1 /* Command line parsing.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
3 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012 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 { "accept-regex", 0, OPT_VALUE, "acceptregex", -1 },
162 { "adjust-extension", 'E', OPT_BOOLEAN, "adjustextension", -1 },
163 { "append-output", 'a', OPT__APPEND_OUTPUT, NULL, required_argument },
164 { "ask-password", 0, OPT_BOOLEAN, "askpassword", -1 },
165 { "auth-no-challenge", 0, OPT_BOOLEAN, "authnochallenge", -1 },
166 { "background", 'b', OPT_BOOLEAN, "background", -1 },
167 { "backup-converted", 'K', OPT_BOOLEAN, "backupconverted", -1 },
168 { "backups", 0, OPT_BOOLEAN, "backups", -1 },
169 { "base", 'B', OPT_VALUE, "base", -1 },
170 { "bind-address", 0, OPT_VALUE, "bindaddress", -1 },
171 { IF_SSL ("ca-certificate"), 0, OPT_VALUE, "cacertificate", -1 },
172 { IF_SSL ("ca-directory"), 0, OPT_VALUE, "cadirectory", -1 },
173 { "cache", 0, OPT_BOOLEAN, "cache", -1 },
174 { IF_SSL ("certificate"), 0, OPT_VALUE, "certificate", -1 },
175 { IF_SSL ("certificate-type"), 0, OPT_VALUE, "certificatetype", -1 },
176 { IF_SSL ("check-certificate"), 0, OPT_BOOLEAN, "checkcertificate", -1 },
177 { "clobber", 0, OPT__CLOBBER, NULL, optional_argument },
178 { "config", 0, OPT_VALUE, "chooseconfig", -1 },
179 { "connect-timeout", 0, OPT_VALUE, "connecttimeout", -1 },
180 { "continue", 'c', OPT_BOOLEAN, "continue", -1 },
181 { "convert-links", 'k', OPT_BOOLEAN, "convertlinks", -1 },
182 { "content-disposition", 0, OPT_BOOLEAN, "contentdisposition", -1 },
183 { "content-on-error", 0, OPT_BOOLEAN, "contentonerror", -1 },
184 { "cookies", 0, OPT_BOOLEAN, "cookies", -1 },
185 { "cut-dirs", 0, OPT_VALUE, "cutdirs", -1 },
186 { WHEN_DEBUG ("debug"), 'd', OPT_BOOLEAN, "debug", -1 },
187 { "default-page", 0, OPT_VALUE, "defaultpage", -1 },
188 { "delete-after", 0, OPT_BOOLEAN, "deleteafter", -1 },
189 { "directories", 0, OPT_BOOLEAN, "dirstruct", -1 },
190 { "directory-prefix", 'P', OPT_VALUE, "dirprefix", -1 },
191 { "dns-cache", 0, OPT_BOOLEAN, "dnscache", -1 },
192 { "dns-timeout", 0, OPT_VALUE, "dnstimeout", -1 },
193 { "domains", 'D', OPT_VALUE, "domains", -1 },
194 { "dont-remove-listing", 0, OPT__DONT_REMOVE_LISTING, NULL, no_argument },
195 { "dot-style", 0, OPT_VALUE, "dotstyle", -1 }, /* deprecated */
196 { "egd-file", 0, OPT_VALUE, "egdfile", -1 },
197 { "exclude-directories", 'X', OPT_VALUE, "excludedirectories", -1 },
198 { "exclude-domains", 0, OPT_VALUE, "excludedomains", -1 },
199 { "execute", 'e', OPT__EXECUTE, NULL, required_argument },
200 { "follow-ftp", 0, OPT_BOOLEAN, "followftp", -1 },
201 { "follow-tags", 0, OPT_VALUE, "followtags", -1 },
202 { "force-directories", 'x', OPT_BOOLEAN, "dirstruct", -1 },
203 { "force-html", 'F', OPT_BOOLEAN, "forcehtml", -1 },
204 { "ftp-password", 0, OPT_VALUE, "ftppassword", -1 },
206 { "ftp-stmlf", 0, OPT_BOOLEAN, "ftpstmlf", -1 },
207 #endif /* def __VMS */
208 { "ftp-user", 0, OPT_VALUE, "ftpuser", -1 },
209 { "glob", 0, OPT_BOOLEAN, "glob", -1 },
210 { "header", 0, OPT_VALUE, "header", -1 },
211 { "help", 'h', OPT_FUNCALL, (void *)print_help, no_argument },
212 { "host-directories", 0, OPT_BOOLEAN, "addhostdir", -1 },
213 { "html-extension", 'E', OPT_BOOLEAN, "adjustextension", -1 }, /* deprecated */
214 { "htmlify", 0, OPT_BOOLEAN, "htmlify", -1 },
215 { "http-keep-alive", 0, OPT_BOOLEAN, "httpkeepalive", -1 },
216 { "http-passwd", 0, OPT_VALUE, "httppassword", -1 }, /* deprecated */
217 { "http-password", 0, OPT_VALUE, "httppassword", -1 },
218 { "http-user", 0, OPT_VALUE, "httpuser", -1 },
219 { "ignore-case", 0, OPT_BOOLEAN, "ignorecase", -1 },
220 { "ignore-length", 0, OPT_BOOLEAN, "ignorelength", -1 },
221 { "ignore-tags", 0, OPT_VALUE, "ignoretags", -1 },
222 { "include-directories", 'I', OPT_VALUE, "includedirectories", -1 },
224 { "inet4-only", '4', OPT_BOOLEAN, "inet4only", -1 },
225 { "inet6-only", '6', OPT_BOOLEAN, "inet6only", -1 },
227 { "input-file", 'i', OPT_VALUE, "input", -1 },
228 { "iri", 0, OPT_BOOLEAN, "iri", -1 },
229 { "keep-session-cookies", 0, OPT_BOOLEAN, "keepsessioncookies", -1 },
230 { "level", 'l', OPT_VALUE, "reclevel", -1 },
231 { "limit-rate", 0, OPT_VALUE, "limitrate", -1 },
232 { "load-cookies", 0, OPT_VALUE, "loadcookies", -1 },
233 { "local-encoding", 0, OPT_VALUE, "localencoding", -1 },
234 { "max-redirect", 0, OPT_VALUE, "maxredirect", -1 },
235 { "mirror", 'm', OPT_BOOLEAN, "mirror", -1 },
236 { "no", 'n', OPT__NO, NULL, required_argument },
237 { "no-clobber", 0, OPT_BOOLEAN, "noclobber", -1 },
238 { "no-parent", 0, OPT_BOOLEAN, "noparent", -1 },
239 { "output-document", 'O', OPT_VALUE, "outputdocument", -1 },
240 { "output-file", 'o', OPT_VALUE, "logfile", -1 },
241 { "page-requisites", 'p', OPT_BOOLEAN, "pagerequisites", -1 },
242 { "parent", 0, OPT__PARENT, NULL, optional_argument },
243 { "passive-ftp", 0, OPT_BOOLEAN, "passiveftp", -1 },
244 { "password", 0, OPT_VALUE, "password", -1 },
245 { "post-data", 0, OPT_VALUE, "postdata", -1 },
246 { "post-file", 0, OPT_VALUE, "postfile", -1 },
247 { "prefer-family", 0, OPT_VALUE, "preferfamily", -1 },
248 { "preserve-permissions", 0, OPT_BOOLEAN, "preservepermissions", -1 },
249 { IF_SSL ("private-key"), 0, OPT_VALUE, "privatekey", -1 },
250 { IF_SSL ("private-key-type"), 0, OPT_VALUE, "privatekeytype", -1 },
251 { "progress", 0, OPT_VALUE, "progress", -1 },
252 { "protocol-directories", 0, OPT_BOOLEAN, "protocoldirectories", -1 },
253 { "proxy", 0, OPT_BOOLEAN, "useproxy", -1 },
254 { "proxy__compat", 'Y', OPT_VALUE, "useproxy", -1 }, /* back-compatible */
255 { "proxy-passwd", 0, OPT_VALUE, "proxypassword", -1 }, /* deprecated */
256 { "proxy-password", 0, OPT_VALUE, "proxypassword", -1 },
257 { "proxy-user", 0, OPT_VALUE, "proxyuser", -1 },
258 { "quiet", 'q', OPT_BOOLEAN, "quiet", -1 },
259 { "quota", 'Q', OPT_VALUE, "quota", -1 },
260 { "random-file", 0, OPT_VALUE, "randomfile", -1 },
261 { "random-wait", 0, OPT_BOOLEAN, "randomwait", -1 },
262 { "read-timeout", 0, OPT_VALUE, "readtimeout", -1 },
263 { "recursive", 'r', OPT_BOOLEAN, "recursive", -1 },
264 { "referer", 0, OPT_VALUE, "referer", -1 },
265 { "regex-type", 0, OPT_VALUE, "regextype", -1 },
266 { "reject", 'R', OPT_VALUE, "reject", -1 },
267 { "reject-regex", 0, OPT_VALUE, "rejectregex", -1 },
268 { "relative", 'L', OPT_BOOLEAN, "relativeonly", -1 },
269 { "remote-encoding", 0, OPT_VALUE, "remoteencoding", -1 },
270 { "remove-listing", 0, OPT_BOOLEAN, "removelisting", -1 },
271 { "report-speed", 0, OPT_BOOLEAN, "reportspeed", -1 },
272 { "restrict-file-names", 0, OPT_BOOLEAN, "restrictfilenames", -1 },
273 { "retr-symlinks", 0, OPT_BOOLEAN, "retrsymlinks", -1 },
274 { "retry-connrefused", 0, OPT_BOOLEAN, "retryconnrefused", -1 },
275 { "save-cookies", 0, OPT_VALUE, "savecookies", -1 },
276 { "save-headers", 0, OPT_BOOLEAN, "saveheaders", -1 },
277 { IF_SSL ("secure-protocol"), 0, OPT_VALUE, "secureprotocol", -1 },
278 { "server-response", 'S', OPT_BOOLEAN, "serverresponse", -1 },
279 { "span-hosts", 'H', OPT_BOOLEAN, "spanhosts", -1 },
280 { "spider", 0, OPT_BOOLEAN, "spider", -1 },
281 { "strict-comments", 0, OPT_BOOLEAN, "strictcomments", -1 },
282 { "timeout", 'T', OPT_VALUE, "timeout", -1 },
283 { "timestamping", 'N', OPT_BOOLEAN, "timestamping", -1 },
284 { "tries", 't', OPT_VALUE, "tries", -1 },
285 { "unlink", 0, OPT_BOOLEAN, "unlink", -1 },
286 { "trust-server-names", 0, OPT_BOOLEAN, "trustservernames", -1 },
287 { "use-server-timestamps", 0, OPT_BOOLEAN, "useservertimestamps", -1 },
288 { "user", 0, OPT_VALUE, "user", -1 },
289 { "user-agent", 'U', OPT_VALUE, "useragent", -1 },
290 { "verbose", 'v', OPT_BOOLEAN, "verbose", -1 },
291 { "verbose", 0, OPT_BOOLEAN, "verbose", -1 },
292 { "version", 'V', OPT_FUNCALL, (void *) print_version, no_argument },
293 { "wait", 'w', OPT_VALUE, "wait", -1 },
294 { "waitretry", 0, OPT_VALUE, "waitretry", -1 },
295 { "warc-cdx", 0, OPT_BOOLEAN, "warccdx", -1 },
297 { "warc-compression", 0, OPT_BOOLEAN, "warccompression", -1 },
299 { "warc-dedup", 0, OPT_VALUE, "warccdxdedup", -1 },
300 { "warc-digests", 0, OPT_BOOLEAN, "warcdigests", -1 },
301 { "warc-file", 0, OPT_VALUE, "warcfile", -1 },
302 { "warc-header", 0, OPT_VALUE, "warcheader", -1 },
303 { "warc-keep-log", 0, OPT_BOOLEAN, "warckeeplog", -1 },
304 { "warc-max-size", 0, OPT_VALUE, "warcmaxsize", -1 },
305 { "warc-tempdir", 0, OPT_VALUE, "warctempdir", -1 },
307 { "wdebug", 0, OPT_BOOLEAN, "wdebug", -1 },
314 /* Return a string that contains S with "no-" prepended. The string
315 is NUL-terminated and allocated off static storage at Wget
319 no_prefix (const char *s)
321 static char buffer[1024];
322 static char *p = buffer;
325 int size = 3 + strlen (s) + 1; /* "no-STRING\0" */
326 if (p + size >= buffer + sizeof (buffer))
329 cp[0] = 'n', cp[1] = 'o', cp[2] = '-';
335 /* The arguments that that main passes to getopt_long. */
336 static struct option long_options[2 * countof (option_data) + 1];
337 static char short_options[128];
339 /* Mapping between short option chars and option_data indices. */
340 static unsigned char optmap[96];
342 /* Marker for `--no-FOO' values in long_options. */
343 #define BOOLEAN_NEG_MARKER 1024
345 /* Initialize the long_options array used by getopt_long from the data
351 char *p = short_options;
353 for (i = 0; i < countof (option_data); i++)
355 struct cmdline_option *opt = &option_data[i];
356 struct option *longopt;
359 /* The option is disabled. */
362 longopt = &long_options[o++];
363 longopt->name = opt->long_name;
367 *p++ = opt->short_name;
368 optmap[opt->short_name - 32] = longopt - long_options;
373 longopt->has_arg = required_argument;
378 /* Specify an optional argument for long options, so that
379 --option=off works the same as --no-option, for
380 compatibility with pre-1.10 Wget. However, don't specify
381 optional arguments short-option booleans because they
382 prevent combining of short options. */
383 longopt->has_arg = optional_argument;
384 /* For Boolean options, add the "--no-FOO" variant, which is
385 identical to "--foo", except it has opposite meaning and
386 it doesn't allow an argument. */
387 longopt = &long_options[o++];
388 longopt->name = no_prefix (opt->long_name);
389 longopt->has_arg = no_argument;
390 /* Mask the value so we'll be able to recognize that we're
391 dealing with the false value. */
392 longopt->val = i | BOOLEAN_NEG_MARKER;
395 assert (opt->argtype != -1);
396 longopt->has_arg = opt->argtype;
399 if (longopt->has_arg == required_argument)
401 /* Don't handle optional_argument */
405 /* Terminate short_options. */
407 /* No need for xzero(long_options[o]) because its storage is static
408 and it will be zeroed by default. */
409 assert (o <= countof (long_options));
412 /* Print the usage message. */
414 print_usage (int error)
416 return fprintf (error ? stderr : stdout,
417 _("Usage: %s [OPTION]... [URL]...\n"), exec_name);
420 /* Print the help message, describing all the available options. If
421 you add an option, be sure to update this list. */
425 /* We split the help text this way to ease translation of individual
427 static const char *help[] = {
430 Mandatory arguments to long options are mandatory for short options too.\n\n"),
434 -V, --version display the version of Wget and exit.\n"),
436 -h, --help print this help.\n"),
438 -b, --background go to background after startup.\n"),
440 -e, --execute=COMMAND execute a `.wgetrc'-style command.\n"),
444 Logging and input file:\n"),
446 -o, --output-file=FILE log messages to FILE.\n"),
448 -a, --append-output=FILE append messages to FILE.\n"),
451 -d, --debug print lots of debugging information.\n"),
455 --wdebug print Watt-32 debug output.\n"),
458 -q, --quiet quiet (no output).\n"),
460 -v, --verbose be verbose (this is the default).\n"),
462 -nv, --no-verbose turn off verboseness, without being quiet.\n"),
464 --report-speed=TYPE Output bandwidth as TYPE. TYPE can be bits.\n"),
466 -i, --input-file=FILE download URLs found in local or external FILE.\n"),
468 -F, --force-html treat input file as HTML.\n"),
470 -B, --base=URL resolves HTML input-file links (-i -F)\n\
471 relative to URL.\n"),
473 --config=FILE Specify config file to use.\n"),
479 -t, --tries=NUMBER set number of retries to NUMBER (0 unlimits).\n"),
481 --retry-connrefused retry even if connection is refused.\n"),
483 -O, --output-document=FILE write documents to FILE.\n"),
485 -nc, --no-clobber skip downloads that would download to\n\
486 existing files (overwriting them).\n"),
488 -c, --continue resume getting a partially-downloaded file.\n"),
490 --progress=TYPE select progress gauge type.\n"),
492 -N, --timestamping don't re-retrieve files unless newer than\n\
495 --no-use-server-timestamps don't set the local file's timestamp by\n\
496 the one on the server.\n"),
498 -S, --server-response print server response.\n"),
500 --spider don't download anything.\n"),
502 -T, --timeout=SECONDS set all timeout values to SECONDS.\n"),
504 --dns-timeout=SECS set the DNS lookup timeout to SECS.\n"),
506 --connect-timeout=SECS set the connect timeout to SECS.\n"),
508 --read-timeout=SECS set the read timeout to SECS.\n"),
510 -w, --wait=SECONDS wait SECONDS between retrievals.\n"),
512 --waitretry=SECONDS wait 1..SECONDS between retries of a retrieval.\n"),
514 --random-wait wait from 0.5*WAIT...1.5*WAIT secs between retrievals.\n"),
516 --no-proxy explicitly turn off proxy.\n"),
518 -Q, --quota=NUMBER set retrieval quota to NUMBER.\n"),
520 --bind-address=ADDRESS bind to ADDRESS (hostname or IP) on local host.\n"),
522 --limit-rate=RATE limit download rate to RATE.\n"),
524 --no-dns-cache disable caching DNS lookups.\n"),
526 --restrict-file-names=OS restrict chars in file names to ones OS allows.\n"),
528 --ignore-case ignore case when matching files/directories.\n"),
531 -4, --inet4-only connect only to IPv4 addresses.\n"),
533 -6, --inet6-only connect only to IPv6 addresses.\n"),
535 --prefer-family=FAMILY connect first to addresses of specified family,\n\
536 one of IPv6, IPv4, or none.\n"),
539 --user=USER set both ftp and http user to USER.\n"),
541 --password=PASS set both ftp and http password to PASS.\n"),
543 --ask-password prompt for passwords.\n"),
545 --no-iri turn off IRI support.\n"),
547 --local-encoding=ENC use ENC as the local encoding for IRIs.\n"),
549 --remote-encoding=ENC use ENC as the default remote encoding.\n"),
551 --unlink remove file before clobber.\n"),
557 -nd, --no-directories don't create directories.\n"),
559 -x, --force-directories force creation of directories.\n"),
561 -nH, --no-host-directories don't create host directories.\n"),
563 --protocol-directories use protocol name in directories.\n"),
565 -P, --directory-prefix=PREFIX save files to PREFIX/...\n"),
567 --cut-dirs=NUMBER ignore NUMBER remote directory components.\n"),
573 --http-user=USER set http user to USER.\n"),
575 --http-password=PASS set http password to PASS.\n"),
577 --no-cache disallow server-cached data.\n"),
579 --default-page=NAME Change the default page name (normally\n\
580 this is `index.html'.).\n"),
582 -E, --adjust-extension save HTML/CSS documents with proper extensions.\n"),
584 --ignore-length ignore `Content-Length' header field.\n"),
586 --header=STRING insert STRING among the headers.\n"),
588 --max-redirect maximum redirections allowed per page.\n"),
590 --proxy-user=USER set USER as proxy username.\n"),
592 --proxy-password=PASS set PASS as proxy password.\n"),
594 --referer=URL include `Referer: URL' header in HTTP request.\n"),
596 --save-headers save the HTTP headers to file.\n"),
598 -U, --user-agent=AGENT identify as AGENT instead of Wget/VERSION.\n"),
600 --no-http-keep-alive disable HTTP keep-alive (persistent connections).\n"),
602 --no-cookies don't use cookies.\n"),
604 --load-cookies=FILE load cookies from FILE before session.\n"),
606 --save-cookies=FILE save cookies to FILE after session.\n"),
608 --keep-session-cookies load and save session (non-permanent) cookies.\n"),
610 --post-data=STRING use the POST method; send STRING as the data.\n"),
612 --post-file=FILE use the POST method; send contents of FILE.\n"),
614 --content-disposition honor the Content-Disposition header when\n\
615 choosing local file names (EXPERIMENTAL).\n"),
617 --content-on-error output the received content on server errors.\n"),
619 --auth-no-challenge send Basic HTTP authentication information\n\
620 without first waiting for the server's\n\
626 HTTPS (SSL/TLS) options:\n"),
628 --secure-protocol=PR choose secure protocol, one of auto, SSLv2,\n\
629 SSLv3, and TLSv1.\n"),
631 --no-check-certificate don't validate the server's certificate.\n"),
633 --certificate=FILE client certificate file.\n"),
635 --certificate-type=TYPE client certificate type, PEM or DER.\n"),
637 --private-key=FILE private key file.\n"),
639 --private-key-type=TYPE private key type, PEM or DER.\n"),
641 --ca-certificate=FILE file with the bundle of CA's.\n"),
643 --ca-directory=DIR directory where hash list of CA's is stored.\n"),
645 --random-file=FILE file with random data for seeding the SSL PRNG.\n"),
647 --egd-file=FILE file naming the EGD socket with random data.\n"),
649 #endif /* HAVE_SSL */
655 --ftp-stmlf Use Stream_LF format for all binary FTP files.\n"),
656 #endif /* def __VMS */
658 --ftp-user=USER set ftp user to USER.\n"),
660 --ftp-password=PASS set ftp password to PASS.\n"),
662 --no-remove-listing don't remove `.listing' files.\n"),
664 --no-glob turn off FTP file name globbing.\n"),
666 --no-passive-ftp disable the \"passive\" transfer mode.\n"),
668 --preserve-permissions preserve remote file permissions.\n"),
670 --retr-symlinks when recursing, get linked-to files (not dir).\n"),
676 --warc-file=FILENAME save request/response data to a .warc.gz file.\n"),
678 --warc-header=STRING insert STRING into the warcinfo record.\n"),
680 --warc-max-size=NUMBER set maximum size of WARC files to NUMBER.\n"),
682 --warc-cdx write CDX index files.\n"),
684 --warc-dedup=FILENAME do not store records listed in this CDX file.\n"),
687 --no-warc-compression do not compress WARC files with GZIP.\n"),
690 --no-warc-digests do not calculate SHA1 digests.\n"),
692 --no-warc-keep-log do not store the log file in a WARC record.\n"),
694 --warc-tempdir=DIRECTORY location for temporary files created by the\n\
699 Recursive download:\n"),
701 -r, --recursive specify recursive download.\n"),
703 -l, --level=NUMBER maximum recursion depth (inf or 0 for infinite).\n"),
705 --delete-after delete files locally after downloading them.\n"),
707 -k, --convert-links make links in downloaded HTML or CSS point to\n\
711 -K, --backup-converted before converting file X, back up as X_orig.\n"),
712 #else /* def __VMS */
714 -K, --backup-converted before converting file X, back up as X.orig.\n"),
715 #endif /* def __VMS [else] */
717 -m, --mirror shortcut for -N -r -l inf --no-remove-listing.\n"),
719 -p, --page-requisites get all images, etc. needed to display HTML page.\n"),
721 --strict-comments turn on strict (SGML) handling of HTML comments.\n"),
725 Recursive accept/reject:\n"),
727 -A, --accept=LIST comma-separated list of accepted extensions.\n"),
729 -R, --reject=LIST comma-separated list of rejected extensions.\n"),
731 --accept-regex=REGEX regex matching accepted URLs.\n"),
733 --reject-regex=REGEX regex matching rejected URLs.\n"),
736 --regex-type=TYPE regex type (posix|pcre).\n"),
739 --regex-type=TYPE regex type (posix).\n"),
742 -D, --domains=LIST comma-separated list of accepted domains.\n"),
744 --exclude-domains=LIST comma-separated list of rejected domains.\n"),
746 --follow-ftp follow FTP links from HTML documents.\n"),
748 --follow-tags=LIST comma-separated list of followed HTML tags.\n"),
750 --ignore-tags=LIST comma-separated list of ignored HTML tags.\n"),
752 -H, --span-hosts go to foreign hosts when recursive.\n"),
754 -L, --relative follow relative links only.\n"),
756 -I, --include-directories=LIST list of allowed directories.\n"),
758 --trust-server-names use the name specified by the redirection\n\
759 url last component.\n"),
761 -X, --exclude-directories=LIST list of excluded directories.\n"),
763 -np, --no-parent don't ascend to the parent directory.\n"),
765 N_("Mail bug reports and suggestions to <bug-wget@gnu.org>.\n")
770 if (printf (_("GNU Wget %s, a non-interactive network retriever.\n"),
773 if (print_usage (0) < 0)
776 for (i = 0; i < countof (help); i++)
777 if (fputs (_(help[i]), stdout) < 0)
783 /* Return a human-readable printed representation of INTERVAL,
784 measured in seconds. */
787 secs_to_human_time (double interval)
790 int secs = (int) (interval + 0.5);
791 int hours, mins, days;
793 days = secs / 86400, secs %= 86400;
794 hours = secs / 3600, secs %= 3600;
795 mins = secs / 60, secs %= 60;
798 sprintf (buf, "%dd %dh %dm %ds", days, hours, mins, secs);
800 sprintf (buf, "%dh %dm %ds", hours, mins, secs);
802 sprintf (buf, "%dm %ds", mins, secs);
804 sprintf (buf, "%ss", print_decimal (interval));
810 prompt_for_password (void)
813 fprintf (stderr, _("Password for user %s: "), quote (opt.user));
815 fprintf (stderr, _("Password: "));
819 /* Function that prints the line argument while limiting it
820 to at most line_length. prefix is printed on the first line
821 and an appropriate number of spaces are added on subsequent
824 format_and_print_line (const char *prefix, const char *line,
828 char *line_dup, *token;
830 assert (prefix != NULL);
831 assert (line != NULL);
833 line_dup = xstrdup (line);
835 if (line_length <= 0)
836 line_length = MAX_CHARS_PER_LINE - TABULATION;
838 if (printf ("%s", prefix) < 0)
840 remaining_chars = line_length;
841 /* We break on spaces. */
842 token = strtok (line_dup, " ");
843 while (token != NULL)
845 /* If however a token is much larger than the maximum
846 line length, all bets are off and we simply print the
847 token on the next line. */
848 if (remaining_chars <= strlen (token))
850 if (printf ("\n%*c", TABULATION, ' ') < 0)
852 remaining_chars = line_length - TABULATION;
854 if (printf ("%s ", token) < 0)
856 remaining_chars -= strlen (token) + 1; /* account for " " */
857 token = strtok (NULL, " ");
860 if (printf ("\n") < 0)
870 const char *wgetrc_title = _("Wgetrc: ");
871 const char *locale_title = _("Locale: ");
872 const char *compile_title = _("Compile: ");
873 const char *link_title = _("Link: ");
874 char *env_wgetrc, *user_wgetrc;
877 if (printf (_("GNU Wget %s built on %s.\n\n"), version_string, OS_TYPE) < 0)
880 for (i = 0; compiled_features[i] != NULL; )
882 int line_length = MAX_CHARS_PER_LINE;
883 while ((line_length > 0) && (compiled_features[i] != NULL))
885 if (printf ("%s ", compiled_features[i]) < 0)
887 line_length -= strlen (compiled_features[i]) + 2;
890 if (printf ("\n") < 0)
893 if (printf ("\n") < 0)
896 /* Handle the case when $WGETRC is unset and $HOME/.wgetrc is
898 if (printf ("%s\n", wgetrc_title) < 0)
901 env_wgetrc = wgetrc_env_file_name ();
902 if (env_wgetrc && *env_wgetrc)
904 if (printf (_(" %s (env)\n"), env_wgetrc) < 0)
908 user_wgetrc = wgetrc_user_file_name ();
911 if (printf (_(" %s (user)\n"), user_wgetrc) < 0)
916 if (printf (_(" %s (system)\n"), SYSTEM_WGETRC) < 0)
921 if (format_and_print_line (locale_title,
923 MAX_CHARS_PER_LINE) < 0)
925 #endif /* def ENABLE_NLS */
927 if (compilation_string != NULL)
928 if (format_and_print_line (compile_title,
930 MAX_CHARS_PER_LINE) < 0)
933 if (link_string != NULL)
934 if (format_and_print_line (link_title,
936 MAX_CHARS_PER_LINE) < 0)
939 if (printf ("\n") < 0)
942 /* TRANSLATORS: When available, an actual copyright character
943 (circle-c) should be used in preference to "(C)". */
945 Copyright (C) 2011 Free Software Foundation, Inc.\n"), stdout) < 0)
948 License GPLv3+: GNU GPL version 3 or later\n\
949 <http://www.gnu.org/licenses/gpl.html>.\n\
950 This is free software: you are free to change and redistribute it.\n\
951 There is NO WARRANTY, to the extent permitted by law.\n"), stdout) < 0)
953 /* TRANSLATORS: When available, please use the proper diacritics for
954 names such as this one. See en_US.po for reference. */
955 if (fputs (_("\nOriginally written by Hrvoje Niksic <hniksic@xemacs.org>.\n"),
958 if (fputs (_("Please send bug reports and questions to <bug-wget@gnu.org>.\n"),
965 char *program_name; /* Needed by lib/error.c. */
966 char *program_argstring; /* Needed by wget_warc.c. */
969 main (int argc, char **argv)
972 int i, ret, longindex;
974 bool append_to_log = false;
976 total_downloaded_bytes = 0;
978 program_name = argv[0];
980 struct ptimer *timer = ptimer_new ();
981 double start_time = ptimer_measure (timer);
985 atexit (close_stdout);
987 /* Construct the name of the executable, without the directory part. */
989 /* On VMS, lose the "dev:[dir]" prefix and the ".EXE;nnn" suffix. */
990 exec_name = vms_basename (argv[0]);
991 #else /* def __VMS */
992 exec_name = strrchr (argv[0], PATH_SEPARATOR);
997 #endif /* def __VMS [else] */
1000 /* Drop extension (typically .EXE) from executable filename. */
1001 windows_main ((char **) &exec_name);
1004 /* Construct the arguments string. */
1005 int argstring_length = 1;
1006 for (i = 1; i < argc; i++)
1007 argstring_length += strlen (argv[i]) + 2 + 1;
1008 char *p = program_argstring = malloc (argstring_length * sizeof (char));
1011 fprintf (stderr, _("Memory allocation problem\n"));
1014 for (i = 1; i < argc; i++)
1017 int arglen = strlen (argv[i]);
1018 memcpy (p, argv[i], arglen);
1025 /* Load the hard-coded defaults. */
1030 /* This separate getopt_long is needed to find the user config file
1031 option ("--config") and parse it before the other user options. */
1034 bool use_userconfig = false;
1036 while ((retconf = getopt_long (argc, argv,
1037 short_options, long_options, &longindex)) != -1)
1040 bool userrc_ret = true;
1041 struct cmdline_option *config_opt;
1043 /* There is no short option for "--config". */
1046 confval = long_options[longindex].val;
1047 config_opt = &option_data[confval & ~BOOLEAN_NEG_MARKER];
1048 if (strcmp (config_opt->long_name, "config") == 0)
1050 userrc_ret &= run_wgetrc (optarg);
1051 use_userconfig = true;
1055 fprintf (stderr, "Exiting due to error in %s\n", optarg);
1063 /* If the user did not specify a config, read the system wgetrc and ~/.wgetrc. */
1064 if (use_userconfig == false)
1071 while ((ret = getopt_long (argc, argv,
1072 short_options, long_options, &longindex)) != -1)
1075 struct cmdline_option *opt;
1077 /* If LONGINDEX is unchanged, it means RET is referring a short
1079 if (longindex == -1)
1084 fprintf (stderr, "\n");
1085 fprintf (stderr, _("Try `%s --help' for more options.\n"),
1089 /* Find the short option character in the mapping. */
1090 longindex = optmap[ret - 32];
1092 val = long_options[longindex].val;
1094 /* Use the retrieved value to locate the option in the
1095 option_data array, and to see if we're dealing with the
1096 negated "--no-FOO" variant of the boolean option "--foo". */
1097 opt = &option_data[val & ~BOOLEAN_NEG_MARKER];
1101 setoptval (opt->data, optarg, opt->long_name);
1105 /* The user has specified a value -- use it. */
1106 setoptval (opt->data, optarg, opt->long_name);
1109 /* NEG is true for `--no-FOO' style boolean options. */
1110 bool neg = !!(val & BOOLEAN_NEG_MARKER);
1111 setoptval (opt->data, neg ? "0" : "1", opt->long_name);
1116 void (*func) (void) = (void (*) (void)) opt->data;
1120 case OPT__APPEND_OUTPUT:
1121 setoptval ("logfile", optarg, opt->long_name);
1122 append_to_log = true;
1125 run_command (optarg);
1129 /* We support real --no-FOO flags now, but keep these
1130 short options for convenience and backward
1133 for (p = optarg; p && *p; p++)
1137 setoptval ("verbose", "0", opt->long_name);
1140 setoptval ("addhostdir", "0", opt->long_name);
1143 setoptval ("dirstruct", "0", opt->long_name);
1146 setoptval ("noclobber", "1", opt->long_name);
1149 setoptval ("noparent", "1", opt->long_name);
1152 fprintf (stderr, _("%s: illegal option -- `-n%c'\n"),
1155 fprintf (stderr, "\n");
1156 fprintf (stderr, _("Try `%s --help' for more options.\n"),
1165 /* The wgetrc commands are named noparent and noclobber,
1166 so we must revert the meaning of the cmdline options
1167 before passing the value to setoptval. */
1170 flag = (*optarg == '1' || c_tolower (*optarg) == 'y'
1171 || (c_tolower (optarg[0]) == 'o'
1172 && c_tolower (optarg[1]) == 'n'));
1173 setoptval (opt->type == OPT__PARENT ? "noparent" : "noclobber",
1174 flag ? "0" : "1", opt->long_name);
1177 case OPT__DONT_REMOVE_LISTING:
1178 setoptval ("removelisting", "0", opt->long_name);
1185 nurl = argc - optind;
1187 /* All user options have now been processed, so it's now safe to do
1188 interoption dependency checks. */
1190 if (opt.noclobber && opt.convert_links)
1193 _("Both --no-clobber and --convert-links were specified,"
1194 "only --convert-links will be used.\n"));
1195 opt.noclobber = false;
1198 if (opt.reclevel == 0)
1199 opt.reclevel = INFINITE_RECURSION; /* see recur.h for commentary */
1201 if (opt.spider || opt.delete_after)
1202 opt.no_dirstruct = true;
1204 if (opt.page_requisites && !opt.recursive)
1206 /* Don't set opt.recursive here because it would confuse the FTP
1207 code. Instead, call retrieve_tree below when either
1208 page_requisites or recursive is requested. */
1210 if (!opt.no_dirstruct)
1211 opt.dirstruct = 1; /* normally handled by cmd_spec_recursive() */
1214 if (opt.verbose == -1)
1215 opt.verbose = !opt.quiet;
1217 /* Sanity checks. */
1218 if (opt.verbose && opt.quiet)
1220 fprintf (stderr, _("Can't be verbose and quiet at the same time.\n"));
1224 if (opt.timestamping && opt.noclobber)
1226 fprintf (stderr, _("\
1227 Can't timestamp and not clobber old files at the same time.\n"));
1232 if (opt.ipv4_only && opt.ipv6_only)
1235 _("Cannot specify both --inet4-only and --inet6-only.\n"));
1240 if (opt.output_document)
1242 if (opt.convert_links
1243 && (nurl > 1 || opt.page_requisites || opt.recursive))
1246 Cannot specify both -k and -O if multiple URLs are given, or in combination\n\
1247 with -p or -r. See the manual for details.\n\n"), stderr);
1251 if (opt.page_requisites
1254 logprintf (LOG_NOTQUIET, "%s", _("\
1255 WARNING: combining -O with -r or -p will mean that all downloaded content\n\
1256 will be placed in the single file you specified.\n\n"));
1258 if (opt.timestamping)
1260 logprintf (LOG_NOTQUIET, "%s", _("\
1261 WARNING: timestamping does nothing in combination with -O. See the manual\n\
1262 for details.\n\n"));
1263 opt.timestamping = false;
1265 if (opt.noclobber && file_exists_p(opt.output_document))
1267 /* Check if output file exists; if it does, exit. */
1268 logprintf (LOG_VERBOSE,
1269 _("File `%s' already there; not retrieving.\n"),
1270 opt.output_document);
1275 if (opt.warc_filename != 0)
1280 _("WARC output does not work with --no-clobber, "
1281 "--no-clobber will be disabled.\n"));
1282 opt.noclobber = false;
1284 if (opt.timestamping)
1287 _("WARC output does not work with timestamping, "
1288 "timestamping will be disabled.\n"));
1289 opt.timestamping = false;
1294 _("WARC output does not work with --spider.\n"));
1297 if (opt.always_rest)
1300 _("WARC output does not work with --continue, "
1301 "--continue will be disabled.\n"));
1302 opt.always_rest = false;
1304 if (opt.warc_cdx_dedup_filename != 0 && !opt.warc_digests_enabled)
1307 _("Digests are disabled; WARC deduplication will "
1308 "not find duplicate records.\n"));
1310 if (opt.warc_keep_log)
1312 opt.progress_type = xstrdup ("dot");
1316 if (opt.ask_passwd && opt.passwd)
1319 _("Cannot specify both --ask-password and --password.\n"));
1324 if (!nurl && !opt.input_filename)
1326 /* No URL specified. */
1327 fprintf (stderr, _("%s: missing URL\n"), exec_name);
1329 fprintf (stderr, "\n");
1330 /* #### Something nicer should be printed here -- similar to the
1331 pre-1.5 `--help' page. */
1332 fprintf (stderr, _("Try `%s --help' for more options.\n"), exec_name);
1336 /* Compile the regular expressions. */
1337 switch (opt.regex_type)
1340 case regex_type_pcre:
1341 opt.regex_compile_fun = compile_pcre_regex;
1342 opt.regex_match_fun = match_pcre_regex;
1346 case regex_type_posix:
1348 opt.regex_compile_fun = compile_posix_regex;
1349 opt.regex_match_fun = match_posix_regex;
1352 if (opt.acceptregex_s)
1354 opt.acceptregex = opt.regex_compile_fun (opt.acceptregex_s);
1355 if (!opt.acceptregex)
1358 if (opt.rejectregex_s)
1360 opt.rejectregex = opt.regex_compile_fun (opt.rejectregex_s);
1361 if (!opt.rejectregex)
1368 if (opt.locale && !check_encoding_name (opt.locale))
1372 opt.locale = find_locale ();
1374 if (opt.encoding_remote && !check_encoding_name (opt.encoding_remote))
1375 opt.encoding_remote = NULL;
1378 memset (&dummy_iri, 0, sizeof (dummy_iri));
1379 if (opt.enable_iri || opt.locale || opt.encoding_remote)
1381 /* sXXXav : be more specific... */
1382 fprintf (stderr, _("This version does not have support for IRIs\n"));
1389 opt.passwd = prompt_for_password ();
1391 if (opt.passwd == NULL || opt.passwd[0] == '\0')
1401 fork_to_background ();
1404 /* Initialize progress. Have to do this after the options are
1405 processed so we know where the log file is. */
1407 set_progress_implementation (opt.progress_type);
1409 /* Fill in the arguments. */
1410 url = alloca_array (char *, nurl + 1);
1413 fprintf (stderr, _("Memory allocation problem\n"));
1416 for (i = 0; i < nurl; i++, optind++)
1418 char *rewritten = rewrite_shorthand_url (argv[optind]);
1422 url[i] = xstrdup (argv[optind]);
1426 /* Initialize logging. */
1427 log_init (opt.lfilename, append_to_log);
1429 /* Open WARC file. */
1430 if (opt.warc_filename != 0)
1433 DEBUGP (("DEBUG output created by Wget %s on %s.\n\n",
1434 version_string, OS_TYPE));
1436 /* Open the output filename if necessary. */
1439 Note that having the output_stream ("-O") file opened here for an FTP
1440 URL rather than in getftp() (ftp.c) (and the http equivalent) rather
1441 limits the ability in VMS to open the file differently for ASCII
1442 versus binary FTP there. (Of course, doing it here allows a open
1443 failure to be detected immediately, without first connecting to the
1446 if (opt.output_document)
1448 if (HYPHENP (opt.output_document))
1451 _setmode (_fileno (stdout), _O_BINARY);
1453 output_stream = stdout;
1460 /* Common fopen() optional arguments:
1461 sequential access only, access callback function.
1463 # define FOPEN_OPT_ARGS , "fop=sqo", "acc", acc_cb, &open_id
1465 #else /* def __VMS */
1466 # define FOPEN_OPT_ARGS
1467 #endif /* def __VMS [else] */
1469 output_stream = fopen (opt.output_document,
1470 opt.always_rest ? "ab" : "wb"
1472 if (output_stream == NULL)
1474 perror (opt.output_document);
1477 if (fstat (fileno (output_stream), &st) == 0 && S_ISREG (st.st_mode))
1478 output_stream_regular = true;
1480 if (!output_stream_regular && opt.convert_links)
1482 fprintf (stderr, _("-k can be used together with -O only if \
1483 outputting to a regular file.\n"));
1490 /* Set global ODS5 flag according to the specified destination (if
1491 any), otherwise according to the current default device.
1493 if (output_stream == NULL)
1494 set_ods5_dest( "SYS$DISK");
1495 else if (output_stream != stdout)
1496 set_ods5_dest( opt.output_document);
1497 #endif /* def __VMS */
1504 /* Setup the signal handler to redirect output when hangup is
1506 if (signal(SIGHUP, SIG_IGN) != SIG_IGN)
1507 signal(SIGHUP, redirect_output_signal);
1509 /* ...and do the same for SIGUSR1. */
1511 signal (SIGUSR1, redirect_output_signal);
1514 /* Writing to a closed socket normally signals SIGPIPE, and the
1515 process exits. What we want is to ignore SIGPIPE and just check
1516 for the return value of write(). */
1517 signal (SIGPIPE, SIG_IGN);
1520 signal (SIGWINCH, progress_handle_sigwinch);
1523 /* Retrieve the URLs from argument list. */
1524 for (t = url; *t; t++)
1526 char *filename = NULL, *redirected_URL = NULL;
1528 /* Need to do a new struct iri every time, because
1529 * retrieve_url may modify it in some circumstances,
1531 struct iri *iri = iri_new ();
1532 struct url *url_parsed;
1534 set_uri_encoding (iri, opt.locale, true);
1535 url_parsed = url_parse (*t, &url_err, iri, true);
1539 char *error = url_error (*t, url_err);
1540 logprintf (LOG_NOTQUIET, "%s: %s.\n",*t, error);
1542 inform_exit_status (URLERROR);
1546 if ((opt.recursive || opt.page_requisites)
1547 && (url_scheme (*t) != SCHEME_FTP || url_uses_proxy (url_parsed)))
1549 int old_follow_ftp = opt.follow_ftp;
1551 /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
1552 if (url_scheme (*t) == SCHEME_FTP)
1555 retrieve_tree (url_parsed, NULL);
1557 opt.follow_ftp = old_follow_ftp;
1561 retrieve_url (url_parsed, *t, &filename, &redirected_URL, NULL,
1562 &dt, opt.recursive, iri, true);
1565 if (opt.delete_after && filename != NULL && file_exists_p (filename))
1567 DEBUGP (("Removing file due to --delete-after in main():\n"));
1568 logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
1569 if (unlink (filename))
1570 logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
1572 xfree_null (redirected_URL);
1573 xfree_null (filename);
1574 url_free (url_parsed);
1579 /* And then from the input file, if any. */
1580 if (opt.input_filename)
1584 status = retrieve_from_file (opt.input_filename, opt.force_html, &count);
1585 inform_exit_status (status);
1587 logprintf (LOG_NOTQUIET, _("No URLs found in %s.\n"),
1588 opt.input_filename);
1591 /* Print broken links. */
1592 if (opt.recursive && opt.spider)
1593 print_broken_links ();
1595 /* Print the downloaded sum. */
1596 if ((opt.recursive || opt.page_requisites
1598 || (opt.input_filename && total_downloaded_bytes != 0))
1600 total_downloaded_bytes != 0)
1602 double end_time = ptimer_measure (timer);
1603 ptimer_destroy (timer);
1605 char *wall_time = xstrdup (secs_to_human_time (end_time - start_time));
1606 char *download_time = xstrdup (secs_to_human_time (total_download_time));
1607 logprintf (LOG_NOTQUIET,
1608 _("FINISHED --%s--\nTotal wall clock time: %s\n"
1609 "Downloaded: %d files, %s in %s (%s)\n"),
1610 datetime_str (time (NULL)),
1613 human_readable (total_downloaded_bytes),
1615 retr_rate (total_downloaded_bytes, total_download_time));
1617 xfree (download_time);
1619 /* Print quota warning, if exceeded. */
1620 if (opt.quota && total_downloaded_bytes > opt.quota)
1621 logprintf (LOG_NOTQUIET,
1622 _("Download quota of %s EXCEEDED!\n"),
1623 human_readable (opt.quota));
1626 if (opt.cookies_output)
1629 if (opt.convert_links && !opt.delete_after)
1630 convert_all_links ();
1632 /* Close WARC file. */
1633 if (opt.warc_filename != 0)
1638 for (i = 0; i < nurl; i++)
1642 exit (get_exit_status ());
1644 #endif /* TESTING */
1646 #if defined(SIGHUP) || defined(SIGUSR1)
1648 /* So the signal_name check doesn't blow when only one is available. */
1656 /* Hangup signal handler. When wget receives SIGHUP or SIGUSR1, it
1657 will proceed operation as usual, trying to write into a log file.
1658 If that is impossible, the output will be turned off. */
1661 redirect_output_signal (int sig)
1663 const char *signal_name = (sig == SIGHUP ? "SIGHUP" :
1664 (sig == SIGUSR1 ? "SIGUSR1" :
1666 log_request_redirect_output (signal_name);
1667 progress_schedule_redirect ();
1668 signal (sig, redirect_output_signal);