2 * gunicode.c: Some Unicode routines
5 * Miguel de Icaza (miguel@novell.com)
7 * (C) 2006 Novell, Inc.
9 * utf8 validation code came from:
10 * libxml2-2.6.26 licensed under the MIT X11 license
12 * Authors credit in libxml's string.c:
13 * William Brack <wbrack@mmm.com.hk>
16 * Permission is hereby granted, free of charge, to any person obtaining
17 * a copy of this software and associated documentation files (the
18 * "Software"), to deal in the Software without restriction, including
19 * without limitation the rights to use, copy, modify, merge, publish,
20 * distribute, sublicense, and/or sell copies of the Software, and to
21 * permit persons to whom the Software is furnished to do so, subject to
22 * the following conditions:
24 * The above copyright notice and this permission notice shall be
25 * included in all copies or substantial portions of the Software.
27 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
28 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
29 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
30 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
31 * LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
32 * OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
33 * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
39 #include <unicode-data.h>
42 #if defined(_MSC_VER) || defined(G_OS_WIN32)
50 # ifdef HAVE_LANGINFO_H
51 # include <langinfo.h>
58 static char *my_charset;
59 static gboolean is_utf8;
62 * Character set conversion
65 * Index into the table below with the first byte of a UTF-8 sequence to
66 * get the number of trailing bytes that are supposed to follow it.
67 * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is
68 * left as-is for anyone who may want to do such conversion, which was
69 * allowed in earlier algorithms.
71 const gchar g_trailingBytesForUTF8 [256] = {
72 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
73 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
74 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
75 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
76 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
77 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
78 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,
79 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,0,0
83 * Magic values subtracted from a buffer value during UTF8 conversion.
84 * This table contains as many values as there might be trailing bytes
85 * in a UTF-8 sequence.
87 static const gulong offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL,
88 0x03C82080UL, 0xFA082080UL, 0x82082080UL };
91 g_unichar_type (gunichar c)
95 guint16 cp = (guint16) c;
96 for (i = 0; i < unicode_category_ranges_count; i++) {
97 if (cp < unicode_category_ranges [i].start)
99 if (unicode_category_ranges [i].end <= cp)
101 return unicode_category [i] [cp - unicode_category_ranges [i].start];
105 // 3400-4DB5: OtherLetter
106 // 4E00-9FC3: OtherLetter
107 // AC00-D7A3: OtherLetter
108 // D800-DFFF: OtherSurrogate
109 // E000-F8FF: OtherPrivateUse
110 // 20000-2A6D6 OtherLetter
111 // F0000-FFFFD OtherPrivateUse
112 // 100000-10FFFD OtherPrivateUse
114 if (0x3400 <= cp && cp < 0x4DB5)
115 return G_UNICODE_OTHER_LETTER;
116 if (0x4E00 <= cp && cp < 0x9FC3)
117 return G_UNICODE_OTHER_LETTER;
118 if (0xAC00<= cp && cp < 0xD7A3)
119 return G_UNICODE_OTHER_LETTER;
120 if (0xD800 <= cp && cp < 0xDFFF)
121 return G_UNICODE_SURROGATE;
122 if (0xE000 <= cp && cp < 0xF8FF)
123 return G_UNICODE_PRIVATE_USE;
124 /* since the argument is UTF-16, we cannot check beyond FFFF */
126 /* It should match any of above */
131 g_unichar_case (gunichar c, gboolean upper)
134 guint32 cp = (guint32) c, v;
136 for (i = 0; i < simple_case_map_ranges_count; i++) {
137 if (cp < simple_case_map_ranges [i].start)
139 if (simple_case_map_ranges [i].end <= cp)
142 const guint16 *tab = upper ? simple_upper_case_mapping_lowarea [i] : simple_lower_case_mapping_lowarea [i];
143 v = tab [cp - simple_case_map_ranges [i].start];
146 i2 = (gint8)(i - (upper ? simple_upper_case_mapping_lowarea_table_count : simple_lower_case_mapping_lowarea_table_count));
147 tab = upper ? simple_upper_case_mapping_higharea [i2] : simple_lower_case_mapping_higharea [i2];
148 v = tab [cp - simple_case_map_ranges [i].start];
150 return v != 0 ? (gunichar) v : c;
156 g_unichar_toupper (gunichar c)
158 return g_unichar_case (c, TRUE);
162 g_unichar_tolower (gunichar c)
164 return g_unichar_case (c, FALSE);
168 g_unichar_totitle (gunichar c)
174 for (i = 0; i < simple_titlecase_mapping_count; i++) {
175 if (simple_titlecase_mapping [i].codepoint == cp)
176 return simple_titlecase_mapping [i].title;
177 if (simple_titlecase_mapping [i].codepoint > cp)
178 /* it is ordered, hence no more match */
181 return g_unichar_toupper (c);
185 g_unichar_isxdigit (gunichar c)
187 return (g_unichar_xdigit_value (c) != -1);
192 g_unichar_xdigit_value (gunichar c)
194 if (c >= 0x30 && c <= 0x39) /*0-9*/
196 if (c >= 0x41 && c <= 0x46) /*A-F*/
198 if (c >= 0x61 && c <= 0x66) /*a-f*/
204 g_convert (const gchar *str, gssize len,
205 const gchar *to_codeset, const gchar *from_codeset,
206 gsize *bytes_read, gsize *bytes_written, GError **error)
212 char *buffer, *output;
213 const char *strptr = (const char *) str;
214 size_t str_len = len == -1 ? strlen (str) : len;
216 size_t left, out_left;
218 convertor = iconv_open (to_codeset, from_codeset);
219 if (convertor == (iconv_t) -1){
225 buffer_size = str_len + 1 + 8;
226 buffer = g_malloc (buffer_size);
231 int res = iconv (convertor, (char **) &strptr, &left, &output, &out_left);
232 if (res == (size_t) -1){
235 size_t extra_space = 8 + left;
236 size_t output_used = output - buffer;
238 buffer_size += extra_space;
240 n = g_realloc (buffer, buffer_size);
244 *error = g_error_new (NULL, G_CONVERT_ERROR_FAILED, "No memory left");
250 out_left += extra_space;
251 output = buffer + output_used;
252 } else if (errno == EILSEQ){
254 *error = g_error_new (NULL, G_CONVERT_ERROR_ILLEGAL_SEQUENCE, "Invalid multi-byte sequence on input");
258 } else if (errno == EINVAL){
260 *error = g_error_new (NULL, G_CONVERT_ERROR_PARTIAL_INPUT, "Partial character sequence");
267 if (bytes_read != NULL)
268 *bytes_read = strptr - str;
269 if (bytes_written != NULL)
270 *bytes_written = output - buffer;
274 iconv_close (convertor);
280 * This is broken, and assumes an UTF8 system, but will do for eglib's first user
283 g_filename_from_utf8 (const gchar *utf8string, gssize len, gsize *bytes_read, gsize *bytes_written, GError **error)
288 len = strlen (utf8string);
290 res = g_malloc (len + 1);
291 g_strlcpy (res, utf8string, len + 1);
296 g_get_charset (G_CONST_RETURN char **charset)
300 sprintf (buf, "CP%u", GetACP ());
304 if (my_charset == NULL){
305 /* These shouldn't be heap allocated */
307 my_charset = nl_langinfo (CODESET);
309 my_charset = "UTF-8";
311 is_utf8 = strcmp (my_charset, "UTF-8") == 0;
315 *charset = my_charset;
322 g_locale_to_utf8 (const gchar *opsysstring, gssize len, gsize *bytes_read, gsize *bytes_written, GError **error)
324 g_get_charset (NULL);
326 return g_convert (opsysstring, len, "UTF-8", my_charset, bytes_read, bytes_written, error);
330 g_locale_from_utf8 (const gchar *utf8string, gssize len, gsize *bytes_read, gsize *bytes_written, GError **error)
332 g_get_charset (NULL);
334 return g_convert (utf8string, len, my_charset, "UTF-8", bytes_read, bytes_written, error);
338 * @utf: Pointer to putative UTF-8 encoded string.
340 * Checks @utf for being valid UTF-8. @utf is assumed to be
341 * null-terminated. This function is not super-strict, as it will
342 * allow longer UTF-8 sequences than necessary. Note that Java is
343 * capable of producing these sequences if provoked. Also note, this
344 * routine checks for the 4-byte maximum size, but does not check for
345 * 0x10ffff maximum value.
347 * Return value: true if @utf is valid.
350 g_utf8_validate (const gchar *str, gssize max_len, const gchar **end)
352 gssize byteCount = 0;
353 gboolean retVal = TRUE;
354 gboolean lastRet = TRUE;
355 guchar* ptr = (guchar*) str;
361 else if (max_len < 0)
363 while (*ptr != 0 && byteCount <= max_len) {
364 length = g_trailingBytesForUTF8 [*ptr] + 1;
365 srcPtr = (guchar*) ptr + length;
367 default: retVal = FALSE;
368 /* Everything else falls through when "TRUE"... */
369 case 4: if ((a = (*--srcPtr)) < (guchar) 0x80 || a > (guchar) 0xBF) retVal = FALSE;
370 if ((a == (guchar) 0xBF || a == (guchar) 0xBE) && *(srcPtr-1) == (guchar) 0xBF) {
371 if (*(srcPtr-2) == (guchar) 0x8F || *(srcPtr-2) == (guchar) 0x9F ||
372 *(srcPtr-2) == (guchar) 0xAF || *(srcPtr-2) == (guchar) 0xBF)
375 case 3: if ((a = (*--srcPtr)) < (guchar) 0x80 || a > (guchar) 0xBF) retVal = FALSE;
376 case 2: if ((a = (*--srcPtr)) < (guchar) 0x80 || a > (guchar) 0xBF) retVal = FALSE;
379 /* no fall-through in this inner switch */
380 case 0xE0: if (a < (guchar) 0xA0) retVal = FALSE; break;
381 case 0xED: if (a > (guchar) 0x9F) retVal = FALSE; break;
382 case 0xEF: if (a == (guchar)0xB7 && (*(srcPtr+1) > (guchar) 0x8F && *(srcPtr+1) < 0xB0)) retVal = FALSE;
383 if (a == (guchar)0xBF && (*(srcPtr+1) == (guchar) 0xBE || *(srcPtr+1) == 0xBF)) retVal = FALSE; break;
384 case 0xF0: if (a < (guchar) 0x90) retVal = FALSE; break;
385 case 0xF4: if (a > (guchar) 0x8F) retVal = FALSE; break;
386 default: if (a < (guchar) 0x80) retVal = FALSE;
389 case 1: if (*ptr >= (guchar ) 0x80 && *ptr < (guchar) 0xC2) retVal = FALSE;
391 if (*ptr > (guchar) 0xF4)
393 //If the string is invalid, set the end to the invalid byte.
394 if (!retVal && lastRet) {
403 if (retVal && end != NULL)
409 * @src: Pointer to UTF-8 encoded character.
411 * Return value: UTF-16 value of @src
414 g_utf8_get_char (const gchar *src)
417 guchar* ptr = (guchar*) src;
418 gushort extraBytesToRead = g_trailingBytesForUTF8 [*ptr];
420 switch (extraBytesToRead) {
421 case 5: ch += *ptr++; ch <<= 6; // remember, illegal UTF-8
422 case 4: ch += *ptr++; ch <<= 6; // remember, illegal UTF-8
423 case 3: ch += *ptr++; ch <<= 6;
424 case 2: ch += *ptr++; ch <<= 6;
425 case 1: ch += *ptr++; ch <<= 6;
428 ch -= offsetsFromUTF8 [extraBytesToRead];
432 g_utf8_strlen (const gchar *str, gssize max)
434 gssize byteCount = 0;
435 guchar* ptr = (guchar*) str;
441 while (*ptr != 0 && byteCount <= max) {
442 gssize cLen = g_trailingBytesForUTF8 [*ptr] + 1;
443 if (max > 0 && (byteCount + cLen) > max)