fetch.c revision 60924 
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 * $FreeBSD: head/lib/libfetch/fetch.c 60924 2000-05-25 16:24:31Z des $
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;
44char	 fetchLastErrString[MAXERRSTRING];
45int	 fetchTimeout;
46
47
48/*** Local data **************************************************************/
49
50/*
51 * Error messages for parser errors
52 */
53#define URL_MALFORMED		1
54#define URL_BAD_SCHEME		2
55#define URL_BAD_PORT		3
56static struct fetcherr _url_errlist[] = {
57    { URL_MALFORMED,	FETCH_URL,	"Malformed URL" },
58    { URL_BAD_SCHEME,	FETCH_URL,	"Invalid URL scheme" },
59    { URL_BAD_PORT,	FETCH_URL,	"Invalid server port" },
60    { -1,		FETCH_UNKNOWN,	"Unknown parser error" }
61};
62
63
64/*** Public API **************************************************************/
65
66/*
67 * Select the appropriate protocol for the URL scheme, and return a
68 * read-only stream connected to the document referenced by the URL.
69 */
70FILE *
71fetchGet(struct url *URL, char *flags)
72{
73    int direct;
74
75    direct = (flags && strchr(flags, 'd'));
76    if (strcasecmp(URL->scheme, "file") == 0)
77	return fetchGetFile(URL, flags);
78    else if (strcasecmp(URL->scheme, "http") == 0)
79	return fetchGetHTTP(URL, flags);
80    else if (strcasecmp(URL->scheme, "ftp") == 0) {
81	if (!direct &&
82	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
83	    return fetchGetHTTP(URL, flags);
84	return fetchGetFTP(URL, flags);
85    } else {
86	_url_seterr(URL_BAD_SCHEME);
87	return NULL;
88    }
89}
90
91/*
92 * Select the appropriate protocol for the URL scheme, and return a
93 * write-only stream connected to the document referenced by the URL.
94 */
95FILE *
96fetchPut(struct url *URL, char *flags)
97{
98    int direct;
99
100    direct = (flags && strchr(flags, 'd'));
101    if (strcasecmp(URL->scheme, "file") == 0)
102	return fetchPutFile(URL, flags);
103    else if (strcasecmp(URL->scheme, "http") == 0)
104	return fetchPutHTTP(URL, flags);
105    else if (strcasecmp(URL->scheme, "ftp") == 0) {
106	if (!direct &&
107	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
108	    return fetchPutHTTP(URL, flags);
109	return fetchPutFTP(URL, flags);
110    } else {
111	_url_seterr(URL_BAD_SCHEME);
112	return NULL;
113    }
114}
115
116/*
117 * Select the appropriate protocol for the URL scheme, and return the
118 * size of the document referenced by the URL if it exists.
119 */
120int
121fetchStat(struct url *URL, struct url_stat *us, char *flags)
122{
123    int direct;
124
125    direct = (flags && strchr(flags, 'd'));
126    if (strcasecmp(URL->scheme, "file") == 0)
127	return fetchStatFile(URL, us, flags);
128    else if (strcasecmp(URL->scheme, "http") == 0)
129	return fetchStatHTTP(URL, us, flags);
130    else if (strcasecmp(URL->scheme, "ftp") == 0) {
131	if (!direct &&
132	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
133	    return fetchStatHTTP(URL, us, flags);
134	return fetchStatFTP(URL, us, flags);
135    } else {
136	_url_seterr(URL_BAD_SCHEME);
137	return -1;
138    }
139}
140
141/*
142 * Select the appropriate protocol for the URL scheme, and return a
143 * list of files in the directory pointed to by the URL.
144 */
145struct url_ent *
146fetchList(struct url *URL, char *flags)
147{
148    int direct;
149
150    direct = (flags && strchr(flags, 'd'));
151    if (strcasecmp(URL->scheme, "file") == 0)
152	return fetchListFile(URL, flags);
153    else if (strcasecmp(URL->scheme, "http") == 0)
154	return fetchListHTTP(URL, flags);
155    else if (strcasecmp(URL->scheme, "ftp") == 0) {
156	if (!direct &&
157	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
158	    return fetchListHTTP(URL, flags);
159	return fetchListFTP(URL, flags);
160    } else {
161	_url_seterr(URL_BAD_SCHEME);
162	return NULL;
163    }
164}
165
166/*
167 * Attempt to parse the given URL; if successful, call fetchGet().
168 */
169FILE *
170fetchGetURL(char *URL, char *flags)
171{
172    struct url *u;
173    FILE *f;
174
175    if ((u = fetchParseURL(URL)) == NULL)
176	return NULL;
177
178    f = fetchGet(u, flags);
179
180    fetchFreeURL(u);
181    return f;
182}
183
184
185/*
186 * Attempt to parse the given URL; if successful, call fetchPut().
187 */
188FILE *
189fetchPutURL(char *URL, char *flags)
190{
191    struct url *u;
192    FILE *f;
193
194    if ((u = fetchParseURL(URL)) == NULL)
195	return NULL;
196
197    f = fetchPut(u, flags);
198
199    fetchFreeURL(u);
200    return f;
201}
202
203/*
204 * Attempt to parse the given URL; if successful, call fetchStat().
205 */
206int
207fetchStatURL(char *URL, struct url_stat *us, char *flags)
208{
209    struct url *u;
210    int s;
211
212    if ((u = fetchParseURL(URL)) == NULL)
213	return -1;
214
215    s = fetchStat(u, us, flags);
216
217    fetchFreeURL(u);
218    return s;
219}
220
221/*
222 * Attempt to parse the given URL; if successful, call fetchList().
223 */
224struct url_ent *
225fetchListURL(char *URL, char *flags)
226{
227    struct url *u;
228    struct url_ent *ue;
229
230    if ((u = fetchParseURL(URL)) == NULL)
231	return NULL;
232
233    ue = fetchList(u, flags);
234
235    fetchFreeURL(u);
236    return ue;
237}
238
239/*
240 * Split an URL into components. URL syntax is:
241 * method:[//[user[:pwd]@]host[:port]]/[document]
242 * This almost, but not quite, RFC1738 URL syntax.
243 */
244struct url *
245fetchParseURL(char *URL)
246{
247    char *p, *q;
248    struct url *u;
249    int i;
250
251    /* allocate struct url */
252    if ((u = calloc(1, sizeof *u)) == NULL) {
253	errno = ENOMEM;
254	_fetch_syserr();
255	return NULL;
256    }
257
258    /* scheme name */
259    for (i = 0; *URL && (*URL != ':'); URL++)
260	if (i < URL_SCHEMELEN)
261	    u->scheme[i++] = *URL;
262    if (!URL[0] || (URL[1] != '/')) {
263	_url_seterr(URL_BAD_SCHEME);
264	goto ouch;
265    }
266    else URL++;
267    if (URL[1] != '/') {
268	p = URL;
269	goto nohost;
270    }
271    else URL += 2;
272
273    p = strpbrk(URL, "/@");
274    if (p && *p == '@') {
275	/* username */
276	for (q = URL, i = 0; (*q != ':') && (*q != '@'); q++)
277	    if (i < URL_USERLEN)
278		u->user[i++] = *q;
279
280	/* password */
281	if (*q == ':')
282	    for (q++, i = 0; (*q != ':') && (*q != '@'); q++)
283		if (i < URL_PWDLEN)
284		    u->pwd[i++] = *q;
285
286	p++;
287    } else p = URL;
288
289    /* hostname */
290#ifdef INET6
291    if (*p == '[' && (q = strchr(p + 1, ']')) != NULL &&
292	(*++q == '\0' || *q == '/' || *q == ':')) {
293	if ((i = q - p - 2) > MAXHOSTNAMELEN)
294	    i = MAXHOSTNAMELEN;
295	strncpy(u->host, ++p, i);
296	p = q;
297    } else
298#endif
299	for (i = 0; *p && (*p != '/') && (*p != ':'); p++)
300	    if (i < MAXHOSTNAMELEN)
301		u->host[i++] = *p;
302
303    /* port */
304    if (*p == ':') {
305	for (q = ++p; *q && (*q != '/'); q++)
306	    if (isdigit(*q))
307		u->port = u->port * 10 + (*q - '0');
308	    else {
309		/* invalid port */
310		_url_seterr(URL_BAD_PORT);
311		goto ouch;
312	    }
313	while (*p && (*p != '/'))
314	    p++;
315    }
316
317nohost:
318    /* document */
319    if (!*p)
320	p = "/";
321
322    if ((u->doc = strdup(p)) == NULL) {
323	errno = ENOMEM;
324	_fetch_syserr();
325	goto ouch;
326    }
327
328    DEBUG(fprintf(stderr,
329		  "scheme:   [\033[1m%s\033[m]\n"
330		  "user:     [\033[1m%s\033[m]\n"
331		  "password: [\033[1m%s\033[m]\n"
332		  "host:     [\033[1m%s\033[m]\n"
333		  "port:     [\033[1m%d\033[m]\n"
334		  "document: [\033[1m%s\033[m]\n",
335		  u->scheme, u->user, u->pwd,
336		  u->host, u->port, u->doc));
337
338    return u;
339
340ouch:
341    free(u);
342    return NULL;
343}
344
345/*
346 * Free a URL
347 */
348void
349fetchFreeURL(struct url *u)
350{
351    free(u->doc);
352    free(u);
353}
354