/* mbutil.c -- readline multibyte character utility functions */
-/* Copyright (C) 2001 Free Software Foundation, Inc.
+/* Copyright (C) 2001-2017 Free Software Foundation, Inc.
- This file is part of the GNU Readline Library, a library for
- reading lines of text with interactive input and history editing.
+ This file is part of the GNU Readline Library (Readline), a library
+ for reading lines of text with interactive input and history editing.
- The GNU Readline Library is free software; you can redistribute it
- and/or modify it under the terms of the GNU General Public License
- as published by the Free Software Foundation; either version 2, or
+ Readline is free software: you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
- The GNU Readline Library is distributed in the hope that it will be
- useful, but WITHOUT ANY WARRANTY; without even the implied warranty
- of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ Readline is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
- The GNU General Public License is often shipped with GNU software, and
- is generally kept in a file called COPYING or LICENSE. If you do not
- have a copy of the license, write to the Free Software Foundation,
- 59 Temple Place, Suite 330, Boston, MA 02111 USA. */
+ You should have received a copy of the GNU General Public License
+ along with Readline. If not, see <http://www.gnu.org/licenses/>.
+*/
+
#define READLINE_LIBRARY
#if defined (HAVE_CONFIG_H)
int rl_byte_oriented = 1;
#endif
+/* Ditto */
+int _rl_utf8locale = 0;
+
/* **************************************************************** */
/* */
/* Multibyte Character Utility Functions */
#if defined(HANDLE_MULTIBYTE)
+/* **************************************************************** */
+/* */
+/* UTF-8 specific Character Utility Functions */
+/* */
+/* **************************************************************** */
+
+/* Return the length in bytes of the possibly-multibyte character beginning
+ at S. Encoding is UTF-8. */
static int
-_rl_find_next_mbchar_internal (string, seed, count, find_non_zero)
- char *string;
- int seed, count, find_non_zero;
+_rl_utf8_mblen (const char *s, size_t n)
{
- size_t tmp = 0;
+ unsigned char c, c1;
+
+ if (s == 0)
+ return (0); /* no shift states */
+ if (n <= 0)
+ return (-1);
+
+ c = (unsigned char)*s;
+ if (c < 0x80)
+ return (c != 0);
+ if (c >= 0xc2)
+ {
+ c1 = (unsigned char)s[1];
+ if (c < 0xe0)
+ {
+ if (n >= 2 && (s[1] ^ 0x80) < 0x40)
+ return 2;
+ }
+ else if (c < 0xf0)
+ {
+ if (n >= 3
+ && (s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
+ && (c >= 0xe1 || c1 >= 0xa0)
+ && (c != 0xed || c1 < 0xa0))
+ return 3;
+ }
+ else if (c < 0xf8)
+ {
+ if (n >= 4
+ && (s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
+ && (s[3] ^ 0x80) < 0x40
+ && (c >= 0xf1 || c1 >= 0x90)
+ && (c < 0xf4 || (c == 0xf4 && c1 < 0x90)))
+ return 4;
+ }
+ }
+ /* invalid or incomplete multibyte character */
+ return -1;
+}
+
+static int
+_rl_find_next_mbchar_internal (char *string, int seed, int count, int find_non_zero)
+{
+ size_t tmp, len;
mbstate_t ps;
- int point = 0;
+ int point;
wchar_t wc;
+ tmp = 0;
+
memset(&ps, 0, sizeof (mbstate_t));
if (seed < 0)
seed = 0;
if (count <= 0)
return seed;
- point = seed + _rl_adjust_point(string, seed, &ps);
- /* if this is true, means that seed was not pointed character
- started byte. So correct the point and consume count */
+ point = seed + _rl_adjust_point (string, seed, &ps);
+ /* if _rl_adjust_point returns -1, the character or string is invalid.
+ treat as a byte. */
+ if (point == seed - 1) /* invalid */
+ return seed + 1;
+
+ /* if this is true, means that seed was not pointing to a byte indicating
+ the beginning of a multibyte character. Correct the point and consume
+ one char. */
if (seed < point)
- count --;
+ count--;
while (count > 0)
{
- tmp = mbrtowc (&wc, string+point, strlen(string + point), &ps);
- if ((size_t)(tmp) == (size_t)-1 || (size_t)(tmp) == (size_t)-2)
+ len = strlen (string + point);
+ if (len == 0)
+ break;
+ if (_rl_utf8locale && UTF8_SINGLEBYTE(string[point]))
{
- /* invalid bytes. asume a byte represents a character */
+ tmp = 1;
+ wc = (wchar_t) string[point];
+ memset(&ps, 0, sizeof(mbstate_t));
+ }
+ else
+ tmp = mbrtowc (&wc, string+point, len, &ps);
+ if (MB_INVALIDCH ((size_t)tmp))
+ {
+ /* invalid bytes. assume a byte represents a character */
point++;
count--;
/* reset states. */
memset(&ps, 0, sizeof(mbstate_t));
}
- else if (tmp == (size_t)0)
- /* found '\0' char */
- break;
+ else if (MB_NULLWCH (tmp))
+ break; /* found wide '\0' */
else
{
/* valid bytes */
point += tmp;
if (find_non_zero)
{
- if (wcwidth (wc) == 0)
+ if (WCWIDTH (wc) == 0)
continue;
else
count--;
if (find_non_zero)
{
tmp = mbrtowc (&wc, string + point, strlen (string + point), &ps);
- while (wcwidth (wc) == 0)
+ while (MB_NULLWCH (tmp) == 0 && MB_INVALIDCH (tmp) == 0 && WCWIDTH (wc) == 0)
{
point += tmp;
tmp = mbrtowc (&wc, string + point, strlen (string + point), &ps);
- if (tmp == (size_t)(0) || tmp == (size_t)(-1) || tmp == (size_t)(-2))
- break;
}
}
- return point;
+
+ return point;
}
-static int
-_rl_find_prev_mbchar_internal (string, seed, find_non_zero)
- char *string;
- int seed, find_non_zero;
+/*static*/ int
+_rl_find_prev_mbchar_internal (char *string, int seed, int find_non_zero)
{
mbstate_t ps;
int prev, non_zero_prev, point, length;
prev = non_zero_prev = point = 0;
while (point < seed)
{
- tmp = mbrtowc (&wc, string + point, length - point, &ps);
- if ((size_t)(tmp) == (size_t)-1 || (size_t)(tmp) == (size_t)-2)
+ if (_rl_utf8locale && UTF8_SINGLEBYTE(string[point]))
{
- /* in this case, bytes are invalid or shorted to compose
+ tmp = 1;
+ wc = (wchar_t) string[point];
+ memset(&ps, 0, sizeof(mbstate_t));
+ }
+ else
+ tmp = mbrtowc (&wc, string + point, length - point, &ps);
+ if (MB_INVALIDCH ((size_t)tmp))
+ {
+ /* in this case, bytes are invalid or too short to compose
multibyte char, so assume that the first byte represents
a single character anyway. */
tmp = 1;
/* clear the state of the byte sequence, because
in this case effect of mbstate is undefined */
memset(&ps, 0, sizeof (mbstate_t));
+
+ /* Since we're assuming that this byte represents a single
+ non-zero-width character, don't forget about it. */
+ prev = point;
}
- else if (tmp == 0)
+ else if (MB_NULLWCH (tmp))
break; /* Found '\0' char. Can this happen? */
else
{
if (find_non_zero)
{
- if (wcwidth (wc) != 0)
+ if (WCWIDTH (wc) != 0)
prev = point;
}
else
if an invalid multibyte sequence was encountered. It returns (size_t)(-2)
if it couldn't parse a complete multibyte character. */
int
-_rl_get_char_len (src, ps)
- char *src;
- mbstate_t *ps;
+_rl_get_char_len (char *src, mbstate_t *ps)
{
- size_t tmp;
+ size_t tmp, l;
+ int mb_cur_max;
- tmp = mbrlen((const char *)src, (size_t)strlen (src), ps);
+ /* Look at no more than MB_CUR_MAX characters */
+ l = (size_t)strlen (src);
+ if (_rl_utf8locale && l > 0 && UTF8_SINGLEBYTE(*src))
+ tmp = (*src != 0) ? 1 : 0;
+ else
+ {
+ mb_cur_max = MB_CUR_MAX;
+ tmp = mbrlen((const char *)src, (l < mb_cur_max) ? l : mb_cur_max, ps);
+ }
if (tmp == (size_t)(-2))
{
- /* shorted to compose multibyte char */
- memset (ps, 0, sizeof(mbstate_t));
+ /* too short to compose multibyte char */
+ if (ps)
+ memset (ps, 0, sizeof(mbstate_t));
return -2;
}
else if (tmp == (size_t)(-1))
{
/* invalid to compose multibyte char */
/* initialize the conversion state */
- memset (ps, 0, sizeof(mbstate_t));
+ if (ps)
+ memset (ps, 0, sizeof(mbstate_t));
return -1;
}
else if (tmp == (size_t)0)
/* compare the specified two characters. If the characters matched,
return 1. Otherwise return 0. */
int
-_rl_compare_chars (buf1, pos1, ps1, buf2, pos2, ps2)
- char *buf1, *buf2;
- mbstate_t *ps1, *ps2;
- int pos1, pos2;
+_rl_compare_chars (char *buf1, int pos1, mbstate_t *ps1, char *buf2, int pos2, mbstate_t *ps2)
{
int i, w1, w2;
/* adjust pointed byte and find mbstate of the point of string.
adjusted point will be point <= adjusted_point, and returns
differences of the byte(adjusted_point - point).
- if point is invalied (point < 0 || more than string length),
+ if point is invalid (point < 0 || more than string length),
it returns -1 */
int
-_rl_adjust_point(string, point, ps)
- char *string;
- int point;
- mbstate_t *ps;
+_rl_adjust_point (char *string, int point, mbstate_t *ps)
{
- size_t tmp = 0;
- int length;
- int pos = 0;
+ size_t tmp;
+ int length, pos;
+ tmp = 0;
+ pos = 0;
length = strlen(string);
if (point < 0)
return -1;
while (pos < point)
{
- tmp = mbrlen (string + pos, length - pos, ps);
- if((size_t)(tmp) == (size_t)-1 || (size_t)(tmp) == (size_t)-2)
+ if (_rl_utf8locale && UTF8_SINGLEBYTE(string[pos]))
+ tmp = 1;
+ else
+ tmp = mbrlen (string + pos, length - pos, ps);
+ if (MB_INVALIDCH ((size_t)tmp))
{
- /* in this case, bytes are invalid or shorted to compose
+ /* in this case, bytes are invalid or too short to compose
multibyte char, so assume that the first byte represents
a single character anyway. */
pos++;
/* clear the state of the byte sequence, because
in this case effect of mbstate is undefined */
- memset (ps, 0, sizeof (mbstate_t));
+ if (ps)
+ memset (ps, 0, sizeof (mbstate_t));
}
+ else if (MB_NULLWCH (tmp))
+ pos++;
else
pos += tmp;
}
}
int
-_rl_is_mbchar_matched (string, seed, end, mbchar, length)
- char *string;
- int seed, end;
- char *mbchar;
- int length;
+_rl_is_mbchar_matched (char *string, int seed, int end, char *mbchar, int length)
{
int i;
return 0;
return 1;
}
+
+wchar_t
+_rl_char_value (char *buf, int ind)
+{
+ size_t tmp;
+ wchar_t wc;
+ mbstate_t ps;
+ int l;
+
+ if (MB_LEN_MAX == 1 || rl_byte_oriented)
+ return ((wchar_t) buf[ind]);
+ if (_rl_utf8locale && UTF8_SINGLEBYTE(buf[ind]))
+ return ((wchar_t) buf[ind]);
+ l = strlen (buf);
+ if (ind >= l - 1)
+ return ((wchar_t) buf[ind]);
+ if (l < ind) /* Sanity check */
+ l = strlen (buf+ind);
+ memset (&ps, 0, sizeof (mbstate_t));
+ tmp = mbrtowc (&wc, buf + ind, l - ind, &ps);
+ if (MB_INVALIDCH (tmp) || MB_NULLWCH (tmp))
+ return ((wchar_t) buf[ind]);
+ return wc;
+}
#endif /* HANDLE_MULTIBYTE */
/* Find next `count' characters started byte point of the specified seed.
characters. */
#undef _rl_find_next_mbchar
int
-_rl_find_next_mbchar (string, seed, count, flags)
- char *string;
- int seed, count, flags;
+_rl_find_next_mbchar (char *string, int seed, int count, int flags)
{
#if defined (HANDLE_MULTIBYTE)
return _rl_find_next_mbchar_internal (string, seed, count, flags);
we look for non-zero-width multibyte characters. */
#undef _rl_find_prev_mbchar
int
-_rl_find_prev_mbchar (string, seed, flags)
- char *string;
- int seed, flags;
+_rl_find_prev_mbchar (char *string, int seed, int flags)
{
#if defined (HANDLE_MULTIBYTE)
return _rl_find_prev_mbchar_internal (string, seed, flags);