/* GNU Emacs case conversion functions.
- Copyright (C) 1985,94,97,98,99, 2001, 2002, 2004
- Free Software Foundation, Inc.
+ Copyright (C) 1985, 1994, 1997, 1998, 1999, 2001, 2002, 2003, 2004,
+ 2005, 2006 Free Software Foundation, Inc.
This file is part of GNU Emacs.
You should have received a copy of the GNU General Public License
along with GNU Emacs; see the file COPYING. If not, write to
-the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA. */
+the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA. */
#include <config.h>
if (STRINGP (obj))
{
int multibyte = STRING_MULTIBYTE (obj);
+ int n;
obj = Fcopy_sequence (obj);
len = SBYTES (obj);
- /* Scan all single-byte characters from start of string. */
- for (i = 0; i < len;)
+ /* I counts bytes, and N counts chars. */
+ for (i = n = 0; i < len; n++)
{
+ int from_len = 1, to_len = 1;
+
c = SREF (obj, i);
if (multibyte && c >= 0x80)
- /* A multibyte character can't be handled in this
- simple loop. */
- break;
+ c = STRING_CHAR_AND_LENGTH (SDATA (obj) + i, len -i, from_len);
if (inword && flag != CASE_CAPITALIZE_UP)
c = DOWNCASE (c);
else if (!UPPERCASEP (c)
&& (!inword || flag != CASE_CAPITALIZE_UP))
c = UPCASE1 (c);
- /* If this char won't fit in a single-byte string.
- fall out to the multibyte case. */
- if (multibyte ? ! ASCII_BYTE_P (c)
- : ! SINGLE_BYTE_CHAR_P (c))
- break;
-
- SSET (obj, i, c);
- if ((int) flag >= (int) CASE_CAPITALIZE)
- inword = SYNTAX (c) == Sword;
- i++;
- }
-
- /* If we didn't do the whole string as single-byte,
- scan the rest in a more complex way. */
- if (i < len)
- {
- /* The work is not yet finished because of a multibyte
- character just encountered. */
- int fromlen, j_byte = i;
- char *buf;
- int bufsize;
- USE_SAFE_ALLOCA;
-
- bufsize = (len - i) * MAX_MULTIBYTE_LENGTH + i;
- SAFE_ALLOCA (buf, char *, bufsize);
-
- /* Copy data already handled. */
- bcopy (SDATA (obj), buf, i);
-
- /* From now on, I counts bytes. */
- while (i < len)
+ if ((ASCII_BYTE_P (c) && from_len == 1)
+ || (! multibyte && SINGLE_BYTE_CHAR_P (c)))
+ SSET (obj, i, c);
+ else
{
- c = STRING_CHAR_AND_LENGTH (SDATA (obj) + i,
- len - i, fromlen);
- if (inword && flag != CASE_CAPITALIZE_UP)
- c = DOWNCASE (c);
- else if (!UPPERCASEP (c)
- && (!inword || flag != CASE_CAPITALIZE_UP))
- c = UPCASE1 (c);
- i += fromlen;
- j_byte += CHAR_STRING (c, buf + j_byte);
- if ((int) flag >= (int) CASE_CAPITALIZE)
- inword = SYNTAX (c) == Sword;
+ to_len = CHAR_BYTES (c);
+ if (from_len == to_len)
+ CHAR_STRING (c, SDATA (obj) + i);
+ else
+ {
+ Faset (obj, make_number (n), make_number (c));
+ len += to_len - from_len;
+ }
}
- obj = make_multibyte_string (buf, SCHARS (obj),
- j_byte);
- SAFE_FREE ();
+ if ((int) flag >= (int) CASE_CAPITALIZE)
+ inword = SYNTAX (c) == Sword;
+ i += to_len;
}
return obj;
}
else if (!UPPERCASEP (c)
&& (!inword || flag != CASE_CAPITALIZE_UP))
c = UPCASE1 (c);
+ if (multibyte && c >= 0x80)
+ /* A multibyte result character can't be handled in this
+ simple loop. */
+ break;
FETCH_BYTE (i) = c;
if (c != c2)
changed = 1;
int opoint_byte = PT_BYTE;
int c2;
- while (i < end_byte)
+ while (start < end)
{
if ((c = FETCH_BYTE (i)) >= 0x80)
c = FETCH_MULTIBYTE_CHAR (i);
tolen = CHAR_STRING (c2, str),
fromlen == tolen)
{
+ /* Length is unchanged. */
for (j = 0; j < tolen; ++j)
FETCH_BYTE (i + j) = str[j];
}
else
{
- error ("Can't casify letters that change length");
-#if 0 /* This is approximately what we'd like to be able to do here */
- if (tolen < fromlen)
- del_range_1 (i + tolen, i + fromlen, 0, 0);
- else if (tolen > fromlen)
- {
- TEMP_SET_PT (i + fromlen);
- insert_1 (str + fromlen, tolen - fromlen, 1, 0, 0);
- }
-#endif
+ /* Replace one character with the other,
+ keeping text properties the same. */
+ replace_range_2 (start, i,
+ start + 1, i + fromlen,
+ str, 1, tolen,
+ 1);
+ if (opoint > start)
+ opoint_byte += tolen - fromlen;
}
}
if ((int) flag >= (int) CASE_CAPITALIZE)