xref: /freebsd/usr.bin/whois/whois.c (revision 8ddb146abcdf061be9f2c0db7e391697dafad85c)
1 /*-
2  * SPDX-License-Identifier: BSD-3-Clause
3  *
4  * Copyright (c) 1980, 1993
5  *	The Regents of the University of California.  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  * 2. Redistributions in binary form must reproduce the above copyright
13  *    notice, this list of conditions and the following disclaimer in the
14  *    documentation and/or other materials provided with the distribution.
15  * 3. Neither the name of the University nor the names of its contributors
16  *    may be used to endorse or promote products derived from this software
17  *    without specific prior written permission.
18  *
19  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29  * SUCH DAMAGE.
30  */
31 
32 #ifndef lint
33 static const char copyright[] =
34 "@(#) Copyright (c) 1980, 1993\n\
35 	The Regents of the University of California.  All rights reserved.\n";
36 #endif /* not lint */
37 
38 #if 0
39 #ifndef lint
40 static char sccsid[] = "@(#)whois.c	8.1 (Berkeley) 6/6/93";
41 #endif /* not lint */
42 #endif
43 
44 #include <sys/cdefs.h>
45 __FBSDID("$FreeBSD$");
46 
47 #include <sys/types.h>
48 #include <sys/socket.h>
49 #include <sys/poll.h>
50 #include <netinet/in.h>
51 #include <arpa/inet.h>
52 #include <ctype.h>
53 #include <err.h>
54 #include <netdb.h>
55 #include <stdarg.h>
56 #include <stdio.h>
57 #include <stdlib.h>
58 #include <string.h>
59 #include <sysexits.h>
60 #include <unistd.h>
61 #include <fcntl.h>
62 #include <errno.h>
63 
64 #define	ABUSEHOST	"whois.abuse.net"
65 #define	ANICHOST	"whois.arin.net"
66 #define	DENICHOST	"whois.denic.de"
67 #define	DKNICHOST	"whois.dk-hostmaster.dk"
68 #define	FNICHOST	"whois.afrinic.net"
69 #define	GNICHOST	"whois.nic.gov"
70 #define	IANAHOST	"whois.iana.org"
71 #define	INICHOST	"whois.internic.net"
72 #define	KNICHOST	"whois.krnic.net"
73 #define	LNICHOST	"whois.lacnic.net"
74 #define	MNICHOST	"whois.ra.net"
75 #define	PDBHOST		"whois.peeringdb.com"
76 #define	PNICHOST	"whois.apnic.net"
77 #define	QNICHOST_TAIL	".whois-servers.net"
78 #define	RNICHOST	"whois.ripe.net"
79 #define	VNICHOST	"whois.verisign-grs.com"
80 
81 #define	DEFAULT_PORT	"whois"
82 
83 #define WHOIS_RECURSE	0x01
84 #define WHOIS_QUICK	0x02
85 #define WHOIS_SPAM_ME	0x04
86 
87 #define CHOPSPAM	">>> Last update of WHOIS database:"
88 
89 #define ishost(h) (isalnum((unsigned char)h) || h == '.' || h == '-')
90 
91 #define SCAN(p, end, check)					\
92 	while ((p) < (end))					\
93 		if (check) ++(p);				\
94 		else break
95 
96 static struct {
97 	const char *suffix, *server;
98 } whoiswhere[] = {
99 	/* Various handles */
100 	{ "-ARIN", ANICHOST },
101 	{ "-NICAT", "at" QNICHOST_TAIL },
102 	{ "-NORID", "no" QNICHOST_TAIL },
103 	{ "-RIPE", RNICHOST },
104 	/* Nominet's whois server doesn't return referrals to JANET */
105 	{ ".ac.uk", "ac.uk" QNICHOST_TAIL },
106 	{ ".gov.uk", "ac.uk" QNICHOST_TAIL },
107 	{ "", IANAHOST }, /* default */
108 	{ NULL, NULL } /* safety belt */
109 };
110 
111 #define WHOIS_REFERRAL(s) { s, sizeof(s) - 1 }
112 static struct {
113 	const char *prefix;
114 	size_t len;
115 } whois_referral[] = {
116 	WHOIS_REFERRAL("whois:"), /* IANA */
117 	WHOIS_REFERRAL("Whois Server:"),
118 	WHOIS_REFERRAL("Registrar WHOIS Server:"), /* corporatedomains.com */
119 	WHOIS_REFERRAL("ReferralServer:  whois://"), /* ARIN */
120 	WHOIS_REFERRAL("ReferralServer:  rwhois://"), /* ARIN */
121 	WHOIS_REFERRAL("descr:          region. Please query"), /* AfriNIC */
122 	{ NULL, 0 }
123 };
124 
125 /*
126  * We have a list of patterns for RIRs that assert ignorance rather than
127  * providing referrals. If that happens, we guess that ARIN will be more
128  * helpful. But, before following a referral to an RIR, we check if we have
129  * asked that RIR already, and if so we make another guess.
130  */
131 static const char *actually_arin[] = {
132 	"netname:        ERX-NETBLOCK\n", /* APNIC */
133 	"netname:        NON-RIPE-NCC-MANAGED-ADDRESS-BLOCK\n",
134 	NULL
135 };
136 
137 static struct {
138 	int loop;
139 	const char *host;
140 } try_rir[] = {
141 	{ 0, ANICHOST },
142 	{ 0, RNICHOST },
143 	{ 0, PNICHOST },
144 	{ 0, FNICHOST },
145 	{ 0, LNICHOST },
146 	{ 0, NULL }
147 };
148 
149 static void
150 reset_rir(void) {
151 	int i;
152 
153 	for (i = 0; try_rir[i].host != NULL; i++)
154 		try_rir[i].loop = 0;
155 }
156 
157 static const char *port = DEFAULT_PORT;
158 
159 static const char *choose_server(char *);
160 static struct addrinfo *gethostinfo(const char *, const char *, int);
161 static void s_asprintf(char **ret, const char *format, ...) __printflike(2, 3);
162 static void usage(void);
163 static void whois(const char *, const char *, const char *, int);
164 
165 int
166 main(int argc, char *argv[])
167 {
168 	const char *country, *host;
169 	int ch, flags;
170 
171 #ifdef	SOCKS
172 	SOCKSinit(argv[0]);
173 #endif
174 
175 	country = host = NULL;
176 	flags = 0;
177 	while ((ch = getopt(argc, argv, "aAbc:fgh:iIklmp:PQrRS")) != -1) {
178 		switch (ch) {
179 		case 'a':
180 			host = ANICHOST;
181 			break;
182 		case 'A':
183 			host = PNICHOST;
184 			break;
185 		case 'b':
186 			host = ABUSEHOST;
187 			break;
188 		case 'c':
189 			country = optarg;
190 			break;
191 		case 'f':
192 			host = FNICHOST;
193 			break;
194 		case 'g':
195 			host = GNICHOST;
196 			break;
197 		case 'h':
198 			host = optarg;
199 			break;
200 		case 'i':
201 			host = INICHOST;
202 			break;
203 		case 'I':
204 			host = IANAHOST;
205 			break;
206 		case 'k':
207 			host = KNICHOST;
208 			break;
209 		case 'l':
210 			host = LNICHOST;
211 			break;
212 		case 'm':
213 			host = MNICHOST;
214 			break;
215 		case 'p':
216 			port = optarg;
217 			break;
218 		case 'P':
219 			host = PDBHOST;
220 			break;
221 		case 'Q':
222 			flags |= WHOIS_QUICK;
223 			break;
224 		case 'r':
225 			host = RNICHOST;
226 			break;
227 		case 'R':
228 			flags |= WHOIS_RECURSE;
229 			break;
230 		case 'S':
231 			flags |= WHOIS_SPAM_ME;
232 			break;
233 		case '?':
234 		default:
235 			usage();
236 			/* NOTREACHED */
237 		}
238 	}
239 	argc -= optind;
240 	argv += optind;
241 
242 	if (!argc || (country != NULL && host != NULL))
243 		usage();
244 
245 	/*
246 	 * If no host or country is specified, rely on referrals from IANA.
247 	 */
248 	if (host == NULL && country == NULL) {
249 		if ((host = getenv("WHOIS_SERVER")) == NULL &&
250 		    (host = getenv("RA_SERVER")) == NULL) {
251 			if (!(flags & WHOIS_QUICK))
252 				flags |= WHOIS_RECURSE;
253 		}
254 	}
255 	while (argc-- > 0) {
256 		if (country != NULL) {
257 			char *qnichost;
258 			s_asprintf(&qnichost, "%s%s", country, QNICHOST_TAIL);
259 			whois(*argv, qnichost, port, flags);
260 			free(qnichost);
261 		} else
262 			whois(*argv, host != NULL ? host :
263 			      choose_server(*argv), port, flags);
264 		reset_rir();
265 		argv++;
266 	}
267 	exit(0);
268 }
269 
270 static const char *
271 choose_server(char *domain)
272 {
273 	size_t len = strlen(domain);
274 	int i;
275 
276 	for (i = 0; whoiswhere[i].suffix != NULL; i++) {
277 		size_t suffix_len = strlen(whoiswhere[i].suffix);
278 		if (len > suffix_len &&
279 		    strcasecmp(domain + len - suffix_len,
280 			       whoiswhere[i].suffix) == 0)
281 			return (whoiswhere[i].server);
282 	}
283 	errx(EX_SOFTWARE, "no default whois server");
284 }
285 
286 static struct addrinfo *
287 gethostinfo(const char *host, const char *hport, int exit_on_noname)
288 {
289 	struct addrinfo hints, *res;
290 	int error;
291 
292 	memset(&hints, 0, sizeof(hints));
293 	hints.ai_flags = AI_CANONNAME;
294 	hints.ai_family = AF_UNSPEC;
295 	hints.ai_socktype = SOCK_STREAM;
296 	res = NULL;
297 	error = getaddrinfo(host, hport, &hints, &res);
298 	if (error && (exit_on_noname || error != EAI_NONAME))
299 		err(EX_NOHOST, "%s: %s", host, gai_strerror(error));
300 	return (res);
301 }
302 
303 /*
304  * Wrapper for asprintf(3) that exits on error.
305  */
306 static void
307 s_asprintf(char **ret, const char *format, ...)
308 {
309 	va_list ap;
310 
311 	va_start(ap, format);
312 	if (vasprintf(ret, format, ap) == -1) {
313 		va_end(ap);
314 		err(EX_OSERR, "vasprintf()");
315 	}
316 	va_end(ap);
317 }
318 
319 static int
320 connect_to_any_host(struct addrinfo *hostres)
321 {
322 	struct addrinfo *res;
323 	nfds_t i, j;
324 	size_t count;
325 	struct pollfd *fds;
326 	int timeout = 180, s = -1;
327 
328 	for (res = hostres, count = 0; res; res = res->ai_next)
329 		count++;
330 	fds = calloc(count, sizeof(*fds));
331 	if (fds == NULL)
332 		err(EX_OSERR, "calloc()");
333 
334 	/*
335 	 * Traverse the result list elements and make non-block
336 	 * connection attempts.
337 	 */
338 	count = i = 0;
339 	for (res = hostres; res != NULL; res = res->ai_next) {
340 		s = socket(res->ai_family, res->ai_socktype | SOCK_NONBLOCK,
341 		    res->ai_protocol);
342 		if (s < 0)
343 			continue;
344 		if (connect(s, res->ai_addr, res->ai_addrlen) < 0) {
345 			if (errno == EINPROGRESS) {
346 				/* Add the socket to poll list */
347 				fds[i].fd = s;
348 				fds[i].events = POLLERR | POLLHUP |
349 						POLLIN | POLLOUT;
350 				/*
351 				 * From here until a socket connects, the
352 				 * socket fd is owned by the fds[] poll array.
353 				 */
354 				s = -1;
355 				count++;
356 				i++;
357 			} else {
358 				close(s);
359 				s = -1;
360 
361 				/*
362 				 * Poll only if we have something to poll,
363 				 * otherwise just go ahead and try next
364 				 * address
365 				 */
366 				if (count == 0)
367 					continue;
368 			}
369 		} else
370 			goto done;
371 
372 		/*
373 		 * If we are at the last address, poll until a connection is
374 		 * established or we failed all connection attempts.
375 		 */
376 		if (res->ai_next == NULL)
377 			timeout = INFTIM;
378 
379 		/*
380 		 * Poll the watched descriptors for successful connections:
381 		 * if we still have more untried resolved addresses, poll only
382 		 * once; otherwise, poll until all descriptors have errors,
383 		 * which will be considered as ETIMEDOUT later.
384 		 */
385 		do {
386 			int n;
387 
388 			n = poll(fds, i, timeout);
389 			if (n == 0) {
390 				/*
391 				 * No event reported in time.  Try with a
392 				 * smaller timeout (but cap at 2-3ms)
393 				 * after a new host have been added.
394 				 */
395 				if (timeout >= 3)
396 					timeout >>= 1;
397 
398 				break;
399 			} else if (n < 0) {
400 				/*
401 				 * errno here can only be EINTR which we would
402 				 * want to clean up and bail out.
403 				 */
404 				s = -1;
405 				goto done;
406 			}
407 
408 			/*
409 			 * Check for the event(s) we have seen.
410 			 */
411 			for (j = 0; j < i; j++) {
412 				if (fds[j].fd == -1 || fds[j].events == 0 ||
413 				    fds[j].revents == 0)
414 					continue;
415 				if (fds[j].revents & ~(POLLIN | POLLOUT)) {
416 					close(fds[j].fd);
417 					fds[j].fd = -1;
418 					fds[j].events = 0;
419 					count--;
420 					continue;
421 				} else if (fds[j].revents & (POLLIN | POLLOUT)) {
422 					/* Connect succeeded. */
423 					s = fds[j].fd;
424 					fds[j].fd = -1;
425 
426 					goto done;
427 				}
428 
429 			}
430 		} while (timeout == INFTIM && count != 0);
431 	}
432 
433 	/* All attempts were failed */
434 	s = -1;
435 	if (count == 0)
436 		errno = ETIMEDOUT;
437 
438 done:
439 	/* Close all watched fds except the succeeded one */
440 	for (j = 0; j < i; j++)
441 		if (fds[j].fd != -1)
442 			close(fds[j].fd);
443 	free(fds);
444 	return (s);
445 }
446 
447 static void
448 whois(const char *query, const char *hostname, const char *hostport, int flags)
449 {
450 	FILE *fp;
451 	struct addrinfo *hostres;
452 	char *buf, *host, *nhost, *nport, *p;
453 	int comment, s, f;
454 	size_t len, i;
455 
456 	hostres = gethostinfo(hostname, hostport, 1);
457 	s = connect_to_any_host(hostres);
458 	if (s == -1)
459 		err(EX_OSERR, "connect()");
460 
461 	/* Restore default blocking behavior.  */
462 	if ((f = fcntl(s, F_GETFL)) == -1)
463 		err(EX_OSERR, "fcntl()");
464 	f &= ~O_NONBLOCK;
465 	if (fcntl(s, F_SETFL, f) == -1)
466 		err(EX_OSERR, "fcntl()");
467 
468 	fp = fdopen(s, "r+");
469 	if (fp == NULL)
470 		err(EX_OSERR, "fdopen()");
471 
472 	if (!(flags & WHOIS_SPAM_ME) &&
473 	    (strcasecmp(hostname, DENICHOST) == 0 ||
474 	     strcasecmp(hostname, "de" QNICHOST_TAIL) == 0)) {
475 		const char *q;
476 		int idn = 0;
477 		for (q = query; *q != '\0'; q++)
478 			if (!isascii(*q))
479 				idn = 1;
480 		fprintf(fp, "-T dn%s %s\r\n", idn ? "" : ",ace", query);
481 	} else if (!(flags & WHOIS_SPAM_ME) &&
482 		   (strcasecmp(hostname, DKNICHOST) == 0 ||
483 		    strcasecmp(hostname, "dk" QNICHOST_TAIL) == 0))
484 		fprintf(fp, "--show-handles %s\r\n", query);
485 	else if ((flags & WHOIS_SPAM_ME) ||
486 		 strchr(query, ' ') != NULL)
487 		fprintf(fp, "%s\r\n", query);
488 	else if (strcasecmp(hostname, ANICHOST) == 0) {
489 		if (strncasecmp(query, "AS", 2) == 0 &&
490 		    strspn(query+2, "0123456789") == strlen(query+2))
491 			fprintf(fp, "+ a %s\r\n", query+2);
492 		else
493 			fprintf(fp, "+ %s\r\n", query);
494 	} else if (strcasecmp(hostres->ai_canonname, VNICHOST) == 0)
495 		fprintf(fp, "domain %s\r\n", query);
496 	else
497 		fprintf(fp, "%s\r\n", query);
498 	fflush(fp);
499 
500 	comment = 0;
501 	if (!(flags & WHOIS_SPAM_ME) &&
502 	    (strcasecmp(hostname, ANICHOST) == 0 ||
503 	     strcasecmp(hostname, RNICHOST) == 0)) {
504 		comment = 2;
505 	}
506 
507 	nhost = NULL;
508 	while ((buf = fgetln(fp, &len)) != NULL) {
509 		/* Nominet */
510 		if (!(flags & WHOIS_SPAM_ME) &&
511 		    len == 5 && strncmp(buf, "-- \r\n", 5) == 0)
512 			break;
513 		/* RIRs */
514 		if (comment == 1 && buf[0] == '#')
515 			break;
516 		else if (comment == 2) {
517 			if (strchr("#%\r\n", buf[0]) != NULL)
518 				continue;
519 			else
520 				comment = 1;
521 		}
522 
523 		printf("%.*s", (int)len, buf);
524 
525 		if ((flags & WHOIS_RECURSE) && nhost == NULL) {
526 			for (i = 0; whois_referral[i].prefix != NULL; i++) {
527 				p = buf;
528 				SCAN(p, buf+len, *p == ' ');
529 				if (strncasecmp(p, whois_referral[i].prefix,
530 					           whois_referral[i].len) != 0)
531 					continue;
532 				p += whois_referral[i].len;
533 				SCAN(p, buf+len, *p == ' ');
534 				host = p;
535 				SCAN(p, buf+len, ishost(*p));
536 				if (p > host) {
537 					char *pstr;
538 
539 					s_asprintf(&nhost, "%.*s",
540 						   (int)(p - host), host);
541 
542 					if (*p != ':') {
543 						s_asprintf(&nport, "%s", port);
544 						break;
545 					}
546 
547 					pstr = ++p;
548 					SCAN(p, buf+len, isdigit(*p));
549 					if (p > pstr && (p - pstr) < 6) {
550 						s_asprintf(&nport, "%.*s",
551 						    (int)(p - pstr), pstr);
552 						break;
553 					}
554 
555 					/* Invalid port; don't recurse */
556 					free(nhost);
557 					nhost = NULL;
558 				}
559 				break;
560 			}
561 			for (i = 0; actually_arin[i] != NULL; i++) {
562 				if (strncmp(buf, actually_arin[i], len) == 0) {
563 					s_asprintf(&nhost, "%s", ANICHOST);
564 					s_asprintf(&nport, "%s", port);
565 					break;
566 				}
567 			}
568 		}
569 		/* Verisign etc. */
570 		if (!(flags & WHOIS_SPAM_ME) &&
571 		    len >= sizeof(CHOPSPAM)-1 &&
572 		    (strncasecmp(buf, CHOPSPAM, sizeof(CHOPSPAM)-1) == 0 ||
573 		     strncasecmp(buf, CHOPSPAM+4, sizeof(CHOPSPAM)-5) == 0)) {
574 			printf("\n");
575 			break;
576 		}
577 	}
578 	fclose(fp);
579 	freeaddrinfo(hostres);
580 
581 	f = 0;
582 	for (i = 0; try_rir[i].host != NULL; i++) {
583 		/* Remember visits to RIRs */
584 		if (try_rir[i].loop == 0 &&
585 		    strcasecmp(try_rir[i].host, hostname) == 0)
586 			try_rir[i].loop = 1;
587 		/* Do we need to find an alternative RIR? */
588 		if (try_rir[i].loop != 0 && nhost != NULL &&
589 		    strcasecmp(try_rir[i].host, nhost) == 0) {
590 			free(nhost);
591 			nhost = NULL;
592 			free(nport);
593 			nport = NULL;
594 			f = 1;
595 		}
596 	}
597 	if (f) {
598 		/* Find a replacement RIR */
599 		for (i = 0; try_rir[i].host != NULL; i++) {
600 			if (try_rir[i].loop == 0) {
601 				s_asprintf(&nhost, "%s", try_rir[i].host);
602 				s_asprintf(&nport, "%s", port);
603 				break;
604 			}
605 		}
606 	}
607 	if (nhost != NULL) {
608 		/* Ignore self-referrals */
609 		if (strcasecmp(hostname, nhost) != 0) {
610 			printf("# %s\n\n", nhost);
611 			whois(query, nhost, nport, flags);
612 		}
613 		free(nhost);
614 		free(nport);
615 	}
616 }
617 
618 static void
619 usage(void)
620 {
621 	fprintf(stderr,
622 	    "usage: whois [-aAbfgiIklmPQrRS] [-c country-code | -h hostname] "
623 	    "[-p port] name ...\n");
624 	exit(EX_USAGE);
625 }
626