X-Git-Url: http://git.hcoop.net/bpt/emacs.git/blobdiff_plain/e6c3da2065ac72cc4e1a2bef22d367cd75401892..e4920bc99dfcee02c3bb83b46a761b0893f76626:/src/character.c?ds=sidebyside
diff --git a/src/character.c b/src/character.c
index 4087e8984d..88b1f11b96 100644
--- a/src/character.c
+++ b/src/character.c
@@ -35,6 +35,7 @@ along with GNU Emacs. If not, see . */
#include
#include
+#include
#include "lisp.h"
#include "character.h"
#include "buffer.h"
@@ -122,7 +123,7 @@ char_string (unsigned int c, unsigned char *p)
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;
}
@@ -162,11 +163,11 @@ char_string (unsigned int c, unsigned char *p)
}
-/* 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
@@ -206,11 +207,10 @@ string_char (const unsigned char *p, const unsigned char **advanced, int *len)
}
-/* 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)
@@ -258,7 +258,8 @@ multibyte_char_to_unibyte_safe (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);
@@ -325,7 +326,7 @@ usage: (char-width CHAR) */)
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);
@@ -357,7 +358,7 @@ c_string_width (const unsigned char *str, EMACS_INT len, int precision,
{
val = DISP_CHAR_VECTOR (dp, c);
if (VECTORP (val))
- thiswidth = XVECTOR (val)->size;
+ thiswidth = sanitize_char_width (ASIZE (val));
else
thiswidth = CHAR_WIDTH (c);
}
@@ -405,7 +406,7 @@ strwidth (const char *str, EMACS_INT len)
in *NCHARS and *NBYTES respectively. */
EMACS_INT
-lisp_string_width (Lisp_Object string, int precision,
+lisp_string_width (Lisp_Object string, EMACS_INT precision,
EMACS_INT *nchars, EMACS_INT *nbytes)
{
EMACS_INT len = SCHARS (string);
@@ -420,9 +421,9 @@ lisp_string_width (Lisp_Object string, int precision,
while (i < len)
{
- int chars, bytes, thiswidth;
+ EMACS_INT chars, bytes, thiswidth;
Lisp_Object val;
- int cmp_id;
+ ptrdiff_t cmp_id;
EMACS_INT ignore, end;
if (find_composition (i, -1, &ignore, &end, &val, string)
@@ -438,7 +439,11 @@ lisp_string_width (Lisp_Object string, int precision,
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;
@@ -446,7 +451,7 @@ lisp_string_width (Lisp_Object string, int precision,
{
val = DISP_CHAR_VECTOR (dp, c);
if (VECTORP (val))
- thiswidth = XVECTOR (val)->size;
+ thiswidth = sanitize_char_width (ASIZE (val));
else
thiswidth = CHAR_WIDTH (c);
}
@@ -456,8 +461,14 @@ lisp_string_width (Lisp_Object string, int precision,
}
}
- 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;
@@ -466,7 +477,7 @@ lisp_string_width (Lisp_Object string, int precision,
i += chars;
i_byte += bytes;
width += thiswidth;
- }
+ }
if (precision > 0)
{
@@ -494,19 +505,6 @@ usage: (string-width STRING) */)
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.
@@ -590,7 +588,7 @@ parse_str_as_multibyte (const unsigned char *str, EMACS_INT len,
}
/* Arrange unibyte text at STR of NBYTES bytes as a multibyte text.
- It actually converts only such 8-bit characters that don't contruct
+ It actually converts only such 8-bit characters that don't construct
a multibyte sequence to multibyte forms of Latin-1 characters. If
NCHARS is nonzero, set *NCHARS to the number of characters in the
text. It is assured that we can use LEN bytes at STR as a work
@@ -671,22 +669,27 @@ str_as_multibyte (unsigned char *str, EMACS_INT len, EMACS_INT nbytes,
}
/* Parse unibyte string at STR of LEN bytes, and return the number of
- bytes it may ocupy when converted to multibyte string by
+ bytes it may occupy when converted to multibyte string by
`str_to_multibyte'. */
EMACS_INT
-parse_str_to_multibyte (const unsigned char *str, EMACS_INT len)
+count_size_as_multibyte (const unsigned char *str, EMACS_INT len)
{
const unsigned char *endp = str + len;
EMACS_INT 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
@@ -836,8 +839,8 @@ string_escape_byte8 (Lisp_Object string)
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,
@@ -845,8 +848,9 @@ string_escape_byte8 (Lisp_Object string)
}
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);
}
@@ -890,15 +894,15 @@ DEFUN ("string", Fstring, Sstring, 0, MANY, 0,
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++)
@@ -916,9 +920,9 @@ usage: (string &rest CHARACTERS) */)
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;
@@ -965,7 +969,7 @@ character is a target to get a byte value. In this case, POSITION, if
non-nil, is an index of a target character in the string.
If the current buffer (or STRING) is multibyte, and the target
-character is not ASCII nor 8-bit character, an error is signalled. */)
+character is not ASCII nor 8-bit character, an error is signaled. */)
(Lisp_Object position, Lisp_Object string)
{
int c;
@@ -1038,7 +1042,6 @@ syms_of_character (void)
defsubr (&Smultibyte_char_to_unibyte);
defsubr (&Schar_width);
defsubr (&Sstring_width);
- defsubr (&Schar_direction);
defsubr (&Sstring);
defsubr (&Sunibyte_string);
defsubr (&Schar_resolve_modifiers);
@@ -1067,10 +1070,6 @@ A char-table for width (columns) of each character. */);
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);