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 { "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 { "bits", 0, OPT_BOOLEAN, "bits", -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 { "reject", 'R', OPT_VALUE, "reject", -1 },
266 { "relative", 'L', OPT_BOOLEAN, "relativeonly", -1 },
267 { "remote-encoding", 0, OPT_VALUE, "remoteencoding", -1 },
268 { "remove-listing", 0, OPT_BOOLEAN, "removelisting", -1 },
269 { "restrict-file-names", 0, OPT_BOOLEAN, "restrictfilenames", -1 },
270 { "retr-symlinks", 0, OPT_BOOLEAN, "retrsymlinks", -1 },
271 { "retry-connrefused", 0, OPT_BOOLEAN, "retryconnrefused", -1 },
272 { "save-cookies", 0, OPT_VALUE, "savecookies", -1 },
273 { "save-headers", 0, OPT_BOOLEAN, "saveheaders", -1 },
274 { IF_SSL ("secure-protocol"), 0, OPT_VALUE, "secureprotocol", -1 },
275 { "server-response", 'S', OPT_BOOLEAN, "serverresponse", -1 },
276 { "span-hosts", 'H', OPT_BOOLEAN, "spanhosts", -1 },
277 { "spider", 0, OPT_BOOLEAN, "spider", -1 },
278 { "strict-comments", 0, OPT_BOOLEAN, "strictcomments", -1 },
279 { "timeout", 'T', OPT_VALUE, "timeout", -1 },
280 { "timestamping", 'N', OPT_BOOLEAN, "timestamping", -1 },
281 { "tries", 't', OPT_VALUE, "tries", -1 },
282 { "unlink", 0, OPT_BOOLEAN, "unlink", -1 },
283 { "trust-server-names", 0, OPT_BOOLEAN, "trustservernames", -1 },
284 { "use-server-timestamps", 0, OPT_BOOLEAN, "useservertimestamps", -1 },
285 { "user", 0, OPT_VALUE, "user", -1 },
286 { "user-agent", 'U', OPT_VALUE, "useragent", -1 },
287 { "verbose", 'v', OPT_BOOLEAN, "verbose", -1 },
288 { "verbose", 0, OPT_BOOLEAN, "verbose", -1 },
289 { "version", 'V', OPT_FUNCALL, (void *) print_version, no_argument },
290 { "wait", 'w', OPT_VALUE, "wait", -1 },
291 { "waitretry", 0, OPT_VALUE, "waitretry", -1 },
292 { "warc-cdx", 0, OPT_BOOLEAN, "warccdx", -1 },
294 { "warc-compression", 0, OPT_BOOLEAN, "warccompression", -1 },
296 { "warc-dedup", 0, OPT_VALUE, "warccdxdedup", -1 },
297 { "warc-digests", 0, OPT_BOOLEAN, "warcdigests", -1 },
298 { "warc-file", 0, OPT_VALUE, "warcfile", -1 },
299 { "warc-header", 0, OPT_VALUE, "warcheader", -1 },
300 { "warc-keep-log", 0, OPT_BOOLEAN, "warckeeplog", -1 },
301 { "warc-max-size", 0, OPT_VALUE, "warcmaxsize", -1 },
302 { "warc-tempdir", 0, OPT_VALUE, "warctempdir", -1 },
304 { "wdebug", 0, OPT_BOOLEAN, "wdebug", -1 },
311 /* Return a string that contains S with "no-" prepended. The string
312 is NUL-terminated and allocated off static storage at Wget
316 no_prefix (const char *s)
318 static char buffer[1024];
319 static char *p = buffer;
322 int size = 3 + strlen (s) + 1; /* "no-STRING\0" */
323 if (p + size >= buffer + sizeof (buffer))
326 cp[0] = 'n', cp[1] = 'o', cp[2] = '-';
332 /* The arguments that that main passes to getopt_long. */
333 static struct option long_options[2 * countof (option_data) + 1];
334 static char short_options[128];
336 /* Mapping between short option chars and option_data indices. */
337 static unsigned char optmap[96];
339 /* Marker for `--no-FOO' values in long_options. */
340 #define BOOLEAN_NEG_MARKER 1024
342 /* Initialize the long_options array used by getopt_long from the data
348 char *p = short_options;
350 for (i = 0; i < countof (option_data); i++)
352 struct cmdline_option *opt = &option_data[i];
353 struct option *longopt;
356 /* The option is disabled. */
359 longopt = &long_options[o++];
360 longopt->name = opt->long_name;
364 *p++ = opt->short_name;
365 optmap[opt->short_name - 32] = longopt - long_options;
370 longopt->has_arg = required_argument;
375 /* Specify an optional argument for long options, so that
376 --option=off works the same as --no-option, for
377 compatibility with pre-1.10 Wget. However, don't specify
378 optional arguments short-option booleans because they
379 prevent combining of short options. */
380 longopt->has_arg = optional_argument;
381 /* For Boolean options, add the "--no-FOO" variant, which is
382 identical to "--foo", except it has opposite meaning and
383 it doesn't allow an argument. */
384 longopt = &long_options[o++];
385 longopt->name = no_prefix (opt->long_name);
386 longopt->has_arg = no_argument;
387 /* Mask the value so we'll be able to recognize that we're
388 dealing with the false value. */
389 longopt->val = i | BOOLEAN_NEG_MARKER;
392 assert (opt->argtype != -1);
393 longopt->has_arg = opt->argtype;
396 if (longopt->has_arg == required_argument)
398 /* Don't handle optional_argument */
402 /* Terminate short_options. */
404 /* No need for xzero(long_options[o]) because its storage is static
405 and it will be zeroed by default. */
406 assert (o <= countof (long_options));
409 /* Print the usage message. */
411 print_usage (int error)
413 return fprintf (error ? stderr : stdout,
414 _("Usage: %s [OPTION]... [URL]...\n"), exec_name);
417 /* Print the help message, describing all the available options. If
418 you add an option, be sure to update this list. */
422 /* We split the help text this way to ease translation of individual
424 static const char *help[] = {
427 Mandatory arguments to long options are mandatory for short options too.\n\n"),
431 -V, --version display the version of Wget and exit.\n"),
433 -h, --help print this help.\n"),
435 -b, --background go to background after startup.\n"),
437 -e, --execute=COMMAND execute a `.wgetrc'-style command.\n"),
441 Logging and input file:\n"),
443 -o, --output-file=FILE log messages to FILE.\n"),
445 -a, --append-output=FILE append messages to FILE.\n"),
448 -d, --debug print lots of debugging information.\n"),
452 --wdebug print Watt-32 debug output.\n"),
455 -q, --quiet quiet (no output).\n"),
457 -v, --verbose be verbose (this is the default).\n"),
459 -nv, --no-verbose turn off verboseness, without being quiet.\n"),
461 -i, --input-file=FILE download URLs found in local or external FILE.\n"),
463 -F, --force-html treat input file as HTML.\n"),
465 -B, --base=URL resolves HTML input-file links (-i -F)\n\
466 relative to URL.\n"),
468 --config=FILE Specify config file to use.\n"),
474 -t, --tries=NUMBER set number of retries to NUMBER (0 unlimits).\n"),
476 --retry-connrefused retry even if connection is refused.\n"),
478 -O, --output-document=FILE write documents to FILE.\n"),
480 -nc, --no-clobber skip downloads that would download to\n\
481 existing files (overwriting them).\n"),
483 -c, --continue resume getting a partially-downloaded file.\n"),
485 --progress=TYPE select progress gauge type.\n"),
487 -N, --timestamping don't re-retrieve files unless newer than\n\
490 --no-use-server-timestamps don't set the local file's timestamp by\n\
491 the one on the server.\n"),
493 -S, --server-response print server response.\n"),
495 --spider don't download anything.\n"),
497 -T, --timeout=SECONDS set all timeout values to SECONDS.\n"),
499 --dns-timeout=SECS set the DNS lookup timeout to SECS.\n"),
501 --connect-timeout=SECS set the connect timeout to SECS.\n"),
503 --read-timeout=SECS set the read timeout to SECS.\n"),
505 -w, --wait=SECONDS wait SECONDS between retrievals.\n"),
507 --waitretry=SECONDS wait 1..SECONDS between retries of a retrieval.\n"),
509 --random-wait wait from 0.5*WAIT...1.5*WAIT secs between retrievals.\n"),
511 --no-proxy explicitly turn off proxy.\n"),
513 -Q, --quota=NUMBER set retrieval quota to NUMBER.\n"),
515 --bind-address=ADDRESS bind to ADDRESS (hostname or IP) on local host.\n"),
517 --limit-rate=RATE limit download rate to RATE.\n"),
519 --no-dns-cache disable caching DNS lookups.\n"),
521 --restrict-file-names=OS restrict chars in file names to ones OS allows.\n"),
523 --ignore-case ignore case when matching files/directories.\n"),
526 -4, --inet4-only connect only to IPv4 addresses.\n"),
528 -6, --inet6-only connect only to IPv6 addresses.\n"),
530 --prefer-family=FAMILY connect first to addresses of specified family,\n\
531 one of IPv6, IPv4, or none.\n"),
534 --user=USER set both ftp and http user to USER.\n"),
536 --password=PASS set both ftp and http password to PASS.\n"),
538 --ask-password prompt for passwords.\n"),
540 --no-iri turn off IRI support.\n"),
542 --local-encoding=ENC use ENC as the local encoding for IRIs.\n"),
544 --remote-encoding=ENC use ENC as the default remote encoding.\n"),
546 --unlink remove file before clobber.\n"),
552 -nd, --no-directories don't create directories.\n"),
554 -x, --force-directories force creation of directories.\n"),
556 -nH, --no-host-directories don't create host directories.\n"),
558 --protocol-directories use protocol name in directories.\n"),
560 -P, --directory-prefix=PREFIX save files to PREFIX/...\n"),
562 --cut-dirs=NUMBER ignore NUMBER remote directory components.\n"),
568 --http-user=USER set http user to USER.\n"),
570 --http-password=PASS set http password to PASS.\n"),
572 --no-cache disallow server-cached data.\n"),
574 --default-page=NAME Change the default page name (normally\n\
575 this is `index.html'.).\n"),
577 -E, --adjust-extension save HTML/CSS documents with proper extensions.\n"),
579 --ignore-length ignore `Content-Length' header field.\n"),
581 --header=STRING insert STRING among the headers.\n"),
583 --max-redirect maximum redirections allowed per page.\n"),
585 --proxy-user=USER set USER as proxy username.\n"),
587 --proxy-password=PASS set PASS as proxy password.\n"),
589 --referer=URL include `Referer: URL' header in HTTP request.\n"),
591 --save-headers save the HTTP headers to file.\n"),
593 -U, --user-agent=AGENT identify as AGENT instead of Wget/VERSION.\n"),
595 --no-http-keep-alive disable HTTP keep-alive (persistent connections).\n"),
597 --no-cookies don't use cookies.\n"),
599 --load-cookies=FILE load cookies from FILE before session.\n"),
601 --save-cookies=FILE save cookies to FILE after session.\n"),
603 --keep-session-cookies load and save session (non-permanent) cookies.\n"),
605 --post-data=STRING use the POST method; send STRING as the data.\n"),
607 --post-file=FILE use the POST method; send contents of FILE.\n"),
609 --content-disposition honor the Content-Disposition header when\n\
610 choosing local file names (EXPERIMENTAL).\n"),
612 --content-on-error output the received content on server errors.\n"),
614 --auth-no-challenge send Basic HTTP authentication information\n\
615 without first waiting for the server's\n\
621 HTTPS (SSL/TLS) options:\n"),
623 --secure-protocol=PR choose secure protocol, one of auto, SSLv2,\n\
624 SSLv3, and TLSv1.\n"),
626 --no-check-certificate don't validate the server's certificate.\n"),
628 --certificate=FILE client certificate file.\n"),
630 --certificate-type=TYPE client certificate type, PEM or DER.\n"),
632 --private-key=FILE private key file.\n"),
634 --private-key-type=TYPE private key type, PEM or DER.\n"),
636 --ca-certificate=FILE file with the bundle of CA's.\n"),
638 --ca-directory=DIR directory where hash list of CA's is stored.\n"),
640 --random-file=FILE file with random data for seeding the SSL PRNG.\n"),
642 --egd-file=FILE file naming the EGD socket with random data.\n"),
644 #endif /* HAVE_SSL */
650 --ftp-stmlf Use Stream_LF format for all binary FTP files.\n"),
651 #endif /* def __VMS */
653 --ftp-user=USER set ftp user to USER.\n"),
655 --ftp-password=PASS set ftp password to PASS.\n"),
657 --no-remove-listing don't remove `.listing' files.\n"),
659 --no-glob turn off FTP file name globbing.\n"),
661 --no-passive-ftp disable the \"passive\" transfer mode.\n"),
663 --preserve-permissions preserve remote file permissions.\n"),
665 --retr-symlinks when recursing, get linked-to files (not dir).\n"),
671 --warc-file=FILENAME save request/response data to a .warc.gz file.\n"),
673 --warc-header=STRING insert STRING into the warcinfo record.\n"),
675 --warc-max-size=NUMBER set maximum size of WARC files to NUMBER.\n"),
677 --warc-cdx write CDX index files.\n"),
679 --warc-dedup=FILENAME do not store records listed in this CDX file.\n"),
682 --no-warc-compression do not compress WARC files with GZIP.\n"),
685 --no-warc-digests do not calculate SHA1 digests.\n"),
687 --no-warc-keep-log do not store the log file in a WARC record.\n"),
689 --warc-tempdir=DIRECTORY location for temporary files created by the\n\
694 Recursive download:\n"),
696 -r, --recursive specify recursive download.\n"),
698 -l, --level=NUMBER maximum recursion depth (inf or 0 for infinite).\n"),
700 --delete-after delete files locally after downloading them.\n"),
702 -k, --convert-links make links in downloaded HTML or CSS point to\n\
706 -K, --backup-converted before converting file X, back up as X_orig.\n"),
707 #else /* def __VMS */
709 -K, --backup-converted before converting file X, back up as X.orig.\n"),
710 #endif /* def __VMS [else] */
712 -m, --mirror shortcut for -N -r -l inf --no-remove-listing.\n"),
714 -p, --page-requisites get all images, etc. needed to display HTML page.\n"),
716 --strict-comments turn on strict (SGML) handling of HTML comments.\n"),
720 Recursive accept/reject:\n"),
722 -A, --accept=LIST comma-separated list of accepted extensions.\n"),
724 -R, --reject=LIST comma-separated list of rejected extensions.\n"),
726 -D, --domains=LIST comma-separated list of accepted domains.\n"),
728 --exclude-domains=LIST comma-separated list of rejected domains.\n"),
730 --follow-ftp follow FTP links from HTML documents.\n"),
732 --follow-tags=LIST comma-separated list of followed HTML tags.\n"),
734 --ignore-tags=LIST comma-separated list of ignored HTML tags.\n"),
736 -H, --span-hosts go to foreign hosts when recursive.\n"),
738 -L, --relative follow relative links only.\n"),
740 -I, --include-directories=LIST list of allowed directories.\n"),
742 --trust-server-names use the name specified by the redirection\n\
743 url last component.\n"),
745 -X, --exclude-directories=LIST list of excluded directories.\n"),
747 -np, --no-parent don't ascend to the parent directory.\n"),
753 --bits Output bandwidth in bits.\n"),
755 N_("Mail bug reports and suggestions to <bug-wget@gnu.org>.\n")
760 if (printf (_("GNU Wget %s, a non-interactive network retriever.\n"),
763 if (print_usage (0) < 0)
766 for (i = 0; i < countof (help); i++)
767 if (fputs (_(help[i]), stdout) < 0)
773 /* Return a human-readable printed representation of INTERVAL,
774 measured in seconds. */
777 secs_to_human_time (double interval)
780 int secs = (int) (interval + 0.5);
781 int hours, mins, days;
783 days = secs / 86400, secs %= 86400;
784 hours = secs / 3600, secs %= 3600;
785 mins = secs / 60, secs %= 60;
788 sprintf (buf, "%dd %dh %dm %ds", days, hours, mins, secs);
790 sprintf (buf, "%dh %dm %ds", hours, mins, secs);
792 sprintf (buf, "%dm %ds", mins, secs);
794 sprintf (buf, "%ss", print_decimal (interval));
800 prompt_for_password (void)
803 fprintf (stderr, _("Password for user %s: "), quote (opt.user));
805 fprintf (stderr, _("Password: "));
809 /* Function that prints the line argument while limiting it
810 to at most line_length. prefix is printed on the first line
811 and an appropriate number of spaces are added on subsequent
814 format_and_print_line (const char *prefix, const char *line,
818 char *line_dup, *token;
820 assert (prefix != NULL);
821 assert (line != NULL);
823 line_dup = xstrdup (line);
825 if (line_length <= 0)
826 line_length = MAX_CHARS_PER_LINE - TABULATION;
828 if (printf ("%s", prefix) < 0)
830 remaining_chars = line_length;
831 /* We break on spaces. */
832 token = strtok (line_dup, " ");
833 while (token != NULL)
835 /* If however a token is much larger than the maximum
836 line length, all bets are off and we simply print the
837 token on the next line. */
838 if (remaining_chars <= strlen (token))
840 if (printf ("\n%*c", TABULATION, ' ') < 0)
842 remaining_chars = line_length - TABULATION;
844 if (printf ("%s ", token) < 0)
846 remaining_chars -= strlen (token) + 1; /* account for " " */
847 token = strtok (NULL, " ");
850 if (printf ("\n") < 0)
860 const char *wgetrc_title = _("Wgetrc: ");
861 const char *locale_title = _("Locale: ");
862 const char *compile_title = _("Compile: ");
863 const char *link_title = _("Link: ");
864 char *env_wgetrc, *user_wgetrc;
867 if (printf (_("GNU Wget %s built on %s.\n\n"), version_string, OS_TYPE) < 0)
870 for (i = 0; compiled_features[i] != NULL; )
872 int line_length = MAX_CHARS_PER_LINE;
873 while ((line_length > 0) && (compiled_features[i] != NULL))
875 if (printf ("%s ", compiled_features[i]) < 0)
877 line_length -= strlen (compiled_features[i]) + 2;
880 if (printf ("\n") < 0)
883 if (printf ("\n") < 0)
886 /* Handle the case when $WGETRC is unset and $HOME/.wgetrc is
888 if (printf ("%s\n", wgetrc_title) < 0)
891 env_wgetrc = wgetrc_env_file_name ();
892 if (env_wgetrc && *env_wgetrc)
894 if (printf (_(" %s (env)\n"), env_wgetrc) < 0)
898 user_wgetrc = wgetrc_user_file_name ();
901 if (printf (_(" %s (user)\n"), user_wgetrc) < 0)
906 if (printf (_(" %s (system)\n"), SYSTEM_WGETRC) < 0)
911 if (format_and_print_line (locale_title,
913 MAX_CHARS_PER_LINE) < 0)
915 #endif /* def ENABLE_NLS */
917 if (compilation_string != NULL)
918 if (format_and_print_line (compile_title,
920 MAX_CHARS_PER_LINE) < 0)
923 if (link_string != NULL)
924 if (format_and_print_line (link_title,
926 MAX_CHARS_PER_LINE) < 0)
929 if (printf ("\n") < 0)
932 /* TRANSLATORS: When available, an actual copyright character
933 (circle-c) should be used in preference to "(C)". */
935 Copyright (C) 2011 Free Software Foundation, Inc.\n"), stdout) < 0)
938 License GPLv3+: GNU GPL version 3 or later\n\
939 <http://www.gnu.org/licenses/gpl.html>.\n\
940 This is free software: you are free to change and redistribute it.\n\
941 There is NO WARRANTY, to the extent permitted by law.\n"), stdout) < 0)
943 /* TRANSLATORS: When available, please use the proper diacritics for
944 names such as this one. See en_US.po for reference. */
945 if (fputs (_("\nOriginally written by Hrvoje Niksic <hniksic@xemacs.org>.\n"),
948 if (fputs (_("Please send bug reports and questions to <bug-wget@gnu.org>.\n"),
955 char *program_name; /* Needed by lib/error.c. */
956 char *program_argstring; /* Needed by wget_warc.c. */
959 main (int argc, char **argv)
962 int i, ret, longindex;
964 bool append_to_log = false;
966 total_downloaded_bytes = 0;
968 program_name = argv[0];
970 struct ptimer *timer = ptimer_new ();
971 double start_time = ptimer_measure (timer);
975 atexit (close_stdout);
977 /* Construct the name of the executable, without the directory part. */
979 /* On VMS, lose the "dev:[dir]" prefix and the ".EXE;nnn" suffix. */
980 exec_name = vms_basename (argv[0]);
981 #else /* def __VMS */
982 exec_name = strrchr (argv[0], PATH_SEPARATOR);
987 #endif /* def __VMS [else] */
990 /* Drop extension (typically .EXE) from executable filename. */
991 windows_main ((char **) &exec_name);
994 /* Construct the arguments string. */
995 int argstring_length = 1;
996 for (i = 1; i < argc; i++)
997 argstring_length += strlen (argv[i]) + 2 + 1;
998 char *p = program_argstring = malloc (argstring_length * sizeof (char));
1001 fprintf (stderr, _("Memory allocation problem\n"));
1004 for (i = 1; i < argc; i++)
1007 int arglen = strlen (argv[i]);
1008 memcpy (p, argv[i], arglen);
1015 /* Load the hard-coded defaults. */
1020 /* This separate getopt_long is needed to find the user config file
1021 option ("--config") and parse it before the other user options. */
1024 bool use_userconfig = false;
1026 while ((retconf = getopt_long (argc, argv,
1027 short_options, long_options, &longindex)) != -1)
1030 bool userrc_ret = true;
1031 struct cmdline_option *config_opt;
1033 /* There is no short option for "--config". */
1036 confval = long_options[longindex].val;
1037 config_opt = &option_data[confval & ~BOOLEAN_NEG_MARKER];
1038 if (strcmp (config_opt->long_name, "config") == 0)
1040 userrc_ret &= run_wgetrc (optarg);
1041 use_userconfig = true;
1045 fprintf (stderr, "Exiting due to error in %s\n", optarg);
1053 /* If the user did not specify a config, read the system wgetrc and ~/.wgetrc. */
1054 if (use_userconfig == false)
1061 while ((ret = getopt_long (argc, argv,
1062 short_options, long_options, &longindex)) != -1)
1065 struct cmdline_option *opt;
1067 /* If LONGINDEX is unchanged, it means RET is referring a short
1069 if (longindex == -1)
1074 fprintf (stderr, "\n");
1075 fprintf (stderr, _("Try `%s --help' for more options.\n"),
1079 /* Find the short option character in the mapping. */
1080 longindex = optmap[ret - 32];
1082 val = long_options[longindex].val;
1084 /* Use the retrieved value to locate the option in the
1085 option_data array, and to see if we're dealing with the
1086 negated "--no-FOO" variant of the boolean option "--foo". */
1087 opt = &option_data[val & ~BOOLEAN_NEG_MARKER];
1091 setoptval (opt->data, optarg, opt->long_name);
1095 /* The user has specified a value -- use it. */
1096 setoptval (opt->data, optarg, opt->long_name);
1099 /* NEG is true for `--no-FOO' style boolean options. */
1100 bool neg = !!(val & BOOLEAN_NEG_MARKER);
1101 setoptval (opt->data, neg ? "0" : "1", opt->long_name);
1106 void (*func) (void) = (void (*) (void)) opt->data;
1110 case OPT__APPEND_OUTPUT:
1111 setoptval ("logfile", optarg, opt->long_name);
1112 append_to_log = true;
1115 run_command (optarg);
1119 /* We support real --no-FOO flags now, but keep these
1120 short options for convenience and backward
1123 for (p = optarg; p && *p; p++)
1127 setoptval ("verbose", "0", opt->long_name);
1130 setoptval ("addhostdir", "0", opt->long_name);
1133 setoptval ("dirstruct", "0", opt->long_name);
1136 setoptval ("noclobber", "1", opt->long_name);
1139 setoptval ("noparent", "1", opt->long_name);
1142 fprintf (stderr, _("%s: illegal option -- `-n%c'\n"),
1145 fprintf (stderr, "\n");
1146 fprintf (stderr, _("Try `%s --help' for more options.\n"),
1155 /* The wgetrc commands are named noparent and noclobber,
1156 so we must revert the meaning of the cmdline options
1157 before passing the value to setoptval. */
1160 flag = (*optarg == '1' || c_tolower (*optarg) == 'y'
1161 || (c_tolower (optarg[0]) == 'o'
1162 && c_tolower (optarg[1]) == 'n'));
1163 setoptval (opt->type == OPT__PARENT ? "noparent" : "noclobber",
1164 flag ? "0" : "1", opt->long_name);
1167 case OPT__DONT_REMOVE_LISTING:
1168 setoptval ("removelisting", "0", opt->long_name);
1175 nurl = argc - optind;
1177 /* All user options have now been processed, so it's now safe to do
1178 interoption dependency checks. */
1180 if (opt.noclobber && opt.convert_links)
1183 _("Both --no-clobber and --convert-links were specified,"
1184 "only --convert-links will be used.\n"));
1185 opt.noclobber = false;
1188 if (opt.reclevel == 0)
1189 opt.reclevel = INFINITE_RECURSION; /* see recur.h for commentary */
1191 if (opt.spider || opt.delete_after)
1192 opt.no_dirstruct = true;
1194 if (opt.page_requisites && !opt.recursive)
1196 /* Don't set opt.recursive here because it would confuse the FTP
1197 code. Instead, call retrieve_tree below when either
1198 page_requisites or recursive is requested. */
1200 if (!opt.no_dirstruct)
1201 opt.dirstruct = 1; /* normally handled by cmd_spec_recursive() */
1204 if (opt.verbose == -1)
1205 opt.verbose = !opt.quiet;
1207 /* Sanity checks. */
1208 if (opt.verbose && opt.quiet)
1210 fprintf (stderr, _("Can't be verbose and quiet at the same time.\n"));
1214 if (opt.timestamping && opt.noclobber)
1216 fprintf (stderr, _("\
1217 Can't timestamp and not clobber old files at the same time.\n"));
1222 if (opt.ipv4_only && opt.ipv6_only)
1225 _("Cannot specify both --inet4-only and --inet6-only.\n"));
1230 if (opt.output_document)
1232 if (opt.convert_links
1233 && (nurl > 1 || opt.page_requisites || opt.recursive))
1236 Cannot specify both -k and -O if multiple URLs are given, or in combination\n\
1237 with -p or -r. See the manual for details.\n\n"), stderr);
1241 if (opt.page_requisites
1244 logprintf (LOG_NOTQUIET, "%s", _("\
1245 WARNING: combining -O with -r or -p will mean that all downloaded content\n\
1246 will be placed in the single file you specified.\n\n"));
1248 if (opt.timestamping)
1250 logprintf (LOG_NOTQUIET, "%s", _("\
1251 WARNING: timestamping does nothing in combination with -O. See the manual\n\
1252 for details.\n\n"));
1253 opt.timestamping = false;
1255 if (opt.noclobber && file_exists_p(opt.output_document))
1257 /* Check if output file exists; if it does, exit. */
1258 logprintf (LOG_VERBOSE,
1259 _("File `%s' already there; not retrieving.\n"),
1260 opt.output_document);
1265 if (opt.warc_filename != 0)
1270 _("WARC output does not work with --no-clobber, "
1271 "--no-clobber will be disabled.\n"));
1272 opt.noclobber = false;
1274 if (opt.timestamping)
1277 _("WARC output does not work with timestamping, "
1278 "timestamping will be disabled.\n"));
1279 opt.timestamping = false;
1284 _("WARC output does not work with --spider.\n"));
1287 if (opt.always_rest)
1290 _("WARC output does not work with --continue, "
1291 "--continue will be disabled.\n"));
1292 opt.always_rest = false;
1294 if (opt.warc_cdx_dedup_filename != 0 && !opt.warc_digests_enabled)
1297 _("Digests are disabled; WARC deduplication will "
1298 "not find duplicate records.\n"));
1300 if (opt.warc_keep_log)
1302 opt.progress_type = "dot";
1306 if (opt.ask_passwd && opt.passwd)
1309 _("Cannot specify both --ask-password and --password.\n"));
1314 if (!nurl && !opt.input_filename)
1316 /* No URL specified. */
1317 fprintf (stderr, _("%s: missing URL\n"), exec_name);
1319 fprintf (stderr, "\n");
1320 /* #### Something nicer should be printed here -- similar to the
1321 pre-1.5 `--help' page. */
1322 fprintf (stderr, _("Try `%s --help' for more options.\n"), exec_name);
1329 if (opt.locale && !check_encoding_name (opt.locale))
1333 opt.locale = find_locale ();
1335 if (opt.encoding_remote && !check_encoding_name (opt.encoding_remote))
1336 opt.encoding_remote = NULL;
1339 memset (&dummy_iri, 0, sizeof (dummy_iri));
1340 if (opt.enable_iri || opt.locale || opt.encoding_remote)
1342 /* sXXXav : be more specific... */
1343 fprintf (stderr, _("This version does not have support for IRIs\n"));
1350 opt.passwd = prompt_for_password ();
1352 if (opt.passwd == NULL || opt.passwd[0] == '\0')
1362 fork_to_background ();
1365 /* Initialize progress. Have to do this after the options are
1366 processed so we know where the log file is. */
1368 set_progress_implementation (opt.progress_type);
1370 /* Fill in the arguments. */
1371 url = alloca_array (char *, nurl + 1);
1374 fprintf (stderr, _("Memory allocation problem\n"));
1377 for (i = 0; i < nurl; i++, optind++)
1379 char *rewritten = rewrite_shorthand_url (argv[optind]);
1383 url[i] = xstrdup (argv[optind]);
1387 /* Initialize logging. */
1388 log_init (opt.lfilename, append_to_log);
1390 /* Open WARC file. */
1391 if (opt.warc_filename != 0)
1394 DEBUGP (("DEBUG output created by Wget %s on %s.\n\n",
1395 version_string, OS_TYPE));
1397 /* Open the output filename if necessary. */
1400 Note that having the output_stream ("-O") file opened here for an FTP
1401 URL rather than in getftp() (ftp.c) (and the http equivalent) rather
1402 limits the ability in VMS to open the file differently for ASCII
1403 versus binary FTP there. (Of course, doing it here allows a open
1404 failure to be detected immediately, without first connecting to the
1407 if (opt.output_document)
1409 if (HYPHENP (opt.output_document))
1412 _setmode (_fileno (stdout), _O_BINARY);
1414 output_stream = stdout;
1421 /* Common fopen() optional arguments:
1422 sequential access only, access callback function.
1424 # define FOPEN_OPT_ARGS , "fop=sqo", "acc", acc_cb, &open_id
1426 #else /* def __VMS */
1427 # define FOPEN_OPT_ARGS
1428 #endif /* def __VMS [else] */
1430 output_stream = fopen (opt.output_document,
1431 opt.always_rest ? "ab" : "wb"
1433 if (output_stream == NULL)
1435 perror (opt.output_document);
1438 if (fstat (fileno (output_stream), &st) == 0 && S_ISREG (st.st_mode))
1439 output_stream_regular = true;
1441 if (!output_stream_regular && opt.convert_links)
1443 fprintf (stderr, _("-k can be used together with -O only if \
1444 outputting to a regular file.\n"));
1451 /* Set global ODS5 flag according to the specified destination (if
1452 any), otherwise according to the current default device.
1454 if (output_stream == NULL)
1455 set_ods5_dest( "SYS$DISK");
1456 else if (output_stream != stdout)
1457 set_ods5_dest( opt.output_document);
1458 #endif /* def __VMS */
1465 /* Setup the signal handler to redirect output when hangup is
1467 if (signal(SIGHUP, SIG_IGN) != SIG_IGN)
1468 signal(SIGHUP, redirect_output_signal);
1470 /* ...and do the same for SIGUSR1. */
1472 signal (SIGUSR1, redirect_output_signal);
1475 /* Writing to a closed socket normally signals SIGPIPE, and the
1476 process exits. What we want is to ignore SIGPIPE and just check
1477 for the return value of write(). */
1478 signal (SIGPIPE, SIG_IGN);
1481 signal (SIGWINCH, progress_handle_sigwinch);
1484 /* Retrieve the URLs from argument list. */
1485 for (t = url; *t; t++)
1487 char *filename = NULL, *redirected_URL = NULL;
1489 /* Need to do a new struct iri every time, because
1490 * retrieve_url may modify it in some circumstances,
1492 struct iri *iri = iri_new ();
1493 struct url *url_parsed;
1495 set_uri_encoding (iri, opt.locale, true);
1496 url_parsed = url_parse (*t, &url_err, iri, true);
1500 char *error = url_error (*t, url_err);
1501 logprintf (LOG_NOTQUIET, "%s: %s.\n",*t, error);
1503 inform_exit_status (URLERROR);
1507 if ((opt.recursive || opt.page_requisites)
1508 && (url_scheme (*t) != SCHEME_FTP || url_uses_proxy (url_parsed)))
1510 int old_follow_ftp = opt.follow_ftp;
1512 /* Turn opt.follow_ftp on in case of recursive FTP retrieval */
1513 if (url_scheme (*t) == SCHEME_FTP)
1516 retrieve_tree (url_parsed, NULL);
1518 opt.follow_ftp = old_follow_ftp;
1522 retrieve_url (url_parsed, *t, &filename, &redirected_URL, NULL,
1523 &dt, opt.recursive, iri, true);
1526 if (opt.delete_after && file_exists_p(filename))
1528 DEBUGP (("Removing file due to --delete-after in main():\n"));
1529 logprintf (LOG_VERBOSE, _("Removing %s.\n"), filename);
1530 if (unlink (filename))
1531 logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno));
1533 xfree_null (redirected_URL);
1534 xfree_null (filename);
1535 url_free (url_parsed);
1540 /* And then from the input file, if any. */
1541 if (opt.input_filename)
1545 status = retrieve_from_file (opt.input_filename, opt.force_html, &count);
1546 inform_exit_status (status);
1548 logprintf (LOG_NOTQUIET, _("No URLs found in %s.\n"),
1549 opt.input_filename);
1552 /* Print broken links. */
1553 if (opt.recursive && opt.spider)
1554 print_broken_links ();
1556 /* Print the downloaded sum. */
1557 if ((opt.recursive || opt.page_requisites
1559 || (opt.input_filename && total_downloaded_bytes != 0))
1561 total_downloaded_bytes != 0)
1563 double end_time = ptimer_measure (timer);
1564 ptimer_destroy (timer);
1566 char *wall_time = xstrdup (secs_to_human_time (end_time - start_time));
1567 char *download_time = xstrdup (secs_to_human_time (total_download_time));
1568 logprintf (LOG_NOTQUIET,
1569 _("FINISHED --%s--\nTotal wall clock time: %s\n"
1570 "Downloaded: %d files, %s in %s (%s)\n"),
1571 datetime_str (time (NULL)),
1574 human_readable (total_downloaded_bytes),
1576 retr_rate (total_downloaded_bytes, total_download_time));
1578 xfree (download_time);
1580 /* Print quota warning, if exceeded. */
1581 if (opt.quota && total_downloaded_bytes > opt.quota)
1582 logprintf (LOG_NOTQUIET,
1583 _("Download quota of %s EXCEEDED!\n"),
1584 human_readable (opt.quota));
1587 if (opt.cookies_output)
1590 if (opt.convert_links && !opt.delete_after)
1591 convert_all_links ();
1593 /* Close WARC file. */
1594 if (opt.warc_filename != 0)
1599 for (i = 0; i < nurl; i++)
1603 exit (get_exit_status ());
1605 #endif /* TESTING */
1607 #if defined(SIGHUP) || defined(SIGUSR1)
1609 /* So the signal_name check doesn't blow when only one is available. */
1617 /* Hangup signal handler. When wget receives SIGHUP or SIGUSR1, it
1618 will proceed operation as usual, trying to write into a log file.
1619 If that is impossible, the output will be turned off. */
1622 redirect_output_signal (int sig)
1624 const char *signal_name = (sig == SIGHUP ? "SIGHUP" :
1625 (sig == SIGUSR1 ? "SIGUSR1" :
1627 log_request_redirect_output (signal_name);
1628 progress_schedule_redirect ();
1629 signal (sig, redirect_output_signal);