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.
25 #include <sys/ioctl.h>
28 #include <sys/types.h>
30 #include <sys/socket.h>
31 #include <netinet/in.h>
32 #include <arpa/inet.h>
36 void parse_url(char *url, char **uri_host, int *uri_port, char **uri_path);
37 FILE *open_socket(char *host, int port);
38 char *gethdr(char *buf, size_t bufsiz, FILE *fp, int *istrunc);
39 void progressmeter(int flag);
41 /* Globals (can be accessed from signal handlers */
42 static off_t filesize = 0; /* content-length of the file */
43 #ifdef BB_FEATURE_STATUSBAR
44 static char *curfile; /* Name of current file being transferred. */
45 static struct timeval start; /* Time a transfer started. */
46 volatile unsigned long statbytes; /* Number of bytes transferred so far. */
47 /* For progressmeter() -- number of seconds before xfer considered "stalled" */
51 int wget_main(int argc, char **argv)
53 FILE *sfp; /* socket to web server */
54 char *uri_host, *uri_path; /* parsed from command line url */
60 char *fname_out = NULL; /* where to direct output (-O) */
61 int do_continue = 0; /* continue a prev transfer (-c) */
62 long beg_range = 0L; /* range at which continue begins */
63 int got_clen = 0; /* got content-length: from server */
64 FILE *output; /* socket to web server */
69 while ((n = getopt(argc, argv, "cO:")) != EOF) {
75 /* can't set fname_out to NULL if outputting to stdout, because
76 * this gets interpreted as the auto-gen output filename
77 * case below - tausq@debian.org
79 fname_out = (strcmp(optarg, "-") == 0 ? (char *)1 : optarg);
86 if (argc - optind != 1)
89 /* Guess an output filename */
92 #ifdef BB_FEATURE_STATUSBAR
95 get_last_path_component(argv[optind]);
96 #ifdef BB_FEATURE_STATUSBAR
103 if (do_continue && !fname_out)
104 error_msg_and_die("cannot specify continue (-c) without a filename (-O)\n");
107 * Use the proxy if necessary.
109 if ((proxy = getenv("http_proxy")) != NULL) {
110 proxy = xstrdup(proxy);
111 parse_url(proxy, &uri_host, &uri_port, &uri_path);
112 uri_path = argv[optind];
115 * Parse url into components.
117 parse_url(argv[optind], &uri_host, &uri_port, &uri_path);
121 * Open socket to server.
123 sfp = open_socket(uri_host, uri_port);
126 * Open the output stream.
128 if (fname_out != (char *)1) {
129 if ( (output=fopen(fname_out, (do_continue ? "a" : "w")))
131 perror_msg_and_die("fopen(%s)", fname_out);
137 * Determine where to start transfer.
141 if (fstat(fileno(output), &sbuf) < 0)
142 error_msg_and_die("fstat()");
143 if (sbuf.st_size > 0)
144 beg_range = sbuf.st_size;
152 fprintf(sfp, "GET %s HTTP/1.1\r\nHost: %s\r\n", uri_path, uri_host);
154 fprintf(sfp, "Range: bytes=%ld-\r\n", beg_range);
155 fputs("Connection: close\r\n\r\n", sfp);
158 * Retrieve HTTP response line and check for "200" status code.
160 if (fgets(buf, sizeof(buf), sfp) == NULL)
161 error_msg_and_die("no response from server\n");
162 for (s = buf ; *s != '\0' && !isspace(*s) ; ++s)
164 for ( ; isspace(*s) ; ++s)
170 error_msg_and_die("server does not support ranges\n");
176 error_msg_and_die("server returned error: %s", buf);
180 * Retrieve HTTP headers.
182 while ((s = gethdr(buf, sizeof(buf), sfp, &n)) != NULL) {
183 if (strcmp(buf, "content-length") == 0) {
188 if (strcmp(buf, "transfer-encoding") == 0) {
189 error_msg_and_die("server wants to do %s transfer encoding\n", s);
195 * Retrieve HTTP body.
197 #ifdef BB_FEATURE_STATUSBAR
201 while (filesize > 0 && (n = fread(buf, 1, sizeof(buf), sfp)) > 0) {
202 fwrite(buf, 1, n, output);
203 #ifdef BB_FEATURE_STATUSBAR
210 if (n == 0 && ferror(sfp))
211 perror_msg_and_die("network read error");
217 void parse_url(char *url, char **uri_host, int *uri_port, char **uri_path)
220 static char *defaultpath = "/";
224 if (strncmp(url, "http://", 7) != 0)
225 error_msg_and_die("not an http url: %s\n", url);
227 /* pull the host portion to the front of the buffer */
228 for (s = url, h = url+7 ; *h != '/' && *h != 0; ++h) {
230 *uri_port = atoi(h+1);
237 if (*h == 0) h = defaultpath;
244 FILE *open_socket(char *host, int port)
246 struct sockaddr_in sin;
251 memzero(&sin, sizeof(sin));
252 sin.sin_family = AF_INET;
253 if ((hp = (struct hostent *) gethostbyname(host)) == NULL)
254 error_msg_and_die("cannot resolve %s\n", host);
255 memcpy(&sin.sin_addr, hp->h_addr_list[0], hp->h_length);
256 sin.sin_port = htons(port);
259 * Get the server onto a stdio stream.
261 if ((fd = socket(AF_INET, SOCK_STREAM, 0)) < 0)
262 perror_msg_and_die("socket()");
263 if (connect(fd, (struct sockaddr *) &sin, sizeof(sin)) < 0)
264 perror_msg_and_die("connect(%s)", host);
265 if ((fp = fdopen(fd, "r+")) == NULL)
266 perror_msg_and_die("fdopen()");
272 char *gethdr(char *buf, size_t bufsiz, FILE *fp, int *istrunc)
279 /* retrieve header line */
280 if (fgets(buf, bufsiz, fp) == NULL)
283 /* see if we are at the end of the headers */
284 for (s = buf ; *s == '\r' ; ++s)
289 /* convert the header name to lower case */
290 for (s = buf ; isalnum(*s) || *s == '-' ; ++s)
293 /* verify we are at the end of the header name */
295 error_msg_and_die("bad header line: %s\n", buf);
297 /* locate the start of the header value */
298 for (*s++ = '\0' ; *s == ' ' || *s == '\t' ; ++s)
302 /* locate the end of header */
303 while (*s != '\0' && *s != '\r' && *s != '\n')
306 /* end of header found */
312 /* Rats! The buffer isn't big enough to hold the entire header value. */
313 while (c = getc(fp), c != EOF && c != '\n')
319 #ifdef BB_FEATURE_STATUSBAR
320 /* Stuff below is from BSD rcp util.c, as added to openshh.
321 * Original copyright notice is retained at the end of this file.
329 struct winsize winsize;
331 if (ioctl(fileno(stdout), TIOCGWINSZ, &winsize) != -1)
332 return (winsize.ws_col ? winsize.ws_col : 80);
338 updateprogressmeter(int ignore)
340 int save_errno = errno;
349 struct itimerval itv;
351 itv.it_value.tv_sec = wait;
352 itv.it_value.tv_usec = 0;
353 itv.it_interval = itv.it_value;
354 setitimer(ITIMER_REAL, &itv, NULL);
359 progressmeter(int flag)
361 static const char prefixes[] = " KMGTP";
362 static struct timeval lastupdate;
363 static off_t lastsize;
364 struct timeval now, td, wait;
365 off_t cursize, abbrevsize;
367 int ratio, barlength, i, remaining;
371 (void) gettimeofday(&start, (struct timezone *) 0);
376 (void) gettimeofday(&now, (struct timezone *) 0);
379 ratio = 100.0 * cursize / filesize;
380 ratio = MAX(ratio, 0);
381 ratio = MIN(ratio, 100);
385 snprintf(buf, sizeof(buf), "\r%-20.20s %3d%% ", curfile, ratio);
387 barlength = getttywidth() - 51;
389 i = barlength * ratio / 100;
390 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
392 "*****************************************************************************"
393 "*****************************************************************************",
397 abbrevsize = cursize;
398 while (abbrevsize >= 100000 && i < sizeof(prefixes)) {
402 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), " %5d %c%c ",
403 (int) abbrevsize, prefixes[i], prefixes[i] == ' ' ? ' ' :
406 timersub(&now, &lastupdate, &wait);
407 if (cursize > lastsize) {
410 if (wait.tv_sec >= STALLTIME) {
411 start.tv_sec += wait.tv_sec;
412 start.tv_usec += wait.tv_usec;
416 timersub(&now, &start, &td);
417 elapsed = td.tv_sec + (td.tv_usec / 1000000.0);
419 if (statbytes <= 0 || elapsed <= 0.0 || cursize > filesize) {
420 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
422 } else if (wait.tv_sec >= STALLTIME) {
423 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
426 remaining = (int) (filesize / (statbytes / elapsed) - elapsed);
427 i = remaining / 3600;
429 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
432 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
434 i = remaining % 3600;
435 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
436 "%02d:%02d ETA", i / 60, i % 60);
438 write(fileno(stderr), buf, strlen(buf));
442 sa.sa_handler = updateprogressmeter;
443 sigemptyset(&sa.sa_mask);
444 sa.sa_flags = SA_RESTART;
445 sigaction(SIGALRM, &sa, NULL);
447 } else if (flag == 1) {
454 /* Original copyright notice which applies to the BB_FEATURE_STATUSBAR stuff,
455 * much of which was blatently stolen from openssh. */
458 * Copyright (c) 1992, 1993
459 * The Regents of the University of California. All rights reserved.
461 * Redistribution and use in source and binary forms, with or without
462 * modification, are permitted provided that the following conditions
464 * 1. Redistributions of source code must retain the above copyright
465 * notice, this list of conditions and the following disclaimer.
466 * 2. Redistributions in binary form must reproduce the above copyright
467 * notice, this list of conditions and the following disclaimer in the
468 * documentation and/or other materials provided with the distribution.
470 * 3. <BSD Advertising Clause omitted per the July 22, 1999 licensing change
471 * ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change>
473 * 4. Neither the name of the University nor the names of its contributors
474 * may be used to endorse or promote products derived from this software
475 * without specific prior written permission.
477 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
478 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
479 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
480 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
481 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
482 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
483 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
484 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
485 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
486 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
489 * $Id: wget.c,v 1.11 2000/12/07 22:42:11 andersen Exp $
496 c-file-style: "linux"