1 /* vi: set sw=4 ts=4: */
3 * wget - retrieve a file using HTTP
5 * Chip Rosenthal Covad Communications <chip@laserlink.net>
7 * Note: According to RFC2616 section 3.6.1, "All HTTP/1.1 applications MUST be
8 * able to receive and decode the "chunked" transfer-coding, and MUST ignore
9 * chunk-extension extensions they do not understand."
11 * This prevents this particular wget app from completely RFC compliant, and as
12 * such, prevents it from being used as a general purpose web browser... This
13 * is a design decision, since it makes the code smaller.
17 #warning This applet has moved to netkit-tiny. After BusyBox 0.49, this
18 #warning applet will be removed from BusyBox. All maintenance efforts
19 #warning should be done in the netkit-tiny source tree.
29 #include <sys/ioctl.h>
32 #include <sys/types.h>
34 #include <sys/socket.h>
35 #include <netinet/in.h>
36 #include <arpa/inet.h>
40 void parse_url(char *url, char **uri_host, int *uri_port, char **uri_path);
41 FILE *open_socket(char *host, int port);
42 char *gethdr(char *buf, size_t bufsiz, FILE *fp, int *istrunc);
43 void progressmeter(int flag);
45 /* Globals (can be accessed from signal handlers */
46 static off_t filesize = 0; /* content-length of the file */
47 #ifdef BB_FEATURE_WGET_STATUSBAR
48 static char *curfile; /* Name of current file being transferred. */
49 static struct timeval start; /* Time a transfer started. */
50 volatile unsigned long statbytes; /* Number of bytes transferred so far. */
51 /* For progressmeter() -- number of seconds before xfer considered "stalled" */
52 static const int STALLTIME = 5;
55 int wget_main(int argc, char **argv)
58 char *proxy, *proxy_host;
59 int uri_port, proxy_port;
63 FILE *sfp; /* socket to web server */
64 char *uri_host, *uri_path; /* parsed from command line url */
65 char *fname_out = NULL; /* where to direct output (-O) */
66 int do_continue = 0; /* continue a prev transfer (-c) */
67 long beg_range = 0L; /* range at which continue begins */
68 int got_clen = 0; /* got content-length: from server */
69 FILE *output; /* socket to web server */
70 int quiet_flag = FALSE; /* Be verry, verry quiet... */
75 while ((n = getopt(argc, argv, "cqO:")) != EOF) {
84 /* can't set fname_out to NULL if outputting to stdout, because
85 * this gets interpreted as the auto-gen output filename
86 * case below - tausq@debian.org
88 fname_out = (strcmp(optarg, "-") == 0 ? (char *)1 : optarg);
95 if (argc - optind != 1)
98 if (do_continue && !fname_out)
99 error_msg_and_die("cannot specify continue (-c) without a filename (-O)\n");
102 * Use the proxy if necessary.
104 if ((proxy = getenv("http_proxy")) != NULL) {
105 proxy = xstrdup(proxy);
106 parse_url(proxy, &proxy_host, &proxy_port, &uri_path);
107 parse_url(argv[optind], &uri_host, &uri_port, &uri_path);
110 * Parse url into components.
112 parse_url(argv[optind], &uri_host, &uri_port, &uri_path);
117 /* Guess an output filename */
120 #ifdef BB_FEATURE_WGET_STATUSBAR
123 get_last_path_component(uri_path);
124 if (fname_out==NULL || strlen(fname_out)<1) {
126 #ifdef BB_FEATURE_WGET_STATUSBAR
131 #ifdef BB_FEATURE_WGET_STATUSBAR
133 curfile=argv[optind];
139 * Open socket to server.
141 sfp = open_socket(proxy_host, proxy_port);
143 /* Make the assumption that if the file already exists
144 * on disk that the intention is to continue downloading
145 * a previously aborted download -Erik */
146 if (stat(fname_out, &sbuf) == 0) {
151 * Open the output file stream.
153 if (fname_out != (char *)1) {
154 if ( (output=fopen(fname_out, (do_continue ? "a" : "w")))
156 perror_msg_and_die("fopen(%s)", fname_out);
162 * Determine where to start transfer.
165 if (fstat(fileno(output), &sbuf) < 0)
166 perror_msg_and_die("fstat()");
167 if (sbuf.st_size > 0)
168 beg_range = sbuf.st_size;
176 fprintf(sfp, "GET http://%s:%d/%s HTTP/1.1\r\n",
177 uri_host, uri_port, uri_path);
178 fprintf(sfp, "Host: %s\r\nUser-Agent: Wget\r\n", uri_host);
180 fprintf(sfp, "Range: bytes=%ld-\r\n", beg_range);
181 fprintf(sfp,"Connection: close\r\n\r\n");
184 * Retrieve HTTP response line and check for "200" status code.
186 if (fgets(buf, sizeof(buf), sfp) == NULL) {
187 error_msg_and_die("no response from server\n");
189 for (s = buf ; *s != '\0' && !isspace(*s) ; ++s)
191 for ( ; isspace(*s) ; ++s)
197 error_msg_and_die("server does not support ranges\n");
203 error_msg_and_die("server returned error: %s", buf);
207 * Retrieve HTTP headers.
209 while ((s = gethdr(buf, sizeof(buf), sfp, &n)) != NULL) {
210 if (strcmp(buf, "content-length") == 0) {
215 if (strcmp(buf, "transfer-encoding") == 0) {
216 error_msg_and_die("server wants to do %s transfer encoding\n", s);
222 * Retrieve HTTP body.
224 #ifdef BB_FEATURE_WGET_STATUSBAR
226 if (quiet_flag==FALSE)
229 while (filesize > 0 && (n = fread(buf, 1, sizeof(buf), sfp)) > 0) {
230 fwrite(buf, 1, n, output);
231 #ifdef BB_FEATURE_WGET_STATUSBAR
233 if (quiet_flag==FALSE)
239 if (n == 0 && ferror(sfp))
240 perror_msg_and_die("network read error");
246 void parse_url(char *url, char **uri_host, int *uri_port, char **uri_path)
252 if (strncmp(url, "http://", 7) != 0)
253 error_msg_and_die("not an http url: %s\n", url);
257 cp = strchr(*uri_host, ':');
258 sp = strchr(*uri_host, '/');
260 if (cp != NULL && (sp == NULL || cp < sp)) {
262 *uri_port = atoi(cp);
273 FILE *open_socket(char *host, int port)
275 struct sockaddr_in sin;
280 memzero(&sin, sizeof(sin));
281 sin.sin_family = AF_INET;
282 if ((hp = (struct hostent *) gethostbyname(host)) == NULL)
283 error_msg_and_die("cannot resolve %s\n", host);
284 memcpy(&sin.sin_addr, hp->h_addr_list[0], hp->h_length);
285 sin.sin_port = htons(port);
288 * Get the server onto a stdio stream.
290 if ((fd = socket(AF_INET, SOCK_STREAM, 0)) < 0)
291 perror_msg_and_die("socket()");
292 if (connect(fd, (struct sockaddr *) &sin, sizeof(sin)) < 0)
293 perror_msg_and_die("connect(%s)", host);
294 if ((fp = fdopen(fd, "r+")) == NULL)
295 perror_msg_and_die("fdopen()");
301 char *gethdr(char *buf, size_t bufsiz, FILE *fp, int *istrunc)
308 /* retrieve header line */
309 if (fgets(buf, bufsiz, fp) == NULL)
312 /* see if we are at the end of the headers */
313 for (s = buf ; *s == '\r' ; ++s)
318 /* convert the header name to lower case */
319 for (s = buf ; isalnum(*s) || *s == '-' ; ++s)
322 /* verify we are at the end of the header name */
324 error_msg_and_die("bad header line: %s\n", buf);
326 /* locate the start of the header value */
327 for (*s++ = '\0' ; *s == ' ' || *s == '\t' ; ++s)
331 /* locate the end of header */
332 while (*s != '\0' && *s != '\r' && *s != '\n')
335 /* end of header found */
341 /* Rats! The buffer isn't big enough to hold the entire header value. */
342 while (c = getc(fp), c != EOF && c != '\n')
348 #ifdef BB_FEATURE_WGET_STATUSBAR
349 /* Stuff below is from BSD rcp util.c, as added to openshh.
350 * Original copyright notice is retained at the end of this file.
358 struct winsize winsize;
360 if (ioctl(fileno(stdout), TIOCGWINSZ, &winsize) != -1)
361 return (winsize.ws_col ? winsize.ws_col : 80);
367 updateprogressmeter(int ignore)
369 int save_errno = errno;
378 struct itimerval itv;
380 itv.it_value.tv_sec = wait;
381 itv.it_value.tv_usec = 0;
382 itv.it_interval = itv.it_value;
383 setitimer(ITIMER_REAL, &itv, NULL);
388 progressmeter(int flag)
390 static const char prefixes[] = " KMGTP";
391 static struct timeval lastupdate;
392 static off_t lastsize;
393 struct timeval now, td, wait;
394 off_t cursize, abbrevsize;
396 int ratio, barlength, i, remaining;
400 (void) gettimeofday(&start, (struct timezone *) 0);
405 (void) gettimeofday(&now, (struct timezone *) 0);
408 ratio = 100.0 * cursize / filesize;
409 ratio = MAX(ratio, 0);
410 ratio = MIN(ratio, 100);
414 snprintf(buf, sizeof(buf), "\r%-20.20s %3d%% ", curfile, ratio);
416 barlength = getttywidth() - 51;
418 i = barlength * ratio / 100;
419 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
421 "*****************************************************************************"
422 "*****************************************************************************",
426 abbrevsize = cursize;
427 while (abbrevsize >= 100000 && i < sizeof(prefixes)) {
431 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), " %5d %c%c ",
432 (int) abbrevsize, prefixes[i], prefixes[i] == ' ' ? ' ' :
435 timersub(&now, &lastupdate, &wait);
436 if (cursize > lastsize) {
439 if (wait.tv_sec >= STALLTIME) {
440 start.tv_sec += wait.tv_sec;
441 start.tv_usec += wait.tv_usec;
445 timersub(&now, &start, &td);
446 elapsed = td.tv_sec + (td.tv_usec / 1000000.0);
448 if (statbytes <= 0 || elapsed <= 0.0 || cursize > filesize) {
449 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
451 } else if (wait.tv_sec >= STALLTIME) {
452 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
455 remaining = (int) (filesize / (statbytes / elapsed) - elapsed);
456 i = remaining / 3600;
458 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
461 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
463 i = remaining % 3600;
464 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
465 "%02d:%02d ETA", i / 60, i % 60);
467 write(fileno(stderr), buf, strlen(buf));
471 sa.sa_handler = updateprogressmeter;
472 sigemptyset(&sa.sa_mask);
473 sa.sa_flags = SA_RESTART;
474 sigaction(SIGALRM, &sa, NULL);
476 } else if (flag == 1) {
483 /* Original copyright notice which applies to the BB_FEATURE_WGET_STATUSBAR stuff,
484 * much of which was blatently stolen from openssh. */
487 * Copyright (c) 1992, 1993
488 * The Regents of the University of California. All rights reserved.
490 * Redistribution and use in source and binary forms, with or without
491 * modification, are permitted provided that the following conditions
493 * 1. Redistributions of source code must retain the above copyright
494 * notice, this list of conditions and the following disclaimer.
495 * 2. Redistributions in binary form must reproduce the above copyright
496 * notice, this list of conditions and the following disclaimer in the
497 * documentation and/or other materials provided with the distribution.
499 * 3. <BSD Advertising Clause omitted per the July 22, 1999 licensing change
500 * ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change>
502 * 4. Neither the name of the University nor the names of its contributors
503 * may be used to endorse or promote products derived from this software
504 * without specific prior written permission.
506 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
507 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
508 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
509 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
510 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
511 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
512 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
513 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
514 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
515 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
518 * $Id: wget.c,v 1.19 2001/01/24 17:37:07 andersen Exp $
525 c-file-style: "linux"