1 /* Command line parsing.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
3 2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation,
6 This file is part of GNU Wget.
8 GNU Wget is free software; you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation; either version 3 of the License, or
11 (at your option) any later version.
13 GNU Wget is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
18 You should have received a copy of the GNU General Public License
19 along with Wget. If not, see <http://www.gnu.org/licenses/>.
21 Additional permission under GNU GPL version 3 section 7
23 If you modify this program, or any covered work, by linking or
24 combining it with the OpenSSL project's OpenSSL library (or a
25 modified version of that library), containing parts covered by the
26 terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
27 grants you additional permission to convey the resulting work.
28 Corresponding Source for a non-source form of such a combination
29 shall include the source code for the parts of OpenSSL used as well
30 as that of the covered work. */
53 #include "progress.h" /* for progress_handle_sigwinch */
56 #include "http.h" /* for save_cookies */
73 #ifndef PATH_SEPARATOR
74 # define PATH_SEPARATOR '/'
83 /* defined in version.c */
84 extern char *version_string;
85 extern char *compilation_string;
86 extern char *system_getrc;
87 extern char *link_string;
88 /* defined in build_info.c */
89 extern const char *compiled_features[];
90 /* Used for --version output in print_version */
91 #define MAX_CHARS_PER_LINE 72
94 #if defined(SIGHUP) || defined(SIGUSR1)
95 static void redirect_output_signal (int);
98 const char *exec_name;
100 /* Number of successfully downloaded URLs */
104 /* Initialize I18N/L10N. That amounts to invoking setlocale, and
105 setting up gettext's message catalog using bindtextdomain and
106 textdomain. Does nothing if NLS is disabled or missing. */
109 i18n_initialize (void)
111 /* ENABLE_NLS implies existence of functions invoked here. */
113 /* Set the current locale. */
114 setlocale (LC_ALL, "");
115 /* Set the text message domain. */
116 bindtextdomain ("wget", LOCALEDIR);
118 #endif /* ENABLE_NLS */
121 /* Definition of command-line options. */
123 static void print_help (void);
124 static void print_version (void);
129 # define IF_SSL(x) NULL
133 # define WHEN_DEBUG(x) x
135 # define WHEN_DEBUG(x) NULL
138 struct cmdline_option {
139 const char *long_name;
145 /* Non-standard options that have to be handled specially in
149 OPT__DONT_REMOVE_LISTING,
154 const void *data; /* for standard options */
155 int argtype; /* for non-standard options */
158 static struct cmdline_option option_data[] =
160 { "accept", 'A', OPT_VALUE, "accept", -1 },
161 { "adjust-extension", 'E', OPT_BOOLEAN, "adjustextension", -1 },
162 { "append-output", 'a', OPT__APPEND_OUTPUT, NULL, required_argument },
163 { "ask-password", 0, OPT_BOOLEAN, "askpassword", -1 },
164 { "auth-no-challenge", 0, OPT_BOOLEAN, "authnochallenge", -1 },
165 { "background", 'b', OPT_BOOLEAN, "background", -1 },
166 { "backup-converted", 'K', OPT_BOOLEAN, "backupconverted", -1 },
167 { "backups", 0, OPT_BOOLEAN, "backups", -1 },
168 { "base", 'B', OPT_VALUE, "base", -1 },
169 { "bind-address", 0, OPT_VALUE, "bindaddress", -1 },
170 { IF_SSL ("ca-certificate"), 0, OPT_VALUE, "cacertificate", -1 },
171 { IF_SSL ("ca-directory"), 0, OPT_VALUE, "cadirectory", -1 },
172 { "cache", 0, OPT_BOOLEAN, "cache", -1 },
173 { IF_SSL ("certificate"), 0, OPT_VALUE, "certificate", -1 },
174 { IF_SSL ("certificate-type"), 0, OPT_VALUE, "certificatetype", -1 },
175 { IF_SSL ("check-certificate"), 0, OPT_BOOLEAN, "checkcertificate", -1 },
176 { "clobber", 0, OPT__CLOBBER, NULL, optional_argument },
177 { "config", 0, OPT_VALUE, "chooseconfig", -1 },
178 { "connect-timeout", 0, OPT_VALUE, "connecttimeout", -1 },
179 { "continue", 'c', OPT_BOOLEAN, "continue", -1 },
180 { "convert-links", 'k', OPT_BOOLEAN, "convertlinks", -1 },
181 { "content-disposition", 0, OPT_BOOLEAN, "contentdisposition", -1 },
182 { "content-on-error", 0, OPT_BOOLEAN, "contentonerror", -1 },
183 { "cookies", 0, OPT_BOOLEAN, "cookies", -1 },
184 { "cut-dirs", 0, OPT_VALUE, "cutdirs", -1 },
185 { WHEN_DEBUG ("debug"), 'd', OPT_BOOLEAN, "debug", -1 },
186 { "default-page", 0, OPT_VALUE, "defaultpage", -1 },
187 { "delete-after", 0, OPT_BOOLEAN, "deleteafter", -1 },
188 { "directories", 0, OPT_BOOLEAN, "dirstruct", -1 },
189 { "directory-prefix", 'P', OPT_VALUE, "dirprefix", -1 },
190 { "dns-cache", 0, OPT_BOOLEAN, "dnscache", -1 },
191 { "dns-timeout", 0, OPT_VALUE, "dnstimeout", -1 },
192 { "domains", 'D', OPT_VALUE, "domains", -1 },
193 { "dont-remove-listing", 0, OPT__DONT_REMOVE_LISTING, NULL, no_argument },
194 { "dot-style", 0, OPT_VALUE, "dotstyle", -1 }, /* deprecated */
195 { "egd-file", 0, OPT_VALUE, "egdfile", -1 },
196 { "exclude-directories", 'X', OPT_VALUE, "excludedirectories", -1 },
197 { "exclude-domains", 0, OPT_VALUE, "excludedomains", -1 },
198 { "execute", 'e', OPT__EXECUTE, NULL, required_argument },
199 { "follow-ftp", 0, OPT_BOOLEAN, "followftp", -1 },
200 { "follow-tags", 0, OPT_VALUE, "followtags", -1 },
201 { "force-directories", 'x', OPT_BOOLEAN, "dirstruct", -1 },
202 { "force-html", 'F', OPT_BOOLEAN, "forcehtml", -1 },
203 { "ftp-password", 0, OPT_VALUE, "ftppassword", -1 },
205 { "ftp-stmlf", 0, OPT_BOOLEAN, "ftpstmlf", -1 },
206 #endif /* def __VMS */
207 { "ftp-user", 0, OPT_VALUE, "ftpuser", -1 },
208 { "glob", 0, OPT_BOOLEAN, "glob", -1 },
209 { "header", 0, OPT_VALUE, "header", -1 },
210 { "help", 'h', OPT_FUNCALL, (void *)print_help, no_argument },
211 { "host-directories", 0, OPT_BOOLEAN, "addhostdir", -1 },
212 { "html-extension", 'E', OPT_BOOLEAN, "adjustextension", -1 }, /* deprecated */
213 { "htmlify", 0, OPT_BOOLEAN, "htmlify", -1 },
214 { "http-keep-alive", 0, OPT_BOOLEAN, "httpkeepalive", -1 },
215 { "http-passwd", 0, OPT_VALUE, "httppassword", -1 }, /* deprecated */
216 { "http-password", 0, OPT_VALUE, "httppassword", -1 },
217 { "http-user", 0, OPT_VALUE, "httpuser", -1 },
218 { "ignore-case", 0, OPT_BOOLEAN, "ignorecase", -1 },
219 { "ignore-length", 0, OPT_BOOLEAN, "ignorelength", -1 },
220 { "ignore-tags", 0, OPT_VALUE, "ignoretags", -1 },
221 { "include-directories", 'I', OPT_VALUE, "includedirectories", -1 },
223 { "inet4-only", '4', OPT_BOOLEAN, "inet4only", -1 },
224 { "inet6-only", '6', OPT_BOOLEAN, "inet6only", -1 },
226 { "input-file", 'i', OPT_VALUE, "input", -1 },
227 { "iri", 0, OPT_BOOLEAN, "iri", -1 },
228 { "keep-session-cookies", 0, OPT_BOOLEAN, "keepsessioncookies", -1 },
229 { "level", 'l', OPT_VALUE, "reclevel", -1 },
230 { "limit-rate", 0, OPT_VALUE, "limitrate", -1 },
231 { "load-cookies", 0, OPT_VALUE, "loadcookies", -1 },
232 { "local-encoding", 0, OPT_VALUE, "localencoding", -1 },
233 { "max-redirect", 0, OPT_VALUE, "maxredirect", -1 },
234 { "mirror", 'm', OPT_BOOLEAN, "mirror", -1 },
235 { "no", 'n', OPT__NO, NULL, required_argument },
236 { "no-clobber", 0, OPT_BOOLEAN, "noclobber", -1 },
237 { "no-parent", 0, OPT_BOOLEAN, "noparent", -1 },
238 { "output-document", 'O', OPT_VALUE, "outputdocument", -1 },
239 { "output-file", 'o', OPT_VALUE, "logfile", -1 },
240 { "page-requisites", 'p', OPT_BOOLEAN, "pagerequisites", -1 },
241 { "parent", 0, OPT__PARENT, NULL, optional_argument },
242 { "passive-ftp", 0, OPT_BOOLEAN, "passiveftp", -1 },
243 { "password", 0, OPT_VALUE, "password", -1 },
244 { "post-data", 0, OPT_VALUE, "postdata", -1 },
245 { "post-file", 0, OPT_VALUE, "postfile", -1 },
246 { "prefer-family", 0, OPT_VALUE, "preferfamily", -1 },
247 { "preserve-permissions", 0, OPT_BOOLEAN, "preservepermissions", -1 },
248 { IF_SSL ("private-key"), 0, OPT_VALUE, "privatekey", -1 },
249 { IF_SSL ("private-key-type"), 0, OPT_VALUE, "privatekeytype", -1 },
250 { "progress", 0, OPT_VALUE, "progress", -1 },
251 { "protocol-directories", 0, OPT_BOOLEAN, "protocoldirectories", -1 },
252 { "proxy", 0, OPT_BOOLEAN, "useproxy", -1 },
253 { "proxy__compat", 'Y', OPT_VALUE, "useproxy", -1 }, /* back-compatible */
254 { "proxy-passwd", 0, OPT_VALUE, "proxypassword", -1 }, /* deprecated */
255 { "proxy-password", 0, OPT_VALUE, "proxypassword", -1 },
256 { "proxy-user", 0, OPT_VALUE, "proxyuser", -1 },
257 { "quiet", 'q', OPT_BOOLEAN, "quiet", -1 },
258 { "quota", 'Q', OPT_VALUE, "quota", -1 },
259 { "random-file", 0, OPT_VALUE, "randomfile", -1 },
260 { "random-wait", 0, OPT_BOOLEAN, "randomwait", -1 },
261 { "read-timeout", 0, OPT_VALUE, "readtimeout", -1 },
262 { "recursive", 'r', OPT_BOOLEAN, "recursive", -1 },
263 { "referer", 0, OPT_VALUE, "referer", -1 },
264 { "reject", 'R', OPT_VALUE, "reject", -1 },
265 { "relative", 'L', OPT_BOOLEAN, "relativeonly", -1 },
266 { "remote-encoding", 0, OPT_VALUE, "remoteencoding", -1 },
267 { "remove-listing", 0, OPT_BOOLEAN, "removelisting", -1 },
268 { "restrict-file-names", 0, OPT_BOOLEAN, "restrictfilenames", -1 },
269 { "retr-symlinks", 0, OPT_BOOLEAN, "retrsymlinks", -1 },
270 { "retry-connrefused", 0, OPT_BOOLEAN, "retryconnrefused", -1 },
271 { "save-cookies", 0, OPT_VALUE, "savecookies", -1 },
272 { "save-headers", 0, OPT_BOOLEAN, "saveheaders", -1 },
273 { IF_SSL ("secure-protocol"), 0, OPT_VALUE, "secureprotocol", -1 },
274 { "server-response", 'S', OPT_BOOLEAN, "serverresponse", -1 },
275 { "span-hosts", 'H', OPT_BOOLEAN, "spanhosts", -1 },
276 { "spider", 0, OPT_BOOLEAN, "spider", -1 },
277 { "strict-comments", 0, OPT_BOOLEAN, "strictcomments", -1 },
278 { "timeout", 'T', OPT_VALUE, "timeout", -1 },
279 { "timestamping", 'N', OPT_BOOLEAN, "timestamping", -1 },
280 { "tries", 't', OPT_VALUE, "tries", -1 },
281 { "unlink", 0, OPT_BOOLEAN, "unlink", -1 },
282 { "trust-server-names", 0, OPT_BOOLEAN, "trustservernames", -1 },
283 { "use-server-timestamps", 0, OPT_BOOLEAN, "useservertimestamps", -1 },
284 { "user", 0, OPT_VALUE, "user", -1 },
285 { "user-agent", 'U', OPT_VALUE, "useragent", -1 },
286 { "verbose", 'v', OPT_BOOLEAN, "verbose", -1 },
287 { "verbose", 0, OPT_BOOLEAN, "verbose", -1 },
288 { "version", 'V', OPT_FUNCALL, (void *) print_version, no_argument },
289 { "wait", 'w', OPT_VALUE, "wait", -1 },
290 { "waitretry", 0, OPT_VALUE, "waitretry", -1 },
291 { "warc-cdx", 0, OPT_BOOLEAN, "warccdx", -1 },
292 { "warc-compression", 0, OPT_BOOLEAN, "warccompression", -1 },
293 { "warc-dedup", 0, OPT_VALUE, "warccdxdedup", -1 },
294 { "warc-digests", 0, OPT_BOOLEAN, "warcdigests", -1 },
295 { "warc-file", 0, OPT_VALUE, "warcfile", -1 },
296 { "warc-header", 0, OPT_VALUE, "warcheader", -1 },
297 { "warc-keep-log", 0, OPT_BOOLEAN, "warckeeplog", -1 },
298 { "warc-max-size", 0, OPT_VALUE, "warcmaxsize", -1 },
299 { "warc-tempdir", 0, OPT_VALUE, "warctempdir", -1 },
301 { "wdebug", 0, OPT_BOOLEAN, "wdebug", -1 },
308 /* Return a string that contains S with "no-" prepended. The string
309 is NUL-terminated and allocated off static storage at Wget
313 no_prefix (const char *s)
315 static char buffer[1024];
316 static char *p = buffer;
319 int size = 3 + strlen (s) + 1; /* "no-STRING\0" */
320 if (p + size >= buffer + sizeof (buffer))
323 cp[0] = 'n', cp[1] = 'o', cp[2] = '-';
329 /* The arguments that that main passes to getopt_long. */
330 static struct option long_options[2 * countof (option_data) + 1];
331 static char short_options[128];
333 /* Mapping between short option chars and option_data indices. */
334 static unsigned char optmap[96];
336 /* Marker for `--no-FOO' values in long_options. */
337 #define BOOLEAN_NEG_MARKER 1024
339 /* Initialize the long_options array used by getopt_long from the data
345 char *p = short_options;
347 for (i = 0; i < countof (option_data); i++)
349 struct cmdline_option *opt = &option_data[i];
350 struct option *longopt;
353 /* The option is disabled. */
356 longopt = &long_options[o++];
357 longopt->name = opt->long_name;
361 *p++ = opt->short_name;
362 optmap[opt->short_name - 32] = longopt - long_options;
367 longopt->has_arg = required_argument;
372 /* Specify an optional argument for long options, so that
373 --option=off works the same as --no-option, for
374 compatibility with pre-1.10 Wget. However, don't specify
375 optional arguments short-option booleans because they
376 prevent combining of short options. */
377 longopt->has_arg = optional_argument;
378 /* For Boolean options, add the "--no-FOO" variant, which is
379 identical to "--foo", except it has opposite meaning and
380 it doesn't allow an argument. */
381 longopt = &long_options[o++];
382 longopt->name = no_prefix (opt->long_name);
383 longopt->has_arg = no_argument;
384 /* Mask the value so we'll be able to recognize that we're
385 dealing with the false value. */
386 longopt->val = i | BOOLEAN_NEG_MARKER;
389 assert (opt->argtype != -1);
390 longopt->has_arg = opt->argtype;
393 if (longopt->has_arg == required_argument)
395 /* Don't handle optional_argument */
399 /* Terminate short_options. */
401 /* No need for xzero(long_options[o]) because its storage is static
402 and it will be zeroed by default. */
403 assert (o <= countof (long_options));
406 /* Print the usage message. */
408 print_usage (int error)
410 return fprintf (error ? stderr : stdout,
411 _("Usage: %s [OPTION]... [URL]...\n"), exec_name);
414 /* Print the help message, describing all the available options. If
415 you add an option, be sure to update this list. */
419 /* We split the help text this way to ease translation of individual
421 static const char *help[] = {
424 Mandatory arguments to long options are mandatory for short options too.\n\n"),
428 -V, --version display the version of Wget and exit.\n"),
430 -h, --help print this help.\n"),
432 -b, --background go to background after startup.\n"),
434 -e, --execute=COMMAND execute a `.wgetrc'-style command.\n"),
438 Logging and input file:\n"),
440 -o, --output-file=FILE log messages to FILE.\n"),
442 -a, --append-output=FILE append messages to FILE.\n"),
445 -d, --debug print lots of debugging information.\n"),
449 --wdebug print Watt-32 debug output.\n"),
452 -q, --quiet quiet (no output).\n"),
454 -v, --verbose be verbose (this is the default).\n"),
456 -nv, --no-verbose turn off verboseness, without being quiet.\n"),
458 -i, --input-file=FILE download URLs found in local or external FILE.\n"),
460 -F, --force-html treat input file as HTML.\n"),
462 -B, --base=URL resolves HTML input-file links (-i -F)\n\
463 relative to URL.\n"),
465 --config=FILE Specify config file to use.\n"),
471 -t, --tries=NUMBER set number of retries to NUMBER (0 unlimits).\n"),
473 --retry-connrefused retry even if connection is refused.\n"),
475 -O, --output-document=FILE write documents to FILE.\n"),
477 -nc, --no-clobber skip downloads that would download to\n\
478 existing files (overwriting them).\n"),
480 -c, --continue resume getting a partially-downloaded file.\n"),
482 --progress=TYPE select progress gauge type.\n"),
484 -N, --timestamping don't re-retrieve files unless newer than\n\
487 --no-use-server-timestamps don't set the local file's timestamp by\n\
488 the one on the server.\n"),
490 -S, --server-response print server response.\n"),
492 --spider don't download anything.\n"),
494 -T, --timeout=SECONDS set all timeout values to SECONDS.\n"),
496 --dns-timeout=SECS set the DNS lookup timeout to SECS.\n"),
498 --connect-timeout=SECS set the connect timeout to SECS.\n"),
500 --read-timeout=SECS set the read timeout to SECS.\n"),
502 -w, --wait=SECONDS wait SECONDS between retrievals.\n"),
504 --waitretry=SECONDS wait 1..SECONDS between retries of a retrieval.\n"),
506 --random-wait wait from 0.5*WAIT...1.5*WAIT secs between retrievals.\n"),
508 --no-proxy explicitly turn off proxy.\n"),
510 -Q, --quota=NUMBER set retrieval quota to NUMBER.\n"),
512 --bind-address=ADDRESS bind to ADDRESS (hostname or IP) on local host.\n"),
514 --limit-rate=RATE limit download rate to RATE.\n"),
516 --no-dns-cache disable caching DNS lookups.\n"),
518 --restrict-file-names=OS restrict chars in file names to ones OS allows.\n"),
520 --ignore-case ignore case when matching files/directories.\n"),
523 -4, --inet4-only connect only to IPv4 addresses.\n"),
525 -6, --inet6-only connect only to IPv6 addresses.\n"),
527 --prefer-family=FAMILY connect first to addresses of specified family,\n\
528 one of IPv6, IPv4, or none.\n"),
531 --user=USER set both ftp and http user to USER.\n"),
533 --password=PASS set both ftp and http password to PASS.\n"),
535 --ask-password prompt for passwords.\n"),
537 --no-iri turn off IRI support.\n"),
539 --local-encoding=ENC use ENC as the local encoding for IRIs.\n"),
541 --remote-encoding=ENC use ENC as the default remote encoding.\n"),
543 --unlink remove file before clobber.\n"),
549 -nd, --no-directories don't create directories.\n"),
551 -x, --force-directories force creation of directories.\n"),
553 -nH, --no-host-directories don't create host directories.\n"),
555 --protocol-directories use protocol name in directories.\n"),
557 -P, --directory-prefix=PREFIX save files to PREFIX/...\n"),
559 --cut-dirs=NUMBER ignore NUMBER remote directory components.\n"),
565 --http-user=USER set http user to USER.\n"),
567 --http-password=PASS set http password to PASS.\n"),
569 --no-cache disallow server-cached data.\n"),
571 --default-page=NAME Change the default page name (normally\n\
572 this is `index.html'.).\n"),
574 -E, --adjust-extension save HTML/CSS documents with proper extensions.\n"),
576 --ignore-length ignore `Content-Length' header field.\n"),
578 --header=STRING insert STRING among the headers.\n"),
580 --max-redirect maximum redirections allowed per page.\n"),
582 --proxy-user=USER set USER as proxy username.\n"),
584 --proxy-password=PASS set PASS as proxy password.\n"),
586 --referer=URL include `Referer: URL' header in HTTP request.\n"),
588 --save-headers save the HTTP headers to file.\n"),
590 -U, --user-agent=AGENT identify as AGENT instead of Wget/VERSION.\n"),
592 --no-http-keep-alive disable HTTP keep-alive (persistent connections).\n"),
594 --no-cookies don't use cookies.\n"),
596 --load-cookies=FILE load cookies from FILE before session.\n"),
598 --save-cookies=FILE save cookies to FILE after session.\n"),
600 --keep-session-cookies load and save session (non-permanent) cookies.\n"),
602 --post-data=STRING use the POST method; send STRING as the data.\n"),
604 --post-file=FILE use the POST method; send contents of FILE.\n"),
606 --content-disposition honor the Content-Disposition header when\n\
607 choosing local file names (EXPERIMENTAL).\n"),
609 --content-on-error output the received content on server errors.\n"),
611 --auth-no-challenge send Basic HTTP authentication information\n\
612 without first waiting for the server's\n\
618 HTTPS (SSL/TLS) options:\n"),
620 --secure-protocol=PR choose secure protocol, one of auto, SSLv2,\n\
621 SSLv3, and TLSv1.\n"),
623 --no-check-certificate don't validate the server's certificate.\n"),
625 --certificate=FILE client certificate file.\n"),
627 --certificate-type=TYPE client certificate type, PEM or DER.\n"),
629 --private-key=FILE private key file.\n"),
631 --private-key-type=TYPE private key type, PEM or DER.\n"),
633 --ca-certificate=FILE file with the bundle of CA's.\n"),
635 --ca-directory=DIR directory where hash list of CA's is stored.\n"),
637 --random-file=FILE file with random data for seeding the SSL PRNG.\n"),
639 --egd-file=FILE file naming the EGD socket with random data.\n"),
641 #endif /* HAVE_SSL */
647 --ftp-stmlf Use Stream_LF format for all binary FTP files.\n"),
648 #endif /* def __VMS */
650 --ftp-user=USER set ftp user to USER.\n"),
652 --ftp-password=PASS set ftp password to PASS.\n"),
654 --no-remove-listing don't remove `.listing' files.\n"),
656 --no-glob turn off FTP file name globbing.\n"),
658 --no-passive-ftp disable the \"passive\" transfer mode.\n"),
660 --preserve-permissions preserve remote file permissions.\n"),
662 --retr-symlinks when recursing, get linked-to files (not dir).\n"),
668 --warc-file=FILENAME save request/response data to a .warc.gz file.\n"),
670 --warc-header=STRING insert STRING into the warcinfo record.\n"),
672 --warc-max-size=NUMBER set maximum size of WARC files to NUMBER.\n"),
674 --warc-cdx write CDX index files.\n"),
676 --warc-dedup=FILENAME do not store records listed in this CDX file.\n"),
678 --no-warc-compression do not compress WARC files with GZIP.\n"),
680 --no-warc-digests do not calculate SHA1 digests.\n"),
682 --no-warc-keep-log do not store the log file in a WARC record.\n"),
684 --warc-tempdir=DIRECTORY location for temporary files created by the\n\
689 Recursive download:\n"),
691 -r, --recursive specify recursive download.\n"),
693 -l, --level=NUMBER maximum recursion depth (inf or 0 for infinite).\n"),
695 --delete-after delete files locally after downloading them.\n"),
697 -k, --convert-links make links in downloaded HTML or CSS point to\n\
701 -K, --backup-converted before converting file X, back up as X_orig.\n"),
702 #else /* def __VMS */
704 -K, --backup-converted before converting file X, back up as X.orig.\n"),
705 #endif /* def __VMS [else] */
707 -m, --mirror shortcut for -N -r -l inf --no-remove-listing.\n"),
709 -p, --page-requisites get all images, etc. needed to display HTML page.\n"),
711 --strict-comments turn on strict (SGML) handling of HTML comments.\n"),
715 Recursive accept/reject:\n"),
717 -A, --accept=LIST comma-separated list of accepted extensions.\n"),
719 -R, --reject=LIST comma-separated list of rejected extensions.\n"),
721 -D, --domains=LIST comma-separated list of accepted domains.\n"),
723 --exclude-domains=LIST comma-separated list of rejected domains.\n"),
725 --follow-ftp follow FTP links from HTML documents.\n"),
727 --follow-tags=LIST comma-separated list of followed HTML tags.\n"),
729 --ignore-tags=LIST comma-separated list of ignored HTML tags.\n"),
731 -H, --span-hosts go to foreign hosts when recursive.\n"),
733 -L, --relative follow relative links only.\n"),
735 -I, --include-directories=LIST list of allowed directories.\n"),
737 --trust-server-names use the name specified by the redirection\n\
738 url last component.\n"),
740 -X, --exclude-directories=LIST list of excluded directories.\n"),
742 -np, --no-parent don't ascend to the parent directory.\n"),
745 N_("Mail bug reports and suggestions to <bug-wget@gnu.org>.\n")
750 if (printf (_("GNU Wget %s, a non-interactive network retriever.\n"),
753 if (print_usage (0) < 0)
756 for (i = 0; i < countof (help); i++)
757 if (fputs (_(help[i]), stdout) < 0)
763 /* Return a human-readable printed representation of INTERVAL,
764 measured in seconds. */
767 secs_to_human_time (double interval)
770 int secs = (int) (interval + 0.5);
771 int hours, mins, days;
773 days = secs / 86400, secs %= 86400;
774 hours = secs / 3600, secs %= 3600;
775 mins = secs / 60, secs %= 60;
778 sprintf (buf, "%dd %dh %dm %ds", days, hours, mins, secs);
780 sprintf (buf, "%dh %dm %ds", hours, mins, secs);
782 sprintf (buf, "%dm %ds", mins, secs);
784 sprintf (buf, "%ss", print_decimal (interval));
790 prompt_for_password (void)
793 fprintf (stderr, _("Password for user %s: "), quote (opt.user));
795 fprintf (stderr, _("Password: "));
799 /* Function that prints the line argument while limiting it
800 to at most line_length. prefix is printed on the first line
801 and an appropriate number of spaces are added on subsequent
804 format_and_print_line (const char *prefix, const char *line,
808 char *line_dup, *token;
810 assert (prefix != NULL);
811 assert (line != NULL);
813 line_dup = xstrdup (line);
815 if (line_length <= 0)
816 line_length = MAX_CHARS_PER_LINE - TABULATION;
818 if (printf ("%s", prefix) < 0)
820 remaining_chars = line_length;
821 /* We break on spaces. */
822 token = strtok (line_dup, " ");
823 while (token != NULL)
825 /* If however a token is much larger than the maximum
826 line length, all bets are off and we simply print the
827 token on the next line. */
828 if (remaining_chars <= strlen (token))
830 if (printf ("\n%*c", TABULATION, ' ') < 0)
832 remaining_chars = line_length - TABULATION;
834 if (printf ("%s ", token) < 0)
836 remaining_chars -= strlen (token) + 1; /* account for " " */
837 token = strtok (NULL, " ");
840 if (printf ("\n") < 0)
850 const char *wgetrc_title = _("Wgetrc: ");
851 const char *locale_title = _("Locale: ");
852 const char *compile_title = _("Compile: ");
853 const char *link_title = _("Link: ");
854 char *env_wgetrc, *user_wgetrc;
857 if (printf (_("GNU Wget %s built on %s.\n\n"), version_string, OS_TYPE) < 0)
860 for (i = 0; compiled_features[i] != NULL; )
862 int line_length = MAX_CHARS_PER_LINE;
863 while ((line_length > 0) && (compiled_features[i] != NULL))
865 if (printf ("%s ", compiled_features[i]) < 0)
867 line_length -= strlen (compiled_features[i]) + 2;
870 if (printf ("\n") < 0)
873 if (printf ("\n") < 0)
876 /* Handle the case when $WGETRC is unset and $HOME/.wgetrc is
878 if (printf ("%s\n", wgetrc_title) < 0)
881 env_wgetrc = wgetrc_env_file_name ();
882 if (env_wgetrc && *env_wgetrc)
884 if (printf (_(" %s (env)\n"), env_wgetrc) < 0)
888 user_wgetrc = wgetrc_user_file_name ();
891 if (printf (_(" %s (user)\n"), user_wgetrc) < 0)
896 if (printf (_(" %s (system)\n"), SYSTEM_WGETRC) < 0)
901 if (format_and_print_line (locale_title,
903 MAX_CHARS_PER_LINE) < 0)
905 #endif /* def ENABLE_NLS */
907 if (compilation_string != NULL)
908 if (format_and_print_line (compile_title,
910 MAX_CHARS_PER_LINE) < 0)
913 if (link_string != NULL)
914 if (format_and_print_line (link_title,
916 MAX_CHARS_PER_LINE) < 0)
919 if (printf ("\n") < 0)
922 /* TRANSLATORS: When available, an actual copyright character
923 (circle-c) should be used in preference to "(C)". */
925 Copyright (C) 2011 Free Software Foundation, Inc.\n"), stdout) < 0)
928 License GPLv3+: GNU GPL version 3 or later\n\
929 <http://www.gnu.org/licenses/gpl.html>.\n\
930 This is free software: you are free to change and redistribute it.\n\
931 There is NO WARRANTY, to the extent permitted by law.\n"), stdout) < 0)
933 /* TRANSLATORS: When available, please use the proper diacritics for
934 names such as this one. See en_US.po for reference. */
935 if (fputs (_("\nOriginally written by Hrvoje Niksic <hniksic@xemacs.org>.\n"),
938 if (fputs (_("Please send bug reports and questions to <bug-wget@gnu.org>.\n"),
945 char *program_name; /* Needed by lib/error.c. */
946 char *program_argstring; /* Needed by wget_warc.c. */
949 main (int argc, char **argv)
952 int i, ret, longindex;
954 bool append_to_log = false;
956 total_downloaded_bytes = 0;
958 program_name = argv[0];
960 struct ptimer *timer = ptimer_new ();
961 double start_time = ptimer_measure (timer);
965 /* Construct the name of the executable, without the directory part. */
967 /* On VMS, lose the "dev:[dir]" prefix and the ".EXE;nnn" suffix. */
968 exec_name = vms_basename (argv[0]);
969 #else /* def __VMS */
970 exec_name = strrchr (argv[0], PATH_SEPARATOR);
975 #endif /* def __VMS [else] */
978 /* Drop extension (typically .EXE) from executable filename. */
979 windows_main ((char **) &exec_name);
982 /* Construct the arguments string. */
983 int argstring_length = 1;
984 for (i = 1; i < argc; i++)
985 argstring_length += strlen (argv[i]) + 2 + 1;
986 char *p = program_argstring = malloc (argstring_length * sizeof (char));
987 for (i = 1; i < argc; i++)
990 int arglen = strlen (argv[i]);
991 memcpy (p, argv[i], arglen);
998 /* Load the hard-coded defaults. */
1003 /* This separate getopt_long is needed to find the user config file
1004 option ("--config") and parse it before the other user options. */
1007 bool use_userconfig = false;
1009 while ((retconf = getopt_long (argc, argv,
1010 short_options, long_options, &longindex)) != -1)
1013 bool userrc_ret = true;
1014 struct cmdline_option *config_opt;
1016 /* There is no short option for "--config". */
1019 confval = long_options[longindex].val;
1020 config_opt = &option_data[confval & ~BOOLEAN_NEG_MARKER];
1021 if (strcmp (config_opt->long_name, "config") == 0)
1023 userrc_ret &= run_wgetrc (optarg);
1024 use_userconfig = true;
1028 printf ("Exiting due to error in %s\n", optarg);
1036 /* If the user did not specify a config, read the system wgetrc and ~/.wgetrc. */
1037 if (use_userconfig == false)
1044 while ((ret = getopt_long (argc, argv,
1045 short_options, long_options, &longindex)) != -1)
1048 struct cmdline_option *opt;
1050 /* If LONGINDEX is unchanged, it means RET is referring a short
1052 if (longindex == -1)
1058 printf (_("Try `%s --help' for more options.\n"), exec_name);
1061 /* Find the short option character in the mapping. */
1062 longindex = optmap[ret - 32];
1064 val = long_options[longindex].val;
1066 /* Use the retrieved value to locate the option in the
1067 option_data array, and to see if we're dealing with the
1068 negated "--no-FOO" variant of the boolean option "--foo". */
1069 opt = &option_data[val & ~BOOLEAN_NEG_MARKER];
1073 setoptval (opt->data, optarg, opt->long_name);
1077 /* The user has specified a value -- use it. */
1078 setoptval (opt->data, optarg, opt->long_name);
1081 /* NEG is true for `--no-FOO' style boolean options. */
1082 bool neg = !!(val & BOOLEAN_NEG_MARKER);
1083 setoptval (opt->data, neg ? "0" : "1", opt->long_name);
1088 void (*func) (void) = (void (*) (void)) opt->data;
1092 case OPT__APPEND_OUTPUT:
1093 setoptval ("logfile", optarg, opt->long_name);
1094 append_to_log = true;
1097 run_command (optarg);
1101 /* We support real --no-FOO flags now, but keep these
1102 short options for convenience and backward
1105 for (p = optarg; p && *p; p++)
1109 setoptval ("verbose", "0", opt->long_name);
1112 setoptval ("addhostdir", "0", opt->long_name);
1115 setoptval ("dirstruct", "0", opt->long_name);
1118 setoptval ("noclobber", "1", opt->long_name);
1121 setoptval ("noparent", "1", opt->long_name);
1124 fprintf (stderr, _("%s: illegal option -- `-n%c'\n"),
1127 fprintf (stderr, "\n");
1128 fprintf (stderr, _("Try `%s --help' for more options.\n"),
1137 /* The wgetrc commands are named noparent and noclobber,
1138 so we must revert the meaning of the cmdline options
1139 before passing the value to setoptval. */
1142 flag = (*optarg == '1' || c_tolower (*optarg) == 'y'
1143 || (c_tolower (optarg[0]) == 'o'
1144 && c_tolower (optarg[1]) == 'n'));
1145 setoptval (opt->type == OPT__PARENT ? "noparent" : "noclobber",
1146 flag ? "0" : "1", opt->long_name);
1149 case OPT__DONT_REMOVE_LISTING:
1150 setoptval ("removelisting", "0", opt->long_name);
1157 nurl = argc - optind;
1159 /* All user options have now been processed, so it's now safe to do
1160 interoption dependency checks. */
1162 if (opt.noclobber && opt.convert_links)
1165 _("Both --no-clobber and --convert-links were specified,"
1166 "only --convert-links will be used.\n"));
1167 opt.noclobber = false;
1170 if (opt.reclevel == 0)
1171 opt.reclevel = INFINITE_RECURSION; /* see recur.h for commentary */
1173 if (opt.spider || opt.delete_after)
1174 opt.no_dirstruct = true;
1176 if (opt.page_requisites && !opt.recursive)
1178 /* Don't set opt.recursive here because it would confuse the FTP
1179 code. Instead, call retrieve_tree below when either
1180 page_requisites or recursive is requested. */
1182 if (!opt.no_dirstruct)
1183 opt.dirstruct = 1; /* normally handled by cmd_spec_recursive() */
1186 if (opt.verbose == -1)
1187 opt.verbose = !opt.quiet;
1189 /* Sanity checks. */
1190 if (opt.verbose && opt.quiet)
1192 fprintf (stderr, _("Can't be verbose and quiet at the same time.\n"));
1196 if (opt.timestamping && opt.noclobber)
1198 fprintf (stderr, _("\
1199 Can't timestamp and not clobber old files at the same time.\n"));
1204 if (opt.ipv4_only && opt.ipv6_only)
1207 _("Cannot specify both --inet4-only and --inet6-only.\n"));
1212 if (opt.output_document)
1214 if (opt.convert_links
1215 && (nurl > 1 || opt.page_requisites || opt.recursive))
1218 Cannot specify both -k and -O if multiple URLs are given, or in combination\n\
1219 with -p or -r. See the manual for details.\n\n"), stderr);
1223 if (opt.page_requisites
1226 logprintf (LOG_NOTQUIET, "%s", _("\
1227 WARNING: combining -O with -r or -p will mean that all downloaded content\n\
1228 will be placed in the single file you specified.\n\n"));
1230 if (opt.timestamping)
1232 logprintf (LOG_NOTQUIET, "%s", _("\
1233 WARNING: timestamping does nothing in combination with -O. See the manual\n\
1234 for details.\n\n"));
1235 opt.timestamping = false;
1237 if (opt.noclobber && file_exists_p(opt.output_document))
1239 /* Check if output file exists; if it does, exit. */
1240 logprintf (LOG_VERBOSE,
1241 _("File `%s' already there; not retrieving.\n"),
1242 opt.output_document);
1247 if (opt.warc_filename != 0)
1252 _("WARC output does not work with --no-clobber, "
1253 "--no-clobber will be disabled.\n"));
1254 opt.noclobber = false;
1256 if (opt.timestamping)
1259 _("WARC output does not work with timestamping, "
1260 "timestamping will be disabled.\n"));
1261 opt.timestamping = false;
1266 _("WARC output does not work with --spider.\n"));
1269 if (opt.always_rest)
1272 _("WARC output does not work with --continue, "
1273 "--continue will be disabled.\n"));
1274 opt.always_rest = false;
1276 if (opt.warc_cdx_dedup_filename != 0 && !opt.warc_digests_enabled)
1279 _("Digests are disabled; WARC deduplication will "
1280 "not find duplicate records.\n"));
1282 if (opt.warc_keep_log)
1284 opt.progress_type = "dot";
1288 if (opt.ask_passwd && opt.passwd)
1291 _("Cannot specify both --ask-password and --password.\n"));
1296 if (!nurl && !opt.input_filename)
1298 /* No URL specified. */
1299 fprintf (stderr, _("%s: missing URL\n"), exec_name);
1302 /* #### Something nicer should be printed here -- similar to the
1303 pre-1.5 `--help' page. */
1304 fprintf (stderr, _("Try `%s --help' for more options.\n"), exec_name);
1311 if (opt.locale && !check_encoding_name (opt.locale))
1315 opt.locale = find_locale ();
1317 if (opt.encoding_remote && !check_encoding_name (opt.encoding_remote))
1318 opt.encoding_remote = NULL;
1321 memset (&dummy_iri, 0, sizeof (dummy_iri));
1322 if (opt.enable_iri || opt.locale || opt.encoding_remote)
1324 /* sXXXav : be more specific... */
1325 fprintf (stderr, _("This version does not have support for IRIs\n"));
1332 opt.passwd = prompt_for_password ();
1334 if (opt.passwd == NULL || opt.passwd[0] == '\0')
1344 fork_to_background ();
1347 /* Initialize progress. Have to do this after the options are
1348 processed so we know where the log file is. */
1350 set_progress_implementation (opt.progress_type);
1352 /* Fill in the arguments. */
1353 url = alloca_array (char *, nurl + 1);
1354 for (i = 0; i < nurl; i++, optind++)
1356 char *rewritten = rewrite_shorthand_url (argv[optind]);
1360 url[i] = xstrdup (argv[optind]);
1364 /* Initialize logging. */
1365 log_init (opt.lfilename, append_to_log);
1367 /* Open WARC file. */
1368 if (opt.warc_filename != 0)
1371 DEBUGP (("DEBUG output created by Wget %s on %s.\n\n",
1372 version_string, OS_TYPE));
1374 /* Open the output filename if necessary. */
1377 Note that having the output_stream ("-O") file opened here for an FTP
1378 URL rather than in getftp() (ftp.c) (and the http equivalent) rather
1379 limits the ability in VMS to open the file differently for ASCII
1380 versus binary FTP there. (Of course, doing it here allows a open
1381 failure to be detected immediately, without first connecting to the
1384 if (opt.output_document)
1386 if (HYPHENP (opt.output_document))
1389 _setmode (_fileno (stdout), _O_BINARY);
1391 output_stream = stdout;
1398 /* Common fopen() optional arguments:
1399 sequential access only, access callback function.
1401 # define FOPEN_OPT_ARGS , "fop=sqo", "acc", acc_cb, &open_id
1403 #else /* def __VMS */
1404 # define FOPEN_OPT_ARGS
1405 #endif /* def __VMS [else] */
1407 output_stream = fopen (opt.output_document,
1408 opt.always_rest ? "ab" : "wb"
1410 if (output_stream == NULL)
1412 perror (opt.output_document);
1415 if (fstat (fileno (output_stream), &st) == 0 && S_ISREG (st.st_mode))
1416 output_stream_regular = true;
1418 if (!output_stream_regular && opt.convert_links)
1420 fprintf (stderr, _("-k can be used together with -O only if \
1421 outputting to a regular file.\n"));
1428 /* Set global ODS5 flag according to the specified destination (if
1429 any), otherwise according to the current default device.
1431 if (output_stream == NULL)
1432 set_ods5_dest( "SYS$DISK");
1433 else if (output_stream != stdout)
1434 set_ods5_dest( opt.output_document);
1435 #endif /* def __VMS */
1442 /* Setup the signal handler to redirect output when hangup is
1444 if (signal(SIGHUP, SIG_IGN) != SIG_IGN)
1445 signal(SIGHUP, redirect_output_signal);
1447 /* ...and do the same for SIGUSR1. */
1449 signal (SIGUSR1, redirect_output_signal);
1452 /* Writing to a closed socket normally signals SIGPIPE, and the
1453 process exits. What we want is to ignore SIGPIPE and just check
1454 for the return value of write(). */
1455 signal (SIGPIPE, SIG_IGN);
1458 signal (SIGWINCH, progress_handle_sigwinch);
1461 /* Retrieve the URLs from argument list. */
1462 for (t = url; *t; t++)
1464 char *filename = NULL, *redirected_URL = NULL;
1466 /* Need to do a new struct iri every time, because
1467 * retrieve_url may modify it in some circumstances,
1469 struct iri *iri = iri_new ();
1470 struct url *url_parsed;
1472 set_uri_encoding (iri, opt.locale, true);
1473 url_parsed = url_parse (*t, &url_err, iri, true);
1477 char *error = url_error (*t, url_err);
1478 logprintf (LOG_NOTQUIET, "%s: %s.\n",*t, error);
1480 inform_exit_status (URLERROR);
1484 if ((opt.recursive || opt.page_requisites)
1485 && (url_scheme (*t) != SCHEME_FTP || url_uses_proxy (url_parsed)))
1487 int old_follow_ftp = opt.follow_ftp;
1489 /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
1490 if (url_scheme (*t) == SCHEME_FTP)
1493 retrieve_tree (url_parsed, NULL);
1495 opt.follow_ftp = old_follow_ftp;
1499 retrieve_url (url_parsed, *t, &filename, &redirected_URL, NULL,
1500 &dt, opt.recursive, iri, true);
1503 if (opt.delete_after && file_exists_p(filename))
1505 DEBUGP (("Removing file due to --delete-after in main():\n"));
1506 logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
1507 if (unlink (filename))
1508 logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
1510 xfree_null (redirected_URL);
1511 xfree_null (filename);
1512 url_free (url_parsed);
1517 /* And then from the input file, if any. */
1518 if (opt.input_filename)
1522 status = retrieve_from_file (opt.input_filename, opt.force_html, &count);
1523 inform_exit_status (status);
1525 logprintf (LOG_NOTQUIET, _("No URLs found in %s.\n"),
1526 opt.input_filename);
1529 /* Print broken links. */
1530 if (opt.recursive && opt.spider)
1531 print_broken_links ();
1533 /* Print the downloaded sum. */
1534 if ((opt.recursive || opt.page_requisites
1536 || (opt.input_filename && total_downloaded_bytes != 0))
1538 total_downloaded_bytes != 0)
1540 double end_time = ptimer_measure (timer);
1541 ptimer_destroy (timer);
1543 char *wall_time = xstrdup (secs_to_human_time (end_time - start_time));
1544 char *download_time = xstrdup (secs_to_human_time (total_download_time));
1545 logprintf (LOG_NOTQUIET,
1546 _("FINISHED --%s--\nTotal wall clock time: %s\n"
1547 "Downloaded: %d files, %s in %s (%s)\n"),
1548 datetime_str (time (NULL)),
1551 human_readable (total_downloaded_bytes),
1553 retr_rate (total_downloaded_bytes, total_download_time));
1555 xfree (download_time);
1557 /* Print quota warning, if exceeded. */
1558 if (opt.quota && total_downloaded_bytes > opt.quota)
1559 logprintf (LOG_NOTQUIET,
1560 _("Download quota of %s EXCEEDED!\n"),
1561 human_readable (opt.quota));
1564 if (opt.cookies_output)
1567 if (opt.convert_links && !opt.delete_after)
1568 convert_all_links ();
1570 /* Close WARC file. */
1571 if (opt.warc_filename != 0)
1576 for (i = 0; i < nurl; i++)
1580 exit (get_exit_status ());
1582 #endif /* TESTING */
1584 #if defined(SIGHUP) || defined(SIGUSR1)
1586 /* So the signal_name check doesn't blow when only one is available. */
1594 /* Hangup signal handler. When wget receives SIGHUP or SIGUSR1, it
1595 will proceed operation as usual, trying to write into a log file.
1596 If that is impossible, the output will be turned off. */
1599 redirect_output_signal (int sig)
1601 const char *signal_name = (sig == SIGHUP ? "SIGHUP" :
1602 (sig == SIGUSR1 ? "SIGUSR1" :
1604 log_request_redirect_output (signal_name);
1605 progress_schedule_redirect ();
1606 signal (sig, redirect_output_signal);