fetch.c revision 41862 
1/*-
2 * Copyright (c) 1998 Dag-Erling Co�dan Sm�rgrav
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 *    notice, this list of conditions and the following disclaimer
10 *    in this position and unchanged.
11 * 2. Redistributions in binary form must reproduce the above copyright
12 *    notice, this list of conditions and the following disclaimer in the
13 *    documentation and/or other materials provided with the distribution.
14 * 3. The name of the author may not be used to endorse or promote products
15 *    derived from this software without specific prior written permission
16 *
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
18 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
19 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
20 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
21 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
22 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
23 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
24 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
25 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
26 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
27 *
28 *	$Id: fetch.c,v 1.6 1998/11/06 22:14:08 des Exp $
29 */
30
31#include <sys/param.h>
32#include <sys/errno.h>
33
34#include <ctype.h>
35#include <stdio.h>
36#include <stdlib.h>
37#include <string.h>
38
39#include "fetch.h"
40#include "common.h"
41
42
43int fetchLastErrCode;
44
45
46/*** Local data **************************************************************/
47
48/*
49 * Error messages for parser errors
50 */
51#define URL_MALFORMED		1
52#define URL_BAD_SCHEME		2
53#define URL_BAD_PORT		3
54static struct fetcherr _url_errlist[] = {
55    { URL_MALFORMED,	FETCH_URL,	"Malformed URL" },
56    { URL_BAD_SCHEME,	FETCH_URL,	"Invalid URL scheme" },
57    { URL_BAD_PORT,	FETCH_URL,	"Invalid server port" },
58    { -1,		FETCH_UNKNOWN,	"Unknown parser error" }
59};
60
61
62/*** Public API **************************************************************/
63
64/*
65 * Select the appropriate protocol for the URL scheme, and return a
66 * read-only stream connected to the document referenced by the URL.
67 */
68FILE *
69fetchGet(struct url *URL, char *flags)
70{
71    if (strcasecmp(URL->scheme, "file") == 0)
72	return fetchGetFile(URL, flags);
73    else if (strcasecmp(URL->scheme, "http") == 0)
74	return fetchGetHTTP(URL, flags);
75    else if (strcasecmp(URL->scheme, "ftp") == 0)
76	return fetchGetFTP(URL, flags);
77    else {
78	_url_seterr(URL_BAD_SCHEME);
79	return NULL;
80    }
81}
82
83/*
84 * Select the appropriate protocol for the URL scheme, and return a
85 * write-only stream connected to the document referenced by the URL.
86 */
87FILE *
88fetchPut(struct url *URL, char *flags)
89{
90    if (strcasecmp(URL->scheme, "file") == 0)
91	return fetchPutFile(URL, flags);
92    else if (strcasecmp(URL->scheme, "http") == 0)
93	return fetchPutHTTP(URL, flags);
94    else if (strcasecmp(URL->scheme, "ftp") == 0)
95	return fetchPutFTP(URL, flags);
96    else {
97	_url_seterr(URL_BAD_SCHEME);
98	return NULL;
99    }
100}
101
102/*
103 * Select the appropriate protocol for the URL scheme, and return the
104 * size of the document referenced by the URL if it exists.
105 */
106int
107fetchStat(struct url *URL, struct url_stat *us, char *flags)
108{
109    if (strcasecmp(URL->scheme, "file") == 0)
110	return fetchStatFile(URL, us, flags);
111    else if (strcasecmp(URL->scheme, "http") == 0)
112	return fetchStatHTTP(URL, us, flags);
113    else if (strcasecmp(URL->scheme, "ftp") == 0)
114	return fetchStatFTP(URL, us, flags);
115    else {
116	_url_seterr(URL_BAD_SCHEME);
117	return -1;
118    }
119}
120
121/*
122 * Attempt to parse the given URL; if successful, call fetchGet().
123 */
124FILE *
125fetchGetURL(char *URL, char *flags)
126{
127    struct url *u;
128    FILE *f;
129
130    if ((u = fetchParseURL(URL)) == NULL)
131	return NULL;
132
133    f = fetchGet(u, flags);
134
135    free(u);
136    return f;
137}
138
139
140/*
141 * Attempt to parse the given URL; if successful, call fetchPut().
142 */
143FILE *
144fetchPutURL(char *URL, char *flags)
145{
146    struct url *u;
147    FILE *f;
148
149    if ((u = fetchParseURL(URL)) == NULL)
150	return NULL;
151
152    f = fetchPut(u, flags);
153
154    free(u);
155    return f;
156}
157
158/*
159 * Attempt to parse the given URL; if successful, call fetchStat().
160 */
161int
162fetchStatURL(char *URL, struct url_stat *us, char *flags)
163{
164    struct url *u;
165    int s;
166
167    if ((u = fetchParseURL(URL)) == NULL)
168	return -1;
169
170    s = fetchStat(u, us, flags);
171
172    free(u);
173    return s;
174}
175
176/*
177 * Split an URL into components. URL syntax is:
178 * method:[//[user[:pwd]@]host[:port]]/[document]
179 * This almost, but not quite, RFC1738 URL syntax.
180 */
181struct url *
182fetchParseURL(char *URL)
183{
184    char *p, *q;
185    struct url *u;
186    int i;
187
188    /* allocate struct url */
189    if ((u = calloc(1, sizeof(struct url))) == NULL) {
190	errno = ENOMEM;
191	_fetch_syserr();
192	return NULL;
193    }
194
195    /* scheme name */
196    for (i = 0; *URL && (*URL != ':'); URL++)
197	if (i < URL_SCHEMELEN)
198	    u->scheme[i++] = *URL;
199    if (!URL[0] || (URL[1] != '/')) {
200	_url_seterr(URL_BAD_SCHEME);
201	goto ouch;
202    }
203    else URL++;
204    if (URL[1] != '/') {
205	p = URL;
206	goto nohost;
207    }
208    else URL += 2;
209
210    p = strpbrk(URL, "/@");
211    if (p && *p == '@') {
212	/* username */
213	for (q = URL, i = 0; (*q != ':') && (*q != '@'); q++)
214	    if (i < URL_USERLEN)
215		u->user[i++] = *q;
216
217	/* password */
218	if (*q == ':')
219	    for (q++, i = 0; (*q != ':') && (*q != '@'); q++)
220		if (i < URL_PWDLEN)
221		    u->pwd[i++] = *q;
222
223	p++;
224    } else p = URL;
225
226    /* hostname */
227    for (i = 0; *p && (*p != '/') && (*p != ':'); p++)
228	if (i < MAXHOSTNAMELEN)
229	    u->host[i++] = *p;
230
231    /* port */
232    if (*p == ':') {
233	for (q = ++p; *q && (*q != '/'); q++)
234	    if (isdigit(*q))
235		u->port = u->port * 10 + (*q - '0');
236	    else {
237		/* invalid port */
238		_url_seterr(URL_BAD_PORT);
239		goto ouch;
240	    }
241	while (*p && (*p != '/'))
242	    p++;
243    }
244
245nohost:
246    /* document */
247    if (*p) {
248	struct url *t;
249	t = realloc(u, sizeof(*u)+strlen(p)-1);
250	if (t == NULL) {
251	    errno = ENOMEM;
252	    _fetch_syserr();
253	    goto ouch;
254	}
255	u = t;
256	strcpy(u->doc, p);
257    } else {
258	u->doc[0] = '/';
259	u->doc[1] = 0;
260    }
261
262    DEBUG(fprintf(stderr,
263		  "scheme:   [\033[1m%s\033[m]\n"
264		  "user:     [\033[1m%s\033[m]\n"
265		  "password: [\033[1m%s\033[m]\n"
266		  "host:     [\033[1m%s\033[m]\n"
267		  "port:     [\033[1m%d\033[m]\n"
268		  "document: [\033[1m%s\033[m]\n",
269		  u->scheme, u->user, u->pwd,
270		  u->host, u->port, u->doc));
271
272    return u;
273
274ouch:
275    free(u);
276    return NULL;
277}
278