1 /* gutf8.c - Operations on UTF-8 strings.
3 * Copyright (C) 1999 Tom Tromey
4 * Copyright (C) 2000 Red Hat, Inc.
6 * This library is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2 of the License, or (at your option) any later version.
11 * This library is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with this library; if not, write to the
18 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
19 * Boston, MA 02111-1307, USA.
33 #ifdef G_PLATFORM_WIN32
40 #include "libcharset/libcharset.h"
44 #define UTF8_COMPUTE(Char, Mask, Len) \
50 else if ((Char & 0xe0) == 0xc0) \
55 else if ((Char & 0xf0) == 0xe0) \
60 else if ((Char & 0xf8) == 0xf0) \
65 else if ((Char & 0xfc) == 0xf8) \
70 else if ((Char & 0xfe) == 0xfc) \
78 #define UTF8_LENGTH(Char) \
79 ((Char) < 0x80 ? 1 : \
80 ((Char) < 0x800 ? 2 : \
81 ((Char) < 0x10000 ? 3 : \
82 ((Char) < 0x200000 ? 4 : \
83 ((Char) < 0x4000000 ? 5 : 6)))))
86 #define UTF8_GET(Result, Chars, Count, Mask, Len) \
87 (Result) = (Chars)[0] & (Mask); \
88 for ((Count) = 1; (Count) < (Len); ++(Count)) \
90 if (((Chars)[(Count)] & 0xc0) != 0x80) \
96 (Result) |= ((Chars)[(Count)] & 0x3f); \
99 #define UNICODE_VALID(Char) \
100 ((Char) < 0x110000 && \
101 (((Char) & 0xFFFFF800) != 0xD800) && \
102 ((Char) < 0xFDD0 || (Char) > 0xFDEF) && \
103 ((Char) & 0xFFFE) != 0xFFFE)
106 static const gchar utf8_skip_data[256] = {
107 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
108 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
109 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
110 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
111 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
112 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
113 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
114 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6,1,1
117 const gchar * const g_utf8_skip = utf8_skip_data;
120 * g_utf8_find_prev_char:
121 * @str: pointer to the beginning of a UTF-8 encoded string
122 * @p: pointer to some position within @str
124 * Given a position @p with a UTF-8 encoded string @str, find the start
125 * of the previous UTF-8 character starting before @p. Returns %NULL if no
126 * UTF-8 characters are present in @p before @str.
128 * @p does not have to be at the beginning of a UTF-8 character. No check
129 * is made to see if the character found is actually valid other than
130 * it starts with an appropriate byte.
132 * Return value: a pointer to the found character or %NULL.
135 g_utf8_find_prev_char (const char *str,
138 for (--p; p >= str; --p)
140 if ((*p & 0xc0) != 0x80)
147 * g_utf8_find_next_char:
148 * @p: a pointer to a position within a UTF-8 encoded string
149 * @end: a pointer to the end of the string, or %NULL to indicate
150 * that the string is nul-terminated, in which case
151 * the returned value will be
153 * Finds the start of the next UTF-8 character in the string after @p.
155 * @p does not have to be at the beginning of a UTF-8 character. No check
156 * is made to see if the character found is actually valid other than
157 * it starts with an appropriate byte.
159 * Return value: a pointer to the found character or %NULL
162 g_utf8_find_next_char (const gchar *p,
168 for (++p; p < end && (*p & 0xc0) == 0x80; ++p)
171 for (++p; (*p & 0xc0) == 0x80; ++p)
174 return (p == end) ? NULL : (gchar *)p;
179 * @p: a pointer to a position within a UTF-8 encoded string
181 * Finds the previous UTF-8 character in the string before @p.
183 * @p does not have to be at the beginning of a UTF-8 character. No check
184 * is made to see if the character found is actually valid other than
185 * it starts with an appropriate byte. If @p might be the first
186 * character of the string, you must use g_utf8_find_prev_char() instead.
188 * Return value: a pointer to the found character.
191 g_utf8_prev_char (const gchar *p)
196 if ((*p & 0xc0) != 0x80)
203 * @p: pointer to the start of a UTF-8 encoded string.
204 * @max: the maximum number of bytes to examine. If @max
205 * is less than 0, then the string is assumed to be
206 * nul-terminated. If @max is 0, @p will not be examined and
209 * Returns the length of the string in characters.
211 * Return value: the length of the string in characters
214 g_utf8_strlen (const gchar *p,
218 const gchar *start = p;
219 g_return_val_if_fail (p != NULL || max == 0, 0);
225 p = g_utf8_next_char (p);
234 p = g_utf8_next_char (p);
236 while (p - start < max && *p)
239 p = g_utf8_next_char (p);
242 /* only do the last len increment if we got a complete
243 * char (don't count partial chars)
245 if (p - start == max)
254 * @p: a pointer to Unicode character encoded as UTF-8
256 * Converts a sequence of bytes encoded as UTF-8 to a Unicode character.
257 * If @p does not point to a valid UTF-8 encoded character, results are
258 * undefined. If you are not sure that the bytes are complete
259 * valid Unicode characters, you should use g_utf8_get_char_validated()
262 * Return value: the resulting character
265 g_utf8_get_char (const gchar *p)
267 int i, mask = 0, len;
269 unsigned char c = (unsigned char) *p;
271 UTF8_COMPUTE (c, mask, len);
274 UTF8_GET (result, p, i, mask, len);
280 * g_utf8_offset_to_pointer:
281 * @str: a UTF-8 encoded string
282 * @offset: a character offset within @str
284 * Converts from an integer character offset to a pointer to a position
287 * Return value: the resulting pointer
290 g_utf8_offset_to_pointer (const gchar *str,
293 const gchar *s = str;
295 s = g_utf8_next_char (s);
301 * g_utf8_pointer_to_offset:
302 * @str: a UTF-8 encoded string
303 * @pos: a pointer to a position within @str
305 * Converts from a pointer to position within a string to a integer
308 * Return value: the resulting character offset
311 g_utf8_pointer_to_offset (const gchar *str,
314 const gchar *s = str;
319 s = g_utf8_next_char (s);
329 * @dest: buffer to fill with characters from @src
330 * @src: UTF-8 encoded string
331 * @n: character count
333 * Like the standard C strncpy() function, but
334 * copies a given number of characters instead of a given number of
335 * bytes. The @src string must be valid UTF-8 encoded text.
336 * (Use g_utf8_validate() on all text before trying to use UTF-8
337 * utility functions with it.)
339 * Return value: @dest
342 g_utf8_strncpy (gchar *dest,
346 const gchar *s = src;
349 s = g_utf8_next_char(s);
352 strncpy(dest, src, s - src);
357 G_LOCK_DEFINE_STATIC (aliases);
360 get_alias_hash (void)
362 static GHashTable *alias_hash = NULL;
369 alias_hash = g_hash_table_new (g_str_hash, g_str_equal);
371 aliases = _g_locale_get_charset_aliases ();
372 while (*aliases != '\0')
374 const char *canonical;
376 const char **alias_array;
380 aliases += strlen (aliases) + 1;
382 aliases += strlen (aliases) + 1;
384 alias_array = g_hash_table_lookup (alias_hash, canonical);
387 while (alias_array[count])
391 alias_array = g_renew (const char *, alias_array, count + 2);
392 alias_array[count] = alias;
393 alias_array[count + 1] = NULL;
395 g_hash_table_insert (alias_hash, (char *)canonical, alias_array);
404 /* As an abuse of the alias table, the following routines gets
405 * the charsets that are aliases for the canonical name.
408 _g_charset_get_aliases (const char *canonical_name)
410 GHashTable *alias_hash = get_alias_hash ();
412 return g_hash_table_lookup (alias_hash, canonical_name);
416 g_utf8_get_charset_internal (const char *raw_data,
419 const char *charset = getenv("CHARSET");
421 if (charset && *charset)
425 if (charset && strstr (charset, "UTF-8"))
431 /* The libcharset code tries to be thread-safe without
432 * a lock, but has a memory leak and a missing memory
433 * barrier, so we lock for it
436 charset = _g_locale_charset_unalias (raw_data);
439 if (charset && *charset)
443 if (charset && strstr (charset, "UTF-8"))
449 /* Assume this for compatibility at present. */
455 typedef struct _GCharsetCache GCharsetCache;
457 struct _GCharsetCache {
464 charset_cache_free (gpointer data)
466 GCharsetCache *cache = data;
468 g_free (cache->charset);
474 * @charset: return location for character set name
476 * Obtains the character set for the current locale; you might use
477 * this character set as an argument to g_convert(), to convert from
478 * the current locale's encoding to some other encoding. (Frequently
479 * g_locale_to_utf8() and g_locale_from_utf8() are nice shortcuts,
482 * The return value is %TRUE if the locale's encoding is UTF-8, in that
483 * case you can perhaps avoid calling g_convert().
485 * The string returned in @charset is not allocated, and should not be
488 * Return value: %TRUE if the returned charset is UTF-8
491 g_get_charset (G_CONST_RETURN char **charset)
493 static GStaticPrivate cache_private = G_STATIC_PRIVATE_INIT;
494 GCharsetCache *cache = g_static_private_get (&cache_private);
499 cache = g_new0 (GCharsetCache, 1);
500 g_static_private_set (&cache_private, cache, charset_cache_free);
503 raw = _g_locale_charset_raw ();
505 if (!(cache->raw && strcmp (cache->raw, raw) == 0))
507 const gchar *new_charset;
510 g_free (cache->charset);
511 cache->raw = g_strdup (raw);
512 cache->is_utf8 = g_utf8_get_charset_internal (raw, &new_charset);
513 cache->charset = g_strdup (new_charset);
517 *charset = cache->charset;
519 return cache->is_utf8;
526 * @c: a ISO10646 character code
527 * @outbuf: output buffer, must have at least 6 bytes of space.
528 * If %NULL, the length will be computed and returned
529 * and nothing will be written to @outbuf.
531 * Converts a single character to UTF-8.
533 * Return value: number of bytes written
536 g_unichar_to_utf8 (gunichar c,
553 else if (c < 0x10000)
558 else if (c < 0x200000)
563 else if (c < 0x4000000)
576 for (i = len - 1; i > 0; --i)
578 outbuf[i] = (c & 0x3f) | 0x80;
581 outbuf[0] = c | first;
589 * @p: a nul-terminated UTF-8 encoded string
590 * @len: the maximum length of @p
591 * @c: a ISO10646 character
593 * Finds the leftmost occurrence of the given ISO10646 character
594 * in a UTF-8 encoded string, while limiting the search to @len bytes.
595 * If @len is -1, allow unbounded search.
597 * Return value: %NULL if the string does not contain the character,
598 * otherwise, a pointer to the start of the leftmost occurrence of
599 * the character in the string.
602 g_utf8_strchr (const char *p,
608 gint charlen = g_unichar_to_utf8 (c, ch);
611 return g_strstr_len (p, len, ch);
617 * @p: a nul-terminated UTF-8 encoded string
618 * @len: the maximum length of @p
619 * @c: a ISO10646 character
621 * Find the rightmost occurrence of the given ISO10646 character
622 * in a UTF-8 encoded string, while limiting the search to @len bytes.
623 * If @len is -1, allow unbounded search.
625 * Return value: %NULL if the string does not contain the character,
626 * otherwise, a pointer to the start of the rightmost occurrence of the
627 * character in the string.
630 g_utf8_strrchr (const char *p,
636 gint charlen = g_unichar_to_utf8 (c, ch);
639 return g_strrstr_len (p, len, ch);
643 /* Like g_utf8_get_char, but take a maximum length
644 * and return (gunichar)-2 on incomplete trailing character
646 static inline gunichar
647 g_utf8_get_char_extended (const gchar *p,
651 gunichar wc = (guchar) *p;
691 if (max_len >= 0 && len > max_len)
693 for (i = 1; i < max_len; i++)
695 if ((((guchar *)p)[i] & 0xc0) != 0x80)
701 for (i = 1; i < len; ++i)
703 gunichar ch = ((guchar *)p)[i];
705 if ((ch & 0xc0) != 0x80)
717 if (UTF8_LENGTH(wc) != len)
724 * g_utf8_get_char_validated:
725 * @p: a pointer to Unicode character encoded as UTF-8
726 * @max_len: the maximum number of bytes to read, or -1, for no maximum.
728 * Convert a sequence of bytes encoded as UTF-8 to a Unicode character.
729 * This function checks for incomplete characters, for invalid characters
730 * such as characters that are out of the range of Unicode, and for
731 * overlong encodings of valid characters.
733 * Return value: the resulting character. If @p points to a partial
734 * sequence at the end of a string that could begin a valid
735 * character, returns (gunichar)-2; otherwise, if @p does not point
736 * to a valid UTF-8 encoded Unicode character, returns (gunichar)-1.
739 g_utf8_get_char_validated (const gchar *p,
742 gunichar result = g_utf8_get_char_extended (p, max_len);
744 if (result & 0x80000000)
746 else if (!UNICODE_VALID (result))
753 * g_utf8_to_ucs4_fast:
754 * @str: a UTF-8 encoded string
755 * @len: the maximum length of @str to use. If @len < 0, then
756 * the string is nul-terminated.
757 * @items_written: location to store the number of characters in the
760 * Convert a string from UTF-8 to a 32-bit fixed width
761 * representation as UCS-4, assuming valid UTF-8 input.
762 * This function is roughly twice as fast as g_utf8_to_ucs4()
763 * but does no error checking on the input.
765 * Return value: a pointer to a newly allocated UCS-4 string.
766 * This value must be freed with g_free().
769 g_utf8_to_ucs4_fast (const gchar *str,
771 glong *items_written)
778 g_return_val_if_fail (str != NULL, NULL);
786 p = g_utf8_next_char (p);
792 while (p < str + len && *p)
794 p = g_utf8_next_char (p);
799 result = g_new (gunichar, n_chars + 1);
802 for (i=0; i < n_chars; i++)
804 gunichar wc = ((unsigned char *)p)[0];
839 for (j = 1; j < charlen; j++)
842 wc |= ((unsigned char *)p)[j] & 0x3f;
859 * @str: a UTF-8 encoded string
860 * @len: the maximum length of @str to use. If @len < 0, then
861 * the string is nul-terminated.
862 * @items_read: location to store number of bytes read, or %NULL.
863 * If %NULL, then %G_CONVERT_ERROR_PARTIAL_INPUT will be
864 * returned in case @str contains a trailing partial
865 * character. If an error occurs then the index of the
866 * invalid input is stored here.
867 * @items_written: location to store number of characters written or %NULL.
868 * The value here stored does not include the trailing 0
870 * @error: location to store the error occuring, or %NULL to ignore
871 * errors. Any of the errors in #GConvertError other than
872 * %G_CONVERT_ERROR_NO_CONVERSION may occur.
874 * Convert a string from UTF-8 to a 32-bit fixed width
875 * representation as UCS-4. A trailing 0 will be added to the
876 * string after the converted text.
878 * Return value: a pointer to a newly allocated UCS-4 string.
879 * This value must be freed with g_free(). If an
880 * error occurs, %NULL will be returned and
884 g_utf8_to_ucs4 (const gchar *str,
887 glong *items_written,
890 gunichar *result = NULL;
896 while ((len < 0 || str + len - in > 0) && *in)
898 gunichar wc = g_utf8_get_char_extended (in, str + len - in);
901 if (wc == (gunichar)-2)
906 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_PARTIAL_INPUT,
907 _("Partial character sequence at end of input"));
910 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
911 _("Invalid byte sequence in conversion input"));
918 in = g_utf8_next_char (in);
921 result = g_new (gunichar, n_chars + 1);
924 for (i=0; i < n_chars; i++)
926 result[i] = g_utf8_get_char (in);
927 in = g_utf8_next_char (in);
932 *items_written = n_chars;
936 *items_read = in - str;
943 * @str: a UCS-4 encoded string
944 * @len: the maximum length of @str to use. If @len < 0, then
945 * the string is terminated with a 0 character.
946 * @items_read: location to store number of characters read read, or %NULL.
947 * @items_written: location to store number of bytes written or %NULL.
948 * The value here stored does not include the trailing 0
950 * @error: location to store the error occuring, or %NULL to ignore
951 * errors. Any of the errors in #GConvertError other than
952 * %G_CONVERT_ERROR_NO_CONVERSION may occur.
954 * Convert a string from a 32-bit fixed width representation as UCS-4.
955 * to UTF-8. The result will be terminated with a 0 byte.
957 * Return value: a pointer to a newly allocated UTF-8 string.
958 * This value must be freed with g_free(). If an
959 * error occurs, %NULL will be returned and
963 g_ucs4_to_utf8 (const gunichar *str,
966 glong *items_written,
970 gchar *result = NULL;
975 for (i = 0; len < 0 || i < len ; i++)
980 if (str[i] >= 0x80000000)
985 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
986 _("Character out of range for UTF-8"));
990 result_length += UTF8_LENGTH (str[i]);
993 result = g_malloc (result_length + 1);
997 while (p < result + result_length)
998 p += g_unichar_to_utf8 (str[i++], p);
1003 *items_written = p - result;
1012 #define SURROGATE_VALUE(h,l) (((h) - 0xd800) * 0x400 + (l) - 0xdc00 + 0x10000)
1016 * @str: a UTF-16 encoded string
1017 * @len: the maximum length of @str to use. If @len < 0, then
1018 * the string is terminated with a 0 character.
1019 * @items_read: location to store number of words read, or %NULL.
1020 * If %NULL, then %G_CONVERT_ERROR_PARTIAL_INPUT will be
1021 * returned in case @str contains a trailing partial
1022 * character. If an error occurs then the index of the
1023 * invalid input is stored here.
1024 * @items_written: location to store number of bytes written, or %NULL.
1025 * The value stored here does not include the trailing
1027 * @error: location to store the error occuring, or %NULL to ignore
1028 * errors. Any of the errors in #GConvertError other than
1029 * %G_CONVERT_ERROR_NO_CONVERSION may occur.
1031 * Convert a string from UTF-16 to UTF-8. The result will be
1032 * terminated with a 0 byte.
1034 * Return value: a pointer to a newly allocated UTF-8 string.
1035 * This value must be freed with g_free(). If an
1036 * error occurs, %NULL will be returned and
1040 g_utf16_to_utf8 (const gunichar2 *str,
1043 glong *items_written,
1046 /* This function and g_utf16_to_ucs4 are almost exactly identical - The lines that differ
1049 const gunichar2 *in;
1051 gchar *result = NULL;
1053 gunichar high_surrogate;
1055 g_return_val_if_fail (str != 0, NULL);
1060 while ((len < 0 || in - str < len) && *in)
1065 if (c >= 0xdc00 && c < 0xe000) /* low surrogate */
1069 wc = SURROGATE_VALUE (high_surrogate, c);
1074 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1075 _("Invalid sequence in conversion input"));
1083 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1084 _("Invalid sequence in conversion input"));
1088 if (c >= 0xd800 && c < 0xdc00) /* high surrogate */
1097 /********** DIFFERENT for UTF8/UCS4 **********/
1098 n_bytes += UTF8_LENGTH (wc);
1104 if (high_surrogate && !items_read)
1106 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_PARTIAL_INPUT,
1107 _("Partial character sequence at end of input"));
1111 /* At this point, everything is valid, and we just need to convert
1113 /********** DIFFERENT for UTF8/UCS4 **********/
1114 result = g_malloc (n_bytes + 1);
1119 while (out < result + n_bytes)
1124 if (c >= 0xdc00 && c < 0xe000) /* low surrogate */
1126 wc = SURROGATE_VALUE (high_surrogate, c);
1129 else if (c >= 0xd800 && c < 0xdc00) /* high surrogate */
1137 /********** DIFFERENT for UTF8/UCS4 **********/
1138 out += g_unichar_to_utf8 (wc, out);
1144 /********** DIFFERENT for UTF8/UCS4 **********/
1148 /********** DIFFERENT for UTF8/UCS4 **********/
1149 *items_written = out - result;
1153 *items_read = in - str;
1160 * @str: a UTF-16 encoded string
1161 * @len: the maximum length of @str to use. If @len < 0, then
1162 * the string is terminated with a 0 character.
1163 * @items_read: location to store number of words read, or %NULL.
1164 * If %NULL, then %G_CONVERT_ERROR_PARTIAL_INPUT will be
1165 * returned in case @str contains a trailing partial
1166 * character. If an error occurs then the index of the
1167 * invalid input is stored here.
1168 * @items_written: location to store number of characters written, or %NULL.
1169 * The value stored here does not include the trailing
1171 * @error: location to store the error occuring, or %NULL to ignore
1172 * errors. Any of the errors in #GConvertError other than
1173 * %G_CONVERT_ERROR_NO_CONVERSION may occur.
1175 * Convert a string from UTF-16 to UCS-4. The result will be
1176 * terminated with a 0 character.
1178 * Return value: a pointer to a newly allocated UCS-4 string.
1179 * This value must be freed with g_free(). If an
1180 * error occurs, %NULL will be returned and
1184 g_utf16_to_ucs4 (const gunichar2 *str,
1187 glong *items_written,
1190 const gunichar2 *in;
1192 gchar *result = NULL;
1194 gunichar high_surrogate;
1196 g_return_val_if_fail (str != 0, NULL);
1201 while ((len < 0 || in - str < len) && *in)
1206 if (c >= 0xdc00 && c < 0xe000) /* low surrogate */
1210 wc = SURROGATE_VALUE (high_surrogate, c);
1215 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1216 _("Invalid sequence in conversion input"));
1224 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1225 _("Invalid sequence in conversion input"));
1229 if (c >= 0xd800 && c < 0xdc00) /* high surrogate */
1238 /********** DIFFERENT for UTF8/UCS4 **********/
1239 n_bytes += sizeof (gunichar);
1245 if (high_surrogate && !items_read)
1247 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_PARTIAL_INPUT,
1248 _("Partial character sequence at end of input"));
1252 /* At this point, everything is valid, and we just need to convert
1254 /********** DIFFERENT for UTF8/UCS4 **********/
1255 result = g_malloc (n_bytes + 4);
1260 while (out < result + n_bytes)
1265 if (c >= 0xdc00 && c < 0xe000) /* low surrogate */
1267 wc = SURROGATE_VALUE (high_surrogate, c);
1270 else if (c >= 0xd800 && c < 0xdc00) /* high surrogate */
1278 /********** DIFFERENT for UTF8/UCS4 **********/
1279 *(gunichar *)out = wc;
1280 out += sizeof (gunichar);
1286 /********** DIFFERENT for UTF8/UCS4 **********/
1287 *(gunichar *)out = 0;
1290 /********** DIFFERENT for UTF8/UCS4 **********/
1291 *items_written = (out - result) / sizeof (gunichar);
1295 *items_read = in - str;
1297 return (gunichar *)result;
1302 * @str: a UTF-8 encoded string
1303 * @len: the maximum length of @str to use. If @len < 0, then
1304 * the string is nul-terminated.
1305 * @items_read: location to store number of bytes read, or %NULL.
1306 * If %NULL, then %G_CONVERT_ERROR_PARTIAL_INPUT will be
1307 * returned in case @str contains a trailing partial
1308 * character. If an error occurs then the index of the
1309 * invalid input is stored here.
1310 * @items_written: location to store number of words written, or %NULL.
1311 * The value stored here does not include the trailing
1313 * @error: location to store the error occuring, or %NULL to ignore
1314 * errors. Any of the errors in #GConvertError other than
1315 * %G_CONVERT_ERROR_NO_CONVERSION may occur.
1317 * Convert a string from UTF-8 to UTF-16. A 0 word will be
1318 * added to the result after the converted text.
1320 * Return value: a pointer to a newly allocated UTF-16 string.
1321 * This value must be freed with g_free(). If an
1322 * error occurs, %NULL will be returned and
1326 g_utf8_to_utf16 (const gchar *str,
1329 glong *items_written,
1332 gunichar2 *result = NULL;
1337 g_return_val_if_fail (str != NULL, NULL);
1341 while ((len < 0 || str + len - in > 0) && *in)
1343 gunichar wc = g_utf8_get_char_extended (in, str + len - in);
1344 if (wc & 0x80000000)
1346 if (wc == (gunichar)-2)
1351 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_PARTIAL_INPUT,
1352 _("Partial character sequence at end of input"));
1355 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1356 _("Invalid byte sequence in conversion input"));
1363 else if (wc < 0xe000)
1365 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1366 _("Invalid sequence in conversion input"));
1370 else if (wc < 0x10000)
1372 else if (wc < 0x110000)
1376 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1377 _("Character out of range for UTF-16"));
1382 in = g_utf8_next_char (in);
1385 result = g_new (gunichar2, n16 + 1);
1388 for (i = 0; i < n16;)
1390 gunichar wc = g_utf8_get_char (in);
1398 result[i++] = (wc - 0x10000) / 0x400 + 0xd800;
1399 result[i++] = (wc - 0x10000) % 0x400 + 0xdc00;
1402 in = g_utf8_next_char (in);
1408 *items_written = n16;
1412 *items_read = in - str;
1419 * @str: a UCS-4 encoded string
1420 * @len: the maximum length of @str to use. If @len < 0, then
1421 * the string is terminated with a 0 character.
1422 * @items_read: location to store number of bytes read, or %NULL.
1423 * If an error occurs then the index of the invalid input
1425 * @items_written: location to store number of words written, or %NULL.
1426 * The value stored here does not include the trailing
1428 * @error: location to store the error occuring, or %NULL to ignore
1429 * errors. Any of the errors in #GConvertError other than
1430 * %G_CONVERT_ERROR_NO_CONVERSION may occur.
1432 * Convert a string from UCS-4 to UTF-16. A 0 word will be
1433 * added to the result after the converted text.
1435 * Return value: a pointer to a newly allocated UTF-16 string.
1436 * This value must be freed with g_free(). If an
1437 * error occurs, %NULL will be returned and
1441 g_ucs4_to_utf16 (const gunichar *str,
1444 glong *items_written,
1447 gunichar2 *result = NULL;
1453 while ((len < 0 || i < len) && str[i])
1455 gunichar wc = str[i];
1459 else if (wc < 0xe000)
1461 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1462 _("Invalid sequence in conversion input"));
1466 else if (wc < 0x10000)
1468 else if (wc < 0x110000)
1472 g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
1473 _("Character out of range for UTF-16"));
1481 result = g_new (gunichar2, n16 + 1);
1483 for (i = 0, j = 0; j < n16; i++)
1485 gunichar wc = str[i];
1493 result[j++] = (wc - 0x10000) / 0x400 + 0xd800;
1494 result[j++] = (wc - 0x10000) % 0x400 + 0xdc00;
1500 *items_written = n16;
1511 * @str: a pointer to character data
1512 * @max_len: max bytes to validate, or -1 to go until nul
1513 * @end: return location for end of valid data
1515 * Validates UTF-8 encoded text. @str is the text to validate;
1516 * if @str is nul-terminated, then @max_len can be -1, otherwise
1517 * @max_len should be the number of bytes to validate.
1518 * If @end is non-%NULL, then the end of the valid range
1519 * will be stored there (i.e. the address of the first invalid byte
1520 * if some bytes were invalid, or the end of the text being validated
1523 * Returns %TRUE if all of @str was valid. Many GLib and GTK+
1524 * routines <emphasis>require</emphasis> valid UTF-8 as input;
1525 * so data read from a file or the network should be checked
1526 * with g_utf8_validate() before doing anything else with it.
1528 * Return value: %TRUE if the text was valid UTF-8
1531 g_utf8_validate (const gchar *str,
1538 g_return_val_if_fail (str != NULL, FALSE);
1545 while ((max_len < 0 || (p - str) < max_len) && *p)
1547 int i, mask = 0, len;
1549 unsigned char c = (unsigned char) *p;
1551 UTF8_COMPUTE (c, mask, len);
1556 /* check that the expected number of bytes exists in str */
1558 ((max_len - (p - str)) < len))
1561 UTF8_GET (result, p, i, mask, len);
1563 if (UTF8_LENGTH (result) != len) /* Check for overlong UTF-8 */
1566 if (result == (gunichar)-1)
1569 if (!UNICODE_VALID (result))
1578 /* See that we covered the entire length if a length was
1579 * passed in, or that we ended on a nul if not
1582 p != (str + max_len))
1584 else if (max_len < 0 &&
1592 * g_unichar_validate:
1593 * @ch: a Unicode character
1595 * Checks whether @ch is a valid Unicode character. Some possible
1596 * integer values of @ch will not be valid. 0 is considered a valid
1597 * character, though it's normally a string terminator.
1599 * Return value: %TRUE if @ch is a valid Unicode character
1602 g_unichar_validate (gunichar ch)
1604 return UNICODE_VALID (ch);
1608 * g_utf8_strreverse:
1609 * @str: a UTF-8 encoded string
1610 * @len: the maximum length of @str to use. If @len < 0, then
1611 * the string is nul-terminated.
1613 * Reverses a UTF-8 string. @str must be valid UTF-8 encoded text.
1614 * (Use g_utf8_validate() on all text before trying to use UTF-8
1615 * utility functions with it.)
1617 * Note that unlike g_strreverse(), this function returns
1618 * newly-allocated memory, which should be freed with g_free() when
1621 * Returns: a newly-allocated string which is the reverse of @str.
1626 g_utf8_strreverse (const gchar *str,
1636 result = g_new (gchar, len + 1);
1641 skip = g_utf8_skip[*(guchar*)p];
1643 for (m = r; skip; skip--)