[821] | 1 | /* vi: set sw=4 ts=4: */
|
---|
| 2 | /*
|
---|
| 3 | * wget - retrieve a file using HTTP or FTP
|
---|
| 4 | *
|
---|
| 5 | * Chip Rosenthal Covad Communications <chip@laserlink.net>
|
---|
| 6 | *
|
---|
| 7 | */
|
---|
| 8 |
|
---|
[1765] | 9 | /* We want libc to give us xxx64 functions also */
|
---|
| 10 | /* http://www.unix.org/version2/whatsnew/lfs20mar.html */
|
---|
| 11 | //#define _LARGEFILE64_SOURCE 1
|
---|
[821] | 12 |
|
---|
[1765] | 13 | #include <getopt.h> /* for struct option */
|
---|
| 14 | #include "libbb.h"
|
---|
[821] | 15 |
|
---|
| 16 | struct host_info {
|
---|
[1765] | 17 | // May be used if we ever will want to free() all xstrdup()s...
|
---|
| 18 | /* char *allocated; */
|
---|
[821] | 19 | char *host;
|
---|
| 20 | int port;
|
---|
| 21 | char *path;
|
---|
| 22 | int is_ftp;
|
---|
| 23 | char *user;
|
---|
| 24 | };
|
---|
| 25 |
|
---|
| 26 | static void parse_url(char *url, struct host_info *h);
|
---|
[1765] | 27 | static FILE *open_socket(len_and_sockaddr *lsa);
|
---|
[821] | 28 | static char *gethdr(char *buf, size_t bufsiz, FILE *fp, int *istrunc);
|
---|
[1765] | 29 | static int ftpcmd(const char *s1, const char *s2, FILE *fp, char *buf);
|
---|
[821] | 30 |
|
---|
| 31 | /* Globals (can be accessed from signal handlers */
|
---|
[1765] | 32 | static off_t content_len; /* Content-length of the file */
|
---|
| 33 | static off_t beg_range; /* Range at which continue begins */
|
---|
| 34 | #if ENABLE_FEATURE_WGET_STATUSBAR
|
---|
| 35 | static off_t transferred; /* Number of bytes transferred so far */
|
---|
| 36 | #endif
|
---|
| 37 | static bool chunked; /* chunked transfer encoding */
|
---|
| 38 | #if ENABLE_FEATURE_WGET_STATUSBAR
|
---|
[821] | 39 | static void progressmeter(int flag);
|
---|
[1765] | 40 | static const char *curfile; /* Name of current file being transferred */
|
---|
[821] | 41 | enum {
|
---|
[1765] | 42 | STALLTIME = 5 /* Seconds when xfer considered "stalled" */
|
---|
[821] | 43 | };
|
---|
| 44 | #else
|
---|
[1765] | 45 | static ALWAYS_INLINE void progressmeter(int flag) {}
|
---|
[821] | 46 | #endif
|
---|
| 47 |
|
---|
[1765] | 48 | /* Read NMEMB bytes into PTR from STREAM. Returns the number of bytes read,
|
---|
| 49 | * and a short count if an eof or non-interrupt error is encountered. */
|
---|
| 50 | static size_t safe_fread(void *ptr, size_t nmemb, FILE *stream)
|
---|
[821] | 51 | {
|
---|
[1765] | 52 | size_t ret;
|
---|
| 53 | char *p = (char*)ptr;
|
---|
[821] | 54 |
|
---|
| 55 | do {
|
---|
| 56 | clearerr(stream);
|
---|
[1765] | 57 | ret = fread(p, 1, nmemb, stream);
|
---|
| 58 | p += ret;
|
---|
| 59 | nmemb -= ret;
|
---|
| 60 | } while (nmemb && ferror(stream) && errno == EINTR);
|
---|
[821] | 61 |
|
---|
[1765] | 62 | return p - (char*)ptr;
|
---|
[821] | 63 | }
|
---|
| 64 |
|
---|
[1765] | 65 | /* Read a line or SIZE-1 bytes into S, whichever is less, from STREAM.
|
---|
[821] | 66 | * Returns S, or NULL if an eof or non-interrupt error is encountered. */
|
---|
| 67 | static char *safe_fgets(char *s, int size, FILE *stream)
|
---|
| 68 | {
|
---|
| 69 | char *ret;
|
---|
| 70 |
|
---|
| 71 | do {
|
---|
| 72 | clearerr(stream);
|
---|
| 73 | ret = fgets(s, size, stream);
|
---|
| 74 | } while (ret == NULL && ferror(stream) && errno == EINTR);
|
---|
| 75 |
|
---|
| 76 | return ret;
|
---|
| 77 | }
|
---|
| 78 |
|
---|
[1765] | 79 | #if ENABLE_FEATURE_WGET_AUTHENTICATION
|
---|
| 80 | /* Base64-encode character string. buf is assumed to be char buf[512]. */
|
---|
| 81 | static char *base64enc_512(char buf[512], const char *str)
|
---|
| 82 | {
|
---|
| 83 | unsigned len = strlen(str);
|
---|
| 84 | if (len > 512/4*3 - 10) /* paranoia */
|
---|
| 85 | len = 512/4*3 - 10;
|
---|
| 86 | bb_uuencode(buf, str, len, bb_uuenc_tbl_base64);
|
---|
[821] | 87 | return buf;
|
---|
| 88 | }
|
---|
| 89 | #endif
|
---|
| 90 |
|
---|
[1765] | 91 | int wget_main(int argc, char **argv);
|
---|
[821] | 92 | int wget_main(int argc, char **argv)
|
---|
| 93 | {
|
---|
[1765] | 94 | char buf[512];
|
---|
| 95 | struct host_info server, target;
|
---|
| 96 | len_and_sockaddr *lsa;
|
---|
| 97 | int n, status;
|
---|
[821] | 98 | int port;
|
---|
[1765] | 99 | int try = 5;
|
---|
| 100 | unsigned opt;
|
---|
| 101 | char *str;
|
---|
[821] | 102 | char *proxy = 0;
|
---|
[1765] | 103 | char *dir_prefix = NULL;
|
---|
| 104 | #if ENABLE_FEATURE_WGET_LONG_OPTIONS
|
---|
| 105 | char *extra_headers = NULL;
|
---|
[821] | 106 | llist_t *headers_llist = NULL;
|
---|
[1765] | 107 | #endif
|
---|
[821] | 108 |
|
---|
[1765] | 109 | FILE *sfp = NULL; /* socket to web/ftp server */
|
---|
| 110 | FILE *dfp = NULL; /* socket to ftp server (data) */
|
---|
| 111 | char *fname_out = NULL; /* where to direct output (-O) */
|
---|
| 112 | bool got_clen = 0; /* got content-length: from server */
|
---|
| 113 | int output_fd = -1;
|
---|
| 114 | bool use_proxy = 1; /* Use proxies if env vars are set */
|
---|
| 115 | const char *proxy_flag = "on"; /* Use proxies if env vars are set */
|
---|
| 116 | const char *user_agent = "Wget";/* "User-Agent" header field */
|
---|
| 117 | static const char keywords[] ALIGN1 =
|
---|
| 118 | "content-length\0""transfer-encoding\0""chunked\0""location\0";
|
---|
| 119 | enum {
|
---|
| 120 | KEY_content_length = 1, KEY_transfer_encoding, KEY_chunked, KEY_location
|
---|
| 121 | };
|
---|
| 122 | enum {
|
---|
| 123 | WGET_OPT_CONTINUE = 0x1,
|
---|
| 124 | WGET_OPT_SPIDER = 0x2,
|
---|
| 125 | WGET_OPT_QUIET = 0x4,
|
---|
| 126 | WGET_OPT_OUTNAME = 0x8,
|
---|
| 127 | WGET_OPT_PREFIX = 0x10,
|
---|
| 128 | WGET_OPT_PROXY = 0x20,
|
---|
| 129 | WGET_OPT_USER_AGENT = 0x40,
|
---|
| 130 | WGET_OPT_PASSIVE = 0x80,
|
---|
| 131 | WGET_OPT_HEADER = 0x100,
|
---|
| 132 | };
|
---|
| 133 | #if ENABLE_FEATURE_WGET_LONG_OPTIONS
|
---|
| 134 | static const char wget_longopts[] ALIGN1 =
|
---|
| 135 | /* name, has_arg, val */
|
---|
| 136 | "continue\0" No_argument "c"
|
---|
| 137 | "spider\0" No_argument "s"
|
---|
| 138 | "quiet\0" No_argument "q"
|
---|
| 139 | "output-document\0" Required_argument "O"
|
---|
| 140 | "directory-prefix\0" Required_argument "P"
|
---|
| 141 | "proxy\0" Required_argument "Y"
|
---|
| 142 | "user-agent\0" Required_argument "U"
|
---|
| 143 | "passive-ftp\0" No_argument "\xff"
|
---|
| 144 | "header\0" Required_argument "\xfe"
|
---|
| 145 | ;
|
---|
| 146 | applet_long_options = wget_longopts;
|
---|
[821] | 147 | #endif
|
---|
[1765] | 148 | /* server.allocated = target.allocated = NULL; */
|
---|
| 149 | opt_complementary = "-1" USE_FEATURE_WGET_LONG_OPTIONS(":\xfe::");
|
---|
| 150 | opt = getopt32(argv, "csqO:P:Y:U:",
|
---|
| 151 | &fname_out, &dir_prefix,
|
---|
| 152 | &proxy_flag, &user_agent
|
---|
| 153 | USE_FEATURE_WGET_LONG_OPTIONS(, &headers_llist)
|
---|
| 154 | );
|
---|
[821] | 155 | if (strcmp(proxy_flag, "off") == 0) {
|
---|
[1765] | 156 | /* Use the proxy if necessary */
|
---|
[821] | 157 | use_proxy = 0;
|
---|
| 158 | }
|
---|
[1765] | 159 | #if ENABLE_FEATURE_WGET_LONG_OPTIONS
|
---|
| 160 | if (headers_llist) {
|
---|
| 161 | int size = 1;
|
---|
| 162 | char *cp;
|
---|
| 163 | llist_t *ll = headers_llist;
|
---|
| 164 | while (ll) {
|
---|
| 165 | size += strlen(ll->data) + 2;
|
---|
| 166 | ll = ll->link;
|
---|
| 167 | }
|
---|
| 168 | extra_headers = cp = xmalloc(size);
|
---|
[821] | 169 | while (headers_llist) {
|
---|
[1765] | 170 | cp += sprintf(cp, "%s\r\n", headers_llist->data);
|
---|
[821] | 171 | headers_llist = headers_llist->link;
|
---|
| 172 | }
|
---|
| 173 | }
|
---|
[1765] | 174 | #endif
|
---|
[821] | 175 |
|
---|
| 176 | parse_url(argv[optind], &target);
|
---|
| 177 | server.host = target.host;
|
---|
| 178 | server.port = target.port;
|
---|
| 179 |
|
---|
[1765] | 180 | /* Use the proxy if necessary */
|
---|
[821] | 181 | if (use_proxy) {
|
---|
| 182 | proxy = getenv(target.is_ftp ? "ftp_proxy" : "http_proxy");
|
---|
| 183 | if (proxy && *proxy) {
|
---|
[1765] | 184 | parse_url(proxy, &server);
|
---|
[821] | 185 | } else {
|
---|
| 186 | use_proxy = 0;
|
---|
| 187 | }
|
---|
| 188 | }
|
---|
| 189 |
|
---|
| 190 | /* Guess an output filename */
|
---|
| 191 | if (!fname_out) {
|
---|
| 192 | // Dirty hack. Needed because bb_get_last_path_component
|
---|
| 193 | // will destroy trailing / by storing '\0' in last byte!
|
---|
[1765] | 194 | if (!last_char_is(target.path, '/')) {
|
---|
| 195 | fname_out = bb_get_last_path_component(target.path);
|
---|
| 196 | #if ENABLE_FEATURE_WGET_STATUSBAR
|
---|
| 197 | curfile = fname_out;
|
---|
[821] | 198 | #endif
|
---|
| 199 | }
|
---|
[1765] | 200 | if (!fname_out || !fname_out[0]) {
|
---|
| 201 | /* bb_get_last_path_component writes
|
---|
| 202 | * to last '/' only. We don't have one here... */
|
---|
| 203 | fname_out = (char*)"index.html";
|
---|
| 204 | #if ENABLE_FEATURE_WGET_STATUSBAR
|
---|
| 205 | curfile = fname_out;
|
---|
[821] | 206 | #endif
|
---|
| 207 | }
|
---|
| 208 | if (dir_prefix != NULL)
|
---|
| 209 | fname_out = concat_path_file(dir_prefix, fname_out);
|
---|
[1765] | 210 | #if ENABLE_FEATURE_WGET_STATUSBAR
|
---|
[821] | 211 | } else {
|
---|
| 212 | curfile = bb_get_last_path_component(fname_out);
|
---|
| 213 | #endif
|
---|
| 214 | }
|
---|
[1765] | 215 | /* Impossible?
|
---|
| 216 | if ((opt & WGET_OPT_CONTINUE) && !fname_out)
|
---|
| 217 | bb_error_msg_and_die("cannot specify continue (-c) without a filename (-O)"); */
|
---|
[821] | 218 |
|
---|
[1765] | 219 | /* Determine where to start transfer */
|
---|
| 220 | if (LONE_DASH(fname_out)) {
|
---|
| 221 | output_fd = 1;
|
---|
| 222 | opt &= ~WGET_OPT_CONTINUE;
|
---|
[821] | 223 | }
|
---|
[1765] | 224 | if (opt & WGET_OPT_CONTINUE) {
|
---|
| 225 | output_fd = open(fname_out, O_WRONLY);
|
---|
| 226 | if (output_fd >= 0) {
|
---|
| 227 | beg_range = xlseek(output_fd, 0, SEEK_END);
|
---|
| 228 | }
|
---|
| 229 | /* File doesn't exist. We do not create file here yet.
|
---|
| 230 | We are not sure it exists on remove side */
|
---|
[821] | 231 | }
|
---|
| 232 |
|
---|
| 233 | /* We want to do exactly _one_ DNS lookup, since some
|
---|
| 234 | * sites (i.e. ftp.us.debian.org) use round-robin DNS
|
---|
| 235 | * and we want to connect to only one IP... */
|
---|
[1765] | 236 | lsa = xhost2sockaddr(server.host, server.port);
|
---|
| 237 | if (!(opt & WGET_OPT_QUIET)) {
|
---|
| 238 | fprintf(stderr, "Connecting to %s (%s)\n", server.host,
|
---|
| 239 | xmalloc_sockaddr2dotted(&lsa->sa));
|
---|
| 240 | /* We leak result of xmalloc_sockaddr2dotted */
|
---|
[821] | 241 | }
|
---|
| 242 |
|
---|
| 243 | if (use_proxy || !target.is_ftp) {
|
---|
| 244 | /*
|
---|
| 245 | * HTTP session
|
---|
| 246 | */
|
---|
| 247 | do {
|
---|
| 248 | got_clen = chunked = 0;
|
---|
| 249 |
|
---|
[1765] | 250 | if (!--try)
|
---|
| 251 | bb_error_msg_and_die("too many redirections");
|
---|
[821] | 252 |
|
---|
[1765] | 253 | /* Open socket to http server */
|
---|
[821] | 254 | if (sfp) fclose(sfp);
|
---|
[1765] | 255 | sfp = open_socket(lsa);
|
---|
[821] | 256 |
|
---|
[1765] | 257 | /* Send HTTP request. */
|
---|
[821] | 258 | if (use_proxy) {
|
---|
[1765] | 259 | fprintf(sfp, "GET %stp://%s/%s HTTP/1.1\r\n",
|
---|
[821] | 260 | target.is_ftp ? "f" : "ht", target.host,
|
---|
[1765] | 261 | target.path);
|
---|
[821] | 262 | } else {
|
---|
| 263 | fprintf(sfp, "GET /%s HTTP/1.1\r\n", target.path);
|
---|
| 264 | }
|
---|
| 265 |
|
---|
[1765] | 266 | fprintf(sfp, "Host: %s\r\nUser-Agent: %s\r\n",
|
---|
| 267 | target.host, user_agent);
|
---|
[821] | 268 |
|
---|
[1765] | 269 | #if ENABLE_FEATURE_WGET_AUTHENTICATION
|
---|
[821] | 270 | if (target.user) {
|
---|
[1765] | 271 | fprintf(sfp, "Proxy-Authorization: Basic %s\r\n"+6,
|
---|
| 272 | base64enc_512(buf, target.user));
|
---|
[821] | 273 | }
|
---|
| 274 | if (use_proxy && server.user) {
|
---|
| 275 | fprintf(sfp, "Proxy-Authorization: Basic %s\r\n",
|
---|
[1765] | 276 | base64enc_512(buf, server.user));
|
---|
[821] | 277 | }
|
---|
| 278 | #endif
|
---|
| 279 |
|
---|
[1765] | 280 | if (beg_range)
|
---|
| 281 | fprintf(sfp, "Range: bytes=%"OFF_FMT"d-\r\n", beg_range);
|
---|
| 282 | #if ENABLE_FEATURE_WGET_LONG_OPTIONS
|
---|
| 283 | if (extra_headers)
|
---|
| 284 | fputs(extra_headers, sfp);
|
---|
| 285 | #endif
|
---|
| 286 | fprintf(sfp, "Connection: close\r\n\r\n");
|
---|
[821] | 287 |
|
---|
| 288 | /*
|
---|
| 289 | * Retrieve HTTP response line and check for "200" status code.
|
---|
| 290 | */
|
---|
[1765] | 291 | read_response:
|
---|
[821] | 292 | if (fgets(buf, sizeof(buf), sfp) == NULL)
|
---|
[1765] | 293 | bb_error_msg_and_die("no response from server");
|
---|
[821] | 294 |
|
---|
[1765] | 295 | str = buf;
|
---|
| 296 | str = skip_non_whitespace(str);
|
---|
| 297 | str = skip_whitespace(str);
|
---|
| 298 | // FIXME: no error check
|
---|
| 299 | // xatou wouldn't work: "200 OK"
|
---|
| 300 | status = atoi(str);
|
---|
| 301 | switch (status) {
|
---|
| 302 | case 0:
|
---|
| 303 | case 100:
|
---|
| 304 | while (gethdr(buf, sizeof(buf), sfp, &n) != NULL)
|
---|
| 305 | /* eat all remaining headers */;
|
---|
| 306 | goto read_response;
|
---|
| 307 | case 200:
|
---|
| 308 | break;
|
---|
| 309 | case 300: /* redirection */
|
---|
| 310 | case 301:
|
---|
| 311 | case 302:
|
---|
| 312 | case 303:
|
---|
| 313 | break;
|
---|
| 314 | case 206:
|
---|
| 315 | if (beg_range)
|
---|
[821] | 316 | break;
|
---|
[1765] | 317 | /*FALLTHRU*/
|
---|
| 318 | default:
|
---|
| 319 | /* Show first line only and kill any ESC tricks */
|
---|
| 320 | buf[strcspn(buf, "\n\r\x1b")] = '\0';
|
---|
| 321 | bb_error_msg_and_die("server returned error: %s", buf);
|
---|
[821] | 322 | }
|
---|
| 323 |
|
---|
| 324 | /*
|
---|
| 325 | * Retrieve HTTP headers.
|
---|
| 326 | */
|
---|
[1765] | 327 | while ((str = gethdr(buf, sizeof(buf), sfp, &n)) != NULL) {
|
---|
| 328 | /* gethdr did already convert the "FOO:" string to lowercase */
|
---|
| 329 | smalluint key = index_in_strings(keywords, *&buf) + 1;
|
---|
| 330 | if (key == KEY_content_length) {
|
---|
| 331 | content_len = BB_STRTOOFF(str, NULL, 10);
|
---|
| 332 | if (errno || content_len < 0) {
|
---|
| 333 | bb_error_msg_and_die("content-length %s is garbage", str);
|
---|
[821] | 334 | }
|
---|
| 335 | got_clen = 1;
|
---|
| 336 | continue;
|
---|
| 337 | }
|
---|
[1765] | 338 | if (key == KEY_transfer_encoding) {
|
---|
| 339 | if (index_in_strings(keywords, str_tolower(str)) + 1 != KEY_chunked)
|
---|
| 340 | bb_error_msg_and_die("server wants to do %s transfer encoding", str);
|
---|
| 341 | chunked = got_clen = 1;
|
---|
[821] | 342 | }
|
---|
[1765] | 343 | if (key == KEY_location) {
|
---|
| 344 | if (str[0] == '/')
|
---|
| 345 | /* free(target.allocated); */
|
---|
| 346 | target.path = /* target.allocated = */ xstrdup(str+1);
|
---|
[821] | 347 | else {
|
---|
[1765] | 348 | parse_url(str, &target);
|
---|
[821] | 349 | if (use_proxy == 0) {
|
---|
| 350 | server.host = target.host;
|
---|
| 351 | server.port = target.port;
|
---|
| 352 | }
|
---|
[1765] | 353 | free(lsa);
|
---|
| 354 | lsa = xhost2sockaddr(server.host, server.port);
|
---|
[821] | 355 | break;
|
---|
| 356 | }
|
---|
| 357 | }
|
---|
| 358 | }
|
---|
[1765] | 359 | } while (status >= 300);
|
---|
[821] | 360 |
|
---|
| 361 | dfp = sfp;
|
---|
[1765] | 362 |
|
---|
| 363 | } else {
|
---|
| 364 |
|
---|
[821] | 365 | /*
|
---|
| 366 | * FTP session
|
---|
| 367 | */
|
---|
[1765] | 368 | if (!target.user)
|
---|
| 369 | target.user = xstrdup("anonymous:busybox@");
|
---|
[821] | 370 |
|
---|
[1765] | 371 | sfp = open_socket(lsa);
|
---|
[821] | 372 | if (ftpcmd(NULL, NULL, sfp, buf) != 220)
|
---|
[1765] | 373 | bb_error_msg_and_die("%s", buf+4);
|
---|
[821] | 374 |
|
---|
| 375 | /*
|
---|
| 376 | * Splitting username:password pair,
|
---|
| 377 | * trying to log in
|
---|
| 378 | */
|
---|
[1765] | 379 | str = strchr(target.user, ':');
|
---|
| 380 | if (str)
|
---|
| 381 | *(str++) = '\0';
|
---|
| 382 | switch (ftpcmd("USER ", target.user, sfp, buf)) {
|
---|
| 383 | case 230:
|
---|
| 384 | break;
|
---|
| 385 | case 331:
|
---|
| 386 | if (ftpcmd("PASS ", str, sfp, buf) == 230)
|
---|
[821] | 387 | break;
|
---|
[1765] | 388 | /* FALLTHRU (failed login) */
|
---|
| 389 | default:
|
---|
| 390 | bb_error_msg_and_die("ftp login: %s", buf+4);
|
---|
[821] | 391 | }
|
---|
| 392 |
|
---|
| 393 | ftpcmd("TYPE I", NULL, sfp, buf);
|
---|
| 394 |
|
---|
| 395 | /*
|
---|
| 396 | * Querying file size
|
---|
| 397 | */
|
---|
| 398 | if (ftpcmd("SIZE ", target.path, sfp, buf) == 213) {
|
---|
[1765] | 399 | content_len = BB_STRTOOFF(buf+4, NULL, 10);
|
---|
| 400 | if (errno || content_len < 0) {
|
---|
| 401 | bb_error_msg_and_die("SIZE value is garbage");
|
---|
[821] | 402 | }
|
---|
| 403 | got_clen = 1;
|
---|
| 404 | }
|
---|
| 405 |
|
---|
| 406 | /*
|
---|
| 407 | * Entering passive mode
|
---|
| 408 | */
|
---|
[1765] | 409 | if (ftpcmd("PASV", NULL, sfp, buf) != 227) {
|
---|
| 410 | pasv_error:
|
---|
| 411 | bb_error_msg_and_die("bad response to %s: %s", "PASV", buf);
|
---|
| 412 | }
|
---|
| 413 | // Response is "227 garbageN1,N2,N3,N4,P1,P2[)garbage]
|
---|
| 414 | // Server's IP is N1.N2.N3.N4 (we ignore it)
|
---|
| 415 | // Server's port for data connection is P1*256+P2
|
---|
| 416 | str = strrchr(buf, ')');
|
---|
| 417 | if (str) str[0] = '\0';
|
---|
| 418 | str = strrchr(buf, ',');
|
---|
| 419 | if (!str) goto pasv_error;
|
---|
| 420 | port = xatou_range(str+1, 0, 255);
|
---|
| 421 | *str = '\0';
|
---|
| 422 | str = strrchr(buf, ',');
|
---|
| 423 | if (!str) goto pasv_error;
|
---|
| 424 | port += xatou_range(str+1, 0, 255) * 256;
|
---|
| 425 | set_nport(lsa, htons(port));
|
---|
| 426 | dfp = open_socket(lsa);
|
---|
[821] | 427 |
|
---|
[1765] | 428 | if (beg_range) {
|
---|
| 429 | sprintf(buf, "REST %"OFF_FMT"d", beg_range);
|
---|
| 430 | if (ftpcmd(buf, NULL, sfp, buf) == 350)
|
---|
| 431 | content_len -= beg_range;
|
---|
[821] | 432 | }
|
---|
| 433 |
|
---|
| 434 | if (ftpcmd("RETR ", target.path, sfp, buf) > 150)
|
---|
[1765] | 435 | bb_error_msg_and_die("bad response to RETR: %s", buf);
|
---|
[821] | 436 | }
|
---|
[1765] | 437 | if (opt & WGET_OPT_SPIDER) {
|
---|
| 438 | if (ENABLE_FEATURE_CLEAN_UP)
|
---|
| 439 | fclose(sfp);
|
---|
| 440 | goto done;
|
---|
| 441 | }
|
---|
[821] | 442 |
|
---|
| 443 | /*
|
---|
| 444 | * Retrieve file
|
---|
| 445 | */
|
---|
| 446 | if (chunked) {
|
---|
| 447 | fgets(buf, sizeof(buf), dfp);
|
---|
[1765] | 448 | content_len = STRTOOFF(buf, NULL, 16);
|
---|
| 449 | /* FIXME: error check?? */
|
---|
[821] | 450 | }
|
---|
| 451 |
|
---|
[1765] | 452 | /* Do it before progressmeter (want to have nice error message) */
|
---|
| 453 | if (output_fd < 0)
|
---|
| 454 | output_fd = xopen(fname_out,
|
---|
| 455 | O_WRONLY|O_CREAT|O_EXCL|O_TRUNC);
|
---|
| 456 |
|
---|
| 457 | if (!(opt & WGET_OPT_QUIET))
|
---|
[821] | 458 | progressmeter(-1);
|
---|
| 459 |
|
---|
| 460 | do {
|
---|
[1765] | 461 | while (content_len > 0 || !got_clen) {
|
---|
| 462 | unsigned rdsz = sizeof(buf);
|
---|
| 463 | if (content_len < sizeof(buf) && (chunked || got_clen))
|
---|
| 464 | rdsz = (unsigned)content_len;
|
---|
| 465 | n = safe_fread(buf, rdsz, dfp);
|
---|
| 466 | if (n <= 0)
|
---|
| 467 | break;
|
---|
| 468 | if (full_write(output_fd, buf, n) != n) {
|
---|
[821] | 469 | bb_perror_msg_and_die(bb_msg_write_error);
|
---|
| 470 | }
|
---|
[1765] | 471 | #if ENABLE_FEATURE_WGET_STATUSBAR
|
---|
[821] | 472 | transferred += n;
|
---|
| 473 | #endif
|
---|
| 474 | if (got_clen) {
|
---|
[1765] | 475 | content_len -= n;
|
---|
[821] | 476 | }
|
---|
| 477 | }
|
---|
| 478 |
|
---|
| 479 | if (chunked) {
|
---|
| 480 | safe_fgets(buf, sizeof(buf), dfp); /* This is a newline */
|
---|
| 481 | safe_fgets(buf, sizeof(buf), dfp);
|
---|
[1765] | 482 | content_len = STRTOOFF(buf, NULL, 16);
|
---|
| 483 | /* FIXME: error check? */
|
---|
| 484 | if (content_len == 0) {
|
---|
[821] | 485 | chunked = 0; /* all done! */
|
---|
| 486 | }
|
---|
| 487 | }
|
---|
| 488 |
|
---|
| 489 | if (n == 0 && ferror(dfp)) {
|
---|
| 490 | bb_perror_msg_and_die(bb_msg_read_error);
|
---|
| 491 | }
|
---|
| 492 | } while (chunked);
|
---|
| 493 |
|
---|
[1765] | 494 | if (!(opt & WGET_OPT_QUIET))
|
---|
[821] | 495 | progressmeter(1);
|
---|
| 496 |
|
---|
| 497 | if ((use_proxy == 0) && target.is_ftp) {
|
---|
| 498 | fclose(dfp);
|
---|
| 499 | if (ftpcmd(NULL, NULL, sfp, buf) != 226)
|
---|
| 500 | bb_error_msg_and_die("ftp error: %s", buf+4);
|
---|
| 501 | ftpcmd("QUIT", NULL, sfp, buf);
|
---|
| 502 | }
|
---|
[1765] | 503 | done:
|
---|
[821] | 504 | exit(EXIT_SUCCESS);
|
---|
| 505 | }
|
---|
| 506 |
|
---|
| 507 |
|
---|
[1765] | 508 | static void parse_url(char *src_url, struct host_info *h)
|
---|
[821] | 509 | {
|
---|
[1765] | 510 | char *url, *p, *sp;
|
---|
[821] | 511 |
|
---|
[1765] | 512 | /* h->allocated = */ url = xstrdup(src_url);
|
---|
| 513 |
|
---|
[821] | 514 | if (strncmp(url, "http://", 7) == 0) {
|
---|
| 515 | h->port = bb_lookup_port("http", "tcp", 80);
|
---|
| 516 | h->host = url + 7;
|
---|
| 517 | h->is_ftp = 0;
|
---|
| 518 | } else if (strncmp(url, "ftp://", 6) == 0) {
|
---|
[1765] | 519 | h->port = bb_lookup_port("ftp", "tcp", 21);
|
---|
[821] | 520 | h->host = url + 6;
|
---|
| 521 | h->is_ftp = 1;
|
---|
| 522 | } else
|
---|
| 523 | bb_error_msg_and_die("not an http or ftp url: %s", url);
|
---|
| 524 |
|
---|
[1765] | 525 | // FYI:
|
---|
| 526 | // "Real" wget 'http://busybox.net?var=a/b' sends this request:
|
---|
| 527 | // 'GET /?var=a/b HTTP 1.0'
|
---|
| 528 | // and saves 'index.html?var=a%2Fb' (we save 'b')
|
---|
| 529 | // wget 'http://busybox.net?login=john@doe':
|
---|
| 530 | // request: 'GET /?login=john@doe HTTP/1.0'
|
---|
| 531 | // saves: 'index.html?login=john@doe' (we save '?login=john@doe')
|
---|
| 532 | // wget 'http://busybox.net#test/test':
|
---|
| 533 | // request: 'GET / HTTP/1.0'
|
---|
| 534 | // saves: 'index.html' (we save 'test')
|
---|
| 535 | //
|
---|
| 536 | // We also don't add unique .N suffix if file exists...
|
---|
[821] | 537 | sp = strchr(h->host, '/');
|
---|
[1765] | 538 | p = strchr(h->host, '?'); if (!sp || (p && sp > p)) sp = p;
|
---|
| 539 | p = strchr(h->host, '#'); if (!sp || (p && sp > p)) sp = p;
|
---|
| 540 | if (!sp) {
|
---|
| 541 | /* must be writable because of bb_get_last_path_component() */
|
---|
| 542 | static char nullstr[] ALIGN1 = "";
|
---|
| 543 | h->path = nullstr;
|
---|
| 544 | } else if (*sp == '/') {
|
---|
| 545 | *sp = '\0';
|
---|
| 546 | h->path = sp + 1;
|
---|
| 547 | } else { // '#' or '?'
|
---|
| 548 | // http://busybox.net?login=john@doe is a valid URL
|
---|
| 549 | // memmove converts to:
|
---|
| 550 | // http:/busybox.nett?login=john@doe...
|
---|
| 551 | memmove(h->host-1, h->host, sp - h->host);
|
---|
| 552 | h->host--;
|
---|
| 553 | sp[-1] = '\0';
|
---|
[821] | 554 | h->path = sp;
|
---|
[1765] | 555 | }
|
---|
[821] | 556 |
|
---|
[1765] | 557 | sp = strrchr(h->host, '@');
|
---|
| 558 | h->user = NULL;
|
---|
| 559 | if (sp != NULL) {
|
---|
[821] | 560 | h->user = h->host;
|
---|
[1765] | 561 | *sp = '\0';
|
---|
| 562 | h->host = sp + 1;
|
---|
[821] | 563 | }
|
---|
| 564 |
|
---|
[1765] | 565 | sp = h->host;
|
---|
[821] | 566 | }
|
---|
| 567 |
|
---|
| 568 |
|
---|
[1765] | 569 | static FILE *open_socket(len_and_sockaddr *lsa)
|
---|
[821] | 570 | {
|
---|
| 571 | FILE *fp;
|
---|
| 572 |
|
---|
[1765] | 573 | /* glibc 2.4 seems to try seeking on it - ??! */
|
---|
| 574 | /* hopefully it understands what ESPIPE means... */
|
---|
| 575 | fp = fdopen(xconnect_stream(lsa), "r+");
|
---|
[821] | 576 | if (fp == NULL)
|
---|
[1765] | 577 | bb_perror_msg_and_die("fdopen");
|
---|
[821] | 578 |
|
---|
| 579 | return fp;
|
---|
| 580 | }
|
---|
| 581 |
|
---|
| 582 |
|
---|
[1765] | 583 | static char *gethdr(char *buf, size_t bufsiz, FILE *fp, int *istrunc)
|
---|
[821] | 584 | {
|
---|
| 585 | char *s, *hdrval;
|
---|
| 586 | int c;
|
---|
| 587 |
|
---|
| 588 | *istrunc = 0;
|
---|
| 589 |
|
---|
| 590 | /* retrieve header line */
|
---|
| 591 | if (fgets(buf, bufsiz, fp) == NULL)
|
---|
| 592 | return NULL;
|
---|
| 593 |
|
---|
| 594 | /* see if we are at the end of the headers */
|
---|
[1765] | 595 | for (s = buf; *s == '\r'; ++s)
|
---|
[821] | 596 | ;
|
---|
| 597 | if (s[0] == '\n')
|
---|
| 598 | return NULL;
|
---|
| 599 |
|
---|
| 600 | /* convert the header name to lower case */
|
---|
[1765] | 601 | for (s = buf; isalnum(*s) || *s == '-'; ++s)
|
---|
[821] | 602 | *s = tolower(*s);
|
---|
| 603 |
|
---|
| 604 | /* verify we are at the end of the header name */
|
---|
| 605 | if (*s != ':')
|
---|
| 606 | bb_error_msg_and_die("bad header line: %s", buf);
|
---|
| 607 |
|
---|
| 608 | /* locate the start of the header value */
|
---|
[1765] | 609 | for (*s++ = '\0'; *s == ' ' || *s == '\t'; ++s)
|
---|
[821] | 610 | ;
|
---|
| 611 | hdrval = s;
|
---|
| 612 |
|
---|
| 613 | /* locate the end of header */
|
---|
| 614 | while (*s != '\0' && *s != '\r' && *s != '\n')
|
---|
| 615 | ++s;
|
---|
| 616 |
|
---|
| 617 | /* end of header found */
|
---|
| 618 | if (*s != '\0') {
|
---|
| 619 | *s = '\0';
|
---|
| 620 | return hdrval;
|
---|
| 621 | }
|
---|
| 622 |
|
---|
| 623 | /* Rats! The buffer isn't big enough to hold the entire header value. */
|
---|
| 624 | while (c = getc(fp), c != EOF && c != '\n')
|
---|
| 625 | ;
|
---|
| 626 | *istrunc = 1;
|
---|
| 627 | return hdrval;
|
---|
| 628 | }
|
---|
| 629 |
|
---|
[1765] | 630 | static int ftpcmd(const char *s1, const char *s2, FILE *fp, char *buf)
|
---|
[821] | 631 | {
|
---|
[1765] | 632 | int result;
|
---|
[821] | 633 | if (s1) {
|
---|
[1765] | 634 | if (!s2) s2 = "";
|
---|
[821] | 635 | fprintf(fp, "%s%s\r\n", s1, s2);
|
---|
| 636 | fflush(fp);
|
---|
| 637 | }
|
---|
| 638 |
|
---|
| 639 | do {
|
---|
| 640 | char *buf_ptr;
|
---|
| 641 |
|
---|
| 642 | if (fgets(buf, 510, fp) == NULL) {
|
---|
[1765] | 643 | bb_perror_msg_and_die("error getting response");
|
---|
[821] | 644 | }
|
---|
| 645 | buf_ptr = strstr(buf, "\r\n");
|
---|
| 646 | if (buf_ptr) {
|
---|
| 647 | *buf_ptr = '\0';
|
---|
| 648 | }
|
---|
[1765] | 649 | } while (!isdigit(buf[0]) || buf[3] != ' ');
|
---|
[821] | 650 |
|
---|
[1765] | 651 | buf[3] = '\0';
|
---|
| 652 | result = xatoi_u(buf);
|
---|
| 653 | buf[3] = ' ';
|
---|
| 654 | return result;
|
---|
[821] | 655 | }
|
---|
| 656 |
|
---|
[1765] | 657 | #if ENABLE_FEATURE_WGET_STATUSBAR
|
---|
[821] | 658 | /* Stuff below is from BSD rcp util.c, as added to openshh.
|
---|
| 659 | * Original copyright notice is retained at the end of this file.
|
---|
| 660 | */
|
---|
| 661 | static int
|
---|
| 662 | getttywidth(void)
|
---|
| 663 | {
|
---|
[1765] | 664 | int width;
|
---|
[821] | 665 | get_terminal_width_height(0, &width, NULL);
|
---|
[1765] | 666 | return width;
|
---|
[821] | 667 | }
|
---|
| 668 |
|
---|
| 669 | static void
|
---|
| 670 | updateprogressmeter(int ignore)
|
---|
| 671 | {
|
---|
| 672 | int save_errno = errno;
|
---|
| 673 |
|
---|
| 674 | progressmeter(0);
|
---|
| 675 | errno = save_errno;
|
---|
| 676 | }
|
---|
| 677 |
|
---|
[1765] | 678 | static void alarmtimer(int iwait)
|
---|
[821] | 679 | {
|
---|
| 680 | struct itimerval itv;
|
---|
| 681 |
|
---|
[1765] | 682 | itv.it_value.tv_sec = iwait;
|
---|
[821] | 683 | itv.it_value.tv_usec = 0;
|
---|
| 684 | itv.it_interval = itv.it_value;
|
---|
| 685 | setitimer(ITIMER_REAL, &itv, NULL);
|
---|
| 686 | }
|
---|
| 687 |
|
---|
| 688 | static void
|
---|
| 689 | progressmeter(int flag)
|
---|
| 690 | {
|
---|
[1765] | 691 | static unsigned lastupdate_sec;
|
---|
| 692 | static unsigned start_sec;
|
---|
[821] | 693 | static off_t lastsize, totalsize;
|
---|
| 694 |
|
---|
| 695 | off_t abbrevsize;
|
---|
[1765] | 696 | unsigned since_last_update, elapsed;
|
---|
| 697 | unsigned ratio;
|
---|
| 698 | int barlength, i;
|
---|
[821] | 699 |
|
---|
[1765] | 700 | if (flag == -1) { /* first call to progressmeter */
|
---|
| 701 | start_sec = monotonic_sec();
|
---|
| 702 | lastupdate_sec = start_sec;
|
---|
[821] | 703 | lastsize = 0;
|
---|
[1765] | 704 | totalsize = content_len + beg_range; /* as content_len changes.. */
|
---|
[821] | 705 | }
|
---|
| 706 |
|
---|
| 707 | ratio = 100;
|
---|
| 708 | if (totalsize != 0 && !chunked) {
|
---|
[1765] | 709 | /* long long helps to have it working even if !LFS */
|
---|
| 710 | ratio = (unsigned) (100ULL * (transferred+beg_range) / totalsize);
|
---|
| 711 | if (ratio > 100) ratio = 100;
|
---|
[821] | 712 | }
|
---|
| 713 |
|
---|
| 714 | fprintf(stderr, "\r%-20.20s%4d%% ", curfile, ratio);
|
---|
| 715 |
|
---|
[1765] | 716 | barlength = getttywidth() - 49;
|
---|
| 717 | if (barlength > 0) {
|
---|
| 718 | /* god bless gcc for variable arrays :) */
|
---|
[821] | 719 | i = barlength * ratio / 100;
|
---|
[1765] | 720 | {
|
---|
| 721 | char buf[i+1];
|
---|
| 722 | memset(buf, '*', i);
|
---|
| 723 | buf[i] = '\0';
|
---|
| 724 | fprintf(stderr, "|%s%*s|", buf, barlength - i, "");
|
---|
| 725 | }
|
---|
[821] | 726 | }
|
---|
| 727 | i = 0;
|
---|
[1765] | 728 | abbrevsize = transferred + beg_range;
|
---|
[821] | 729 | while (abbrevsize >= 100000) {
|
---|
| 730 | i++;
|
---|
| 731 | abbrevsize >>= 10;
|
---|
| 732 | }
|
---|
[1765] | 733 | /* see http://en.wikipedia.org/wiki/Tera */
|
---|
| 734 | fprintf(stderr, "%6d%c ", (int)abbrevsize, " kMGTPEZY"[i]);
|
---|
[821] | 735 |
|
---|
[1765] | 736 | // Nuts! Ain't it easier to update progress meter ONLY when we transferred++?
|
---|
| 737 | // FIXME: get rid of alarmtimer + updateprogressmeter mess
|
---|
| 738 |
|
---|
| 739 | elapsed = monotonic_sec();
|
---|
| 740 | since_last_update = elapsed - lastupdate_sec;
|
---|
[821] | 741 | if (transferred > lastsize) {
|
---|
[1765] | 742 | lastupdate_sec = elapsed;
|
---|
[821] | 743 | lastsize = transferred;
|
---|
[1765] | 744 | if (since_last_update >= STALLTIME) {
|
---|
| 745 | /* We "cut off" these seconds from elapsed time
|
---|
| 746 | * by adjusting start time */
|
---|
| 747 | start_sec += since_last_update;
|
---|
| 748 | }
|
---|
| 749 | since_last_update = 0; /* we are un-stalled now */
|
---|
[821] | 750 | }
|
---|
[1765] | 751 | elapsed -= start_sec; /* now it's "elapsed since start" */
|
---|
[821] | 752 |
|
---|
[1765] | 753 | if (since_last_update >= STALLTIME) {
|
---|
[821] | 754 | fprintf(stderr, " - stalled -");
|
---|
| 755 | } else {
|
---|
[1765] | 756 | off_t to_download = totalsize - beg_range;
|
---|
| 757 | if (transferred <= 0 || (int)elapsed <= 0 || transferred > to_download || chunked) {
|
---|
| 758 | fprintf(stderr, "--:--:-- ETA");
|
---|
| 759 | } else {
|
---|
| 760 | /* to_download / (transferred/elapsed) - elapsed: */
|
---|
| 761 | int eta = (int) ((unsigned long long)to_download*elapsed/transferred - elapsed);
|
---|
| 762 | /* (long long helps to have working ETA even if !LFS) */
|
---|
| 763 | i = eta % 3600;
|
---|
| 764 | fprintf(stderr, "%02d:%02d:%02d ETA", eta / 3600, i / 60, i % 60);
|
---|
| 765 | }
|
---|
[821] | 766 | }
|
---|
| 767 |
|
---|
[1765] | 768 | if (flag == -1) { /* first call to progressmeter */
|
---|
[821] | 769 | struct sigaction sa;
|
---|
| 770 | sa.sa_handler = updateprogressmeter;
|
---|
| 771 | sigemptyset(&sa.sa_mask);
|
---|
| 772 | sa.sa_flags = SA_RESTART;
|
---|
| 773 | sigaction(SIGALRM, &sa, NULL);
|
---|
| 774 | alarmtimer(1);
|
---|
[1765] | 775 | } else if (flag == 1) { /* last call to progressmeter */
|
---|
[821] | 776 | alarmtimer(0);
|
---|
| 777 | transferred = 0;
|
---|
| 778 | putc('\n', stderr);
|
---|
| 779 | }
|
---|
| 780 | }
|
---|
[1765] | 781 | #endif /* FEATURE_WGET_STATUSBAR */
|
---|
[821] | 782 |
|
---|
| 783 | /* Original copyright notice which applies to the CONFIG_FEATURE_WGET_STATUSBAR stuff,
|
---|
| 784 | * much of which was blatantly stolen from openssh. */
|
---|
| 785 |
|
---|
| 786 | /*-
|
---|
| 787 | * Copyright (c) 1992, 1993
|
---|
| 788 | * The Regents of the University of California. All rights reserved.
|
---|
| 789 | *
|
---|
| 790 | * Redistribution and use in source and binary forms, with or without
|
---|
| 791 | * modification, are permitted provided that the following conditions
|
---|
| 792 | * are met:
|
---|
| 793 | * 1. Redistributions of source code must retain the above copyright
|
---|
| 794 | * notice, this list of conditions and the following disclaimer.
|
---|
| 795 | * 2. Redistributions in binary form must reproduce the above copyright
|
---|
| 796 | * notice, this list of conditions and the following disclaimer in the
|
---|
| 797 | * documentation and/or other materials provided with the distribution.
|
---|
| 798 | *
|
---|
| 799 | * 3. <BSD Advertising Clause omitted per the July 22, 1999 licensing change
|
---|
| 800 | * ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change>
|
---|
| 801 | *
|
---|
| 802 | * 4. Neither the name of the University nor the names of its contributors
|
---|
| 803 | * may be used to endorse or promote products derived from this software
|
---|
| 804 | * without specific prior written permission.
|
---|
| 805 | *
|
---|
| 806 | * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
---|
| 807 | * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
---|
| 808 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
---|
| 809 | * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
---|
| 810 | * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
---|
| 811 | * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
---|
| 812 | * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
---|
| 813 | * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
---|
| 814 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
---|
| 815 | * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
---|
| 816 | * SUCH DAMAGE.
|
---|
| 817 | *
|
---|
| 818 | */
|
---|