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>
44 /* RFC3987 section 3.1 mandates STD3 ASCII RULES */
45 #define IDNA_FLAGS IDNA_USE_STD3_ASCII_RULES
47 /* Note: locale encoding is kept in options struct (opt.locale) */
49 static iconv_t locale2utf8;
51 static bool open_locale_to_utf8 (void);
52 static bool do_conversion (iconv_t cd, char *in, size_t inlen, char **out);
55 /* Given a string containing "charset=XXX", return the encoding if found,
58 parse_charset (char *str)
65 str = strcasestr (str, "charset=");
72 /* sXXXav: which chars should be banned ??? */
73 while (*charset && !c_isspace (*charset))
76 /* sXXXav: could strdupdelim return NULL ? */
77 charset = strdupdelim (str, charset);
79 /* Do a minimum check on the charset value */
80 if (!check_encoding_name (charset))
86 /*logprintf (LOG_VERBOSE, "parse_charset: %s\n", quote (charset));*/
91 /* Find the locale used, or fall back on a default value */
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 /* Try converting string str from locale to UTF-8. Return a new string
146 on success, or str on error or if conversion isn't needed. */
148 locale_to_utf8 (const char *str)
152 if (!strcasecmp (opt.locale, "utf-8"))
155 if (!open_locale_to_utf8 ())
158 if (do_conversion (locale2utf8, (char *) str, strlen ((char *) str), &new))
159 return (const char *) new;
164 /* Do the conversion according to the passed conversion descriptor cd. *out
165 will containes the transcoded string on success. *out content is
166 unspecified otherwise. */
168 do_conversion (iconv_t cd, char *in, size_t inlen, char **out)
170 /* sXXXav : hummm hard to guess... */
171 size_t len, done, outlen = inlen * 2;
172 int invalid = 0, tooshort = 0;
175 s = xmalloc (outlen + 1);
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 (struct iri *i, char *host)
235 /* Encode to UTF-8 if not done */
238 if (!remote_to_utf8 (i, (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, IDNA_FLAGS);
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
261 on success or NULL on error. */
263 idn_decode (char *host)
268 ret = idna_to_unicode_8zlz (host, &new, IDNA_FLAGS);
269 if (ret != IDNA_SUCCESS)
271 logprintf (LOG_VERBOSE, "idn_decode failed (%d): %s\n", ret,
272 quote (idna_strerror (ret)));
279 /* Try to transcode string str from remote encoding to UTF-8. On success, *new
280 contains the transcoded string. *new content is unspecified otherwise. */
282 remote_to_utf8 (struct iri *i, const char *str, const char **new)
288 if (opt.encoding_remote)
289 r = opt.encoding_remote;
290 else if (i->uri_encoding)
295 cd = iconv_open ("UTF-8", r);
296 if (cd == (iconv_t)(-1))
299 if (do_conversion (cd, (char *) str, strlen ((char *) str), (char **) new))
304 /* Test if something was converted */
305 if (!strcmp (str, *new))
307 xfree ((char *) *new);
317 struct iri *i = xmalloc (sizeof (struct iri));
318 i->uri_encoding = opt.encoding_remote ? xstrdup (opt.encoding_remote) : NULL;
319 i->content_encoding = NULL;
320 i->utf8_encode = opt.enable_iri;
324 iri_free (struct iri *i)
326 xfree_null (i->uri_encoding);
327 xfree_null (i->content_encoding);
332 set_uri_encoding (struct iri *i, char *charset)
334 logprintf (LOG_VERBOSE, "[ uri = `%s'\n", charset);
335 if (opt.encoding_remote)
339 if (!strcasecmp (i->uri_encoding, charset))
341 xfree (i->uri_encoding);
344 i->uri_encoding = charset ? xstrdup (charset) : NULL;
348 set_content_encoding (struct iri *i, char *charset)
350 logprintf (LOG_VERBOSE, "[ content = `%s'\n", charset);
351 if (opt.encoding_remote)
353 if (i->content_encoding)
355 if (!strcasecmp (i->content_encoding, charset))
357 xfree (i->content_encoding);
360 i->content_encoding = charset ? xstrdup (charset) : NULL;