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>
47 static iconv_t locale2utf8;
50 static bool open_locale_to_utf8 (void);
51 static bool do_conversion (iconv_t cd, char *in, size_t inlen, char **out);
54 /* Given a string containing "charset=XXX", return the encoding if found,
57 parse_charset (char *str)
64 str = strcasestr (str, "charset=");
71 /* sXXXav: which chars should be banned ??? */
72 while (*charset && !c_isspace (*charset))
75 /* sXXXav: could strdupdelim return NULL ? */
76 charset = strdupdelim (str, charset);
78 /* Do a minimum check on the charset value */
79 if (!check_encoding_name (charset))
85 /*logprintf (LOG_VERBOSE, "parse_charset: %s\n", quote (charset));*/
90 /* Find the locale used, or fall back on a default value */
94 /* sXXXav, made our own function or use libidn one ?! */
95 return (char *) stringprep_locale_charset ();
98 /* Basic check of an encoding name. */
100 check_encoding_name (char *encoding)
106 if (!c_isascii (*s) || c_isspace (*s))
108 logprintf (LOG_VERBOSE, "Encoding %s isn't valid\n", quote (encoding));
118 /* Try opening an iconv_t descriptor for conversion from locale to UTF-8 */
120 open_locale_to_utf8 (void)
125 /* sXXXav : That shouldn't happen, just in case */
128 logprintf (LOG_VERBOSE, "open_locale_to_utf8: locale is unset\n");
129 opt.locale = find_locale ();
135 locale2utf8 = iconv_open ("UTF-8", opt.locale);
136 if (locale2utf8 != (iconv_t)(-1))
139 logprintf (LOG_VERBOSE, "Conversion from %s to %s isn't supported\n",
140 quote (opt.locale), quote ("UTF-8"));
145 /* Return a new string */
147 locale_to_utf8 (const char *str)
151 if (!strcasecmp (opt.locale, "utf-8"))
154 if (!open_locale_to_utf8 ())
157 if (do_conversion (locale2utf8, (char *) str, strlen ((char *) str), &new))
158 return (const char *) new;
165 do_conversion (iconv_t cd, char *in, size_t inlen, char **out)
167 /* sXXXav : hummm hard to guess... */
168 size_t len, done, outlen = inlen * 2;
169 int invalid = 0, tooshort = 0;
172 s = xmalloc (outlen + 1);
177 /* sXXXav : put a maximum looping factor ??? */
180 if (iconv (cd, &in, &inlen, out, &outlen) != (size_t)(-1))
183 *(s + len - outlen - done) = '\0';
187 /* Incomplete or invalid multibyte sequence */
188 if (errno == EINVAL || errno == EILSEQ)
191 logprintf (LOG_VERBOSE,
192 "Incomplete or invalide multibyte sequence encountered\n");
201 else if (errno == E2BIG) /* Output buffer full */
207 outlen = done + inlen * 2;
208 new = xmalloc (outlen + 1);
209 memcpy (new, s, done);
215 else /* Weird, we got an unspecified error */
217 logprintf (LOG_VERBOSE, "Unhandled errno %d\n", errno);
225 /* Try to ASCII encode UTF-8 host. Return the new domain on success or NULL
228 idn_encode (char *host, bool utf8_encoded)
233 /* Encode to UTF-8 if not done using current remote */
236 if (!remote_to_utf8 ((const char *) host, (const char **) &new))
238 /* Nothing to encode or an error occured */
245 /* toASCII UTF-8 NULL terminated string */
246 ret = idna_to_ascii_8z (host, &new, 0);
247 if (ret != IDNA_SUCCESS)
249 /* sXXXav : free new when needed ! */
250 logprintf (LOG_VERBOSE, "idn_encode failed (%d): %s\n", ret,
251 quote (idna_strerror (ret)));
258 /* Try to decode an ASCII encoded host. Return the new domain in the locale on
259 success or NULL on error. */
261 idn_decode (char *host)
266 ret = idna_to_unicode_8zlz (host, &new, 0);
267 if (ret != IDNA_SUCCESS)
269 logprintf (LOG_VERBOSE, "idn_decode failed (%d): %s\n", ret,
270 quote (idna_strerror (ret)));
277 /* Return a new string */
279 remote_to_utf8 (const char *str, const char **new)
285 if (opt.encoding_remote)
286 remote = opt.encoding_remote;
292 cd = iconv_open ("UTF-8", remote);
293 if (cd == (iconv_t)(-1))
296 if (do_conversion (cd, (char *) str, strlen ((char *) str), (char **) new))
301 /* Test if something was converted */
302 if (!strcmp (str, *new))
304 xfree ((char *) *new);
311 char *get_remote_charset (void)
316 char *get_current_charset (void)
321 void set_current_charset (char *charset)
323 /*printf("[ current = `%s'\n", charset);*/
328 current = charset ? xstrdup (charset) : NULL;
331 void set_current_as_locale (void)
333 /*printf("[ current = locale = `%s'\n", opt.locale);*/
337 /* sXXXav : assert opt.locale NULL ? */
338 current = xstrdup (opt.locale);
342 set_remote_charset (char *charset)
344 /*printf("[ remote = `%s'\n", charset);*/
348 remote = charset ? xstrdup (charset) : NULL;
352 set_remote_as_current (void)
354 /*printf("[ remote = current = `%s'\n", current);*/
358 remote = current ? xstrdup (current) : NULL;