1 /* -*- Mode: c; c-basic-offset: 4; indent-tabs-mode: t; tab-width: 8; -*- */
2 /* cairo - a vector graphics library with display and print output
4 * The code in this file is derived from GLib's gutf8.c and
5 * ultimately from libunicode. It is relicensed under the
6 * dual LGPL/MPL with permission of the original authors.
8 * Copyright © 1999 Tom Tromey
9 * Copyright © 2005 Red Hat, Inc
11 * This library is free software; you can redistribute it and/or
12 * modify it either under the terms of the GNU Lesser General Public
13 * License version 2.1 as published by the Free Software Foundation
14 * (the "LGPL") or, at your option, under the terms of the Mozilla
15 * Public License Version 1.1 (the "MPL"). If you do not alter this
16 * notice, a recipient may use your version of this file under either
17 * the MPL or the LGPL.
19 * You should have received a copy of the LGPL along with this library
20 * in the file COPYING-LGPL-2.1; if not, write to the Free Software
21 * Foundation, Inc., 51 Franklin Street, Suite 500, Boston, MA 02110-1335, USA
22 * You should have received a copy of the MPL along with this library
23 * in the file COPYING-MPL-1.1
25 * The contents of this file are subject to the Mozilla Public License
26 * Version 1.1 (the "License"); you may not use this file except in
27 * compliance with the License. You may obtain a copy of the License at
28 * http://www.mozilla.org/MPL/
30 * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY
31 * OF ANY KIND, either express or implied. See the LGPL or the MPL for
32 * the specific language governing rights and limitations.
34 * The Original Code is the cairo graphics library.
36 * The Initial Developer of the Original Code is Tom Tromey.
40 * Owen Taylor <otaylor@redhat.com>
44 #include "cairo-error-private.h"
46 #define UTF8_COMPUTE(Char, Mask, Len) \
52 else if ((Char & 0xe0) == 0xc0) \
57 else if ((Char & 0xf0) == 0xe0) \
62 else if ((Char & 0xf8) == 0xf0) \
67 else if ((Char & 0xfc) == 0xf8) \
72 else if ((Char & 0xfe) == 0xfc) \
80 #define UTF8_LENGTH(Char) \
81 ((Char) < 0x80 ? 1 : \
82 ((Char) < 0x800 ? 2 : \
83 ((Char) < 0x10000 ? 3 : \
84 ((Char) < 0x200000 ? 4 : \
85 ((Char) < 0x4000000 ? 5 : 6)))))
87 #define UTF8_GET(Result, Chars, Count, Mask, Len) \
88 (Result) = (Chars)[0] & (Mask); \
89 for ((Count) = 1; (Count) < (Len); ++(Count)) \
91 if (((Chars)[(Count)] & 0xc0) != 0x80) \
97 (Result) |= ((Chars)[(Count)] & 0x3f); \
100 #define UNICODE_VALID(Char) \
101 ((Char) < 0x110000 && \
102 (((Char) & 0xFFFFF800) != 0xD800) && \
103 ((Char) < 0xFDD0 || (Char) > 0xFDEF) && \
104 ((Char) & 0xFFFE) != 0xFFFE)
106 static const char 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 #define UTF8_NEXT_CHAR(p) ((p) + utf8_skip_data[*(unsigned char *)(p)])
119 /* Converts a sequence of bytes encoded as UTF-8 to a Unicode character.
120 * If @p does not point to a valid UTF-8 encoded character, results are
124 _utf8_get_char (const unsigned char *p)
126 int i, mask = 0, len;
128 unsigned char c = (unsigned char) *p;
130 UTF8_COMPUTE (c, mask, len);
133 UTF8_GET (result, p, i, mask, len);
138 /* Like _utf8_get_char, but take a maximum length
139 * and return (uint32_t)-2 on incomplete trailing character
142 _utf8_get_char_extended (const unsigned char *p,
146 uint32_t wc = (unsigned char) *p;
150 } else if (wc < 0xc0) {
152 } else if (wc < 0xe0) {
155 } else if (wc < 0xf0) {
158 } else if (wc < 0xf8) {
161 } else if (wc < 0xfc) {
164 } else if (wc < 0xfe) {
171 if (max_len >= 0 && len > max_len) {
172 for (i = 1; i < max_len; i++) {
173 if ((((unsigned char *)p)[i] & 0xc0) != 0x80)
179 for (i = 1; i < len; ++i) {
180 uint32_t ch = ((unsigned char *)p)[i];
182 if ((ch & 0xc0) != 0x80) {
193 if (UTF8_LENGTH(wc) != len)
200 * _cairo_utf8_get_char_validated:
202 * @unicode: location to store one Unicode character
204 * Decodes the first character of a valid UTF-8 string, and returns
205 * the number of bytes consumed.
207 * Note that the string should be valid. Do not use this without
208 * validating the string first.
210 * Returns: the number of bytes forming the character returned.
213 _cairo_utf8_get_char_validated (const char *p,
216 int i, mask = 0, len;
218 unsigned char c = (unsigned char) *p;
220 UTF8_COMPUTE (c, mask, len);
223 *unicode = (uint32_t)-1;
226 UTF8_GET (result, p, i, mask, len);
234 * _cairo_utf8_to_ucs4:
235 * @str: an UTF-8 string
236 * @len: length of @str in bytes, or -1 if it is nul-terminated.
237 * If @len is supplied and the string has an embedded nul
238 * byte, only the portion before the nul byte is converted.
239 * @result: location to store a pointer to a newly allocated UTF-32
240 * string (always native endian), or %NULL. Free with free(). A 0
241 * word will be written after the last character.
242 * @items_written: location to store number of 32-bit words
243 * written. (Not including the trailing 0)
245 * Converts a UTF-8 string to UCS-4. UCS-4 is an encoding of Unicode
246 * with 1 32-bit word per character. The string is validated to
247 * consist entirely of valid Unicode characters.
249 * Return value: %CAIRO_STATUS_SUCCESS if the entire string was
250 * successfully converted. %CAIRO_STATUS_INVALID_STRING if an
251 * invalid sequence was found.
254 _cairo_utf8_to_ucs4 (const char *str,
259 uint32_t *str32 = NULL;
261 const unsigned char *in;
262 const unsigned char * const ustr = (const unsigned char *) str;
266 while ((len < 0 || ustr + len - in > 0) && *in)
268 uint32_t wc = _utf8_get_char_extended (in, ustr + len - in);
269 if (wc & 0x80000000 || !UNICODE_VALID (wc))
270 return _cairo_error (CAIRO_STATUS_INVALID_STRING);
273 if (n_chars == INT_MAX)
274 return _cairo_error (CAIRO_STATUS_INVALID_STRING);
276 in = UTF8_NEXT_CHAR (in);
280 str32 = _cairo_malloc_ab (n_chars + 1, sizeof (uint32_t));
282 return _cairo_error (CAIRO_STATUS_NO_MEMORY);
285 for (i=0; i < n_chars; i++) {
286 str32[i] = _utf8_get_char (in);
287 in = UTF8_NEXT_CHAR (in);
295 *items_written = n_chars;
297 return CAIRO_STATUS_SUCCESS;
301 * _cairo_ucs4_to_utf8:
302 * @unicode: a UCS-4 character
303 * @utf8: buffer to write utf8 string into. Must have at least 4 bytes
304 * space available. Or %NULL.
306 * This space left intentionally blank.
308 * Return value: Number of bytes in the utf8 string or 0 if an invalid
312 _cairo_ucs4_to_utf8 (uint32_t unicode,
318 if (unicode < 0x80) {
322 } else if (unicode < 0x800) {
324 } else if (unicode < 0x10000) {
326 } else if (unicode < 0x200000) {
337 *--p = 0x80 | (unicode & 0x3f);
340 *p |= 0xf0 << (4 - bytes);
345 #if CAIRO_HAS_UTF8_TO_UTF16
347 * _cairo_utf8_to_utf16:
348 * @str: an UTF-8 string
349 * @len: length of @str in bytes, or -1 if it is nul-terminated.
350 * If @len is supplied and the string has an embedded nul
351 * byte, only the portion before the nul byte is converted.
352 * @result: location to store a pointer to a newly allocated UTF-16
353 * string (always native endian). Free with free(). A 0
354 * word will be written after the last character.
355 * @items_written: location to store number of 16-bit words
356 * written. (Not including the trailing 0)
358 * Converts a UTF-8 string to UTF-16. UTF-16 is an encoding of Unicode
359 * where characters are represented either as a single 16-bit word, or
360 * as a pair of 16-bit "surrogates". The string is validated to
361 * consist entirely of valid Unicode characters.
363 * Return value: %CAIRO_STATUS_SUCCESS if the entire string was
364 * successfully converted. %CAIRO_STATUS_INVALID_STRING if an
365 * an invalid sequence was found.
368 _cairo_utf8_to_utf16 (const char *str,
373 uint16_t *str16 = NULL;
375 const unsigned char *in;
376 const unsigned char * const ustr = (const unsigned char *) str;
380 while ((len < 0 || ustr + len - in > 0) && *in) {
381 uint32_t wc = _utf8_get_char_extended (in, ustr + len - in);
382 if (wc & 0x80000000 || !UNICODE_VALID (wc))
383 return _cairo_error (CAIRO_STATUS_INVALID_STRING);
390 if (n16 == INT_MAX - 1 || n16 == INT_MAX)
391 return _cairo_error (CAIRO_STATUS_INVALID_STRING);
393 in = UTF8_NEXT_CHAR (in);
396 str16 = _cairo_malloc_ab (n16 + 1, sizeof (uint16_t));
398 return _cairo_error (CAIRO_STATUS_NO_MEMORY);
401 for (i = 0; i < n16;) {
402 uint32_t wc = _utf8_get_char (in);
407 str16[i++] = (wc - 0x10000) / 0x400 + 0xd800;
408 str16[i++] = (wc - 0x10000) % 0x400 + 0xdc00;
411 in = UTF8_NEXT_CHAR (in);
418 *items_written = n16;
420 return CAIRO_STATUS_SUCCESS;