#include <sys/types.h>
#include <setjmp.h>
+#include <intprops.h>
#include "lisp.h"
#include "character.h"
#include "buffer.h"
/* If character code C has modifier masks, reflect them to the
character code if possible. Return the resulting code. */
-int
-char_resolve_modifier_mask (int c)
+EMACS_INT
+char_resolve_modifier_mask (EMACS_INT c)
{
/* A non-ASCII character can't reflect modifier bits to the code. */
if (! ASCII_CHAR_P ((c & ~CHAR_MODIFIER_MASK)))
if (c & CHAR_MODIFIER_MASK)
{
- c = (unsigned) char_resolve_modifier_mask ((int) c);
+ c = char_resolve_modifier_mask (c);
/* If C still has any modifier bits, just ignore it. */
c &= ~CHAR_MODIFIER_MASK;
}
bytes = BYTE8_STRING (c, p);
}
else
- error ("Invalid character: %d", c);
+ error ("Invalid character: %x", c);
return bytes;
}
-/* Return a character whose multibyte form is at P. Set LEN is not
+/* Return a character whose multibyte form is at P. If LEN is not
NULL, it must be a pointer to integer. In that case, set *LEN to
- the byte length of the multibyte form. If ADVANCED is not NULL, is
+ the byte length of the multibyte form. If ADVANCED is not NULL, it
must be a pointer to unsigned char. In that case, set *ADVANCED to
- the ending address (i.e. the starting address of the next
+ the ending address (i.e., the starting address of the next
character) of the multibyte form. */
int
}
-/* Translate character C by translation table TABLE. If C is
- negative, translate a character specified by CHARSET and CODE. If
- no translation is found in TABLE, return the untranslated
- character. If TABLE is a list, elements are char tables. In this
- case, translace C by all tables. */
+/* Translate character C by translation table TABLE. If no translation is
+ found in TABLE, return the untranslated character. If TABLE is a list,
+ elements are char tables. In that case, recursively translate C by all the
+ tables in the list. */
int
translate_char (Lisp_Object table, int c)
}
DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0,
- doc: /* Return non-nil if OBJECT is a character. */)
+ doc: /* Return non-nil if OBJECT is a character.
+usage: (characterp OBJECT) */)
(Lisp_Object object, Lisp_Object ignore)
{
return (CHARACTERP (object) ? Qt : Qnil);
disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil;
if (VECTORP (disp))
- width = ASIZE (disp);
+ width = sanitize_char_width (ASIZE (disp));
else
width = CHAR_WIDTH (c);
characters and bytes of the substring in *NCHARS and *NBYTES
respectively. */
-EMACS_INT
-c_string_width (const unsigned char *str, EMACS_INT len, int precision,
- EMACS_INT *nchars, EMACS_INT *nbytes)
+ptrdiff_t
+c_string_width (const unsigned char *str, ptrdiff_t len, int precision,
+ ptrdiff_t *nchars, ptrdiff_t *nbytes)
{
- EMACS_INT i = 0, i_byte = 0;
- EMACS_INT width = 0;
+ ptrdiff_t i = 0, i_byte = 0;
+ ptrdiff_t width = 0;
struct Lisp_Char_Table *dp = buffer_display_table ();
while (i_byte < len)
{
val = DISP_CHAR_VECTOR (dp, c);
if (VECTORP (val))
- thiswidth = XVECTOR (val)->size;
+ thiswidth = sanitize_char_width (ASIZE (val));
else
thiswidth = CHAR_WIDTH (c);
}
current buffer. The width is measured by how many columns it
occupies on the screen. */
-EMACS_INT
-strwidth (const char *str, EMACS_INT len)
+ptrdiff_t
+strwidth (const char *str, ptrdiff_t len)
{
return c_string_width ((const unsigned char *) str, len, -1, NULL, NULL);
}
PRECISION, and set number of characters and bytes of the substring
in *NCHARS and *NBYTES respectively. */
-EMACS_INT
-lisp_string_width (Lisp_Object string, int precision,
- EMACS_INT *nchars, EMACS_INT *nbytes)
+ptrdiff_t
+lisp_string_width (Lisp_Object string, ptrdiff_t precision,
+ ptrdiff_t *nchars, ptrdiff_t *nbytes)
{
- EMACS_INT len = SCHARS (string);
+ ptrdiff_t len = SCHARS (string);
/* This set multibyte to 0 even if STRING is multibyte when it
contains only ascii and eight-bit-graphic, but that's
intentional. */
int multibyte = len < SBYTES (string);
unsigned char *str = SDATA (string);
- EMACS_INT i = 0, i_byte = 0;
- EMACS_INT width = 0;
+ ptrdiff_t i = 0, i_byte = 0;
+ ptrdiff_t width = 0;
struct Lisp_Char_Table *dp = buffer_display_table ();
while (i < len)
{
- int chars, bytes, thiswidth;
+ ptrdiff_t chars, bytes, thiswidth;
Lisp_Object val;
- int cmp_id;
- EMACS_INT ignore, end;
+ ptrdiff_t cmp_id;
+ ptrdiff_t ignore, end;
if (find_composition (i, -1, &ignore, &end, &val, string)
&& ((cmp_id = get_composition_id (i, i_byte, end - i, val, string))
int c;
if (multibyte)
- c = STRING_CHAR_AND_LENGTH (str + i_byte, bytes);
+ {
+ int cbytes;
+ c = STRING_CHAR_AND_LENGTH (str + i_byte, cbytes);
+ bytes = cbytes;
+ }
else
c = str[i_byte], bytes = 1;
chars = 1;
{
val = DISP_CHAR_VECTOR (dp, c);
if (VECTORP (val))
- thiswidth = XVECTOR (val)->size;
+ thiswidth = sanitize_char_width (ASIZE (val));
else
thiswidth = CHAR_WIDTH (c);
}
}
}
- if (precision > 0
- && (width + thiswidth > precision))
+ if (precision <= 0)
+ {
+#ifdef emacs
+ if (INT_ADD_OVERFLOW (width, thiswidth))
+ string_overflow ();
+#endif
+ }
+ else if (precision - width < thiswidth)
{
*nchars = i;
*nbytes = i_byte;
i += chars;
i_byte += bytes;
width += thiswidth;
- }
+ }
if (precision > 0)
{
return val;
}
-DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0,
- doc: /* Return the direction of CHAR.
-The returned value is 0 for left-to-right and 1 for right-to-left.
-usage: (char-direction CHAR) */)
- (Lisp_Object ch)
-{
- int c;
-
- CHECK_CHARACTER (ch);
- c = XINT (ch);
- return CHAR_TABLE_REF (Vchar_direction_table, c);
-}
-
/* Return the number of characters in the NBYTES bytes at PTR.
This works by looking at the contents and checking for multibyte
sequences while assuming that there's no invalid sequence.
However, if the current buffer has enable-multibyte-characters =
nil, we treat each byte as a character. */
-EMACS_INT
-chars_in_text (const unsigned char *ptr, EMACS_INT nbytes)
+ptrdiff_t
+chars_in_text (const unsigned char *ptr, ptrdiff_t nbytes)
{
/* current_buffer is null at early stages of Emacs initialization. */
if (current_buffer == 0
sequences while assuming that there's no invalid sequence. It
ignores enable-multibyte-characters. */
-EMACS_INT
-multibyte_chars_in_text (const unsigned char *ptr, EMACS_INT nbytes)
+ptrdiff_t
+multibyte_chars_in_text (const unsigned char *ptr, ptrdiff_t nbytes)
{
const unsigned char *endp = ptr + nbytes;
- EMACS_INT chars = 0;
+ ptrdiff_t chars = 0;
while (ptr < endp)
{
- EMACS_INT len = MULTIBYTE_LENGTH (ptr, endp);
+ int len = MULTIBYTE_LENGTH (ptr, endp);
if (len == 0)
abort ();
represented by 2-byte in a multibyte text. */
void
-parse_str_as_multibyte (const unsigned char *str, EMACS_INT len,
- EMACS_INT *nchars, EMACS_INT *nbytes)
+parse_str_as_multibyte (const unsigned char *str, ptrdiff_t len,
+ ptrdiff_t *nchars, ptrdiff_t *nbytes)
{
const unsigned char *endp = str + len;
- EMACS_INT n, chars = 0, bytes = 0;
+ int n;
+ ptrdiff_t chars = 0, bytes = 0;
if (len >= MAX_MULTIBYTE_LENGTH)
{
area and that is enough. Return the number of bytes of the
resulting text. */
-EMACS_INT
-str_as_multibyte (unsigned char *str, EMACS_INT len, EMACS_INT nbytes,
- EMACS_INT *nchars)
+ptrdiff_t
+str_as_multibyte (unsigned char *str, ptrdiff_t len, ptrdiff_t nbytes,
+ ptrdiff_t *nchars)
{
unsigned char *p = str, *endp = str + nbytes;
unsigned char *to;
- EMACS_INT chars = 0;
+ ptrdiff_t chars = 0;
int n;
if (nbytes >= MAX_MULTIBYTE_LENGTH)
bytes it may ocupy when converted to multibyte string by
`str_to_multibyte'. */
-EMACS_INT
-parse_str_to_multibyte (const unsigned char *str, EMACS_INT len)
+ptrdiff_t
+count_size_as_multibyte (const unsigned char *str, ptrdiff_t len)
{
const unsigned char *endp = str + len;
- EMACS_INT bytes;
+ ptrdiff_t bytes;
for (bytes = 0; str < endp; str++)
- bytes += (*str < 0x80) ? 1 : 2;
+ {
+ int n = *str < 0x80 ? 1 : 2;
+ if (INT_ADD_OVERFLOW (bytes, n))
+ string_overflow ();
+ bytes += n;
+ }
return bytes;
}
-/* Convert unibyte text at STR of NBYTES bytes to a multibyte text
+/* Convert unibyte text at STR of BYTES bytes to a multibyte text
that contains the same single-byte characters. It actually
converts all 8-bit characters to multibyte forms. It is assured
that we can use LEN bytes at STR as a work area and that is
enough. */
-EMACS_INT
-str_to_multibyte (unsigned char *str, EMACS_INT len, EMACS_INT bytes)
+ptrdiff_t
+str_to_multibyte (unsigned char *str, ptrdiff_t len, ptrdiff_t bytes)
{
unsigned char *p = str, *endp = str + bytes;
unsigned char *to;
actually converts characters in the range 0x80..0xFF to
unibyte. */
-EMACS_INT
-str_as_unibyte (unsigned char *str, EMACS_INT bytes)
+ptrdiff_t
+str_as_unibyte (unsigned char *str, ptrdiff_t bytes)
{
const unsigned char *p = str, *endp = str + bytes;
unsigned char *to;
of that character code.
Note: Currently the arg ACCEPT_LATIN_1 is not used. */
-EMACS_INT
-str_to_unibyte (const unsigned char *src, unsigned char *dst, EMACS_INT chars, int accept_latin_1)
+ptrdiff_t
+str_to_unibyte (const unsigned char *src, unsigned char *dst, ptrdiff_t chars, int accept_latin_1)
{
- EMACS_INT i;
+ ptrdiff_t i;
for (i = 0; i < chars; i++)
{
}
-static EMACS_INT
+static ptrdiff_t
string_count_byte8 (Lisp_Object string)
{
int multibyte = STRING_MULTIBYTE (string);
- EMACS_INT nbytes = SBYTES (string);
+ ptrdiff_t nbytes = SBYTES (string);
unsigned char *p = SDATA (string);
unsigned char *pend = p + nbytes;
- EMACS_INT count = 0;
+ ptrdiff_t count = 0;
int c, len;
if (multibyte)
Lisp_Object
string_escape_byte8 (Lisp_Object string)
{
- EMACS_INT nchars = SCHARS (string);
- EMACS_INT nbytes = SBYTES (string);
+ ptrdiff_t nchars = SCHARS (string);
+ ptrdiff_t nbytes = SBYTES (string);
int multibyte = STRING_MULTIBYTE (string);
- EMACS_INT byte8_count;
+ ptrdiff_t byte8_count;
const unsigned char *src, *src_end;
unsigned char *dst;
Lisp_Object val;
if (multibyte)
{
if ((MOST_POSITIVE_FIXNUM - nchars) / 3 < byte8_count
- || (MOST_POSITIVE_FIXNUM - nbytes) / 2 < byte8_count)
- error ("Maximum string size exceeded");
+ || (STRING_BYTES_BOUND - nbytes) / 2 < byte8_count)
+ string_overflow ();
/* Convert 2-byte sequence of byte8 chars to 4-byte octal. */
val = make_uninit_multibyte_string (nchars + byte8_count * 3,
}
else
{
- if ((MOST_POSITIVE_FIXNUM - nchars) / 3 < byte8_count)
- error ("Maximum string size exceeded");
+ if ((STRING_BYTES_BOUND - nbytes) / 3 < byte8_count)
+ string_overflow ();
+
/* Convert 1-byte sequence of byte8 chars to 4-byte octal. */
val = make_uninit_string (nbytes + byte8_count * 3);
}
doc: /*
Concatenate all the argument characters and make the result a string.
usage: (string &rest CHARACTERS) */)
- (size_t n, Lisp_Object *args)
+ (ptrdiff_t n, Lisp_Object *args)
{
- size_t i;
+ ptrdiff_t i;
int c;
unsigned char *buf, *p;
Lisp_Object str;
USE_SAFE_ALLOCA;
- SAFE_ALLOCA (buf, unsigned char *, MAX_MULTIBYTE_LENGTH * n);
+ SAFE_NALLOCA (buf, MAX_MULTIBYTE_LENGTH, n);
p = buf;
for (i = 0; i < n; i++)
DEFUN ("unibyte-string", Funibyte_string, Sunibyte_string, 0, MANY, 0,
doc: /* Concatenate all the argument bytes and make the result a unibyte string.
usage: (unibyte-string &rest BYTES) */)
- (size_t n, Lisp_Object *args)
+ (ptrdiff_t n, Lisp_Object *args)
{
- size_t i;
+ ptrdiff_t i;
int c;
unsigned char *buf, *p;
Lisp_Object str;
for (i = 0; i < n; i++)
{
- CHECK_NATNUM (args[i]);
+ CHECK_CHARACTER (args[i]);
c = XFASTINT (args[i]);
if (c >= 256)
args_out_of_range_3 (args[i], make_number (0), make_number (255));
usage: (char-resolve-modifiers CHAR) */)
(Lisp_Object character)
{
- int c;
+ EMACS_INT c;
CHECK_NUMBER (character);
c = XINT (character);
(Lisp_Object position, Lisp_Object string)
{
int c;
- EMACS_INT pos;
+ ptrdiff_t pos;
unsigned char *p;
if (NILP (string))
defsubr (&Smultibyte_char_to_unibyte);
defsubr (&Schar_width);
defsubr (&Sstring_width);
- defsubr (&Schar_direction);
defsubr (&Sstring);
defsubr (&Sunibyte_string);
defsubr (&Schar_resolve_modifiers);
char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR,
make_number (4));
- DEFVAR_LISP ("char-direction-table", Vchar_direction_table,
- doc: /* A char-table for direction of each character. */);
- Vchar_direction_table = Fmake_char_table (Qnil, make_number (1));
-
DEFVAR_LISP ("printable-chars", Vprintable_chars,
doc: /* A char-table for each printable character. */);
Vprintable_chars = Fmake_char_table (Qnil, Qnil);
HCoop / bpt / emacs.git / blobdiff