1: /* mbutil.c -- readline multibyte character utility functions */
2:
3: /* Copyright (C) 2001-2009 Free Software Foundation, Inc.
4:
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.
7:
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.
12:
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.
17:
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/>.
20: */
21:
22: #define READLINE_LIBRARY
23:
24: #if defined (HAVE_CONFIG_H)
25: # include <config.h>
26: #endif
27:
28: #include <sys/types.h>
29: #include <fcntl.h>
30: #include "posixjmp.h"
31:
32: #if defined (HAVE_UNISTD_H)
33: # include <unistd.h> /* for _POSIX_VERSION */
34: #endif /* HAVE_UNISTD_H */
35:
36: #if defined (HAVE_STDLIB_H)
37: # include <stdlib.h>
38: #else
39: # include "ansi_stdlib.h"
40: #endif /* HAVE_STDLIB_H */
41:
42: #include <stdio.h>
43: #include <ctype.h>
44:
45: /* System-specific feature definitions and include files. */
46: #include "rldefs.h"
47: #include "rlmbutil.h"
48:
49: #if defined (TIOCSTAT_IN_SYS_IOCTL)
50: # include <sys/ioctl.h>
51: #endif /* TIOCSTAT_IN_SYS_IOCTL */
52:
53: /* Some standard library routines. */
54: #include "readline.h"
55:
56: #include "rlprivate.h"
57: #include "xmalloc.h"
58:
59: /* Declared here so it can be shared between the readline and history
60: libraries. */
61: #if defined (HANDLE_MULTIBYTE)
62: int rl_byte_oriented = 0;
63: #else
64: int rl_byte_oriented = 1;
65: #endif
66:
67: /* Ditto */
68: int _rl_utf8locale = 0;
69:
70: /* **************************************************************** */
71: /* */
72: /* Multibyte Character Utility Functions */
73: /* */
74: /* **************************************************************** */
75:
76: #if defined(HANDLE_MULTIBYTE)
77:
78: static int
79: _rl_find_next_mbchar_internal (string, seed, count, find_non_zero)
80: char *string;
81: int seed, count, find_non_zero;
82: {
83: size_t tmp, len;
84: mbstate_t ps;
85: int point;
86: wchar_t wc;
87:
88: tmp = 0;
89:
90: memset(&ps, 0, sizeof (mbstate_t));
91: if (seed < 0)
92: seed = 0;
93: if (count <= 0)
94: return seed;
95:
96: point = seed + _rl_adjust_point (string, seed, &ps);
97: /* if this is true, means that seed was not pointing to a byte indicating
98: the beginning of a multibyte character. Correct the point and consume
99: one char. */
100: if (seed < point)
101: count--;
102:
103: while (count > 0)
104: {
105: len = strlen (string + point);
106: if (len == 0)
107: break;
108: tmp = mbrtowc (&wc, string+point, len, &ps);
109: if (MB_INVALIDCH ((size_t)tmp))
110: {
111: /* invalid bytes. assume a byte represents a character */
112: point++;
113: count--;
114: /* reset states. */
115: memset(&ps, 0, sizeof(mbstate_t));
116: }
117: else if (MB_NULLWCH (tmp))
118: break; /* found wide '\0' */
119: else
120: {
121: /* valid bytes */
122: point += tmp;
123: if (find_non_zero)
124: {
125: if (WCWIDTH (wc) == 0)
126: continue;
127: else
128: count--;
129: }
130: else
131: count--;
132: }
133: }
134:
135: if (find_non_zero)
136: {
137: tmp = mbrtowc (&wc, string + point, strlen (string + point), &ps);
138: while (MB_NULLWCH (tmp) == 0 && MB_INVALIDCH (tmp) == 0 && WCWIDTH (wc) == 0)
139: {
140: point += tmp;
141: tmp = mbrtowc (&wc, string + point, strlen (string + point), &ps);
142: }
143: }
144:
145: return point;
146: }
147:
148: static int
149: _rl_find_prev_mbchar_internal (string, seed, find_non_zero)
150: char *string;
151: int seed, find_non_zero;
152: {
153: mbstate_t ps;
154: int prev, non_zero_prev, point, length;
155: size_t tmp;
156: wchar_t wc;
157:
158: memset(&ps, 0, sizeof(mbstate_t));
159: length = strlen(string);
160:
161: if (seed < 0)
162: return 0;
163: else if (length < seed)
164: return length;
165:
166: prev = non_zero_prev = point = 0;
167: while (point < seed)
168: {
169: tmp = mbrtowc (&wc, string + point, length - point, &ps);
170: if (MB_INVALIDCH ((size_t)tmp))
171: {
172: /* in this case, bytes are invalid or shorted to compose
173: multibyte char, so assume that the first byte represents
174: a single character anyway. */
175: tmp = 1;
176: /* clear the state of the byte sequence, because
177: in this case effect of mbstate is undefined */
178: memset(&ps, 0, sizeof (mbstate_t));
179:
180: /* Since we're assuming that this byte represents a single
181: non-zero-width character, don't forget about it. */
182: prev = point;
183: }
184: else if (MB_NULLWCH (tmp))
185: break; /* Found '\0' char. Can this happen? */
186: else
187: {
188: if (find_non_zero)
189: {
190: if (WCWIDTH (wc) != 0)
191: prev = point;
192: }
193: else
194: prev = point;
195: }
196:
197: point += tmp;
198: }
199:
200: return prev;
201: }
202:
203: /* return the number of bytes parsed from the multibyte sequence starting
204: at src, if a non-L'\0' wide character was recognized. It returns 0,
205: if a L'\0' wide character was recognized. It returns (size_t)(-1),
206: if an invalid multibyte sequence was encountered. It returns (size_t)(-2)
207: if it couldn't parse a complete multibyte character. */
208: int
209: _rl_get_char_len (src, ps)
210: char *src;
211: mbstate_t *ps;
212: {
213: size_t tmp;
214:
215: tmp = mbrlen((const char *)src, (size_t)strlen (src), ps);
216: if (tmp == (size_t)(-2))
217: {
218: /* shorted to compose multibyte char */
219: if (ps)
220: memset (ps, 0, sizeof(mbstate_t));
221: return -2;
222: }
223: else if (tmp == (size_t)(-1))
224: {
225: /* invalid to compose multibyte char */
226: /* initialize the conversion state */
227: if (ps)
228: memset (ps, 0, sizeof(mbstate_t));
229: return -1;
230: }
231: else if (tmp == (size_t)0)
232: return 0;
233: else
234: return (int)tmp;
235: }
236:
237: /* compare the specified two characters. If the characters matched,
238: return 1. Otherwise return 0. */
239: int
240: _rl_compare_chars (buf1, pos1, ps1, buf2, pos2, ps2)
241: char *buf1;
242: int pos1;
243: mbstate_t *ps1;
244: char *buf2;
245: int pos2;
246: mbstate_t *ps2;
247: {
248: int i, w1, w2;
249:
250: if ((w1 = _rl_get_char_len (&buf1[pos1], ps1)) <= 0 ||
251: (w2 = _rl_get_char_len (&buf2[pos2], ps2)) <= 0 ||
252: (w1 != w2) ||
253: (buf1[pos1] != buf2[pos2]))
254: return 0;
255:
256: for (i = 1; i < w1; i++)
257: if (buf1[pos1+i] != buf2[pos2+i])
258: return 0;
259:
260: return 1;
261: }
262:
263: /* adjust pointed byte and find mbstate of the point of string.
264: adjusted point will be point <= adjusted_point, and returns
265: differences of the byte(adjusted_point - point).
266: if point is invalied (point < 0 || more than string length),
267: it returns -1 */
268: int
269: _rl_adjust_point (string, point, ps)
270: char *string;
271: int point;
272: mbstate_t *ps;
273: {
274: size_t tmp = 0;
275: int length;
276: int pos = 0;
277:
278: length = strlen(string);
279: if (point < 0)
280: return -1;
281: if (length < point)
282: return -1;
283:
284: while (pos < point)
285: {
286: tmp = mbrlen (string + pos, length - pos, ps);
287: if (MB_INVALIDCH ((size_t)tmp))
288: {
289: /* in this case, bytes are invalid or shorted to compose
290: multibyte char, so assume that the first byte represents
291: a single character anyway. */
292: pos++;
293: /* clear the state of the byte sequence, because
294: in this case effect of mbstate is undefined */
295: if (ps)
296: memset (ps, 0, sizeof (mbstate_t));
297: }
298: else if (MB_NULLWCH (tmp))
299: pos++;
300: else
301: pos += tmp;
302: }
303:
304: return (pos - point);
305: }
306:
307: int
308: _rl_is_mbchar_matched (string, seed, end, mbchar, length)
309: char *string;
310: int seed, end;
311: char *mbchar;
312: int length;
313: {
314: int i;
315:
316: if ((end - seed) < length)
317: return 0;
318:
319: for (i = 0; i < length; i++)
320: if (string[seed + i] != mbchar[i])
321: return 0;
322: return 1;
323: }
324:
325: wchar_t
326: _rl_char_value (buf, ind)
327: char *buf;
328: int ind;
329: {
330: size_t tmp;
331: wchar_t wc;
332: mbstate_t ps;
333: int l;
334:
335: if (MB_LEN_MAX == 1 || rl_byte_oriented)
336: return ((wchar_t) buf[ind]);
337: l = strlen (buf);
338: if (ind >= l - 1)
339: return ((wchar_t) buf[ind]);
340: memset (&ps, 0, sizeof (mbstate_t));
341: tmp = mbrtowc (&wc, buf + ind, l - ind, &ps);
342: if (MB_INVALIDCH (tmp) || MB_NULLWCH (tmp))
343: return ((wchar_t) buf[ind]);
344: return wc;
345: }
346: #endif /* HANDLE_MULTIBYTE */
347:
348: /* Find next `count' characters started byte point of the specified seed.
349: If flags is MB_FIND_NONZERO, we look for non-zero-width multibyte
350: characters. */
351: #undef _rl_find_next_mbchar
352: int
353: _rl_find_next_mbchar (string, seed, count, flags)
354: char *string;
355: int seed, count, flags;
356: {
357: #if defined (HANDLE_MULTIBYTE)
358: return _rl_find_next_mbchar_internal (string, seed, count, flags);
359: #else
360: return (seed + count);
361: #endif
362: }
363:
364: /* Find previous character started byte point of the specified seed.
365: Returned point will be point <= seed. If flags is MB_FIND_NONZERO,
366: we look for non-zero-width multibyte characters. */
367: #undef _rl_find_prev_mbchar
368: int
369: _rl_find_prev_mbchar (string, seed, flags)
370: char *string;
371: int seed, flags;
372: {
373: #if defined (HANDLE_MULTIBYTE)
374: return _rl_find_prev_mbchar_internal (string, seed, flags);
375: #else
376: return ((seed == 0) ? seed : seed - 1);
377: #endif
378: }
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/cvsweb/back.gif){kind=icon}
Return to mbutil.c CVS log ![[TXT]](/icons/cvsweb/text.gif){kind=icon}
![[DIR]](/icons/cvsweb/dir.gif){kind=icon}
Up to [ELWIX - Embedded LightWeight unIX -] / embedaddon / readline