/* Basic character support.
-Copyright (C) 2001-2011 Free Software Foundation, Inc.
+Copyright (C) 2001-2012 Free Software Foundation, Inc.
Copyright (C) 1995, 1997, 1998, 2001 Electrotechnical Laboratory, JAPAN.
Licensed to the Free Software Foundation.
Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
#include <config.h>
#endif
+#define CHARACTER_INLINE EXTERN_INLINE
+
#include <stdio.h>
#ifdef emacs
Unicode character. Mainly used by the macro MAYBE_UNIFY_CHAR. */
Lisp_Object Vchar_unify_table;
-/* Variable used locally in the macro FETCH_MULTIBYTE_CHAR. */
-unsigned char *_fetch_multibyte_char_p;
-
static Lisp_Object Qchar_script_table;
\f
DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0,
doc: /* Return non-nil if OBJECT is a character.
+In Emacs Lisp, characters are represented by character codes, which
+are non-negative integers. The function `max-char' returns the
+maximum character code.
usage: (characterp OBJECT) */)
(Lisp_Object object, Lisp_Object ignore)
{
}
}
+
+/* Return width (columns) of C considering the buffer display table DP. */
+
+static ptrdiff_t
+char_width (int c, struct Lisp_Char_Table *dp)
+{
+ ptrdiff_t width = CHAR_WIDTH (c);
+
+ if (dp)
+ {
+ Lisp_Object disp = DISP_CHAR_VECTOR (dp, c), ch;
+ int i;
+
+ if (VECTORP (disp))
+ for (i = 0, width = 0; i < ASIZE (disp); i++)
+ {
+ ch = AREF (disp, i);
+ if (CHARACTERP (ch))
+ {
+ int w = CHAR_WIDTH (XFASTINT (ch));
+ if (INT_ADD_OVERFLOW (width, w))
+ string_overflow ();
+ width += w;
+ }
+ }
+ }
+ return width;
+}
+
+
DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0,
doc: /* Return width of CHAR when displayed in the current buffer.
The width is measured by how many columns it occupies on the screen.
usage: (char-width CHAR) */)
(Lisp_Object ch)
{
- Lisp_Object disp;
- int c, width;
- struct Lisp_Char_Table *dp = buffer_display_table ();
+ int c;
+ ptrdiff_t width;
CHECK_CHARACTER (ch);
c = XINT (ch);
-
- /* Get the way the display table would display it. */
- disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil;
-
- if (VECTORP (disp))
- width = sanitize_char_width (ASIZE (disp));
- else
- width = CHAR_WIDTH (c);
-
+ width = char_width (c, buffer_display_table ());
return make_number (width);
}
while (i_byte < len)
{
- int bytes, thiswidth;
- Lisp_Object val;
+ int bytes;
int c = STRING_CHAR_AND_LENGTH (str + i_byte, bytes);
+ ptrdiff_t thiswidth = char_width (c, dp);
- if (dp)
- {
- val = DISP_CHAR_VECTOR (dp, c);
- if (VECTORP (val))
- thiswidth = sanitize_char_width (ASIZE (val));
- else
- thiswidth = CHAR_WIDTH (c);
- }
- else
+ if (precision <= 0)
{
- thiswidth = CHAR_WIDTH (c);
+ if (INT_ADD_OVERFLOW (width, thiswidth))
+ string_overflow ();
}
-
- if (precision > 0
- && (width + thiswidth > precision))
+ else if (precision - width < thiswidth)
{
*nchars = i;
*nbytes = i_byte;
/* 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);
+ bool multibyte = len < SBYTES (string);
unsigned char *str = SDATA (string);
ptrdiff_t i = 0, i_byte = 0;
ptrdiff_t width = 0;
else
c = str[i_byte], bytes = 1;
chars = 1;
- if (dp)
- {
- val = DISP_CHAR_VECTOR (dp, c);
- if (VECTORP (val))
- thiswidth = sanitize_char_width (ASIZE (val));
- else
- thiswidth = CHAR_WIDTH (c);
- }
- else
- {
- thiswidth = CHAR_WIDTH (c);
- }
+ thiswidth = char_width (c, dp);
}
if (precision <= 0)
int len = MULTIBYTE_LENGTH (ptr, endp);
if (len == 0)
- abort ();
+ emacs_abort ();
ptr += len;
chars++;
}
}
/* 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'. */
ptrdiff_t
corresponding byte and store in DST. CHARS is the number of
characters in SRC. The value is the number of bytes stored in DST.
Usually, the value is the same as CHARS, but is less than it if SRC
- contains a non-ASCII, non-eight-bit character. If ACCEPT_LATIN_1
- is nonzero, a Latin-1 character is accepted and converted to a byte
- of that character code.
- Note: Currently the arg ACCEPT_LATIN_1 is not used. */
+ contains a non-ASCII, non-eight-bit character. */
ptrdiff_t
-str_to_unibyte (const unsigned char *src, unsigned char *dst, ptrdiff_t chars, int accept_latin_1)
+str_to_unibyte (const unsigned char *src, unsigned char *dst, ptrdiff_t chars)
{
ptrdiff_t i;
if (CHAR_BYTE8_P (c))
c = CHAR_TO_BYTE8 (c);
- else if (! ASCII_CHAR_P (c)
- && (! accept_latin_1 || c >= 0x100))
+ else if (! ASCII_CHAR_P (c))
return i;
*dst++ = c;
}
static ptrdiff_t
string_count_byte8 (Lisp_Object string)
{
- int multibyte = STRING_MULTIBYTE (string);
+ bool multibyte = STRING_MULTIBYTE (string);
ptrdiff_t nbytes = SBYTES (string);
unsigned char *p = SDATA (string);
unsigned char *pend = p + nbytes;
{
ptrdiff_t nchars = SCHARS (string);
ptrdiff_t nbytes = SBYTES (string);
- int multibyte = STRING_MULTIBYTE (string);
+ bool multibyte = STRING_MULTIBYTE (string);
ptrdiff_t byte8_count;
const unsigned char *src, *src_end;
unsigned char *dst;
{
c = STRING_CHAR_ADVANCE (src);
c = CHAR_TO_BYTE8 (c);
- sprintf ((char *) dst, "\\%03o", c);
- dst += 4;
+ dst += sprintf ((char *) dst, "\\%03o", c);
}
else
while (len--) *dst++ = *src++;
{
c = *src++;
if (c >= 0x80)
- {
- sprintf ((char *) dst, "\\%03o", c);
- dst += 4;
- }
+ dst += sprintf ((char *) dst, "\\%03o", c);
else
*dst++ = c;
}
(ptrdiff_t n, Lisp_Object *args)
{
ptrdiff_t i;
- unsigned char *buf, *p;
Lisp_Object str;
USE_SAFE_ALLOCA;
-
- SAFE_ALLOCA (buf, unsigned char *, n);
- p = buf;
+ unsigned char *buf = SAFE_ALLOCA (n);
+ unsigned char *p = buf;
for (i = 0; i < n; i++)
{
- CHECK_RANGED_INTEGER (0, args[i], 255);
+ CHECK_RANGED_INTEGER (args[i], 0, 255);
*p++ = XINT (args[i]);
}
return make_number (c);
}
-
-void
-init_character_once (void)
-{
-}
-
#ifdef emacs
void