1 /* mbutil.c -- readline multibyte character utility functions */
3 /* Copyright (C) 2001-2009 Free Software Foundation, Inc.
5 This file is part of the GNU Readline Library (Readline), a library
6 for reading lines of text with interactive input and history editing.
8 Readline is free software: you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation, either version 3 of the License, or
11 (at your option) any later version.
13 Readline is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
18 You should have received a copy of the GNU General Public License
19 along with Readline. If not, see <http://www.gnu.org/licenses/>.
22 #define READLINE_LIBRARY
24 #if defined (HAVE_CONFIG_H)
28 #include <sys/types.h>
32 #if defined (HAVE_UNISTD_H)
33 # include <unistd.h> /* for _POSIX_VERSION */
34 #endif /* HAVE_UNISTD_H */
36 #if defined (HAVE_STDLIB_H)
39 # include "ansi_stdlib.h"
40 #endif /* HAVE_STDLIB_H */
45 /* System-specific feature definitions and include files. */
49 #if defined (TIOCSTAT_IN_SYS_IOCTL)
50 # include <sys/ioctl.h>
51 #endif /* TIOCSTAT_IN_SYS_IOCTL */
53 /* Some standard library routines. */
56 #include "rlprivate.h"
59 /* Declared here so it can be shared between the readline and history
61 #if defined (HANDLE_MULTIBYTE)
62 int rl_byte_oriented = 0;
64 int rl_byte_oriented = 1;
67 /* **************************************************************** */
69 /* Multibyte Character Utility Functions */
71 /* **************************************************************** */
73 #if defined(HANDLE_MULTIBYTE)
76 _rl_find_next_mbchar_internal (string, seed, count, find_non_zero)
78 int seed, count, find_non_zero;
87 memset(&ps, 0, sizeof (mbstate_t));
93 point = seed + _rl_adjust_point (string, seed, &ps);
94 /* if this is true, means that seed was not pointing to a byte indicating
95 the beginning of a multibyte character. Correct the point and consume
102 tmp = mbrtowc (&wc, string+point, strlen(string + point), &ps);
103 if (MB_INVALIDCH ((size_t)tmp))
105 /* invalid bytes. assume a byte represents a character */
109 memset(&ps, 0, sizeof(mbstate_t));
111 else if (MB_NULLWCH (tmp))
112 break; /* found wide '\0' */
119 if (wcwidth (wc) == 0)
131 tmp = mbrtowc (&wc, string + point, strlen (string + point), &ps);
132 while (MB_NULLWCH (tmp) == 0 && MB_INVALIDCH (tmp) == 0 && wcwidth (wc) == 0)
135 tmp = mbrtowc (&wc, string + point, strlen (string + point), &ps);
143 _rl_find_prev_mbchar_internal (string, seed, find_non_zero)
145 int seed, find_non_zero;
148 int prev, non_zero_prev, point, length;
152 memset(&ps, 0, sizeof(mbstate_t));
153 length = strlen(string);
157 else if (length < seed)
160 prev = non_zero_prev = point = 0;
163 tmp = mbrtowc (&wc, string + point, length - point, &ps);
164 if (MB_INVALIDCH ((size_t)tmp))
166 /* in this case, bytes are invalid or shorted to compose
167 multibyte char, so assume that the first byte represents
168 a single character anyway. */
170 /* clear the state of the byte sequence, because
171 in this case effect of mbstate is undefined */
172 memset(&ps, 0, sizeof (mbstate_t));
174 /* Since we're assuming that this byte represents a single
175 non-zero-width character, don't forget about it. */
178 else if (MB_NULLWCH (tmp))
179 break; /* Found '\0' char. Can this happen? */
184 if (wcwidth (wc) != 0)
197 /* return the number of bytes parsed from the multibyte sequence starting
198 at src, if a non-L'\0' wide character was recognized. It returns 0,
199 if a L'\0' wide character was recognized. It returns (size_t)(-1),
200 if an invalid multibyte sequence was encountered. It returns (size_t)(-2)
201 if it couldn't parse a complete multibyte character. */
203 _rl_get_char_len (src, ps)
209 tmp = mbrlen((const char *)src, (size_t)strlen (src), ps);
210 if (tmp == (size_t)(-2))
212 /* shorted to compose multibyte char */
214 memset (ps, 0, sizeof(mbstate_t));
217 else if (tmp == (size_t)(-1))
219 /* invalid to compose multibyte char */
220 /* initialize the conversion state */
222 memset (ps, 0, sizeof(mbstate_t));
225 else if (tmp == (size_t)0)
231 /* compare the specified two characters. If the characters matched,
232 return 1. Otherwise return 0. */
234 _rl_compare_chars (buf1, pos1, ps1, buf2, pos2, ps2)
244 if ((w1 = _rl_get_char_len (&buf1[pos1], ps1)) <= 0 ||
245 (w2 = _rl_get_char_len (&buf2[pos2], ps2)) <= 0 ||
247 (buf1[pos1] != buf2[pos2]))
250 for (i = 1; i < w1; i++)
251 if (buf1[pos1+i] != buf2[pos2+i])
257 /* adjust pointed byte and find mbstate of the point of string.
258 adjusted point will be point <= adjusted_point, and returns
259 differences of the byte(adjusted_point - point).
260 if point is invalied (point < 0 || more than string length),
263 _rl_adjust_point(string, point, ps)
272 length = strlen(string);
280 tmp = mbrlen (string + pos, length - pos, ps);
281 if (MB_INVALIDCH ((size_t)tmp))
283 /* in this case, bytes are invalid or shorted to compose
284 multibyte char, so assume that the first byte represents
285 a single character anyway. */
287 /* clear the state of the byte sequence, because
288 in this case effect of mbstate is undefined */
290 memset (ps, 0, sizeof (mbstate_t));
292 else if (MB_NULLWCH (tmp))
298 return (pos - point);
302 _rl_is_mbchar_matched (string, seed, end, mbchar, length)
310 if ((end - seed) < length)
313 for (i = 0; i < length; i++)
314 if (string[seed + i] != mbchar[i])
320 _rl_char_value (buf, ind)
329 if (MB_LEN_MAX == 1 || rl_byte_oriented)
330 return ((wchar_t) buf[ind]);
333 return ((wchar_t) buf[ind]);
334 memset (&ps, 0, sizeof (mbstate_t));
335 tmp = mbrtowc (&wc, buf + ind, l - ind, &ps);
336 if (MB_INVALIDCH (tmp) || MB_NULLWCH (tmp))
337 return ((wchar_t) buf[ind]);
340 #endif /* HANDLE_MULTIBYTE */
342 /* Find next `count' characters started byte point of the specified seed.
343 If flags is MB_FIND_NONZERO, we look for non-zero-width multibyte
345 #undef _rl_find_next_mbchar
347 _rl_find_next_mbchar (string, seed, count, flags)
349 int seed, count, flags;
351 #if defined (HANDLE_MULTIBYTE)
352 return _rl_find_next_mbchar_internal (string, seed, count, flags);
354 return (seed + count);
358 /* Find previous character started byte point of the specified seed.
359 Returned point will be point <= seed. If flags is MB_FIND_NONZERO,
360 we look for non-zero-width multibyte characters. */
361 #undef _rl_find_prev_mbchar
363 _rl_find_prev_mbchar (string, seed, flags)
367 #if defined (HANDLE_MULTIBYTE)
368 return _rl_find_prev_mbchar_internal (string, seed, flags);
370 return ((seed == 0) ? seed : seed - 1);