1 /* -*- Mode: C; tab-width: 8; indent-tabs-mode: t; c-basic-offset: 8 -*- */
2 /* soup-uri.c : utility functions to parse URLs */
6 * Bertrand Guiheneuf <bertrand@helixcode.com>
7 * Dan Winship <danw@helixcode.com>
9 * Copyright 1999, 2000 Helix Code, Inc. (http://www.helixcode.com)
11 * This program is free software; you can redistribute it and/or
12 * modify it under the terms of the GNU General Public License as
13 * published by the Free Software Foundation; either version 2 of the
14 * License, or (at your option) any later version.
16 * This program is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 * GNU General Public License for more details.
21 * You should have received a copy of the GNU General Public License
22 * along with this program; if not, write to the Free Software
23 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
30 * Here we deal with URLs following the general scheme:
31 * protocol://user;AUTH=mech:password@host:port/name
32 * where name is a path-like string (ie dir1/dir2/....) See RFC 1738
33 * for the complete description of Uniform Resource Locators. The
34 * ";AUTH=mech" addition comes from RFC 2384, "POP URL Scheme".
38 * recover the words between #'s or ?'s after the path
48 soup_uri_get_default_port (gchar *proto)
50 gint len = strlen (proto);
52 if (strncasecmp (proto, "https", len) == 0)
54 else if (strncasecmp (proto, "http", len) == 0)
56 else if (strncasecmp (proto, "mailto", len) == 0 ||
57 strncasecmp (proto, "smtp", len) == 0)
59 else if (strncasecmp (proto, "ftp", len) == 0)
66 * soup_uri_new: create a SoupUri object from a string
68 * @uri_string: The string containing the URL to scan
70 * This routine takes a gchar and parses it as a
72 * protocol://user;AUTH=mech:password@host:port/path?querystring
73 * There is no test on the values. For example,
74 * "port" can be a string, not only a number!
75 * The SoupUri structure fields are filled with
76 * the scan results. When a member of the
77 * general URL can not be found, the corresponding
78 * SoupUri member is NULL.
79 * Fields filled in the SoupUri structure are allocated
80 * and url_string is not modified.
82 * Return value: a SoupUri structure containing the URL items.
84 SoupUri *soup_uri_new (const gchar* uri_string)
87 char *semi, *colon, *at, *slash, *path, *query;
90 g_uri = g_new (SoupUri,1);
92 /* Find protocol: initial substring until "://" */
93 colon = strchr (uri_string, ':');
94 if (colon && !strncmp (colon, "://", 3)) {
95 g_uri->protocol = g_strndup (uri_string, colon - uri_string);
96 uri_string = colon + 3;
98 g_uri->protocol = NULL;
100 /* If there is an @ sign, look for user, authmech, and
101 * password before it.
103 at = strchr (uri_string, '@');
105 colon = strchr (uri_string, ':');
106 if (colon && colon < at)
107 g_uri->passwd = g_strndup (colon + 1, at - colon - 1);
109 g_uri->passwd = NULL;
113 semi = strchr(uri_string, ';');
114 if (semi && semi < colon && !strncasecmp (semi, ";auth=", 6))
115 g_uri->authmech = g_strndup (semi + 6, colon - semi - 6);
117 g_uri->authmech = NULL;
121 g_uri->user = g_strndup (uri_string, semi - uri_string);
124 g_uri->user = g_uri->passwd = g_uri->authmech = NULL;
126 /* Find host (required) and port. */
127 slash = strchr (uri_string, '/');
128 colon = strchr (uri_string, ':');
129 if (slash && colon > slash)
133 g_uri->host = g_strndup (uri_string, colon - uri_string);
135 g_uri->port = atoi(colon + 1);
137 g_uri->port = atoi(colon + 1);
139 g_uri->host = g_strndup (uri_string, slash - uri_string);
140 g_uri->port = soup_uri_get_default_port (g_uri->protocol);
142 g_uri->host = g_strdup (uri_string);
143 g_uri->port = soup_uri_get_default_port (g_uri->protocol);
146 /* setup a fallback, if relative, then empty string, else
147 it will be from root */
151 if (slash && *slash && g_uri->protocol == NULL)
154 split = g_strsplit(slash, " ", 0);
155 path = g_strjoinv("%20", split);
158 query = strchr (path, '?');
161 g_uri->path = g_strndup (path, query - path);
162 g_uri->querystring = g_strdup (++query);
170 soup_uri_to_string (const SoupUri *uri, gboolean show_passwd)
173 return g_strdup_printf(
174 "%s%s%s%s%s%s%s%s%s:%d%s",
175 uri->protocol ? uri->protocol : "",
176 uri->protocol ? "://" : "",
177 uri->user ? uri->user : "",
178 uri->authmech ? ";auth=" : "",
179 uri->authmech ? uri->authmech : "",
180 uri->passwd && show_passwd ? ":" : "",
181 uri->passwd && show_passwd ? uri->passwd : "",
182 uri->user ? "@" : "",
185 uri->path ? uri->path : "");
187 return g_strdup_printf(
188 "%s%s%s%s%s%s%s%s%s%s",
189 uri->protocol ? uri->protocol : "",
190 uri->protocol ? "://" : "",
191 uri->user ? uri->user : "",
192 uri->authmech ? ";auth=" : "",
193 uri->authmech ? uri->authmech : "",
194 uri->passwd && show_passwd ? ":" : "",
195 uri->passwd && show_passwd ? uri->passwd : "",
196 uri->user ? "@" : "",
198 uri->path ? uri->path : "");
202 soup_uri_free (SoupUri *uri)
206 g_free (uri->protocol);
208 g_free (uri->authmech);
209 g_free (uri->passwd);