src/core/uri.c

   1 /*
   2  * Copyright (C) 2007 Michael Brown <mbrown@fensystems.co.uk>.
   3  *
   4  * This program is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU General Public License as
   6  * published by the Free Software Foundation; either version 2 of the
   7  * License, or any later version.
   8  *
   9  * This program is distributed in the hope that it will be useful, but
  10  * WITHOUT ANY WARRANTY; without even the implied warranty of
  11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12  * General Public License for more details.
  13  *
  14  * You should have received a copy of the GNU General Public License
  15  * along with this program; if not, write to the Free Software
  16  * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  17  */
  18
  19 /** @file
  20  *
  21  * Uniform Resource Identifiers
  22  *
  23  */
  24
  25 #include <stdint.h>
  26 #include <stdlib.h>
  27 #include <string.h>
  28 #include <gpxe/uri.h>
  29
  30 /**
  31  * Parse URI
  32  *
  33  * @v uri_string        URI as a string
  34  * @ret uri             URI
  35  *
  36  * Splits a URI into its component parts.  The return URI structure is
  37  * dynamically allocated and must eventually be freed by calling
  38  * free_uri().
  39  */
  40 struct uri * parse_uri ( const char *uri_string ) {
  41         struct uri *uri;
  42         char *raw;
  43         char *tmp;
  44         char *path = NULL;
  45         char *authority = NULL;
  46         size_t raw_len;
  47
  48         /* Allocate space for URI struct and a copy of the string */
  49         raw_len = ( strlen ( uri_string ) + 1 /* NUL */ );
  50         uri = malloc ( sizeof ( *uri ) + raw_len );
  51         if ( ! uri )
  52                 return NULL;
  53         raw = ( ( ( char * ) uri ) + sizeof ( *uri ) );
  54
  55         /* Zero URI struct and copy in the raw string */
  56         memset ( uri, 0, sizeof ( *uri ) );
  57         memcpy ( raw, uri_string, raw_len );
  58
  59         /* Start by chopping off the fragment, if it exists */
  60         if ( ( tmp = strchr ( raw, '#' ) ) ) {
  61                 *(tmp++) = '\0';
  62                 uri->fragment = tmp;
  63         }
  64
  65         /* Identify absolute/relative URI */
  66         if ( ( tmp = strchr ( raw, ':' ) ) ) {
  67                 /* Absolute URI: identify hierarchical/opaque */
  68                 uri->scheme = raw;
  69                 *(tmp++) = '\0';
  70                 if ( *tmp == '/' ) {
  71                         /* Absolute URI with hierarchical part */
  72                         path = tmp;
  73                 } else {
  74                         /* Absolute URI with opaque part */
  75                         uri->opaque = tmp;
  76                 }
  77         } else {
  78                 /* Relative URI */
  79                 path = raw;
  80         }
  81
  82         /* If we don't have a path (i.e. we have an absolute URI with
  83          * an opaque portion, we're already finished processing
  84          */
  85         if ( ! path )
  86                 goto done;
  87
  88         /* Chop off the query, if it exists */
  89         if ( ( tmp = strchr ( path, '?' ) ) ) {
  90                 *(tmp++) = '\0';
  91                 uri->query = tmp;
  92         }
  93
  94         /* Identify net/absolute/relative path */
  95         if ( strncmp ( path, "//", 2 ) == 0 ) {
  96                 /* Net path.  If this is terminated by the first '/'
  97                  * of an absolute path, then we have no space for a
  98                  * terminator after the authority field, so shuffle
  99                  * the authority down by one byte, overwriting one of
 100                  * the two slashes.
 101                  */
 102                 authority = ( path + 2 );
 103                 if ( ( tmp = strchr ( authority, '/' ) ) ) {
 104                         /* Shuffle down */
 105                         uri->path = tmp;
 106                         memmove ( ( authority - 1 ), authority,
 107                                   ( tmp - authority ) );
 108                         authority--;
 109                         *(--tmp) = '\0';
 110                 }
 111         } else {
 112                 /* Absolute/relative path */
 113                 uri->path = path;
 114         }
 115
 116         /* Split authority into user[:password] and host[:port] portions */
 117         if ( ( tmp = strchr ( authority, '@' ) ) ) {
 118                 /* Has user[:password] */
 119                 *(tmp++) = '\0';
 120                 uri->host = tmp;
 121                 uri->user = authority;
 122                 if ( ( tmp = strchr ( authority, ':' ) ) ) {
 123                         /* Has password */
 124                         *(tmp++) = '\0';
 125                         uri->password = tmp;
 126                 }
 127         } else {
 128                 /* No user:password */
 129                 uri->host = authority;
 130         }
 131
 132         /* Split host into host[:port] */
 133         if ( ( tmp = strchr ( uri->host, ':' ) ) ) {
 134                 *(tmp++) = '\0';
 135                 uri->port = tmp;
 136         }
 137
 138  done:
 139         DBG ( "URI \"%s\" split into", raw );
 140         if ( uri->scheme )
 141                 DBG ( " scheme \"%s\"", uri->scheme );
 142         if ( uri->opaque )
 143                 DBG ( " opaque \"%s\"", uri->opaque );
 144         if ( uri->user )
 145                 DBG ( " user \"%s\"", uri->user );
 146         if ( uri->password )
 147                 DBG ( " password \"%s\"", uri->password );
 148         if ( uri->host )
 149                 DBG ( " host \"%s\"", uri->host );
 150         if ( uri->port )
 151                 DBG ( " port \"%s\"", uri->port );
 152         if ( uri->path )
 153                 DBG ( " path \"%s\"", uri->path );
 154         if ( uri->query )
 155                 DBG ( " query \"%s\"", uri->query );
 156         if ( uri->fragment )
 157                 DBG ( " fragment \"%s\"", uri->fragment );
 158         DBG ( "\n" );
 159
 160         return uri;
 161 }
 162
 163 /**
 164  * Get port from URI
 165  *
 166  * @v uri               URI
 167  * @v default_port      Default port to use if none specified in URI
 168  * @ret port            Port
 169  */
 170 unsigned int uri_port ( struct uri *uri, unsigned int default_port ) {
 171         return ( uri->port ? strtoul ( uri->port, NULL, 0 ) : default_port );
 172 }