glib/gurifuncs.c

   1 /* GIO - GLib Input, Output and Streaming Library
   2  *
   3  * Copyright (C) 2006-2007 Red Hat, Inc.
   4  *
   5  * This library is free software; you can redistribute it and/or
   6  * modify it under the terms of the GNU Lesser General Public
   7  * License as published by the Free Software Foundation; either
   8  * version 2 of the License, or (at your option) any later version.
   9  *
  10  * This library is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  * Lesser General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU Lesser General
  16  * Public License along with this library; if not, write to the
  17  * Free Software Foundation, Inc., 59 Temple Place, Suite 330,
  18  * Boston, MA 02111-1307, USA.
  19  *
  20  * Author: Alexander Larsson <alexl@redhat.com>
  21  */
  22
  23 #include "config.h"
  24
  25 #include "gurifuncs.h"
  26
  27 #include <glib/gstrfuncs.h>
  28 #include <glib/gmessages.h>
  29 #include <glib/gstring.h>
  30 #include <glib/gmem.h>
  31
  32 #include <string.h>
  33
  34 #include "config.h"
  35
  36 /**
  37  * SECTION:gurifuncs
  38  * @short_description: URI Functions
  39  *
  40  * Functions for manipulating Universal Resource Identifiers (URIs) as
  41  * defined by <ulink url="http://www.ietf.org/rfc/rfc3986.txt">
  42  * RFC 3986</ulink>. It is highly recommended that you have read and
  43  * understand RFC 3986 for understanding this API.
  44  */
  45
  46 static int
  47 unescape_character (const char *scanner)
  48 {
  49   int first_digit;
  50   int second_digit;
  51
  52   first_digit = g_ascii_xdigit_value (*scanner++);
  53   if (first_digit < 0)
  54     return -1;
  55
  56   second_digit = g_ascii_xdigit_value (*scanner++);
  57   if (second_digit < 0)
  58     return -1;
  59
  60   return (first_digit << 4) | second_digit;
  61 }
  62
  63 /**
  64  * g_uri_unescape_segment:
  65  * @escaped_string: a string.
  66  * @escaped_string_end: a string.
  67  * @illegal_characters: an optional string of illegal characters not to be allowed.
  68  *
  69  * Unescapes a segment of an escaped string.
  70  *
  71  * If any of the characters in @illegal_characters or the character zero appears
  72  * as an escaped character in @escaped_string then that is an error and %NULL
  73  * will be returned. This is useful it you want to avoid for instance having a
  74  * slash being expanded in an escaped path element, which might confuse pathname
  75  * handling.
  76  *
  77  * Returns: an unescaped version of @escaped_string or %NULL on error.
  78  * The returned string should be freed when no longer needed.
  79  *
  80  * Since: 2.16
  81  **/
  82 char *
  83 g_uri_unescape_segment (const char *escaped_string,
  84                         const char *escaped_string_end,
  85                         const char *illegal_characters)
  86 {
  87   const char *in;
  88   char *out, *result;
  89   gint character;
  90
  91   if (escaped_string == NULL)
  92     return NULL;
  93
  94   if (escaped_string_end == NULL)
  95     escaped_string_end = escaped_string + strlen (escaped_string);
  96
  97   result = g_malloc (escaped_string_end - escaped_string + 1);
  98
  99   out = result;
 100   for (in = escaped_string; in < escaped_string_end; in++)
 101     {
 102       character = *in;
 103
 104       if (*in == '%')
 105         {
 106           in++;
 107
 108           if (escaped_string_end - in < 2)
 109             {
 110               /* Invalid escaped char (to short) */
 111               g_free (result);
 112               return NULL;
 113             }
 114
 115           character = unescape_character (in);
 116
 117           /* Check for an illegal character. We consider '\0' illegal here. */
 118           if (character <= 0 ||
 119               (illegal_characters != NULL &&
 120                strchr (illegal_characters, (char)character) != NULL))
 121             {
 122               g_free (result);
 123               return NULL;
 124             }
 125
 126           in++; /* The other char will be eaten in the loop header */
 127         }
 128       *out++ = (char)character;
 129     }
 130
 131   *out = '\0';
 132
 133   return result;
 134 }
 135
 136 /**
 137  * g_uri_unescape_string:
 138  * @escaped_string: an escaped string to be unescaped.
 139  * @illegal_characters: an optional string of illegal characters not to be allowed.
 140  *
 141  * Unescapes a whole escaped string.
 142  *
 143  * If any of the characters in @illegal_characters or the character zero appears
 144  * as an escaped character in @escaped_string then that is an error and %NULL
 145  * will be returned. This is useful it you want to avoid for instance having a
 146  * slash being expanded in an escaped path element, which might confuse pathname
 147  * handling.
 148  *
 149  * Returns: an unescaped version of @escaped_string. The returned string
 150  * should be freed when no longer needed.
 151  *
 152  * Since: 2.16
 153  **/
 154 char *
 155 g_uri_unescape_string (const char *escaped_string,
 156                        const char *illegal_characters)
 157 {
 158   return g_uri_unescape_segment (escaped_string, NULL, illegal_characters);
 159 }
 160
 161 /**
 162  * g_uri_parse_scheme:
 163  * @uri: a valid URI.
 164  *
 165  * Gets the scheme portion of a URI string. RFC 3986 decodes the scheme as:
 166  * <programlisting>
 167  * URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]
 168  * </programlisting>
 169  * Common schemes include "file", "http", "svn+ssh", etc.
 170  *
 171  * Returns: The "Scheme" component of the URI, or %NULL on error.
 172  * The returned string should be freed when no longer needed.
 173  *
 174  * Since: 2.16
 175  **/
 176 char *
 177 g_uri_parse_scheme (const char  *uri)
 178 {
 179   const char *p;
 180   char c;
 181
 182   g_return_val_if_fail (uri != NULL, NULL);
 183
 184   /* From RFC 3986 Decodes:
 185    * URI         = scheme ":" hier-part [ "?" query ] [ "#" fragment ]
 186    */
 187
 188   p = uri;
 189
 190   /* Decode scheme:
 191      scheme      = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
 192   */
 193
 194   if (!g_ascii_isalpha (*p))
 195     return NULL;
 196
 197   while (1)
 198     {
 199       c = *p++;
 200
 201       if (c == ':')
 202         break;
 203
 204       if (!(g_ascii_isalnum(c) ||
 205             c == '+' ||
 206             c == '-' ||
 207             c == '.'))
 208         return NULL;
 209     }
 210
 211   return g_strndup (uri, p - uri - 1);
 212 }
 213
 214 /**
 215  * g_uri_escape_string:
 216  * @unescaped: the unescaped input string.
 217  * @reserved_chars_allowed: a string of reserved characters that are
 218  *      allowed to be used, or %NULL.
 219  * @allow_utf8: %TRUE if the result can include UTF-8 characters.
 220  *
 221  * Escapes a string for use in a URI.
 222  *
 223  * Normally all characters that are not "unreserved" (i.e. ASCII alphanumerical
 224  * characters plus dash, dot, underscore and tilde) are escaped.
 225  * But if you specify characters in @reserved_chars_allowed they are not
 226  * escaped. This is useful for the "reserved" characters in the URI
 227  * specification, since those are allowed unescaped in some portions of
 228  * a URI.
 229  *
 230  * Returns: an escaped version of @unescaped. The returned string should be
 231  * freed when no longer needed.
 232  *
 233  * Since: 2.16
 234  **/
 235 char *
 236 g_uri_escape_string (const char *unescaped,
 237                      const char  *reserved_chars_allowed,
 238                      gboolean     allow_utf8)
 239 {
 240   GString *s;
 241
 242   g_return_val_if_fail (unescaped != NULL, NULL);
 243
 244   s = g_string_sized_new (strlen (unescaped) + 10);
 245
 246   g_string_append_uri_escaped (s, unescaped, reserved_chars_allowed, allow_utf8);
 247
 248   return g_string_free (s, FALSE);
 249 }