fetch.c revision 60587 
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 60587 2000-05-15 09:05:36Z ume $
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;
44int fetchTimeout;
45
46
47/*** Local data **************************************************************/
48
49/*
50 * Error messages for parser errors
51 */
52#define URL_MALFORMED		1
53#define URL_BAD_SCHEME		2
54#define URL_BAD_PORT		3
55static struct fetcherr _url_errlist[] = {
56    { URL_MALFORMED,	FETCH_URL,	"Malformed URL" },
57    { URL_BAD_SCHEME,	FETCH_URL,	"Invalid URL scheme" },
58    { URL_BAD_PORT,	FETCH_URL,	"Invalid server port" },
59    { -1,		FETCH_UNKNOWN,	"Unknown parser error" }
60};
61
62
63/*** Public API **************************************************************/
64
65/*
66 * Select the appropriate protocol for the URL scheme, and return a
67 * read-only stream connected to the document referenced by the URL.
68 */
69FILE *
70fetchGet(struct url *URL, char *flags)
71{
72    int direct;
73
74    direct = (flags && strchr(flags, 'd'));
75    if (strcasecmp(URL->scheme, "file") == 0)
76	return fetchGetFile(URL, flags);
77    else if (strcasecmp(URL->scheme, "http") == 0)
78	return fetchGetHTTP(URL, flags);
79    else if (strcasecmp(URL->scheme, "ftp") == 0) {
80	if (!direct &&
81	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
82	    return fetchGetHTTP(URL, flags);
83	return fetchGetFTP(URL, flags);
84    } else {
85	_url_seterr(URL_BAD_SCHEME);
86	return NULL;
87    }
88}
89
90/*
91 * Select the appropriate protocol for the URL scheme, and return a
92 * write-only stream connected to the document referenced by the URL.
93 */
94FILE *
95fetchPut(struct url *URL, char *flags)
96{
97    int direct;
98
99    direct = (flags && strchr(flags, 'd'));
100    if (strcasecmp(URL->scheme, "file") == 0)
101	return fetchPutFile(URL, flags);
102    else if (strcasecmp(URL->scheme, "http") == 0)
103	return fetchPutHTTP(URL, flags);
104    else if (strcasecmp(URL->scheme, "ftp") == 0) {
105	if (!direct &&
106	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
107	    return fetchPutHTTP(URL, flags);
108	return fetchPutFTP(URL, flags);
109    } else {
110	_url_seterr(URL_BAD_SCHEME);
111	return NULL;
112    }
113}
114
115/*
116 * Select the appropriate protocol for the URL scheme, and return the
117 * size of the document referenced by the URL if it exists.
118 */
119int
120fetchStat(struct url *URL, struct url_stat *us, char *flags)
121{
122    int direct;
123
124    direct = (flags && strchr(flags, 'd'));
125    if (strcasecmp(URL->scheme, "file") == 0)
126	return fetchStatFile(URL, us, flags);
127    else if (strcasecmp(URL->scheme, "http") == 0)
128	return fetchStatHTTP(URL, us, flags);
129    else if (strcasecmp(URL->scheme, "ftp") == 0) {
130	if (!direct &&
131	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
132	    return fetchStatHTTP(URL, us, flags);
133	return fetchStatFTP(URL, us, flags);
134    } else {
135	_url_seterr(URL_BAD_SCHEME);
136	return -1;
137    }
138}
139
140/*
141 * Select the appropriate protocol for the URL scheme, and return a
142 * list of files in the directory pointed to by the URL.
143 */
144struct url_ent *
145fetchList(struct url *URL, char *flags)
146{
147    int direct;
148
149    direct = (flags && strchr(flags, 'd'));
150    if (strcasecmp(URL->scheme, "file") == 0)
151	return fetchListFile(URL, flags);
152    else if (strcasecmp(URL->scheme, "http") == 0)
153	return fetchListHTTP(URL, flags);
154    else if (strcasecmp(URL->scheme, "ftp") == 0) {
155	if (!direct &&
156	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
157	    return fetchListHTTP(URL, flags);
158	return fetchListFTP(URL, flags);
159    } else {
160	_url_seterr(URL_BAD_SCHEME);
161	return NULL;
162    }
163}
164
165/*
166 * Attempt to parse the given URL; if successful, call fetchGet().
167 */
168FILE *
169fetchGetURL(char *URL, char *flags)
170{
171    struct url *u;
172    FILE *f;
173
174    if ((u = fetchParseURL(URL)) == NULL)
175	return NULL;
176
177    f = fetchGet(u, flags);
178
179    fetchFreeURL(u);
180    return f;
181}
182
183
184/*
185 * Attempt to parse the given URL; if successful, call fetchPut().
186 */
187FILE *
188fetchPutURL(char *URL, char *flags)
189{
190    struct url *u;
191    FILE *f;
192
193    if ((u = fetchParseURL(URL)) == NULL)
194	return NULL;
195
196    f = fetchPut(u, flags);
197
198    fetchFreeURL(u);
199    return f;
200}
201
202/*
203 * Attempt to parse the given URL; if successful, call fetchStat().
204 */
205int
206fetchStatURL(char *URL, struct url_stat *us, char *flags)
207{
208    struct url *u;
209    int s;
210
211    if ((u = fetchParseURL(URL)) == NULL)
212	return -1;
213
214    s = fetchStat(u, us, flags);
215
216    fetchFreeURL(u);
217    return s;
218}
219
220/*
221 * Attempt to parse the given URL; if successful, call fetchList().
222 */
223struct url_ent *
224fetchListURL(char *URL, char *flags)
225{
226    struct url *u;
227    struct url_ent *ue;
228
229    if ((u = fetchParseURL(URL)) == NULL)
230	return NULL;
231
232    ue = fetchList(u, flags);
233
234    fetchFreeURL(u);
235    return ue;
236}
237
238/*
239 * Split an URL into components. URL syntax is:
240 * method:[//[user[:pwd]@]host[:port]]/[document]
241 * This almost, but not quite, RFC1738 URL syntax.
242 */
243struct url *
244fetchParseURL(char *URL)
245{
246    char *p, *q;
247    struct url *u;
248    int i;
249
250    /* allocate struct url */
251    if ((u = calloc(1, sizeof *u)) == NULL) {
252	errno = ENOMEM;
253	_fetch_syserr();
254	return NULL;
255    }
256
257    /* scheme name */
258    for (i = 0; *URL && (*URL != ':'); URL++)
259	if (i < URL_SCHEMELEN)
260	    u->scheme[i++] = *URL;
261    if (!URL[0] || (URL[1] != '/')) {
262	_url_seterr(URL_BAD_SCHEME);
263	goto ouch;
264    }
265    else URL++;
266    if (URL[1] != '/') {
267	p = URL;
268	goto nohost;
269    }
270    else URL += 2;
271
272    p = strpbrk(URL, "/@");
273    if (p && *p == '@') {
274	/* username */
275	for (q = URL, i = 0; (*q != ':') && (*q != '@'); q++)
276	    if (i < URL_USERLEN)
277		u->user[i++] = *q;
278
279	/* password */
280	if (*q == ':')
281	    for (q++, i = 0; (*q != ':') && (*q != '@'); q++)
282		if (i < URL_PWDLEN)
283		    u->pwd[i++] = *q;
284
285	p++;
286    } else p = URL;
287
288    /* hostname */
289    for (i = 0; *p && (*p != '/') && (*p != ':'); p++)
290	if (i < MAXHOSTNAMELEN)
291	    u->host[i++] = *p;
292
293    /* port */
294    if (*p == ':') {
295	for (q = ++p; *q && (*q != '/'); q++)
296	    if (isdigit(*q))
297		u->port = u->port * 10 + (*q - '0');
298	    else {
299		/* invalid port */
300		_url_seterr(URL_BAD_PORT);
301		goto ouch;
302	    }
303	while (*p && (*p != '/'))
304	    p++;
305    }
306
307nohost:
308    /* document */
309    if (!*p)
310	p = "/";
311
312    if ((u->doc = strdup(p)) == NULL) {
313	errno = ENOMEM;
314	_fetch_syserr();
315	goto ouch;
316    }
317
318    DEBUG(fprintf(stderr,
319		  "scheme:   [\033[1m%s\033[m]\n"
320		  "user:     [\033[1m%s\033[m]\n"
321		  "password: [\033[1m%s\033[m]\n"
322		  "host:     [\033[1m%s\033[m]\n"
323		  "port:     [\033[1m%d\033[m]\n"
324		  "document: [\033[1m%s\033[m]\n",
325		  u->scheme, u->user, u->pwd,
326		  u->host, u->port, u->doc));
327
328    return u;
329
330ouch:
331    free(u);
332    return NULL;
333}
334
335/*
336 * Free a URL
337 */
338void
339fetchFreeURL(struct url *u)
340{
341    free(u->doc);
342    free(u);
343}
344