1 /*****************************************************************************
2 * unicode.c: Unicode <-> locale functions
3 *****************************************************************************
4 * Copyright (C) 2005-2006 the VideoLAN team
5 * Copyright © 2005-2008 Rémi Denis-Courmont
7 * Authors: Rémi Denis-Courmont <rem # videolan.org>
9 * This program is free software; you can redistribute it and/or modify
10 * it under the terms of the GNU General Public License as published by
11 * the Free Software Foundation; either version 2 of the License, or
12 * (at your option) any later version.
14 * This program is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17 * GNU General Public License for more details.
19 * You should have received a copy of the GNU General Public License
20 * along with this program; if not, write to the Free Software
21 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
22 *****************************************************************************/
24 /*****************************************************************************
26 *****************************************************************************/
31 #include <vlc_common.h>
32 #include <vlc_charset.h>
39 #include <sys/types.h>
45 #if defined (__APPLE__) || defined (HAVE_MAEMO)
46 /* Define this if the OS always use UTF-8 internally */
47 # define ASSUME_UTF8 1
50 #if defined (ASSUME_UTF8)
53 #elif defined (WIN32) || defined (UNDER_CE)
56 static char *locale_dup (const char *string, bool from)
61 len = 1 + MultiByteToWideChar (from ? CP_ACP : CP_UTF8,
62 0, string, -1, NULL, 0);
63 wchar_t *wide = malloc (len * sizeof (wchar_t));
67 MultiByteToWideChar (from ? CP_ACP : CP_UTF8, 0, string, -1, wide, len);
68 len = 1 + WideCharToMultiByte (from ? CP_UTF8 : CP_ACP, 0, wide, -1,
72 WideCharToMultiByte (from ? CP_UTF8 : CP_ACP, 0, wide, -1, out, len,
78 #elif defined (HAVE_ICONV)
81 static char *locale_dup (const char *string, bool from)
83 vlc_iconv_t hd = vlc_iconv_open (from ? "UTF-8" : "",
85 if (hd == (vlc_iconv_t)(-1))
86 return NULL; /* Uho! */
88 const char *iptr = string;
89 size_t inb = strlen (string);
90 size_t outb = inb * 6 + 1;
91 char output[outb], *optr = output;
93 while (vlc_iconv (hd, &iptr, &inb, &optr, &outb) == (size_t)(-1))
99 vlc_iconv (hd, NULL, NULL, NULL, NULL); /* reset */
102 vlc_iconv_close (hd);
105 assert (*iptr == '\0');
106 assert (*optr == '\0');
107 assert (strlen (output) == (size_t)(optr - output));
108 return strdup (output);
112 # error No UTF8 charset conversion implemented on this platform!
117 * Releases (if needed) a localized or uniformized string.
118 * @param str non-NULL return value from FromLocale() or ToLocale().
120 void LocaleFree (const char *str)
131 * Converts a string from the system locale character encoding to UTF-8.
133 * @param locale nul-terminated string to convert
135 * @return a nul-terminated UTF-8 string, or NULL in case of error.
136 * To avoid memory leak, you have to pass the result to LocaleFree()
137 * when it is no longer needed.
139 char *FromLocale (const char *locale)
142 return (char *)locale;
144 return locale ? locale_dup (locale, true) : NULL;
149 * converts a string from the system locale character encoding to utf-8,
150 * the result is always allocated on the heap.
152 * @param locale nul-terminated string to convert
154 * @return a nul-terminated utf-8 string, or null in case of error.
155 * The result must be freed using free() - as with the strdup() function.
157 char *FromLocaleDup (const char *locale)
160 return strdup (locale);
162 return locale_dup (locale, true);
168 * ToLocale: converts an UTF-8 string to local system encoding.
170 * @param utf8 nul-terminated string to be converted
172 * @return a nul-terminated string, or NULL in case of error.
173 * To avoid memory leak, you have to pass the result to LocaleFree()
174 * when it is no longer needed.
176 char *ToLocale (const char *utf8)
181 return utf8 ? locale_fast (utf8, false) : NULL
187 * converts a string from UTF-8 to the system locale character encoding,
188 * the result is always allocated on the heap.
190 * @param utf8 nul-terminated string to convert
192 * @return a nul-terminated string, or null in case of error.
193 * The result must be freed using free() - as with the strdup() function.
195 char *ToLocaleDup (const char *utf8)
198 return strdup (utf8);
200 return locale_dup (utf8, false);
205 * Formats an UTF-8 string as vasprintf(), then print it to stdout, with
206 * appropriate conversion to local encoding.
208 static int utf8_vasprintf( char **str, const char *fmt, va_list ap )
211 int res = vasprintf( &utf8, fmt, ap );
218 *str = ToLocaleDup( utf8 );
225 * Formats an UTF-8 string as vfprintf(), then print it, with
226 * appropriate conversion to local encoding.
228 int utf8_vfprintf( FILE *stream, const char *fmt, va_list ap )
231 int res = utf8_vasprintf( &str, fmt, ap );
235 fputs( str, stream );
241 * Formats an UTF-8 string as fprintf(), then print it, with
242 * appropriate conversion to local encoding.
244 int utf8_fprintf( FILE *stream, const char *fmt, ... )
250 res = utf8_vfprintf( stream, fmt, ap );
256 static char *CheckUTF8( char *str, char rep )
258 uint8_t *ptr = (uint8_t *)str;
259 assert (str != NULL);
271 int charlen = clz8 (c ^ 0xFF);
274 case 0: // 7-bit ASCII character -> OK
278 case 1: // continuation byte -> error
282 assert (charlen >= 2 && charlen <= 4);
284 uint32_t cp = c & ~((0xff >> (7 - charlen)) << (7 - charlen));
285 for (int i = 1; i < charlen; i++)
287 assert (cp < (1 << 26));
290 if ((c >> 6) != 2) // not a continuation byte
293 cp = (cp << 6) | (ptr[i] & 0x3f);
299 if (cp > 0x10FFFF) // beyond Unicode
302 if (cp >= 0xD800 && cp < 0xC000) // UTF-16 surrogate
305 if (cp < 128) // ASCII overlong
307 if (cp < (1u << (5 * charlen - 3))) // overlong
324 * Replaces invalid/overlong UTF-8 sequences with question marks.
325 * Note that it is not possible to convert from Latin-1 to UTF-8 on the fly,
326 * so we don't try that, even though it would be less disruptive.
328 * @return str if it was valid UTF-8, NULL if not.
330 char *EnsureUTF8( char *str )
332 return CheckUTF8( str, '?' );
337 * Checks whether a string is a valid UTF-8 byte sequence.
339 * @param str nul-terminated string to be checked
341 * @return str if it was valid UTF-8, NULL if not.
343 const char *IsUTF8( const char *str )
345 return CheckUTF8( (char *)str, 0 );
349 * Converts a string from the given character encoding to utf-8.
351 * @return a nul-terminated utf-8 string, or null in case of error.
352 * The result must be freed using free().
354 char *FromCharset(const char *charset, const void *data, size_t data_size)
356 vlc_iconv_t handle = vlc_iconv_open ("UTF-8", charset);
357 if (handle == (vlc_iconv_t)(-1))
361 for(unsigned mul = 4; mul < 8; mul++ )
363 size_t in_size = data_size;
364 const char *in = data;
365 size_t out_max = mul * data_size;
366 char *tmp = out = malloc (1 + out_max);
370 if (vlc_iconv (handle, &in, &in_size, &tmp, &out_max) != (size_t)(-1)) {
380 vlc_iconv_close(handle);