include/vlc_url.h

   1 /*****************************************************************************
   2  * vlc_url.h: URL related macros
   3  *****************************************************************************
   4  * Copyright (C) 2002-2006 VLC authors and VideoLAN
   5  * $Id$
   6  *
   7  * Authors: Christophe Massiot <massiot@via.ecp.fr>
   8  *          Rémi Denis-Courmont <rem # videolan.org>
   9  *
  10  * This program is free software; you can redistribute it and/or modify it
  11  * under the terms of the GNU Lesser General Public License as published by
  12  * the Free Software Foundation; either version 2.1 of the License, or
  13  * (at your option) any later version.
  14  *
  15  * This program is distributed in the hope that it will be useful,
  16  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  17  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  18  * GNU Lesser General Public License for more details.
  19  *
  20  * You should have received a copy of the GNU Lesser General Public License
  21  * along with this program; if not, write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
  23  *****************************************************************************/
  24
  25 #ifndef VLC_URL_H
  26 # define VLC_URL_H
  27
  28 /**
  29  * \file
  30  * This file defines functions for manipulating URL in vlc
  31  *
  32  * \ingroup strings
  33  * @{
  34  */
  35
  36 /**
  37  * Converts local path to URL.
  38  *
  39  * Builds a URL representation from a local UTF-8 null-terminated file path.
  40  *
  41  * @param path file path
  42  * @param scheme URI scheme to use (default is auto: "file", "fd" or "smb")
  43  * @return a heap-allocated URI string on success
  44  * or NULL in case of error (errno will be set accordingly)
  45  */
  46 VLC_API char *vlc_path2uri(const char *path, const char *scheme) VLC_MALLOC;
  47
  48 /**
  49  * Converts a URI to a local path.
  50  *
  51  * Builds a local path (UTF-8-encoded null-terminated string) from a URI if
  52  * the URI scheme allows.
  53  *
  54  * @param url URI
  55  * @return a heap-allocated string or success
  56  * or NULL on error
  57  */
  58 VLC_API char *vlc_uri2path(const char *url) VLC_MALLOC;
  59
  60 /**
  61  * Decodes an URI component in place.
  62  *
  63  * Decodes one null-terminated UTF-8 URI component to aa null-terminated UTF-8
  64  * string in place.
  65  *
  66  * See also vlc_uri_decode_duplicate() for the not-in-place variant.
  67  *
  68  * \warning <b>This function does NOT decode entire URIs.</b>
  69  * URI can only be decoded (and encoded) one component at a time
  70  * (e.g. the host name, one directory, the file name).
  71  * Complete URIs are always "encoded" (or they are syntaxically invalid).
  72  * See IETF RFC3986, especially §2.4 for details.
  73  *
  74  * \note URI encoding is <b>different</b> from Javascript escaping. Especially,
  75  * white spaces and Unicode non-ASCII code points are encoded differently.
  76  *
  77  * \param str null-terminated component
  78  * \return str is returned on success. NULL if str was not properly encoded.
  79  */
  80 VLC_API char *vlc_uri_decode(char *str);
  81
  82 /**
  83  * Decodes an URI component.
  84  *
  85  * See also vlc_uri_decode() for the in-place variant.
  86  *
  87  * \return a heap-allocated string on success or NULL on error.
  88  */
  89 VLC_API char *vlc_uri_decode_duplicate(const char *str) VLC_MALLOC;
  90
  91 /**
  92  * Encodes a URI component.
  93  *
  94  * Substitutes URI-unsafe, URI delimiters and non-ASCII characters into their
  95  * URI-encoded URI-safe representation. See also IETF RFC3986 §2.
  96  *
  97  * @param str nul-terminated UTF-8 representation of the component.
  98  * @note Obviously, a URI containing nul bytes cannot be passed.
  99  * @return heap-allocated string, or NULL if out of memory.
 100  */
 101 VLC_API char *vlc_uri_encode(const char *str) VLC_MALLOC;
 102
 103 /**
 104  * Composes an URI.
 105  *
 106  * Converts a decomposed/parsed URI structure (\ref vlc_url_t) into a
 107  * nul-terminated URI literal string.
 108  *
 109  * See also IETF RFC3986 section 5.3 for details.
 110  *
 111  * \bug URI fragments (i.e. HTML anchors) are not handled
 112  *
 113  * \return a heap-allocated nul-terminated string or NULL if out of memory
 114  */
 115 VLC_API char *vlc_uri_compose(const vlc_url_t *) VLC_MALLOC;
 116
 117 /**
 118  * Resolves an URI reference.
 119  *
 120  * Resolves an URI reference relative to a base URI.
 121  * If the reference is an absolute URI, then this function simply returns a
 122  * copy of the URI reference.
 123  *
 124  * \param base base URI (as a nul-terminated string)
 125  * \param ref URI reference (also as a nul-terminated string)
 126  *
 127  * \return a heap-allocated nul-terminated string representing the resolved
 128  * absolute URI, or NULL if out of memory.
 129  */
 130 VLC_API char *vlc_uri_resolve(const char *base, const char *ref) VLC_MALLOC;
 131
 132 /**
 133  * Fixes up a URI string.
 134  *
 135  * Attempts to convert a nul-terminated string into a syntactically valid URI.
 136  * If the string is, or may be, a syntactically valid URI, an exact copy is
 137  * returned. In any case, the result will only contain URI-safe and URI
 138  * delimiter characters (generic delimiters or sub-delimiters) and all percent
 139  * signs will be followed by two hexadecimal characters.
 140  *
 141  * @return a heap-allocated string, or NULL if on out of memory.
 142  */
 143 VLC_API char *vlc_uri_fixup(const char *) VLC_MALLOC;
 144
 145 struct vlc_url_t
 146 {
 147     char *psz_protocol;
 148     char *psz_username;
 149     char *psz_password;
 150     char *psz_host;
 151     unsigned i_port;
 152     char *psz_path;
 153     char *psz_option;
 154     char *psz_fragment;
 155
 156     char *psz_buffer; /* to be freed */
 157     char *psz_pathbuffer; /* to be freed */
 158 };
 159
 160 /**
 161  * Parses an URI or IRI.
 162  *
 163  * Extracts the following parts from an URI string:
 164  *  - scheme (i.e. protocol),
 165  *  - user (deprecated),
 166  *  - password (also deprecated),
 167  *  - host name or IP address literal,
 168  *  - port number,
 169  *  - path (including the filename preceded by any and all directories)
 170  *  - request parameters (excluding the leading question mark '?').
 171  *
 172  * The function accepts URIs, as well as UTF-8-encoded IRIs. For IRIs, the hier
 173  * part (specifically, the host name) is assumed to be an IDN and is decoded to
 174  * ASCII according, so it can be used for DNS resolution. If the host is an
 175  * IPv6 address literal, brackets are stripped.
 176  *
 177  * Any missing part is set to nul. For historical reasons, the target structure
 178  * is always initialized, even if parsing the URI string fails.
 179  *
 180  * On error, errno is set to one of the following value:
 181  *  - ENOMEM in case of memory allocation failure,
 182  *  - EINVAL in case of syntax error in the input string.
 183  *
 184  * \bug The URI fragment is discarded if present.
 185  *
 186  * \note This function allocates memory. vlc_UrlClean() must be used free
 187  * associated the allocations, even if the function fails.
 188  *
 189  * \param url structure of URL parts [OUT]
 190  * \param str nul-terminated URL string to split
 191  * \retval 0 success
 192  * \retval -1 failure
 193  */
 194 VLC_API int vlc_UrlParse(vlc_url_t *url, const char *str);
 195
 196 /**
 197  * Parses an URI or IRI and fix up the path part.
 198  *
 199  * \see vlc_UrlParse
 200  * \see vlc_uri_fixup
 201  */
 202 VLC_API int vlc_UrlParseFixup(vlc_url_t *url, const char *str);
 203
 204 /**
 205  * Releases resources allocated by vlc_UrlParse().
 206  */
 207 VLC_API void vlc_UrlClean(vlc_url_t *);
 208
 209 /** @} */
 210
 211 #endif