1 /*-
2 * SPDX-License-Identifier: BSD-3-Clause
3 *
4 * Copyright (c) 1998-2004 Dag-Erling Smørgrav
5 * All rights reserved.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer
12 * in this position and unchanged.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 * 3. The name of the author may not be used to endorse or promote products
17 * derived from this software without specific prior written permission
18 *
19 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
20 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
21 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
22 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
23 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
24 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
25 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
26 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
27 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
28 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
29 */
30
31 #include <sys/cdefs.h>
32 #include <sys/param.h>
33
34 #include <netinet/in.h>
35
36 #include <errno.h>
37 #include <ctype.h>
38 #include <stdio.h>
39 #include <stdlib.h>
40 #include <string.h>
41
42 #include "fetch.h"
43 #include "common.h"
44
45 auth_t fetchAuthMethod;
46 int fetchLastErrCode;
47 char fetchLastErrString[MAXERRSTRING];
48 int fetchTimeout;
49 int fetchRestartCalls = 1;
50 int fetchDebug;
51
52
53 /*** Local data **************************************************************/
54
55 /*
56 * Error messages for parser errors
57 */
58 #define URL_MALFORMED 1
59 #define URL_BAD_SCHEME 2
60 #define URL_BAD_PORT 3
61 static struct fetcherr url_errlist[] = {
62 { URL_MALFORMED, FETCH_URL, "Malformed URL" },
63 { URL_BAD_SCHEME, FETCH_URL, "Invalid URL scheme" },
64 { URL_BAD_PORT, FETCH_URL, "Invalid server port" },
65 { -1, FETCH_UNKNOWN, "Unknown parser error" }
66 };
67
68
69 /*** Public API **************************************************************/
70
71 /*
72 * Select the appropriate protocol for the URL scheme, and return a
73 * read-only stream connected to the document referenced by the URL.
74 * Also fill out the struct url_stat.
75 */
76 FILE *
fetchXGet(struct url * URL,struct url_stat * us,const char * flags)77 fetchXGet(struct url *URL, struct url_stat *us, const char *flags)
78 {
79
80 if (us != NULL) {
81 us->size = -1;
82 us->atime = us->mtime = 0;
83 }
84 if (strcmp(URL->scheme, SCHEME_FILE) == 0)
85 return (fetchXGetFile(URL, us, flags));
86 else if (strcmp(URL->scheme, SCHEME_FTP) == 0)
87 return (fetchXGetFTP(URL, us, flags));
88 else if (strcmp(URL->scheme, SCHEME_HTTP) == 0)
89 return (fetchXGetHTTP(URL, us, flags));
90 else if (strcmp(URL->scheme, SCHEME_HTTPS) == 0)
91 return (fetchXGetHTTP(URL, us, flags));
92 url_seterr(URL_BAD_SCHEME);
93 return (NULL);
94 }
95
96 /*
97 * Select the appropriate protocol for the URL scheme, and return a
98 * read-only stream connected to the document referenced by the URL.
99 */
100 FILE *
fetchGet(struct url * URL,const char * flags)101 fetchGet(struct url *URL, const char *flags)
102 {
103 return (fetchXGet(URL, NULL, flags));
104 }
105
106 /*
107 * Select the appropriate protocol for the URL scheme, and return a
108 * write-only stream connected to the document referenced by the URL.
109 */
110 FILE *
fetchPut(struct url * URL,const char * flags)111 fetchPut(struct url *URL, const char *flags)
112 {
113
114 if (strcmp(URL->scheme, SCHEME_FILE) == 0)
115 return (fetchPutFile(URL, flags));
116 else if (strcmp(URL->scheme, SCHEME_FTP) == 0)
117 return (fetchPutFTP(URL, flags));
118 else if (strcmp(URL->scheme, SCHEME_HTTP) == 0)
119 return (fetchPutHTTP(URL, flags));
120 else if (strcmp(URL->scheme, SCHEME_HTTPS) == 0)
121 return (fetchPutHTTP(URL, flags));
122 url_seterr(URL_BAD_SCHEME);
123 return (NULL);
124 }
125
126 /*
127 * Select the appropriate protocol for the URL scheme, and return the
128 * size of the document referenced by the URL if it exists.
129 */
130 int
fetchStat(struct url * URL,struct url_stat * us,const char * flags)131 fetchStat(struct url *URL, struct url_stat *us, const char *flags)
132 {
133
134 if (us != NULL) {
135 us->size = -1;
136 us->atime = us->mtime = 0;
137 }
138 if (strcmp(URL->scheme, SCHEME_FILE) == 0)
139 return (fetchStatFile(URL, us, flags));
140 else if (strcmp(URL->scheme, SCHEME_FTP) == 0)
141 return (fetchStatFTP(URL, us, flags));
142 else if (strcmp(URL->scheme, SCHEME_HTTP) == 0)
143 return (fetchStatHTTP(URL, us, flags));
144 else if (strcmp(URL->scheme, SCHEME_HTTPS) == 0)
145 return (fetchStatHTTP(URL, us, flags));
146 url_seterr(URL_BAD_SCHEME);
147 return (-1);
148 }
149
150 /*
151 * Select the appropriate protocol for the URL scheme, and return a
152 * list of files in the directory pointed to by the URL.
153 */
154 struct url_ent *
fetchList(struct url * URL,const char * flags)155 fetchList(struct url *URL, const char *flags)
156 {
157
158 if (strcmp(URL->scheme, SCHEME_FILE) == 0)
159 return (fetchListFile(URL, flags));
160 else if (strcmp(URL->scheme, SCHEME_FTP) == 0)
161 return (fetchListFTP(URL, flags));
162 else if (strcmp(URL->scheme, SCHEME_HTTP) == 0)
163 return (fetchListHTTP(URL, flags));
164 else if (strcmp(URL->scheme, SCHEME_HTTPS) == 0)
165 return (fetchListHTTP(URL, flags));
166 url_seterr(URL_BAD_SCHEME);
167 return (NULL);
168 }
169
170 /*
171 * Attempt to parse the given URL; if successful, call fetchXGet().
172 */
173 FILE *
fetchXGetURL(const char * URL,struct url_stat * us,const char * flags)174 fetchXGetURL(const char *URL, struct url_stat *us, const char *flags)
175 {
176 struct url *u;
177 FILE *f;
178
179 if ((u = fetchParseURL(URL)) == NULL)
180 return (NULL);
181
182 f = fetchXGet(u, us, flags);
183
184 fetchFreeURL(u);
185 return (f);
186 }
187
188 /*
189 * Attempt to parse the given URL; if successful, call fetchGet().
190 */
191 FILE *
fetchGetURL(const char * URL,const char * flags)192 fetchGetURL(const char *URL, const char *flags)
193 {
194 return (fetchXGetURL(URL, NULL, flags));
195 }
196
197 /*
198 * Attempt to parse the given URL; if successful, call fetchPut().
199 */
200 FILE *
fetchPutURL(const char * URL,const char * flags)201 fetchPutURL(const char *URL, const char *flags)
202 {
203 struct url *u;
204 FILE *f;
205
206 if ((u = fetchParseURL(URL)) == NULL)
207 return (NULL);
208
209 f = fetchPut(u, flags);
210
211 fetchFreeURL(u);
212 return (f);
213 }
214
215 /*
216 * Attempt to parse the given URL; if successful, call fetchStat().
217 */
218 int
fetchStatURL(const char * URL,struct url_stat * us,const char * flags)219 fetchStatURL(const char *URL, struct url_stat *us, const char *flags)
220 {
221 struct url *u;
222 int s;
223
224 if ((u = fetchParseURL(URL)) == NULL)
225 return (-1);
226
227 s = fetchStat(u, us, flags);
228
229 fetchFreeURL(u);
230 return (s);
231 }
232
233 /*
234 * Attempt to parse the given URL; if successful, call fetchList().
235 */
236 struct url_ent *
fetchListURL(const char * URL,const char * flags)237 fetchListURL(const char *URL, const char *flags)
238 {
239 struct url *u;
240 struct url_ent *ue;
241
242 if ((u = fetchParseURL(URL)) == NULL)
243 return (NULL);
244
245 ue = fetchList(u, flags);
246
247 fetchFreeURL(u);
248 return (ue);
249 }
250
251 /*
252 * Make a URL
253 */
254 struct url *
fetchMakeURL(const char * scheme,const char * host,int port,const char * doc,const char * user,const char * pwd)255 fetchMakeURL(const char *scheme, const char *host, int port, const char *doc,
256 const char *user, const char *pwd)
257 {
258 struct url *u;
259
260 if (!scheme || (!host && !doc)) {
261 url_seterr(URL_MALFORMED);
262 return (NULL);
263 }
264
265 if (port < 0 || port > 65535) {
266 url_seterr(URL_BAD_PORT);
267 return (NULL);
268 }
269
270 /* allocate struct url */
271 if ((u = calloc(1, sizeof(*u))) == NULL) {
272 fetch_syserr();
273 return (NULL);
274 }
275 u->netrcfd = -1;
276
277 if ((u->doc = strdup(doc ? doc : "/")) == NULL) {
278 fetch_syserr();
279 free(u);
280 return (NULL);
281 }
282
283 #define seturl(x) snprintf(u->x, sizeof(u->x), "%s", x)
284 seturl(scheme);
285 seturl(host);
286 seturl(user);
287 seturl(pwd);
288 #undef seturl
289 u->port = port;
290
291 return (u);
292 }
293
294 /*
295 * Return value of the given hex digit.
296 */
297 static int
fetch_hexval(char ch)298 fetch_hexval(char ch)
299 {
300
301 if (ch >= '0' && ch <= '9')
302 return (ch - '0');
303 else if (ch >= 'a' && ch <= 'f')
304 return (ch - 'a' + 10);
305 else if (ch >= 'A' && ch <= 'F')
306 return (ch - 'A' + 10);
307 return (-1);
308 }
309
310 /*
311 * Decode percent-encoded URL component from src into dst, stopping at end
312 * of string, or at @ or : separators. Returns a pointer to the unhandled
313 * part of the input string (null terminator, @, or :). No terminator is
314 * written to dst (it is the caller's responsibility).
315 */
316 static const char *
fetch_pctdecode(char * dst,const char * src,size_t dlen)317 fetch_pctdecode(char *dst, const char *src, size_t dlen)
318 {
319 int d1, d2;
320 char c;
321 const char *s;
322
323 for (s = src; *s != '\0' && *s != '@' && *s != ':'; s++) {
324 if (s[0] == '%' && (d1 = fetch_hexval(s[1])) >= 0 &&
325 (d2 = fetch_hexval(s[2])) >= 0 && (d1 > 0 || d2 > 0)) {
326 c = d1 << 4 | d2;
327 s += 2;
328 } else if (s[0] == '%') {
329 /* Invalid escape sequence. */
330 return (NULL);
331 } else {
332 c = *s;
333 }
334 if (dlen-- > 0)
335 *dst++ = c;
336 else
337 return (NULL);
338 }
339 return (s);
340 }
341
342 /*
343 * Split an URL into components. URL syntax is:
344 * [method:/][/[user[:pwd]@]host[:port]/][document]
345 * This almost, but not quite, RFC1738 URL syntax.
346 */
347 struct url *
fetchParseURL(const char * URL)348 fetchParseURL(const char *URL)
349 {
350 char *doc;
351 const char *p, *q;
352 struct url *u;
353 int i, n;
354
355 /* allocate struct url */
356 if ((u = calloc(1, sizeof(*u))) == NULL) {
357 fetch_syserr();
358 return (NULL);
359 }
360 u->netrcfd = -1;
361
362 /* scheme name */
363 if ((p = strstr(URL, ":/"))) {
364 if (p - URL > URL_SCHEMELEN)
365 goto ouch;
366 for (i = 0; URL + i < p; i++)
367 u->scheme[i] = tolower((unsigned char)URL[i]);
368 URL = ++p;
369 /*
370 * Only one slash: no host, leave slash as part of document
371 * Two slashes: host follows, strip slashes
372 */
373 if (URL[1] == '/')
374 URL = (p += 2);
375 } else {
376 p = URL;
377 }
378 if (!*URL || *URL == '/' || *URL == '.' ||
379 (u->scheme[0] == '\0' &&
380 strchr(URL, '/') == NULL && strchr(URL, ':') == NULL))
381 goto nohost;
382
383 p = strpbrk(URL, "/@");
384 if (p && *p == '@') {
385 /* username */
386 q = fetch_pctdecode(u->user, URL, URL_USERLEN);
387 if (q == NULL)
388 goto ouch;
389
390 /* password */
391 if (*q == ':') {
392 q = fetch_pctdecode(u->pwd, q + 1, URL_PWDLEN);
393 if (q == NULL)
394 goto ouch;
395 }
396 p++;
397 } else {
398 p = URL;
399 }
400
401 /* hostname */
402 if (*p == '[') {
403 q = p + 1 + strspn(p + 1, ":0123456789ABCDEFabcdef");
404 if (*q++ != ']')
405 goto ouch;
406 } else {
407 /* valid characters in a DNS name */
408 q = p + strspn(p, "-." "0123456789"
409 "ABCDEFGHIJKLMNOPQRSTUVWXYZ" "_"
410 "abcdefghijklmnopqrstuvwxyz");
411 }
412 if ((*q != '\0' && *q != '/' && *q != ':') || q - p > MAXHOSTNAMELEN)
413 goto ouch;
414 for (i = 0; p + i < q; i++)
415 u->host[i] = tolower((unsigned char)p[i]);
416 u->host[i] = '\0';
417 p = q;
418
419 /* port */
420 if (*p == ':') {
421 for (n = 0, q = ++p; *q && (*q != '/'); q++) {
422 if (*q >= '0' && *q <= '9' && n < INT_MAX / 10) {
423 n = n * 10 + (*q - '0');
424 } else {
425 /* invalid port */
426 url_seterr(URL_BAD_PORT);
427 goto ouch;
428 }
429 }
430 if (p != q && (n < 1 || n > IPPORT_MAX))
431 goto ouch;
432 u->port = n;
433 p = q;
434 }
435
436 nohost:
437 /* document */
438 if (!*p)
439 p = "/";
440
441 if (strcmp(u->scheme, SCHEME_HTTP) == 0 ||
442 strcmp(u->scheme, SCHEME_HTTPS) == 0) {
443 const char hexnums[] = "0123456789abcdef";
444
445 /* percent-escape whitespace. */
446 if ((doc = malloc(strlen(p) * 3 + 1)) == NULL) {
447 fetch_syserr();
448 goto ouch;
449 }
450 u->doc = doc;
451 /* fragments are reserved for client-side processing, see
452 * https://www.rfc-editor.org/rfc/rfc9110.html#section-7.1
453 */
454 while (*p != '\0' && *p != '#') {
455 if (!isspace((unsigned char)*p)) {
456 *doc++ = *p++;
457 } else {
458 *doc++ = '%';
459 *doc++ = hexnums[((unsigned int)*p) >> 4];
460 *doc++ = hexnums[((unsigned int)*p) & 0xf];
461 p++;
462 }
463 }
464 *doc = '\0';
465 } else if ((u->doc = strdup(p)) == NULL) {
466 fetch_syserr();
467 goto ouch;
468 }
469
470 DEBUGF("scheme: \"%s\"\n"
471 "user: \"%s\"\n"
472 "password: \"%s\"\n"
473 "host: \"%s\"\n"
474 "port: \"%d\"\n"
475 "document: \"%s\"\n",
476 u->scheme, u->user, u->pwd,
477 u->host, u->port, u->doc);
478
479 return (u);
480
481 ouch:
482 free(u);
483 return (NULL);
484 }
485
486 /*
487 * Free a URL
488 */
489 void
fetchFreeURL(struct url * u)490 fetchFreeURL(struct url *u)
491 {
492 free(u->doc);
493 free(u);
494 }
495