fetch.c revision 41989 
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 1.7 1998/12/16 10:24:54 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 * Select the appropriate protocol for the URL scheme, and return a
123 * list of files in the directory pointed to by the URL.
124 */
125struct url_ent *
126fetchList(struct url *URL, char *flags)
127{
128    if (strcasecmp(URL->scheme, "file") == 0)
129	return fetchListFile(URL, flags);
130    else if (strcasecmp(URL->scheme, "http") == 0)
131	return fetchListHTTP(URL, flags);
132    else if (strcasecmp(URL->scheme, "ftp") == 0)
133	return fetchListFTP(URL, flags);
134    else {
135	_url_seterr(URL_BAD_SCHEME);
136	return NULL;
137    }
138}
139
140/*
141 * Attempt to parse the given URL; if successful, call fetchGet().
142 */
143FILE *
144fetchGetURL(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 = fetchGet(u, flags);
153
154    free(u);
155    return f;
156}
157
158
159/*
160 * Attempt to parse the given URL; if successful, call fetchPut().
161 */
162FILE *
163fetchPutURL(char *URL, char *flags)
164{
165    struct url *u;
166    FILE *f;
167
168    if ((u = fetchParseURL(URL)) == NULL)
169	return NULL;
170
171    f = fetchPut(u, flags);
172
173    free(u);
174    return f;
175}
176
177/*
178 * Attempt to parse the given URL; if successful, call fetchStat().
179 */
180int
181fetchStatURL(char *URL, struct url_stat *us, char *flags)
182{
183    struct url *u;
184    int s;
185
186    if ((u = fetchParseURL(URL)) == NULL)
187	return -1;
188
189    s = fetchStat(u, us, flags);
190
191    free(u);
192    return s;
193}
194
195/*
196 * Attempt to parse the given URL; if successful, call fetchList().
197 */
198struct url_ent *
199fetchListURL(char *URL, char *flags)
200{
201    struct url *u;
202    struct url_ent *ue;
203
204    if ((u = fetchParseURL(URL)) == NULL)
205	return NULL;
206
207    ue = fetchList(u, flags);
208
209    free(u);
210    return ue;
211}
212
213/*
214 * Split an URL into components. URL syntax is:
215 * method:[//[user[:pwd]@]host[:port]]/[document]
216 * This almost, but not quite, RFC1738 URL syntax.
217 */
218struct url *
219fetchParseURL(char *URL)
220{
221    char *p, *q;
222    struct url *u;
223    int i;
224
225    /* allocate struct url */
226    if ((u = calloc(1, sizeof(struct url))) == NULL) {
227	errno = ENOMEM;
228	_fetch_syserr();
229	return NULL;
230    }
231
232    /* scheme name */
233    for (i = 0; *URL && (*URL != ':'); URL++)
234	if (i < URL_SCHEMELEN)
235	    u->scheme[i++] = *URL;
236    if (!URL[0] || (URL[1] != '/')) {
237	_url_seterr(URL_BAD_SCHEME);
238	goto ouch;
239    }
240    else URL++;
241    if (URL[1] != '/') {
242	p = URL;
243	goto nohost;
244    }
245    else URL += 2;
246
247    p = strpbrk(URL, "/@");
248    if (p && *p == '@') {
249	/* username */
250	for (q = URL, i = 0; (*q != ':') && (*q != '@'); q++)
251	    if (i < URL_USERLEN)
252		u->user[i++] = *q;
253
254	/* password */
255	if (*q == ':')
256	    for (q++, i = 0; (*q != ':') && (*q != '@'); q++)
257		if (i < URL_PWDLEN)
258		    u->pwd[i++] = *q;
259
260	p++;
261    } else p = URL;
262
263    /* hostname */
264    for (i = 0; *p && (*p != '/') && (*p != ':'); p++)
265	if (i < MAXHOSTNAMELEN)
266	    u->host[i++] = *p;
267
268    /* port */
269    if (*p == ':') {
270	for (q = ++p; *q && (*q != '/'); q++)
271	    if (isdigit(*q))
272		u->port = u->port * 10 + (*q - '0');
273	    else {
274		/* invalid port */
275		_url_seterr(URL_BAD_PORT);
276		goto ouch;
277	    }
278	while (*p && (*p != '/'))
279	    p++;
280    }
281
282nohost:
283    /* document */
284    if (*p) {
285	struct url *t;
286	t = realloc(u, sizeof(*u)+strlen(p)-1);
287	if (t == NULL) {
288	    errno = ENOMEM;
289	    _fetch_syserr();
290	    goto ouch;
291	}
292	u = t;
293	strcpy(u->doc, p);
294    } else {
295	u->doc[0] = '/';
296	u->doc[1] = 0;
297    }
298
299    DEBUG(fprintf(stderr,
300		  "scheme:   [\033[1m%s\033[m]\n"
301		  "user:     [\033[1m%s\033[m]\n"
302		  "password: [\033[1m%s\033[m]\n"
303		  "host:     [\033[1m%s\033[m]\n"
304		  "port:     [\033[1m%d\033[m]\n"
305		  "document: [\033[1m%s\033[m]\n",
306		  u->scheme, u->user, u->pwd,
307		  u->host, u->port, u->doc));
308
309    return u;
310
311ouch:
312    free(u);
313    return NULL;
314}
315