrdiff_t width
= CHAR_WIDTH (c
);
320 Lisp_Object disp
= DISP_CHAR_VECTOR (dp
, c
), ch
;
324 for (i
= 0, width
= 0; i
< ASIZE (disp
); i
++)
329 int w
= CHAR_WIDTH (XFASTINT (ch
));
330 if (INT_ADD_OVERFLOW (width
, w
))
340 DEFUN ("char-width", Fchar_width
, Schar_width
, 1, 1, 0,
341 doc
: /* Return width of CHAR when displayed in the current buffer.
342 The width is measured by how many columns it occupies on the screen.
343 Tab is taken to occupy `tab-width' columns.
344 usage: (char-width CHAR) */)
350 CHECK_CHARACTER (ch
);
352 width
= char_width (c
, buffer_display_table ());
353 return make_number (width
);
356 /* Return width of string STR of length LEN when displayed in the
357 current buffer. The width is measured by how many columns it
358 occupies on the screen. If PRECISION > 0, return the width of
359 longest substring that doesn't exceed PRECISION, and set number of
360 characters and bytes of the substring in *NCHARS and *NBYTES
364 c_string_width (const unsigned char *str
, ptrdiff_t len
, int precision
,
365 ptrdiff_t *nchars
, ptrdiff_t *nbytes
)
367 ptrdiff_t i
= 0, i_byte
= 0;
369 struct Lisp_Char_Table
*dp
= buffer_display_table ();
374 int c
= STRING_CHAR_AND_LENGTH (str
+ i_byte
, bytes
);
375 ptrdiff_t thiswidth
= char_width (c
, dp
);
379 if (INT_ADD_OVERFLOW (width
, thiswidth
))
382 else if (precision
- width
< thiswidth
)
402 /* Return width of string STR of length LEN when displayed in the
403 current buffer. The width is measured by how many columns it
404 occupies on the screen. */
407 strwidth (const char *str
, ptrdiff_t len
)
409 return c_string_width ((const unsigned char *) str
, len
, -1, NULL
, NULL
);
412 /* Return width of Lisp string STRING when displayed in the current
413 buffer. The width is measured by how many columns it occupies on
414 the screen while paying attention to compositions. If PRECISION >
415 0, return the width of longest substring that doesn't exceed
416 PRECISION, and set number of characters and bytes of the substring
417 in *NCHARS and *NBYTES respectively. */
420 lisp_string_width (Lisp_Object string
, ptrdiff_t precision
,
421 ptrdiff_t *nchars
, ptrdiff_t *nbytes
)
423 ptrdiff_t len
= SCHARS (string
);
424 /* This set multibyte to 0 even if STRING is multibyte when it
425 contains only ascii and eight-bit-graphic, but that's
427 int multibyte
= len
< SBYTES (string
);
428 unsigned char *str
= SDATA (string
);
429 ptrdiff_t i
= 0, i_byte
= 0;
431 struct Lisp_Char_Table
*dp
= buffer_display_table ();
435 ptrdiff_t chars
, bytes
, thiswidth
;
438 ptrdiff_t ignore
, end
;
440 if (find_composition (i
, -1, &ignore
, &end
, &val
, string
)
441 && ((cmp_id
= get_composition_id (i
, i_byte
, end
- i
, val
, string
))
444 thiswidth
= composition_table
[cmp_id
]->width
;
446 bytes
= string_char_to_byte (string
, end
) - i_byte
;
455 c
= STRING_CHAR_AND_LENGTH (str
+ i_byte
, cbytes
);
459 c
= str
[i_byte
], bytes
= 1;
461 thiswidth
= char_width (c
, dp
);
467 if (INT_ADD_OVERFLOW (width
, thiswidth
))
471 else if (precision
- width
< thiswidth
)
491 DEFUN ("string-width", Fstring_width
, Sstring_width
, 1, 1, 0,
492 doc
: /* Return width of STRING when displayed in the current buffer.
493 Width is measured by how many columns it occupies on the screen.
494 When calculating width of a multibyte character in STRING,
495 only the base leading-code is considered; the validity of
496 the following bytes is not checked. Tabs in STRING are always
497 taken to occupy `tab-width' columns.
498 usage: (string-width STRING) */)
504 XSETFASTINT (val
, lisp_string_width (str
, -1, NULL
, NULL
));
508 /* Return the number of characters in the NBYTES bytes at PTR.
509 This works by looking at the contents and checking for multibyte
510 sequences while assuming that there's no invalid sequence.
511 However, if the current buffer has enable-multibyte-characters =
512 nil, we treat each byte as a character. */
515 chars_in_text (const unsigned char *ptr
, ptrdiff_t nbytes
)
517 /* current_buffer is null at early stages of Emacs initialization. */
518 if (current_buffer
== 0
519 || NILP (BVAR (current_buffer
, enable_multibyte_characters
)))
522 return multibyte_chars_in_text (ptr
, nbytes
);
525 /* Return the number of characters in the NBYTES bytes at PTR.
526 This works by looking at the contents and checking for multibyte
527 sequences while assuming that there's no invalid sequence. It
528 ignores enable-multibyte-characters. */
531 multibyte_chars_in_text (const unsigned char *ptr
, ptrdiff_t nbytes
)
533 const unsigned char *endp
= ptr
+ nbytes
;
538 int len
= MULTIBYTE_LENGTH (ptr
, endp
);
549 /* Parse unibyte text at STR of LEN bytes as a multibyte text, count
550 characters and bytes in it, and store them in *NCHARS and *NBYTES
551 respectively. On counting bytes, pay attention to that 8-bit
552 characters not constructing a valid multibyte sequence are
553 represented by 2-byte in a multibyte text. */
556 parse_str_as_multibyte (const unsigned char *str
, ptrdiff_t len
,
557 ptrdiff_t *nchars
, ptrdiff_t *nbytes
)
559 const unsigned char *endp
= str
+ len
;
561 ptrdiff_t chars
= 0, bytes
= 0;
563 if (len
>= MAX_MULTIBYTE_LENGTH
)
565 const unsigned char *adjusted_endp
= endp
- MAX_MULTIBYTE_LENGTH
;
566 while (str
< adjusted_endp
)
568 if (! CHAR_BYTE8_HEAD_P (*str
)
569 && (n
= MULTIBYTE_LENGTH_NO_CHECK (str
)) > 0)
570 str
+= n
, bytes
+= n
;
578 if (! CHAR_BYTE8_HEAD_P (*str
)
579 && (n
= MULTIBYTE_LENGTH (str
, endp
)) > 0)
580 str
+= n
, bytes
+= n
;
591 /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text.
592 It actually converts only such 8-bit characters that don't construct
593 a multibyte sequence to multibyte forms of Latin-1 characters. If
594 NCHARS is nonzero, set *NCHARS to the number of characters in the
595 text. It is assured that we can use LEN bytes at STR as a work
596 area and that is enough. Return the number of bytes of the
600 str_as_multibyte (unsigned char *str
, ptrdiff_t len
, ptrdiff_t nbytes
,
603 unsigned char *p
= str
, *endp
= str
+ nbytes
;
608 if (nbytes
>= MAX_MULTIBYTE_LENGTH
)
610 unsigned char *adjusted_endp
= endp
- MAX_MULTIBYTE_LENGTH
;
611 while (p
< adjusted_endp
612 && ! CHAR_BYTE8_HEAD_P (*p
)
613 && (n
= MULTIBYTE_LENGTH_NO_CHECK (p
)) > 0)
617 && ! CHAR_BYTE8_HEAD_P (*p
)
618 && (n
= MULTIBYTE_LENGTH (p
, endp
)) > 0)
628 memmove (endp
- nbytes
, p
, nbytes
);
631 if (nbytes
>= MAX_MULTIBYTE_LENGTH
)
633 unsigned char *adjusted_endp
= endp
- MAX_MULTIBYTE_LENGTH
;
634 while (p
< adjusted_endp
)
636 if (! CHAR_BYTE8_HEAD_P (*p
)
637 && (n
= MULTIBYTE_LENGTH_NO_CHECK (p
)) > 0)
645 c
= BYTE8_TO_CHAR (c
);
646 to
+= CHAR_STRING (c
, to
);
653 if (! CHAR_BYTE8_HEAD_P (*p
)
654 && (n
= MULTIBYTE_LENGTH (p
, endp
)) > 0)
662 c
= BYTE8_TO_CHAR (c
);
663 to
+= CHAR_STRING (c
, to
);
672 /* Parse unibyte string at STR of LEN bytes, and return the number of
673 bytes it may occupy when converted to multibyte string by
674 `str_to_multibyte'. */
677 count_size_as_multibyte (const unsigned char *str
, ptrdiff_t len
)
679 const unsigned char *endp
= str
+ len
;
682 for (bytes
= 0; str
< endp
; str
++)
684 int n
= *str
< 0x80 ? 1 : 2;
685 if (INT_ADD_OVERFLOW (bytes
, n
))
693 /* Convert unibyte text at STR of BYTES bytes to a multibyte text
694 that contains the same single-byte characters. It actually
695 converts all 8-bit characters to multibyte forms. It is assured
696 that we can use LEN bytes at STR as a work area and that is
700 str_to_multibyte (unsigned char *str
, ptrdiff_t len
, ptrdiff_t bytes
)
702 unsigned char *p
= str
, *endp
= str
+ bytes
;
705 while (p
< endp
&& *p
< 0x80) p
++;
711 memmove (endp
- bytes
, p
, bytes
);
718 c
= BYTE8_TO_CHAR (c
);
719 to
+= CHAR_STRING (c
, to
);
724 /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It
725 actually converts characters in the range 0x80..0xFF to
729 str_as_unibyte (unsigned char *str
, ptrdiff_t bytes
)
731 const unsigned char *p
= str
, *endp
= str
+ bytes
;
738 len
= BYTES_BY_CHAR_HEAD (c
);
739 if (CHAR_BYTE8_HEAD_P (c
))
743 to
= str
+ (p
- str
);
747 len
= BYTES_BY_CHAR_HEAD (c
);
748 if (CHAR_BYTE8_HEAD_P (c
))
750 c
= STRING_CHAR_ADVANCE (p
);
751 *to
++ = CHAR_TO_BYTE8 (c
);
755 while (len
--) *to
++ = *p
++;
761 /* Convert eight-bit chars in SRC (in multibyte form) to the
762 corresponding byte and store in DST. CHARS is the number of
763 characters in SRC. The value is the number of bytes stored in DST.
764 Usually, the value is the same as CHARS, but is less than it if SRC
765 contains a non-ASCII, non-eight-bit character. If ACCEPT_LATIN_1
766 is nonzero, a Latin-1 character is accepted and converted to a byte
767 of that character code.
768 Note: Currently the arg ACCEPT_LATIN_1 is not used. */
771 str_to_unibyte (const unsigned char *src
, unsigned char *dst
, ptrdiff_t chars
, int accept_latin_1
)
775 for (i
= 0; i
< chars
; i
++)
777 int c
= STRING_CHAR_ADVANCE (src
);
779 if (CHAR_BYTE8_P (c
))
780 c
= CHAR_TO_BYTE8 (c
);
781 else if (! ASCII_CHAR_P (c
)
782 && (! accept_latin_1
|| c
>= 0x100))
791 string_count_byte8 (Lisp_Object string
)
793 int multibyte
= STRING_MULTIBYTE (string
);
794 ptrdiff_t nbytes
= SBYTES (string
);
795 unsigned char *p
= SDATA (string
);
796 unsigned char *pend
= p
+ nbytes
;
804 len
= BYTES_BY_CHAR_HEAD (c
);
806 if (CHAR_BYTE8_HEAD_P (c
))
821 string_escape_byte8 (Lisp_Object string
)
823 ptrdiff_t nchars
= SCHARS (string
);
824 ptrdiff_t nbytes
= SBYTES (string
);
825 int multibyte
= STRING_MULTIBYTE (string
);
826 ptrdiff_t byte8_count
;
827 const unsigned char *src
, *src_end
;
832 if (multibyte
&& nchars
== nbytes
)
835 byte8_count
= string_count_byte8 (string
);
837 if (byte8_count
== 0)
842 if ((MOST_POSITIVE_FIXNUM
- nchars
) / 3 < byte8_count
843 || (STRING_BYTES_BOUND
- nbytes
) / 2 < byte8_count
)
846 /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */
847 val
= make_uninit_multibyte_string (nchars
+ byte8_count
* 3,
848 nbytes
+ byte8_count
* 2);
852 if ((STRING_BYTES_BOUND
- nbytes
) / 3 < byte8_count
)
855 /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */
856 val
= make_uninit_string (nbytes
+ byte8_count
* 3);
859 src
= SDATA (string
);
860 src_end
= src
+ nbytes
;
863 while (src
< src_end
)
866 len
= BYTES_BY_CHAR_HEAD (c
);
868 if (CHAR_BYTE8_HEAD_P (c
))
870 c
= STRING_CHAR_ADVANCE (src
);
871 c
= CHAR_TO_BYTE8 (c
);
872 dst
+= sprintf ((char *) dst
, "\\%03o", c
);
875 while (len
--) *dst
++ = *src
++;
878 while (src
< src_end
)
882 dst
+= sprintf ((char *) dst
, "\\%03o", c
);
890 DEFUN ("string", Fstring
, Sstring
, 0, MANY
, 0,
892 Concatenate all the argument characters and make the result a string.
893 usage: (string &rest CHARACTERS) */)
894 (ptrdiff_t n
, Lisp_Object
*args
)
898 unsigned char *buf
, *p
;
902 SAFE_NALLOCA (buf
, MAX_MULTIBYTE_LENGTH
, n
);
905 for (i
= 0; i
< n
; i
++)
907 CHECK_CHARACTER (args
[i
]);
909 p
+= CHAR_STRING (c
, p
);
912 str
= make_string_from_bytes ((char *) buf
, n
, p
- buf
);
917 DEFUN ("unibyte-string", Funibyte_string
, Sunibyte_string
, 0, MANY
, 0,
918 doc
: /* Concatenate all the argument bytes and make the result a unibyte string.
919 usage: (unibyte-string &rest BYTES) */)
920 (ptrdiff_t n
, Lisp_Object
*args
)
925 unsigned char *buf
= SAFE_ALLOCA (n
);
926 unsigned char *p
= buf
;
928 for (i
= 0; i
< n
; i
++)
930 CHECK_RANGED_INTEGER (args
[i
], 0, 255);
931 *p
++ = XINT (args
[i
]);
934 str
= make_string_from_bytes ((char *) buf
, n
, p
- buf
);
939 DEFUN ("char-resolve-modifiers", Fchar_resolve_modifiers
,
940 Schar_resolve_modifiers
, 1, 1, 0,
941 doc
: /* Resolve modifiers in the character CHAR.
942 The value is a character with modifiers resolved into the character
943 code. Unresolved modifiers are kept in the value.
944 usage: (char-resolve-modifiers CHAR) */)
945 (Lisp_Object character
)
949 CHECK_NUMBER (character
);
950 c
= XINT (character
);
951 return make_number (char_resolve_modifier_mask (c
));
954 DEFUN ("get-byte", Fget_byte
, Sget_byte
, 0, 2, 0,
955 doc
: /* Return a byte value of a character at point.
956 Optional 1st arg POSITION, if non-nil, is a position of a character to get
958 Optional 2nd arg STRING, if non-nil, is a string of which first
959 character is a target to get a byte value. In this case, POSITION, if
960 non-nil, is an index of a target character in the string.
962 If the current buffer (or STRING) is multibyte, and the target
963 character is not ASCII nor 8-bit character, an error is signaled. */)
964 (Lisp_Object position
, Lisp_Object string
)
978 CHECK_NUMBER_COERCE_MARKER (position
);
979 if (XINT (position
) < BEGV
|| XINT (position
) >= ZV
)
980 args_out_of_range_3 (position
, make_number (BEGV
), make_number (ZV
));
981 pos
= XFASTINT (position
);
982 p
= CHAR_POS_ADDR (pos
);
984 if (NILP (BVAR (current_buffer
, enable_multibyte_characters
)))
985 return make_number (*p
);
989 CHECK_STRING (string
);
996 CHECK_NATNUM (position
);
997 if (XINT (position
) >= SCHARS (string
))
998 args_out_of_range (string
, position
);
999 pos
= XFASTINT (position
);
1000 p
= SDATA (string
) + string_char_to_byte (string
, pos
);
1002 if (! STRING_MULTIBYTE (string
))
1003 return make_number (*p
);
1005 c
= STRING_CHAR (p
);
1006 if (CHAR_BYTE8_P (c
))
1007 c
= CHAR_TO_BYTE8 (c
);
1008 else if (! ASCII_CHAR_P (c
))
1009 error ("Not an ASCII nor an 8-bit character: %d", c
);
1010 return make_number (c
);
1016 syms_of_character (void)
1018 DEFSYM (Qcharacterp
, "characterp");
1019 DEFSYM (Qauto_fill_chars
, "auto-fill-chars");
1021 staticpro (&Vchar_unify_table
);
1022 Vchar_unify_table
= Qnil
;
1024 defsubr (&Smax_char
);
1025 defsubr (&Scharacterp
);
1026 defsubr (&Sunibyte_char_to_multibyte
);
1027 defsubr (&Smultibyte_char_to_unibyte
);
1028 defsubr (&Schar_width
);
1029 defsubr (&Sstring_width
);
1031 defsubr (&Sunibyte_string
);
1032 defsubr (&Schar_resolve_modifiers
);
1033 defsubr (&Sget_byte
);
1035 DEFVAR_LISP ("translation-table-vector", Vtranslation_table_vector
,
1037 Vector recording all translation tables ever defined.
1038 Each element is a pair (SYMBOL . TABLE) relating the table to the
1039 symbol naming it. The ID of a translation table is an index into this vector. */);
1040 Vtranslation_table_vector
= Fmake_vector (make_number (16), Qnil
);
1042 DEFVAR_LISP ("auto-fill-chars", Vauto_fill_chars
,
1044 A char-table for characters which invoke auto-filling.
1045 Such characters have value t in this table. */);
1046 Vauto_fill_chars
= Fmake_char_table (Qauto_fill_chars
, Qnil
);
1047 CHAR_TABLE_SET (Vauto_fill_chars
, ' ', Qt
);
1048 CHAR_TABLE_SET (Vauto_fill_chars
, '\n', Qt
);
1050 DEFVAR_LISP ("char-width-table", Vchar_width_table
,
1052 A char-table for width (columns) of each character. */);
1053 Vchar_width_table
= Fmake_char_table (Qnil
, make_number (1));
1054 char_table_set_range (Vchar_width_table
, 0x80, 0x9F, make_number (4));
1055 char_table_set_range (Vchar_width_table
, MAX_5_BYTE_CHAR
+ 1, MAX_CHAR
,
1058 DEFVAR_LISP ("printable-chars", Vprintable_chars
,
1059 doc
: /* A char-table for each printable character. */);
1060 Vprintable_chars
= Fmake_char_table (Qnil
, Qnil
);
1061 Fset_char_table_range (Vprintable_chars
,
1062 Fcons (make_number (32), make_number (126)), Qt
);
1063 Fset_char_table_range (Vprintable_chars
,
1064 Fcons (make_number (160),
1065 make_number (MAX_5_BYTE_CHAR
)), Qt
);
1067 DEFVAR_LISP ("char-script-table", Vchar_script_table
,
1068 doc
: /* Char table of script symbols.
1069 It has one extra slot whose value is a list of script symbols. */);
1071 /* Intern this now in case it isn't already done.
1072 Setting this variable twice is harmless.
1073 But don't staticpro it here--that is done in alloc.c. */
1074 Qchar_table_extra_slots
= intern_c_string ("char-table-extra-slots");
1075 DEFSYM (Qchar_script_table
, "char-script-table");
1076 Fput (Qchar_script_table
, Qchar_table_extra_slots
, make_number (1));
1077 Vchar_script_table
= Fmake_char_table (Qchar_script_table
, Qnil
);
1079 DEFVAR_LISP ("script-representative-chars", Vscript_representative_chars
,
1080 doc
: /* Alist of scripts vs the representative characters.
1081 Each element is a cons (SCRIPT . CHARS).
1082 SCRIPT is a symbol representing a script or a subgroup of a script.
1083 CHARS is a list or a vector of characters.
1084 If it is a list, all characters in the list are necessary for supporting SCRIPT.
1085 If it is a vector, one of the characters in the vector is necessary.
1086 This variable is used to find a font for a specific script. */);
1087 Vscript_representative_chars
= Qnil
;
1089 DEFVAR_LISP ("unicode-category-table", Vunicode_category_table
,
1090 doc
: /* Char table of Unicode's "General Category".
1091 All Unicode characters have one of the following values (symbol):
1092 Lu, Ll, Lt, Lm, Lo, Mn, Mc, Me, Nd, Nl, No, Pc, Pd, Ps, Pe, Pi, Pf, Po,
1093 Sm, Sc, Sk, So, Zs, Zl, Zp, Cc, Cf, Cs, Co, Cn
1094 See The Unicode Standard for the meaning of those values. */);
1095 /* The correct char-table is setup in characters.el. */
1096 Vunicode_category_table
= Qnil
;