1 /* IRI related functions.
2 Copyright (C) 2008 Free Software Foundation, Inc.
4 This file is part of GNU Wget.
6 GNU Wget is free software; you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation; either version 3 of the License, or (at
9 your option) any later version.
11 GNU Wget 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
14 GNU General Public License for more details.
16 You should have received a copy of the GNU General Public License
17 along with Wget. If not, see <http://www.gnu.org/licenses/>.
19 Additional permission under GNU GPL version 3 section 7
21 If you modify this program, or any covered work, by linking or
22 combining it with the OpenSSL project's OpenSSL library (or a
23 modified version of that library), containing parts covered by the
24 terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
25 grants you additional permission to convey the resulting work.
26 Corresponding Source for a non-source form of such a combination
27 shall include the source code for the parts of OpenSSL used as well
28 as that of the covered work. */
37 #include <stringprep.h>
49 static iconv_t locale2utf8;
52 static bool open_locale_to_utf8 (void);
53 static bool do_conversion (iconv_t cd, char *in, size_t inlen, char **out);
56 /* Given a string containing "charset=XXX", return the encoding if found,
59 parse_charset (char *str)
66 str = strcasestr (str, "charset=");
73 /* sXXXav: which chars should be banned ??? */
74 while (*charset && !c_isspace (*charset))
77 /* sXXXav: could strdupdelim return NULL ? */
78 charset = strdupdelim (str, charset);
80 /* Do a minimum check on the charset value */
81 if (!check_encoding_name (charset))
87 /*logprintf (LOG_VERBOSE, "parse_charset: %s\n", quote (charset));*/
92 /* Find the locale used, or fall back on a default value */
96 /* sXXXav, made our own function or use libidn one ?! */
97 return (char *) stringprep_locale_charset ();
100 /* Basic check of an encoding name. */
102 check_encoding_name (char *encoding)
108 if (!c_isascii (*s) || c_isspace (*s))
110 logprintf (LOG_VERBOSE, "Encoding %s isn't valid\n", quote (encoding));
120 /* Try opening an iconv_t descriptor for conversion from locale to UTF-8 */
122 open_locale_to_utf8 (void)
127 /* sXXXav : That shouldn't happen, just in case */
130 logprintf (LOG_VERBOSE, "open_locale_to_utf8: locale is unset\n");
131 opt.locale = find_locale ();
137 locale2utf8 = iconv_open ("UTF-8", opt.locale);
138 if (locale2utf8 != (iconv_t)(-1))
141 logprintf (LOG_VERBOSE, "Conversion from %s to %s isn't supported\n",
142 quote (opt.locale), quote ("UTF-8"));
147 /* Return a new string */
149 locale_to_utf8 (const char *str)
153 if (!strcasecmp (opt.locale, "utf-8"))
156 if (!open_locale_to_utf8 ())
159 if (do_conversion (locale2utf8, (char *) str, strlen ((char *) str), &new))
160 return (const char *) new;
167 do_conversion (iconv_t cd, char *in, size_t inlen, char **out)
169 /* sXXXav : hummm hard to guess... */
170 size_t len, done, outlen = inlen * 2;
171 int invalid = 0, tooshort = 0;
174 s = xmalloc (outlen + 1);
179 /* sXXXav : put a maximum looping factor ??? */
182 if (iconv (cd, &in, &inlen, out, &outlen) != (size_t)(-1))
185 *(s + len - outlen - done) = '\0';
189 /* Incomplete or invalid multibyte sequence */
190 if (errno == EINVAL || errno == EILSEQ)
193 logprintf (LOG_VERBOSE,
194 "Incomplete or invalide multibyte sequence encountered\n");
203 else if (errno == E2BIG) /* Output buffer full */
209 outlen = done + inlen * 2;
210 new = xmalloc (outlen + 1);
211 memcpy (new, s, done);
217 else /* Weird, we got an unspecified error */
219 logprintf (LOG_VERBOSE, "Unhandled errno %d\n", errno);
227 /* Try to ASCII encode UTF-8 host. Return the new domain on success or NULL
230 idn_encode (char *host, bool utf8_encoded)
235 /* Encode to UTF-8 if not done using current remote */
238 if (!remote_to_utf8 ((const char *) host, (const char **) &new))
240 /* Nothing to encode or an error occured */
247 /* toASCII UTF-8 NULL terminated string */
248 ret = idna_to_ascii_8z (host, &new, 0);
249 if (ret != IDNA_SUCCESS)
251 /* sXXXav : free new when needed ! */
252 logprintf (LOG_VERBOSE, "idn_encode failed (%d): %s\n", ret,
253 quote (idna_strerror (ret)));
260 /* Try to decode an ASCII encoded host. Return the new domain in the locale on
261 success or NULL on error. */
263 idn_decode (char *host)
268 ret = idna_to_unicode_8zlz (host, &new, 0);
269 if (ret != IDNA_SUCCESS)
271 logprintf (LOG_VERBOSE, "idn_decode failed (%d): %s\n", ret,
272 quote (idna_strerror (ret)));
279 /* Return a new string */
281 remote_to_utf8 (const char *str, const char **new)
287 if (opt.encoding_remote)
288 remote = opt.encoding_remote;
294 cd = iconv_open ("UTF-8", remote);
295 if (cd == (iconv_t)(-1))
298 if (do_conversion (cd, (char *) str, strlen ((char *) str), (char **) new))
303 /* Test if something was converted */
304 if (!strcmp (str, *new))
306 xfree ((char *) *new);
313 char *get_remote_charset (void)
318 char *get_current_charset (void)
323 void set_current_charset (char *charset)
325 /*printf("[ current = `%s'\n", charset);*/
330 current = charset ? xstrdup (charset) : NULL;
333 void set_current_as_locale (void)
335 /*printf("[ current = locale = `%s'\n", opt.locale);*/
339 /* sXXXav : assert opt.locale NULL ? */
340 current = xstrdup (opt.locale);
344 set_remote_charset (char *charset)
346 /*printf("[ remote = `%s'\n", charset);*/
350 remote = charset ? xstrdup (charset) : NULL;
354 set_remote_as_current (void)
356 /*printf("[ remote = current = `%s'\n", current);*/
360 remote = current ? xstrdup (current) : NULL;
363 void reset_utf8_encode (void)
365 set_utf8_encode (opt.enable_iri);
368 void set_utf8_encode (bool encode)
370 utf8_encode = encode;
373 bool get_utf8_encode (void)
378 void set_ugly_no_encode (bool ugly)
380 ugly_no_encode = ugly;
383 bool get_ugly_no_encode (void)
385 return ugly_no_encode;