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 */
59 char *fname_out = NULL; /* where to direct output (-O) */
60 int do_continue = 0; /* continue a prev transfer (-c) */
61 long beg_range = 0L; /* range at which continue begins */
62 int got_clen = 0; /* got content-length: from server */
63 FILE *output; /* socket to web server */
68 while ((n = getopt(argc, argv, "cO:")) != EOF) {
74 /* can't set fname_out to NULL if outputting to stdout, because
75 * this gets interpreted as the auto-gen output filename
76 * case below - tausq@debian.org
78 fname_out = (strcmp(optarg, "-") == 0 ? (char *)1 : optarg);
85 if (argc - optind != 1)
88 /* Guess an output filename */
91 #ifdef BB_FEATURE_STATUSBAR
94 get_last_path_component(argv[optind]);
95 #ifdef BB_FEATURE_STATUSBAR
102 if (do_continue && !fname_out)
103 fatalError("cannot specify continue (-c) without a filename (-O)\n");
105 * Parse url into components.
107 parse_url(argv[optind], &uri_host, &uri_port, &uri_path);
110 * Open socket to server.
112 sfp = open_socket(uri_host, uri_port);
115 * Open the output stream.
117 if (fname_out != (char *)1) {
118 if ( (output=fopen(fname_out, (do_continue ? "a" : "w")))
120 fatalPerror("fopen(%s)", fname_out);
126 * Determine where to start transfer.
130 if (fstat(fileno(output), &sbuf) < 0)
131 fatalError("fstat()");
132 if (sbuf.st_size > 0)
133 beg_range = sbuf.st_size;
141 fprintf(sfp, "GET %s HTTP/1.1\r\nHost: %s\r\n", uri_path, uri_host);
143 fprintf(sfp, "Range: bytes=%ld-\r\n", beg_range);
144 fputs("Connection: close\r\n\r\n", sfp);
147 * Retrieve HTTP response line and check for "200" status code.
149 if (fgets(buf, sizeof(buf), sfp) == NULL)
150 fatalError("no response from server\n");
151 for (s = buf ; *s != '\0' && !isspace(*s) ; ++s)
153 for ( ; isspace(*s) ; ++s)
159 fatalError("server does not support ranges\n");
165 fatalError("server returned error: %s", buf);
169 * Retrieve HTTP headers.
171 while ((s = gethdr(buf, sizeof(buf), sfp, &n)) != NULL) {
172 if (strcmp(buf, "content-length") == 0) {
177 if (strcmp(buf, "transfer-encoding") == 0) {
178 fatalError("server wants to do %s transfer encoding\n", s);
184 * Retrieve HTTP body.
186 #ifdef BB_FEATURE_STATUSBAR
190 while (filesize > 0 && (n = fread(buf, 1, sizeof(buf), sfp)) > 0) {
191 fwrite(buf, 1, n, output);
192 #ifdef BB_FEATURE_STATUSBAR
199 if (n == 0 && ferror(sfp))
200 fatalPerror("network read error");
206 void parse_url(char *url, char **uri_host, int *uri_port, char **uri_path)
209 static char *defaultpath = "/";
213 if (strncmp(url, "http://", 7) != 0)
214 fatalError("not an http url: %s\n", url);
216 /* pull the host portion to the front of the buffer */
217 for (s = url, h = url+7 ; *h != '/' && *h != 0; ++h) {
219 *uri_port = atoi(h+1);
226 if (*h == 0) h = defaultpath;
233 FILE *open_socket(char *host, int port)
235 struct sockaddr_in sin;
240 memzero(&sin, sizeof(sin));
241 sin.sin_family = AF_INET;
242 if ((hp = (struct hostent *) gethostbyname(host)) == NULL)
243 fatalError("cannot resolve %s\n", host);
244 memcpy(&sin.sin_addr, hp->h_addr_list[0], hp->h_length);
245 sin.sin_port = htons(port);
248 * Get the server onto a stdio stream.
250 if ((fd = socket(AF_INET, SOCK_STREAM, 0)) < 0)
251 fatalPerror("socket()");
252 if (connect(fd, (struct sockaddr *) &sin, sizeof(sin)) < 0)
253 fatalPerror("connect(%s)", host);
254 if ((fp = fdopen(fd, "r+")) == NULL)
255 fatalPerror("fdopen()");
261 char *gethdr(char *buf, size_t bufsiz, FILE *fp, int *istrunc)
268 /* retrieve header line */
269 if (fgets(buf, bufsiz, fp) == NULL)
272 /* see if we are at the end of the headers */
273 for (s = buf ; *s == '\r' ; ++s)
278 /* convert the header name to lower case */
279 for (s = buf ; isalnum(*s) || *s == '-' ; ++s)
282 /* verify we are at the end of the header name */
284 fatalError("bad header line: %s\n", buf);
286 /* locate the start of the header value */
287 for (*s++ = '\0' ; *s == ' ' || *s == '\t' ; ++s)
291 /* locate the end of header */
292 while (*s != '\0' && *s != '\r' && *s != '\n')
295 /* end of header found */
301 /* Rats! The buffer isn't big enough to hold the entire header value. */
302 while (c = getc(fp), c != EOF && c != '\n')
308 #ifdef BB_FEATURE_STATUSBAR
309 /* Stuff below is from BSD rcp util.c, as added to openshh.
310 * Original copyright notice is retained at the end of this file.
318 struct winsize winsize;
320 if (ioctl(fileno(stdout), TIOCGWINSZ, &winsize) != -1)
321 return (winsize.ws_col ? winsize.ws_col : 80);
327 updateprogressmeter(int ignore)
329 int save_errno = errno;
338 struct itimerval itv;
340 itv.it_value.tv_sec = wait;
341 itv.it_value.tv_usec = 0;
342 itv.it_interval = itv.it_value;
343 setitimer(ITIMER_REAL, &itv, NULL);
348 progressmeter(int flag)
350 static const char prefixes[] = " KMGTP";
351 static struct timeval lastupdate;
352 static off_t lastsize;
353 struct timeval now, td, wait;
354 off_t cursize, abbrevsize;
356 int ratio, barlength, i, remaining;
360 (void) gettimeofday(&start, (struct timezone *) 0);
365 (void) gettimeofday(&now, (struct timezone *) 0);
368 ratio = 100.0 * cursize / filesize;
369 ratio = MAX(ratio, 0);
370 ratio = MIN(ratio, 100);
374 snprintf(buf, sizeof(buf), "\r%-20.20s %3d%% ", curfile, ratio);
376 barlength = getttywidth() - 51;
378 i = barlength * ratio / 100;
379 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
381 "*****************************************************************************"
382 "*****************************************************************************",
386 abbrevsize = cursize;
387 while (abbrevsize >= 100000 && i < sizeof(prefixes)) {
391 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), " %5d %c%c ",
392 (int) abbrevsize, prefixes[i], prefixes[i] == ' ' ? ' ' :
395 timersub(&now, &lastupdate, &wait);
396 if (cursize > lastsize) {
399 if (wait.tv_sec >= STALLTIME) {
400 start.tv_sec += wait.tv_sec;
401 start.tv_usec += wait.tv_usec;
405 timersub(&now, &start, &td);
406 elapsed = td.tv_sec + (td.tv_usec / 1000000.0);
408 if (statbytes <= 0 || elapsed <= 0.0 || cursize > filesize) {
409 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
411 } else if (wait.tv_sec >= STALLTIME) {
412 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
415 remaining = (int) (filesize / (statbytes / elapsed) - elapsed);
416 i = remaining / 3600;
418 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
421 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
423 i = remaining % 3600;
424 snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf),
425 "%02d:%02d ETA", i / 60, i % 60);
427 write(fileno(stdout), buf, strlen(buf));
431 sa.sa_handler = updateprogressmeter;
432 sigemptyset(&sa.sa_mask);
433 sa.sa_flags = SA_RESTART;
434 sigaction(SIGALRM, &sa, NULL);
436 } else if (flag == 1) {
443 /* Original copyright notice which applies to the BB_FEATURE_STATUSBAR stuff,
444 * much of which was blatently stolen from openssh. */
447 * Copyright (c) 1992, 1993
448 * The Regents of the University of California. All rights reserved.
450 * Redistribution and use in source and binary forms, with or without
451 * modification, are permitted provided that the following conditions
453 * 1. Redistributions of source code must retain the above copyright
454 * notice, this list of conditions and the following disclaimer.
455 * 2. Redistributions in binary form must reproduce the above copyright
456 * notice, this list of conditions and the following disclaimer in the
457 * documentation and/or other materials provided with the distribution.
459 * 3. <BSD Advertising Clause omitted per the July 22, 1999 licensing change
460 * ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change>
462 * 4. Neither the name of the University nor the names of its contributors
463 * may be used to endorse or promote products derived from this software
464 * without specific prior written permission.
466 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
467 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
468 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
469 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
470 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
471 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
472 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
473 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
474 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
475 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
478 * $Id: wget.c,v 1.8 2000/12/07 03:53:47 tausq Exp $
485 c-file-style: "linux"