fetch.c revision 63334 
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 63334 2000-07-17 20:49:39Z 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;
46int	 fetchRestartCalls = 1;
47
48
49/*** Local data **************************************************************/
50
51/*
52 * Error messages for parser errors
53 */
54#define URL_MALFORMED		1
55#define URL_BAD_SCHEME		2
56#define URL_BAD_PORT		3
57static struct fetcherr _url_errlist[] = {
58    { URL_MALFORMED,	FETCH_URL,	"Malformed URL" },
59    { URL_BAD_SCHEME,	FETCH_URL,	"Invalid URL scheme" },
60    { URL_BAD_PORT,	FETCH_URL,	"Invalid server port" },
61    { -1,		FETCH_UNKNOWN,	"Unknown parser error" }
62};
63
64
65/*** Public API **************************************************************/
66
67/*
68 * Select the appropriate protocol for the URL scheme, and return a
69 * read-only stream connected to the document referenced by the URL.
70 */
71FILE *
72fetchGet(struct url *URL, char *flags)
73{
74    int direct;
75
76    direct = (flags && strchr(flags, 'd'));
77    if (strcasecmp(URL->scheme, "file") == 0)
78	return fetchGetFile(URL, flags);
79    else if (strcasecmp(URL->scheme, "http") == 0)
80	return fetchGetHTTP(URL, flags);
81    else if (strcasecmp(URL->scheme, "ftp") == 0) {
82	if (!direct &&
83	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
84	    return fetchGetHTTP(URL, flags);
85	return fetchGetFTP(URL, flags);
86    } else {
87	_url_seterr(URL_BAD_SCHEME);
88	return NULL;
89    }
90}
91
92/*
93 * Select the appropriate protocol for the URL scheme, and return a
94 * write-only stream connected to the document referenced by the URL.
95 */
96FILE *
97fetchPut(struct url *URL, char *flags)
98{
99    int direct;
100
101    direct = (flags && strchr(flags, 'd'));
102    if (strcasecmp(URL->scheme, "file") == 0)
103	return fetchPutFile(URL, flags);
104    else if (strcasecmp(URL->scheme, "http") == 0)
105	return fetchPutHTTP(URL, flags);
106    else if (strcasecmp(URL->scheme, "ftp") == 0) {
107	if (!direct &&
108	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
109	    return fetchPutHTTP(URL, flags);
110	return fetchPutFTP(URL, flags);
111    } else {
112	_url_seterr(URL_BAD_SCHEME);
113	return NULL;
114    }
115}
116
117/*
118 * Select the appropriate protocol for the URL scheme, and return the
119 * size of the document referenced by the URL if it exists.
120 */
121int
122fetchStat(struct url *URL, struct url_stat *us, char *flags)
123{
124    int direct;
125
126    direct = (flags && strchr(flags, 'd'));
127    if (strcasecmp(URL->scheme, "file") == 0)
128	return fetchStatFile(URL, us, flags);
129    else if (strcasecmp(URL->scheme, "http") == 0)
130	return fetchStatHTTP(URL, us, flags);
131    else if (strcasecmp(URL->scheme, "ftp") == 0) {
132	if (!direct &&
133	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
134	    return fetchStatHTTP(URL, us, flags);
135	return fetchStatFTP(URL, us, flags);
136    } else {
137	_url_seterr(URL_BAD_SCHEME);
138	return -1;
139    }
140}
141
142/*
143 * Select the appropriate protocol for the URL scheme, and return a
144 * list of files in the directory pointed to by the URL.
145 */
146struct url_ent *
147fetchList(struct url *URL, char *flags)
148{
149    int direct;
150
151    direct = (flags && strchr(flags, 'd'));
152    if (strcasecmp(URL->scheme, "file") == 0)
153	return fetchListFile(URL, flags);
154    else if (strcasecmp(URL->scheme, "http") == 0)
155	return fetchListHTTP(URL, flags);
156    else if (strcasecmp(URL->scheme, "ftp") == 0) {
157	if (!direct &&
158	    getenv("FTP_PROXY") == NULL && getenv("HTTP_PROXY") != NULL)
159	    return fetchListHTTP(URL, flags);
160	return fetchListFTP(URL, flags);
161    } else {
162	_url_seterr(URL_BAD_SCHEME);
163	return NULL;
164    }
165}
166
167/*
168 * Attempt to parse the given URL; if successful, call fetchGet().
169 */
170FILE *
171fetchGetURL(char *URL, char *flags)
172{
173    struct url *u;
174    FILE *f;
175
176    if ((u = fetchParseURL(URL)) == NULL)
177	return NULL;
178
179    f = fetchGet(u, flags);
180
181    fetchFreeURL(u);
182    return f;
183}
184
185
186/*
187 * Attempt to parse the given URL; if successful, call fetchPut().
188 */
189FILE *
190fetchPutURL(char *URL, char *flags)
191{
192    struct url *u;
193    FILE *f;
194
195    if ((u = fetchParseURL(URL)) == NULL)
196	return NULL;
197
198    f = fetchPut(u, flags);
199
200    fetchFreeURL(u);
201    return f;
202}
203
204/*
205 * Attempt to parse the given URL; if successful, call fetchStat().
206 */
207int
208fetchStatURL(char *URL, struct url_stat *us, char *flags)
209{
210    struct url *u;
211    int s;
212
213    if ((u = fetchParseURL(URL)) == NULL)
214	return -1;
215
216    s = fetchStat(u, us, flags);
217
218    fetchFreeURL(u);
219    return s;
220}
221
222/*
223 * Attempt to parse the given URL; if successful, call fetchList().
224 */
225struct url_ent *
226fetchListURL(char *URL, char *flags)
227{
228    struct url *u;
229    struct url_ent *ue;
230
231    if ((u = fetchParseURL(URL)) == NULL)
232	return NULL;
233
234    ue = fetchList(u, flags);
235
236    fetchFreeURL(u);
237    return ue;
238}
239
240/*
241 * Make a URL
242 */
243struct url *
244fetchMakeURL(char *scheme, char *host, int port, char *doc,
245    char *user, char *pwd)
246{
247    struct url *u;
248
249    if (!scheme || (!host && !doc)) {
250	_url_seterr(URL_MALFORMED);
251	return NULL;
252    }
253
254    if (port < 0 || port > 65535) {
255	_url_seterr(URL_BAD_PORT);
256	return NULL;
257    }
258
259    /* allocate struct url */
260    if ((u = calloc(1, sizeof *u)) == NULL) {
261	_fetch_syserr();
262	return NULL;
263    }
264
265    if ((u->doc = strdup(doc ? doc : "/")) == NULL) {
266	_fetch_syserr();
267	free(u);
268	return NULL;
269    }
270
271#define seturl(x) snprintf(u->x, sizeof u->x, "%s", x)
272    seturl(scheme);
273    seturl(host);
274    seturl(user);
275    seturl(pwd);
276#undef seturl
277    u->port = port;
278
279    return u;
280}
281
282/*
283 * Split an URL into components. URL syntax is:
284 * method:[//[user[:pwd]@]host[:port]]/[document]
285 * This almost, but not quite, RFC1738 URL syntax.
286 */
287struct url *
288fetchParseURL(char *URL)
289{
290    char *p, *q;
291    struct url *u;
292    int i;
293
294    /* allocate struct url */
295    if ((u = calloc(1, sizeof *u)) == NULL) {
296	_fetch_syserr();
297	return NULL;
298    }
299
300    /* scheme name */
301    for (i = 0; *URL && (*URL != ':'); URL++)
302	if (i < URL_SCHEMELEN)
303	    u->scheme[i++] = *URL;
304    if (!URL[0] || (URL[1] != '/')) {
305	_url_seterr(URL_BAD_SCHEME);
306	goto ouch;
307    }
308    else URL++;
309    if (URL[1] != '/') {
310	p = URL;
311	goto nohost;
312    }
313    else URL += 2;
314
315    p = strpbrk(URL, "/@");
316    if (p && *p == '@') {
317	/* username */
318	for (q = URL, i = 0; (*q != ':') && (*q != '@'); q++)
319	    if (i < URL_USERLEN)
320		u->user[i++] = *q;
321
322	/* password */
323	if (*q == ':')
324	    for (q++, i = 0; (*q != ':') && (*q != '@'); q++)
325		if (i < URL_PWDLEN)
326		    u->pwd[i++] = *q;
327
328	p++;
329    } else p = URL;
330
331    /* hostname */
332#ifdef INET6
333    if (*p == '[' && (q = strchr(p + 1, ']')) != NULL &&
334	(*++q == '\0' || *q == '/' || *q == ':')) {
335	if ((i = q - p - 2) > MAXHOSTNAMELEN)
336	    i = MAXHOSTNAMELEN;
337	strncpy(u->host, ++p, i);
338	p = q;
339    } else
340#endif
341	for (i = 0; *p && (*p != '/') && (*p != ':'); p++)
342	    if (i < MAXHOSTNAMELEN)
343		u->host[i++] = *p;
344
345    /* port */
346    if (*p == ':') {
347	for (q = ++p; *q && (*q != '/'); q++)
348	    if (isdigit(*q))
349		u->port = u->port * 10 + (*q - '0');
350	    else {
351		/* invalid port */
352		_url_seterr(URL_BAD_PORT);
353		goto ouch;
354	    }
355	while (*p && (*p != '/'))
356	    p++;
357    }
358
359nohost:
360    /* document */
361    if (!*p)
362	p = "/";
363
364    if ((u->doc = strdup(p)) == NULL) {
365	_fetch_syserr();
366	goto ouch;
367    }
368
369    DEBUG(fprintf(stderr,
370		  "scheme:   [\033[1m%s\033[m]\n"
371		  "user:     [\033[1m%s\033[m]\n"
372		  "password: [\033[1m%s\033[m]\n"
373		  "host:     [\033[1m%s\033[m]\n"
374		  "port:     [\033[1m%d\033[m]\n"
375		  "document: [\033[1m%s\033[m]\n",
376		  u->scheme, u->user, u->pwd,
377		  u->host, u->port, u->doc));
378
379    return u;
380
381ouch:
382    free(u);
383    return NULL;
384}
385
386/*
387 * Free a URL
388 */
389void
390fetchFreeURL(struct url *u)
391{
392    free(u->doc);
393    free(u);
394}
395