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 },
293 { "warc-compression", 0, OPT_BOOLEAN, "warccompression", -1 },
295 { "warc-dedup", 0, OPT_VALUE, "warccdxdedup", -1 },
296 { "warc-digests", 0, OPT_BOOLEAN, "warcdigests", -1 },
297 { "warc-file", 0, OPT_VALUE, "warcfile", -1 },
298 { "warc-header", 0, OPT_VALUE, "warcheader", -1 },
299 { "warc-keep-log", 0, OPT_BOOLEAN, "warckeeplog", -1 },
300 { "warc-max-size", 0, OPT_VALUE, "warcmaxsize", -1 },
301 { "warc-tempdir", 0, OPT_VALUE, "warctempdir", -1 },
303 { "wdebug", 0, OPT_BOOLEAN, "wdebug", -1 },
310 /* Return a string that contains S with "no-" prepended. The string
311 is NUL-terminated and allocated off static storage at Wget
315 no_prefix (const char *s)
317 static char buffer[1024];
318 static char *p = buffer;
321 int size = 3 + strlen (s) + 1; /* "no-STRING\0" */
322 if (p + size >= buffer + sizeof (buffer))
325 cp[0] = 'n', cp[1] = 'o', cp[2] = '-';
331 /* The arguments that that main passes to getopt_long. */
332 static struct option long_options[2 * countof (option_data) + 1];
333 static char short_options[128];
335 /* Mapping between short option chars and option_data indices. */
336 static unsigned char optmap[96];
338 /* Marker for `--no-FOO' values in long_options. */
339 #define BOOLEAN_NEG_MARKER 1024
341 /* Initialize the long_options array used by getopt_long from the data
347 char *p = short_options;
349 for (i = 0; i < countof (option_data); i++)
351 struct cmdline_option *opt = &option_data[i];
352 struct option *longopt;
355 /* The option is disabled. */
358 longopt = &long_options[o++];
359 longopt->name = opt->long_name;
363 *p++ = opt->short_name;
364 optmap[opt->short_name - 32] = longopt - long_options;
369 longopt->has_arg = required_argument;
374 /* Specify an optional argument for long options, so that
375 --option=off works the same as --no-option, for
376 compatibility with pre-1.10 Wget. However, don't specify
377 optional arguments short-option booleans because they
378 prevent combining of short options. */
379 longopt->has_arg = optional_argument;
380 /* For Boolean options, add the "--no-FOO" variant, which is
381 identical to "--foo", except it has opposite meaning and
382 it doesn't allow an argument. */
383 longopt = &long_options[o++];
384 longopt->name = no_prefix (opt->long_name);
385 longopt->has_arg = no_argument;
386 /* Mask the value so we'll be able to recognize that we're
387 dealing with the false value. */
388 longopt->val = i | BOOLEAN_NEG_MARKER;
391 assert (opt->argtype != -1);
392 longopt->has_arg = opt->argtype;
395 if (longopt->has_arg == required_argument)
397 /* Don't handle optional_argument */
401 /* Terminate short_options. */
403 /* No need for xzero(long_options[o]) because its storage is static
404 and it will be zeroed by default. */
405 assert (o <= countof (long_options));
408 /* Print the usage message. */
410 print_usage (int error)
412 return fprintf (error ? stderr : stdout,
413 _("Usage: %s [OPTION]... [URL]...\n"), exec_name);
416 /* Print the help message, describing all the available options. If
417 you add an option, be sure to update this list. */
421 /* We split the help text this way to ease translation of individual
423 static const char *help[] = {
426 Mandatory arguments to long options are mandatory for short options too.\n\n"),
430 -V, --version display the version of Wget and exit.\n"),
432 -h, --help print this help.\n"),
434 -b, --background go to background after startup.\n"),
436 -e, --execute=COMMAND execute a `.wgetrc'-style command.\n"),
440 Logging and input file:\n"),
442 -o, --output-file=FILE log messages to FILE.\n"),
444 -a, --append-output=FILE append messages to FILE.\n"),
447 -d, --debug print lots of debugging information.\n"),
451 --wdebug print Watt-32 debug output.\n"),
454 -q, --quiet quiet (no output).\n"),
456 -v, --verbose be verbose (this is the default).\n"),
458 -nv, --no-verbose turn off verboseness, without being quiet.\n"),
460 -i, --input-file=FILE download URLs found in local or external FILE.\n"),
462 -F, --force-html treat input file as HTML.\n"),
464 -B, --base=URL resolves HTML input-file links (-i -F)\n\
465 relative to URL.\n"),
467 --config=FILE Specify config file to use.\n"),
473 -t, --tries=NUMBER set number of retries to NUMBER (0 unlimits).\n"),
475 --retry-connrefused retry even if connection is refused.\n"),
477 -O, --output-document=FILE write documents to FILE.\n"),
479 -nc, --no-clobber skip downloads that would download to\n\
480 existing files (overwriting them).\n"),
482 -c, --continue resume getting a partially-downloaded file.\n"),
484 --progress=TYPE select progress gauge type.\n"),
486 -N, --timestamping don't re-retrieve files unless newer than\n\
489 --no-use-server-timestamps don't set the local file's timestamp by\n\
490 the one on the server.\n"),
492 -S, --server-response print server response.\n"),
494 --spider don't download anything.\n"),
496 -T, --timeout=SECONDS set all timeout values to SECONDS.\n"),
498 --dns-timeout=SECS set the DNS lookup timeout to SECS.\n"),
500 --connect-timeout=SECS set the connect timeout to SECS.\n"),
502 --read-timeout=SECS set the read timeout to SECS.\n"),
504 -w, --wait=SECONDS wait SECONDS between retrievals.\n"),
506 --waitretry=SECONDS wait 1..SECONDS between retries of a retrieval.\n"),
508 --random-wait wait from 0.5*WAIT...1.5*WAIT secs between retrievals.\n"),
510 --no-proxy explicitly turn off proxy.\n"),
512 -Q, --quota=NUMBER set retrieval quota to NUMBER.\n"),
514 --bind-address=ADDRESS bind to ADDRESS (hostname or IP) on local host.\n"),
516 --limit-rate=RATE limit download rate to RATE.\n"),
518 --no-dns-cache disable caching DNS lookups.\n"),
520 --restrict-file-names=OS restrict chars in file names to ones OS allows.\n"),
522 --ignore-case ignore case when matching files/directories.\n"),
525 -4, --inet4-only connect only to IPv4 addresses.\n"),
527 -6, --inet6-only connect only to IPv6 addresses.\n"),
529 --prefer-family=FAMILY connect first to addresses of specified family,\n\
530 one of IPv6, IPv4, or none.\n"),
533 --user=USER set both ftp and http user to USER.\n"),
535 --password=PASS set both ftp and http password to PASS.\n"),
537 --ask-password prompt for passwords.\n"),
539 --no-iri turn off IRI support.\n"),
541 --local-encoding=ENC use ENC as the local encoding for IRIs.\n"),
543 --remote-encoding=ENC use ENC as the default remote encoding.\n"),
545 --unlink remove file before clobber.\n"),
551 -nd, --no-directories don't create directories.\n"),
553 -x, --force-directories force creation of directories.\n"),
555 -nH, --no-host-directories don't create host directories.\n"),
557 --protocol-directories use protocol name in directories.\n"),
559 -P, --directory-prefix=PREFIX save files to PREFIX/...\n"),
561 --cut-dirs=NUMBER ignore NUMBER remote directory components.\n"),
567 --http-user=USER set http user to USER.\n"),
569 --http-password=PASS set http password to PASS.\n"),
571 --no-cache disallow server-cached data.\n"),
573 --default-page=NAME Change the default page name (normally\n\
574 this is `index.html'.).\n"),
576 -E, --adjust-extension save HTML/CSS documents with proper extensions.\n"),
578 --ignore-length ignore `Content-Length' header field.\n"),
580 --header=STRING insert STRING among the headers.\n"),
582 --max-redirect maximum redirections allowed per page.\n"),
584 --proxy-user=USER set USER as proxy username.\n"),
586 --proxy-password=PASS set PASS as proxy password.\n"),
588 --referer=URL include `Referer: URL' header in HTTP request.\n"),
590 --save-headers save the HTTP headers to file.\n"),
592 -U, --user-agent=AGENT identify as AGENT instead of Wget/VERSION.\n"),
594 --no-http-keep-alive disable HTTP keep-alive (persistent connections).\n"),
596 --no-cookies don't use cookies.\n"),
598 --load-cookies=FILE load cookies from FILE before session.\n"),
600 --save-cookies=FILE save cookies to FILE after session.\n"),
602 --keep-session-cookies load and save session (non-permanent) cookies.\n"),
604 --post-data=STRING use the POST method; send STRING as the data.\n"),
606 --post-file=FILE use the POST method; send contents of FILE.\n"),
608 --content-disposition honor the Content-Disposition header when\n\
609 choosing local file names (EXPERIMENTAL).\n"),
611 --content-on-error output the received content on server errors.\n"),
613 --auth-no-challenge send Basic HTTP authentication information\n\
614 without first waiting for the server's\n\
620 HTTPS (SSL/TLS) options:\n"),
622 --secure-protocol=PR choose secure protocol, one of auto, SSLv2,\n\
623 SSLv3, and TLSv1.\n"),
625 --no-check-certificate don't validate the server's certificate.\n"),
627 --certificate=FILE client certificate file.\n"),
629 --certificate-type=TYPE client certificate type, PEM or DER.\n"),
631 --private-key=FILE private key file.\n"),
633 --private-key-type=TYPE private key type, PEM or DER.\n"),
635 --ca-certificate=FILE file with the bundle of CA's.\n"),
637 --ca-directory=DIR directory where hash list of CA's is stored.\n"),
639 --random-file=FILE file with random data for seeding the SSL PRNG.\n"),
641 --egd-file=FILE file naming the EGD socket with random data.\n"),
643 #endif /* HAVE_SSL */
649 --ftp-stmlf Use Stream_LF format for all binary FTP files.\n"),
650 #endif /* def __VMS */
652 --ftp-user=USER set ftp user to USER.\n"),
654 --ftp-password=PASS set ftp password to PASS.\n"),
656 --no-remove-listing don't remove `.listing' files.\n"),
658 --no-glob turn off FTP file name globbing.\n"),
660 --no-passive-ftp disable the \"passive\" transfer mode.\n"),
662 --preserve-permissions preserve remote file permissions.\n"),
664 --retr-symlinks when recursing, get linked-to files (not dir).\n"),
670 --warc-file=FILENAME save request/response data to a .warc.gz file.\n"),
672 --warc-header=STRING insert STRING into the warcinfo record.\n"),
674 --warc-max-size=NUMBER set maximum size of WARC files to NUMBER.\n"),
676 --warc-cdx write CDX index files.\n"),
678 --warc-dedup=FILENAME do not store records listed in this CDX file.\n"),
681 --no-warc-compression do not compress WARC files with GZIP.\n"),
684 --no-warc-digests do not calculate SHA1 digests.\n"),
686 --no-warc-keep-log do not store the log file in a WARC record.\n"),
688 --warc-tempdir=DIRECTORY location for temporary files created by the\n\
693 Recursive download:\n"),
695 -r, --recursive specify recursive download.\n"),
697 -l, --level=NUMBER maximum recursion depth (inf or 0 for infinite).\n"),
699 --delete-after delete files locally after downloading them.\n"),
701 -k, --convert-links make links in downloaded HTML or CSS point to\n\
705 -K, --backup-converted before converting file X, back up as X_orig.\n"),
706 #else /* def __VMS */
708 -K, --backup-converted before converting file X, back up as X.orig.\n"),
709 #endif /* def __VMS [else] */
711 -m, --mirror shortcut for -N -r -l inf --no-remove-listing.\n"),
713 -p, --page-requisites get all images, etc. needed to display HTML page.\n"),
715 --strict-comments turn on strict (SGML) handling of HTML comments.\n"),
719 Recursive accept/reject:\n"),
721 -A, --accept=LIST comma-separated list of accepted extensions.\n"),
723 -R, --reject=LIST comma-separated list of rejected extensions.\n"),
725 -D, --domains=LIST comma-separated list of accepted domains.\n"),
727 --exclude-domains=LIST comma-separated list of rejected domains.\n"),
729 --follow-ftp follow FTP links from HTML documents.\n"),
731 --follow-tags=LIST comma-separated list of followed HTML tags.\n"),
733 --ignore-tags=LIST comma-separated list of ignored HTML tags.\n"),
735 -H, --span-hosts go to foreign hosts when recursive.\n"),
737 -L, --relative follow relative links only.\n"),
739 -I, --include-directories=LIST list of allowed directories.\n"),
741 --trust-server-names use the name specified by the redirection\n\
742 url last component.\n"),
744 -X, --exclude-directories=LIST list of excluded directories.\n"),
746 -np, --no-parent don't ascend to the parent directory.\n"),
749 N_("Mail bug reports and suggestions to <bug-wget@gnu.org>.\n")
754 if (printf (_("GNU Wget %s, a non-interactive network retriever.\n"),
757 if (print_usage (0) < 0)
760 for (i = 0; i < countof (help); i++)
761 if (fputs (_(help[i]), stdout) < 0)
767 /* Return a human-readable printed representation of INTERVAL,
768 measured in seconds. */
771 secs_to_human_time (double interval)
774 int secs = (int) (interval + 0.5);
775 int hours, mins, days;
777 days = secs / 86400, secs %= 86400;
778 hours = secs / 3600, secs %= 3600;
779 mins = secs / 60, secs %= 60;
782 sprintf (buf, "%dd %dh %dm %ds", days, hours, mins, secs);
784 sprintf (buf, "%dh %dm %ds", hours, mins, secs);
786 sprintf (buf, "%dm %ds", mins, secs);
788 sprintf (buf, "%ss", print_decimal (interval));
794 prompt_for_password (void)
797 fprintf (stderr, _("Password for user %s: "), quote (opt.user));
799 fprintf (stderr, _("Password: "));
803 /* Function that prints the line argument while limiting it
804 to at most line_length. prefix is printed on the first line
805 and an appropriate number of spaces are added on subsequent
808 format_and_print_line (const char *prefix, const char *line,
812 char *line_dup, *token;
814 assert (prefix != NULL);
815 assert (line != NULL);
817 line_dup = xstrdup (line);
819 if (line_length <= 0)
820 line_length = MAX_CHARS_PER_LINE - TABULATION;
822 if (printf ("%s", prefix) < 0)
824 remaining_chars = line_length;
825 /* We break on spaces. */
826 token = strtok (line_dup, " ");
827 while (token != NULL)
829 /* If however a token is much larger than the maximum
830 line length, all bets are off and we simply print the
831 token on the next line. */
832 if (remaining_chars <= strlen (token))
834 if (printf ("\n%*c", TABULATION, ' ') < 0)
836 remaining_chars = line_length - TABULATION;
838 if (printf ("%s ", token) < 0)
840 remaining_chars -= strlen (token) + 1; /* account for " " */
841 token = strtok (NULL, " ");
844 if (printf ("\n") < 0)
854 const char *wgetrc_title = _("Wgetrc: ");
855 const char *locale_title = _("Locale: ");
856 const char *compile_title = _("Compile: ");
857 const char *link_title = _("Link: ");
858 char *env_wgetrc, *user_wgetrc;
861 if (printf (_("GNU Wget %s built on %s.\n\n"), version_string, OS_TYPE) < 0)
864 for (i = 0; compiled_features[i] != NULL; )
866 int line_length = MAX_CHARS_PER_LINE;
867 while ((line_length > 0) && (compiled_features[i] != NULL))
869 if (printf ("%s ", compiled_features[i]) < 0)
871 line_length -= strlen (compiled_features[i]) + 2;
874 if (printf ("\n") < 0)
877 if (printf ("\n") < 0)
880 /* Handle the case when $WGETRC is unset and $HOME/.wgetrc is
882 if (printf ("%s\n", wgetrc_title) < 0)
885 env_wgetrc = wgetrc_env_file_name ();
886 if (env_wgetrc && *env_wgetrc)
888 if (printf (_(" %s (env)\n"), env_wgetrc) < 0)
892 user_wgetrc = wgetrc_user_file_name ();
895 if (printf (_(" %s (user)\n"), user_wgetrc) < 0)
900 if (printf (_(" %s (system)\n"), SYSTEM_WGETRC) < 0)
905 if (format_and_print_line (locale_title,
907 MAX_CHARS_PER_LINE) < 0)
909 #endif /* def ENABLE_NLS */
911 if (compilation_string != NULL)
912 if (format_and_print_line (compile_title,
914 MAX_CHARS_PER_LINE) < 0)
917 if (link_string != NULL)
918 if (format_and_print_line (link_title,
920 MAX_CHARS_PER_LINE) < 0)
923 if (printf ("\n") < 0)
926 /* TRANSLATORS: When available, an actual copyright character
927 (circle-c) should be used in preference to "(C)". */
929 Copyright (C) 2011 Free Software Foundation, Inc.\n"), stdout) < 0)
932 License GPLv3+: GNU GPL version 3 or later\n\
933 <http://www.gnu.org/licenses/gpl.html>.\n\
934 This is free software: you are free to change and redistribute it.\n\
935 There is NO WARRANTY, to the extent permitted by law.\n"), stdout) < 0)
937 /* TRANSLATORS: When available, please use the proper diacritics for
938 names such as this one. See en_US.po for reference. */
939 if (fputs (_("\nOriginally written by Hrvoje Niksic <hniksic@xemacs.org>.\n"),
942 if (fputs (_("Please send bug reports and questions to <bug-wget@gnu.org>.\n"),
949 char *program_name; /* Needed by lib/error.c. */
950 char *program_argstring; /* Needed by wget_warc.c. */
953 main (int argc, char **argv)
956 int i, ret, longindex;
958 bool append_to_log = false;
960 total_downloaded_bytes = 0;
962 program_name = argv[0];
964 struct ptimer *timer = ptimer_new ();
965 double start_time = ptimer_measure (timer);
969 /* Construct the name of the executable, without the directory part. */
971 /* On VMS, lose the "dev:[dir]" prefix and the ".EXE;nnn" suffix. */
972 exec_name = vms_basename (argv[0]);
973 #else /* def __VMS */
974 exec_name = strrchr (argv[0], PATH_SEPARATOR);
979 #endif /* def __VMS [else] */
982 /* Drop extension (typically .EXE) from executable filename. */
983 windows_main ((char **) &exec_name);
986 /* Construct the arguments string. */
987 int argstring_length = 1;
988 for (i = 1; i < argc; i++)
989 argstring_length += strlen (argv[i]) + 2 + 1;
990 char *p = program_argstring = malloc (argstring_length * sizeof (char));
993 fprintf (stderr, _("Memory allocation problem\n"));
996 for (i = 1; i < argc; i++)
999 int arglen = strlen (argv[i]);
1000 memcpy (p, argv[i], arglen);
1007 /* Load the hard-coded defaults. */
1012 /* This separate getopt_long is needed to find the user config file
1013 option ("--config") and parse it before the other user options. */
1016 bool use_userconfig = false;
1018 while ((retconf = getopt_long (argc, argv,
1019 short_options, long_options, &longindex)) != -1)
1022 bool userrc_ret = true;
1023 struct cmdline_option *config_opt;
1025 /* There is no short option for "--config". */
1028 confval = long_options[longindex].val;
1029 config_opt = &option_data[confval & ~BOOLEAN_NEG_MARKER];
1030 if (strcmp (config_opt->long_name, "config") == 0)
1032 userrc_ret &= run_wgetrc (optarg);
1033 use_userconfig = true;
1037 fprintf (stderr, "Exiting due to error in %s\n", optarg);
1045 /* If the user did not specify a config, read the system wgetrc and ~/.wgetrc. */
1046 if (use_userconfig == false)
1053 while ((ret = getopt_long (argc, argv,
1054 short_options, long_options, &longindex)) != -1)
1057 struct cmdline_option *opt;
1059 /* If LONGINDEX is unchanged, it means RET is referring a short
1061 if (longindex == -1)
1066 fprintf (stderr, "\n");
1067 fprintf (stderr, _("Try `%s --help' for more options.\n"),
1071 /* Find the short option character in the mapping. */
1072 longindex = optmap[ret - 32];
1074 val = long_options[longindex].val;
1076 /* Use the retrieved value to locate the option in the
1077 option_data array, and to see if we're dealing with the
1078 negated "--no-FOO" variant of the boolean option "--foo". */
1079 opt = &option_data[val & ~BOOLEAN_NEG_MARKER];
1083 setoptval (opt->data, optarg, opt->long_name);
1087 /* The user has specified a value -- use it. */
1088 setoptval (opt->data, optarg, opt->long_name);
1091 /* NEG is true for `--no-FOO' style boolean options. */
1092 bool neg = !!(val & BOOLEAN_NEG_MARKER);
1093 setoptval (opt->data, neg ? "0" : "1", opt->long_name);
1098 void (*func) (void) = (void (*) (void)) opt->data;
1102 case OPT__APPEND_OUTPUT:
1103 setoptval ("logfile", optarg, opt->long_name);
1104 append_to_log = true;
1107 run_command (optarg);
1111 /* We support real --no-FOO flags now, but keep these
1112 short options for convenience and backward
1115 for (p = optarg; p && *p; p++)
1119 setoptval ("verbose", "0", opt->long_name);
1122 setoptval ("addhostdir", "0", opt->long_name);
1125 setoptval ("dirstruct", "0", opt->long_name);
1128 setoptval ("noclobber", "1", opt->long_name);
1131 setoptval ("noparent", "1", opt->long_name);
1134 fprintf (stderr, _("%s: illegal option -- `-n%c'\n"),
1137 fprintf (stderr, "\n");
1138 fprintf (stderr, _("Try `%s --help' for more options.\n"),
1147 /* The wgetrc commands are named noparent and noclobber,
1148 so we must revert the meaning of the cmdline options
1149 before passing the value to setoptval. */
1152 flag = (*optarg == '1' || c_tolower (*optarg) == 'y'
1153 || (c_tolower (optarg[0]) == 'o'
1154 && c_tolower (optarg[1]) == 'n'));
1155 setoptval (opt->type == OPT__PARENT ? "noparent" : "noclobber",
1156 flag ? "0" : "1", opt->long_name);
1159 case OPT__DONT_REMOVE_LISTING:
1160 setoptval ("removelisting", "0", opt->long_name);
1167 nurl = argc - optind;
1169 /* All user options have now been processed, so it's now safe to do
1170 interoption dependency checks. */
1172 if (opt.noclobber && opt.convert_links)
1175 _("Both --no-clobber and --convert-links were specified,"
1176 "only --convert-links will be used.\n"));
1177 opt.noclobber = false;
1180 if (opt.reclevel == 0)
1181 opt.reclevel = INFINITE_RECURSION; /* see recur.h for commentary */
1183 if (opt.spider || opt.delete_after)
1184 opt.no_dirstruct = true;
1186 if (opt.page_requisites && !opt.recursive)
1188 /* Don't set opt.recursive here because it would confuse the FTP
1189 code. Instead, call retrieve_tree below when either
1190 page_requisites or recursive is requested. */
1192 if (!opt.no_dirstruct)
1193 opt.dirstruct = 1; /* normally handled by cmd_spec_recursive() */
1196 if (opt.verbose == -1)
1197 opt.verbose = !opt.quiet;
1199 /* Sanity checks. */
1200 if (opt.verbose && opt.quiet)
1202 fprintf (stderr, _("Can't be verbose and quiet at the same time.\n"));
1206 if (opt.timestamping && opt.noclobber)
1208 fprintf (stderr, _("\
1209 Can't timestamp and not clobber old files at the same time.\n"));
1214 if (opt.ipv4_only && opt.ipv6_only)
1217 _("Cannot specify both --inet4-only and --inet6-only.\n"));
1222 if (opt.output_document)
1224 if (opt.convert_links
1225 && (nurl > 1 || opt.page_requisites || opt.recursive))
1228 Cannot specify both -k and -O if multiple URLs are given, or in combination\n\
1229 with -p or -r. See the manual for details.\n\n"), stderr);
1233 if (opt.page_requisites
1236 logprintf (LOG_NOTQUIET, "%s", _("\
1237 WARNING: combining -O with -r or -p will mean that all downloaded content\n\
1238 will be placed in the single file you specified.\n\n"));
1240 if (opt.timestamping)
1242 logprintf (LOG_NOTQUIET, "%s", _("\
1243 WARNING: timestamping does nothing in combination with -O. See the manual\n\
1244 for details.\n\n"));
1245 opt.timestamping = false;
1247 if (opt.noclobber && file_exists_p(opt.output_document))
1249 /* Check if output file exists; if it does, exit. */
1250 logprintf (LOG_VERBOSE,
1251 _("File `%s' already there; not retrieving.\n"),
1252 opt.output_document);
1257 if (opt.warc_filename != 0)
1262 _("WARC output does not work with --no-clobber, "
1263 "--no-clobber will be disabled.\n"));
1264 opt.noclobber = false;
1266 if (opt.timestamping)
1269 _("WARC output does not work with timestamping, "
1270 "timestamping will be disabled.\n"));
1271 opt.timestamping = false;
1276 _("WARC output does not work with --spider.\n"));
1279 if (opt.always_rest)
1282 _("WARC output does not work with --continue, "
1283 "--continue will be disabled.\n"));
1284 opt.always_rest = false;
1286 if (opt.warc_cdx_dedup_filename != 0 && !opt.warc_digests_enabled)
1289 _("Digests are disabled; WARC deduplication will "
1290 "not find duplicate records.\n"));
1292 if (opt.warc_keep_log)
1294 opt.progress_type = "dot";
1298 if (opt.ask_passwd && opt.passwd)
1301 _("Cannot specify both --ask-password and --password.\n"));
1306 if (!nurl && !opt.input_filename)
1308 /* No URL specified. */
1309 fprintf (stderr, _("%s: missing URL\n"), exec_name);
1311 fprintf (stderr, "\n");
1312 /* #### Something nicer should be printed here -- similar to the
1313 pre-1.5 `--help' page. */
1314 fprintf (stderr, _("Try `%s --help' for more options.\n"), exec_name);
1321 if (opt.locale && !check_encoding_name (opt.locale))
1325 opt.locale = find_locale ();
1327 if (opt.encoding_remote && !check_encoding_name (opt.encoding_remote))
1328 opt.encoding_remote = NULL;
1331 memset (&dummy_iri, 0, sizeof (dummy_iri));
1332 if (opt.enable_iri || opt.locale || opt.encoding_remote)
1334 /* sXXXav : be more specific... */
1335 fprintf (stderr, _("This version does not have support for IRIs\n"));
1342 opt.passwd = prompt_for_password ();
1344 if (opt.passwd == NULL || opt.passwd[0] == '\0')
1354 fork_to_background ();
1357 /* Initialize progress. Have to do this after the options are
1358 processed so we know where the log file is. */
1360 set_progress_implementation (opt.progress_type);
1362 /* Fill in the arguments. */
1363 url = alloca_array (char *, nurl + 1);
1366 fprintf (stderr, _("Memory allocation problem\n"));
1369 for (i = 0; i < nurl; i++, optind++)
1371 char *rewritten = rewrite_shorthand_url (argv[optind]);
1375 url[i] = xstrdup (argv[optind]);
1379 /* Initialize logging. */
1380 log_init (opt.lfilename, append_to_log);
1382 /* Open WARC file. */
1383 if (opt.warc_filename != 0)
1386 DEBUGP (("DEBUG output created by Wget %s on %s.\n\n",
1387 version_string, OS_TYPE));
1389 /* Open the output filename if necessary. */
1392 Note that having the output_stream ("-O") file opened here for an FTP
1393 URL rather than in getftp() (ftp.c) (and the http equivalent) rather
1394 limits the ability in VMS to open the file differently for ASCII
1395 versus binary FTP there. (Of course, doing it here allows a open
1396 failure to be detected immediately, without first connecting to the
1399 if (opt.output_document)
1401 if (HYPHENP (opt.output_document))
1404 _setmode (_fileno (stdout), _O_BINARY);
1406 output_stream = stdout;
1413 /* Common fopen() optional arguments:
1414 sequential access only, access callback function.
1416 # define FOPEN_OPT_ARGS , "fop=sqo", "acc", acc_cb, &open_id
1418 #else /* def __VMS */
1419 # define FOPEN_OPT_ARGS
1420 #endif /* def __VMS [else] */
1422 output_stream = fopen (opt.output_document,
1423 opt.always_rest ? "ab" : "wb"
1425 if (output_stream == NULL)
1427 perror (opt.output_document);
1430 if (fstat (fileno (output_stream), &st) == 0 && S_ISREG (st.st_mode))
1431 output_stream_regular = true;
1433 if (!output_stream_regular && opt.convert_links)
1435 fprintf (stderr, _("-k can be used together with -O only if \
1436 outputting to a regular file.\n"));
1443 /* Set global ODS5 flag according to the specified destination (if
1444 any), otherwise according to the current default device.
1446 if (output_stream == NULL)
1447 set_ods5_dest( "SYS$DISK");
1448 else if (output_stream != stdout)
1449 set_ods5_dest( opt.output_document);
1450 #endif /* def __VMS */
1457 /* Setup the signal handler to redirect output when hangup is
1459 if (signal(SIGHUP, SIG_IGN) != SIG_IGN)
1460 signal(SIGHUP, redirect_output_signal);
1462 /* ...and do the same for SIGUSR1. */
1464 signal (SIGUSR1, redirect_output_signal);
1467 /* Writing to a closed socket normally signals SIGPIPE, and the
1468 process exits. What we want is to ignore SIGPIPE and just check
1469 for the return value of write(). */
1470 signal (SIGPIPE, SIG_IGN);
1473 signal (SIGWINCH, progress_handle_sigwinch);
1476 /* Retrieve the URLs from argument list. */
1477 for (t = url; *t; t++)
1479 char *filename = NULL, *redirected_URL = NULL;
1481 /* Need to do a new struct iri every time, because
1482 * retrieve_url may modify it in some circumstances,
1484 struct iri *iri = iri_new ();
1485 struct url *url_parsed;
1487 set_uri_encoding (iri, opt.locale, true);
1488 url_parsed = url_parse (*t, &url_err, iri, true);
1492 char *error = url_error (*t, url_err);
1493 logprintf (LOG_NOTQUIET, "%s: %s.\n",*t, error);
1495 inform_exit_status (URLERROR);
1499 if ((opt.recursive || opt.page_requisites)
1500 && (url_scheme (*t) != SCHEME_FTP || url_uses_proxy (url_parsed)))
1502 int old_follow_ftp = opt.follow_ftp;
1504 /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
1505 if (url_scheme (*t) == SCHEME_FTP)
1508 retrieve_tree (url_parsed, NULL);
1510 opt.follow_ftp = old_follow_ftp;
1514 retrieve_url (url_parsed, *t, &filename, &redirected_URL, NULL,
1515 &dt, opt.recursive, iri, true);
1518 if (opt.delete_after && file_exists_p(filename))
1520 DEBUGP (("Removing file due to --delete-after in main():\n"));
1521 logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
1522 if (unlink (filename))
1523 logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
1525 xfree_null (redirected_URL);
1526 xfree_null (filename);
1527 url_free (url_parsed);
1532 /* And then from the input file, if any. */
1533 if (opt.input_filename)
1537 status = retrieve_from_file (opt.input_filename, opt.force_html, &count);
1538 inform_exit_status (status);
1540 logprintf (LOG_NOTQUIET, _("No URLs found in %s.\n"),
1541 opt.input_filename);
1544 /* Print broken links. */
1545 if (opt.recursive && opt.spider)
1546 print_broken_links ();
1548 /* Print the downloaded sum. */
1549 if ((opt.recursive || opt.page_requisites
1551 || (opt.input_filename && total_downloaded_bytes != 0))
1553 total_downloaded_bytes != 0)
1555 double end_time = ptimer_measure (timer);
1556 ptimer_destroy (timer);
1558 char *wall_time = xstrdup (secs_to_human_time (end_time - start_time));
1559 char *download_time = xstrdup (secs_to_human_time (total_download_time));
1560 logprintf (LOG_NOTQUIET,
1561 _("FINISHED --%s--\nTotal wall clock time: %s\n"
1562 "Downloaded: %d files, %s in %s (%s)\n"),
1563 datetime_str (time (NULL)),
1566 human_readable (total_downloaded_bytes),
1568 retr_rate (total_downloaded_bytes, total_download_time));
1570 xfree (download_time);
1572 /* Print quota warning, if exceeded. */
1573 if (opt.quota && total_downloaded_bytes > opt.quota)
1574 logprintf (LOG_NOTQUIET,
1575 _("Download quota of %s EXCEEDED!\n"),
1576 human_readable (opt.quota));
1579 if (opt.cookies_output)
1582 if (opt.convert_links && !opt.delete_after)
1583 convert_all_links ();
1585 /* Close WARC file. */
1586 if (opt.warc_filename != 0)
1591 for (i = 0; i < nurl; i++)
1595 exit (get_exit_status ());
1597 #endif /* TESTING */
1599 #if defined(SIGHUP) || defined(SIGUSR1)
1601 /* So the signal_name check doesn't blow when only one is available. */
1609 /* Hangup signal handler. When wget receives SIGHUP or SIGUSR1, it
1610 will proceed operation as usual, trying to write into a log file.
1611 If that is impossible, the output will be turned off. */
1614 redirect_output_signal (int sig)
1616 const char *signal_name = (sig == SIGHUP ? "SIGHUP" :
1617 (sig == SIGUSR1 ? "SIGUSR1" :
1619 log_request_redirect_output (signal_name);
1620 progress_schedule_redirect ();
1621 signal (sig, redirect_output_signal);